Factors Influencing the Prediction of Pile Driveability Using CPT

energies

Article Factors Inﬂuencing the Prediction of Pile Driveability Using CPT-Based Approaches

Luke J. Prendergast 1 , Putri Gandina 2 and Kenneth Gavin 2,*

1 Department of Civil Engineering, Faculty of Engineering, University of Nottingham, Nottingham NG7 2RD, UK; [email protected] 2 Department of Civil Engineering and Geosciences, Delft University of Technology, Building 23, Stevinweg 1, PO Box 5048, 2628 CN Delft/2600 GA Delft, The Netherlands; [email protected] * Correspondence: [email protected]

Received: 1 May 2020; Accepted: 11 June 2020; Published: 16 June 2020

Abstract: This paper investigates the applicability of Cone Penetration Test (CPT)-based axial capacity approaches, used for estimating pile static capacity, to the prediction of pile driveability. An investigation of the influence of various operational parameters in a driveability study is conducted. A variety of axial capacity approaches (IC-05, UWA-05 and Fugro-05) are assessed in unmodified and modified form to appraise their ability to be used in estimating the driveability of open-ended steel piles used to support, for example, offshore jackets or bridge piers. Modifications to the CPT-based design approaches include alterations to the proposed base resistance to account for the resistance mobilized under discrete hammer impacts and the presence of residual stresses, as well as accounting for the effects of static capacity increases over time, namely ageing. Furthermore, a study on the influence of various operational parameters within a wave equation solver is conducted to ascertain the relative impact of uncertain data in this respect. The purpose of the paper is not to suggest a new design procedure for estimating pile driveability, rather to investigate the influence of the various operating parameters in a driveability analysis and how they affect the magnitude of the resulting predictions. The study will be of interest to geotechnical design of piles using CPT data.

Keywords: piles; driveability; CPT; pile installation; pile design

1. Introduction Pile installation by driving is a high-risk activity in any construction project. Inefficient pile driving can potentially cause material damage to the piles and project delays that may have great financial implications. Selected driving equipment must be capable of installing piles to a target depth within a given time-frame without overstressing the pile. A comprehensive driveability analysis is therefore essential to any project involving pile foundations. In the offshore environment, this is even more important as delays can have significant financial costs and associated risks [1]. Driveability analyses must consider all aspects of pile installation, such as soil conditions and soil-structure interaction, driving equipment performance, and pile specifications including geometrical and material properties. Driveability analyses require calculation of Static Resistance to Driving (SRD) profiles, which are a measure of the soil resistance to pile installation. SRD is analogous to the axial capacity of a pile and represents the cumulative increase in shaft capacity associated with further pile penetration. This also encompasses a base resistance that is associated with each driving increment. Schneider and Harmon [2] claim that SRD is similar to pile static axial capacity except that the resistance often differs due to consolidation, stress equalization, and ageing effects (capacity increases over time). The total resistance of a pile to driving is commonly presented in terms of the blow counts (hammer impacts) per 0.25 m penetration required to drive the pile to its target installation depth.

Energies 2020, 13, 3128; doi:10.3390/en13123128 www.mdpi.com/journal/energies Energies 2020, 13, 3128 2 of 19

This can be estimated using wave equation analysis whereby the inputs include a combination of SRD, pile properties, hammer properties, and dynamic resistance components. Solving the wave equation enables incorporation of the dynamic components including inertia and viscous rate effects, which contribute to the resistance. These are represented by damping and quake values, as discussed in more detail in Section2. The solution to the wave equation allows estimation of the blow counts required to drive a pile, the installation-related stresses, and an estimate of the driving time. It should be noted that an over-estimation of blow counts is considered conservative, unlike an over-estimation of axial capacity, which is considered unconservative. It is desirable that a driveability study would lead to a slight over-estimation of the required blow-counts to install a pile. Arguably the most challenging aspect of performing driveability studies lies with the accurate modelling of soil-structure interaction [1,3,4]. Several models have been put forward to derive SRD profiles, see for example [2,5–8]. Many of these approaches are highly empirical and are biased to the dataset from which they are derived. Due to the analogous relationship between a pile being installed and a penetrating cone, Cone Penetration Tests (CPTs) have become popular in recent years as a way to more accurately encapsulate pile-soil interaction behavior. There are various CPT-based methods already developed to determine the axial static capacity of piles [9,10]. Due to the similarity between axially loaded piles and piles being installed by driving, it is of interest to assess the applicability of these approaches to estimating pile driveability [1]. However, there are some notable differences in the behavior. Firstly, several of these CPT-based axial capacity methods have been developed based on pile load tests conducted between 10 and 30 days after pile installation. Previous studies have indicated that static pile capacity may increase over time after pile driving [11–15]. Based on this premise, the pile resistance experienced during driving should be less than the capacity measured after some time has passed. The application of a time factor to account for this ageing is therefore of interest. Secondly, during pile driving, the shear resistance measured at a given point below the ground surface reduces as the pile tip penetration increases, a phenomenon known as friction fatigue. Schneider and Harmon [2] suggest calculating the pseudo average shaft friction to accommodate changes in the shape of shear resistance distribution during pile driving. More details on this are provided in subsequent sections. Thirdly, the pile penetration per blow during driving is typically less than the failure criteria used to derive the base capacity in CPT-based axial capacity approaches including the IC-05, Fugro-05 and UWA-05 methods. These models incorporate a base resistance corresponding to a pile tip displacement of 0.1D, where D is the pile diameter. A reduction factor is therefore required to consider the actual pile tip displacement encountered during driving [1,16]. Finally, residual stresses on the pile base may be significantly miss-represented, which could lead to considerable error particularly in the case of the UWA-05 base resistance [1,17]. Ignoring residual stresses could lead to an underestimation of the base resistance during driving. In this paper, factors influencing the prediction of pile driveability using CPT-based approaches are investigated. Three CPT-based static capacity models are trialed, namely IC-05, UWA-05 and Fugro-05. The effects of pile ageing on the derived SRD profiles, base-mobilization and residual stresses are investigated with a view to understanding their influence on the perceived driveability of piles. Models are compared to measured pile installation records on relatively slender piles at Blessington, Ireland to appraise the performance of the methods used. Finally, the influence of operating parameters within the wave equation analysis including damping (toe and skin), quake, hammer stroke height, and efficiency is assessed.

2. Background to Pile Driveability

2.1. Wave Equation Analysis The main parameters of interest from a driveability study are the blow counts required to install the pile, the driving stresses experienced by the pile during installation and the time required to drive a pile to a target installation depth. These can be estimated using wave equation analysis. The factors Energies 2020, 13, x FOR PEER REVIEW 3 of 19 Energies 2020, 13, 3128 3 of 19 affecting a pile’s resistance to driving include the soil’s SRD profile, increases in the pile capacity due toaff inertiaecting a(mass) pile’s and resistance viscous to rate driving effects include (damping). the soil’s Methods SRD profile, to define increases the SRD in as the an pile input capacity to the pile due driveabilityto inertia (mass) analysis and viscousare discussed rate eff inects Section (damping). 2.2. The Methods inertia toand define viscous the SRDrate aseffects an input are accounted to the pile fordriveability automatically analysis within are discussedthe wave equation in Section solution 2.2. The procedure. inertia and It viscous is necessary rate e fftoects specify are accounted damping (energyfor automatically dissipation) within and thequake wave (displacement equation solution to achieve procedure. yield) values It is necessary for a given to specify analysis. damping A 1-D (energycommercially dissipation) available and finite-difference quake (displacement wave equati to achieveon solver, yield) GRLWEAP values for [18], a given is used analysis. in the present A 1-D studycommercially to estimate available the blow-counts finite-difference required wave to equation install piles solver, and GRLWEAP to investigate [18], how is used variations in the present in the inputstudy parameters to estimate theinfluence blow-counts the predictions required to using install CPT-based piles and toapproaches. investigate GRLWEAP how variations is a commonly- in the input parametersused commercially influence available the predictions software, using created CPT-based by Pile approaches.Dynamics [18]. GRLWEAP It enables is aestimation commonly-used of soil resistancecommercially and available dynamic software, stresses in created piles, byselectio Pile Dynamicsn of appropriate [18]. It enablesdriving estimationequipment, of determination soil resistance ofand whether dynamic a pile stresses will inbe piles, over-stressed, selection ofdeterminatio appropriaten driving of the likelihood equipment, of determination refusal, and estimation of whether of a piledriving will time. be over-stressed, The response determination of a pile can of be the predicted likelihood by of GRLWEAP refusal, and through estimation solving of driving the wave time. Theequation response shown of in a pileEquation can be (1): predicted by GRLWEAP through solving the wave equation shown in Equation (1): ! ! ∂2u ∂2u (1) =ρ =E (1) ∂t2 ∂x2 where uu isis the the displacement displacement of of the the pile pile (m), (m),ρ is theρ is density the density of the of pile the material pile material (kg/m3 ),(kg/m E is the3), Young’sE is the 0.5 . Young’s modulus2 (N/m=2) andE = is the wave speed. modulus (N/m ) and c ρ is the wave speed. The various inputsinputs andand outputsoutputs of of a a pile pile driveability driveability study study in in the the context context of theof the present present paper paper are areshown shown in Figure in Figure1. This 1. paperThis paper focusses focusses on the on influence the influence of various of various factors factors a ffecting affecting the SRD the profile, SRD profile,derived derived by appropriate by approp modificationriate modification of CPT-based of CPT-based axial static axial capacity static approachescapacity approaches for piles, asfor well piles, as asthe well influence as the ofinfluence uncertainty of uncertainty in the damping in the anddamping quake and values, quake used values, for a used given for calculation. a given calculation.

Figure 1. Flow-chartFlow-chart of driveability analysis procedure procedure.. 2.2. Static Resistance to Driving (SRD) 2.2. Static Resistance to Driving (SRD) Soil resistance to pile driving comprises both static and dynamic components. The SRD is Soil resistance to pile driving comprises both static and dynamic components. The SRD is analogous to axial static capacity and includes shaft friction and base resistance. Unlike static capacity analogous to axial static capacity and includes shaft friction and base resistance. Unlike static capacity however, which only has a single base resistance and shaft capacity distribution, a SRD profile has a however, which only has a single base resistance and shaft capacity distribution, a SRD profile has a base resistance for each driving increment. Moreover, the shaft capacity also varies according to pile base resistance for each driving increment. Moreover, the shaft capacity also varies according to pile penetration. Further differences between pile static capacity and SRD arise due to time effects (ageing), penetration. Further differences between pile static capacity and SRD arise due to time effects consolidation, stress equalisation, and the definition of soil failure used to derive base resistance in (ageing), consolidation, stress equalisation, and the definition of soil failure used to derive base static load tests [2]. resistance in static load tests [2]. The occurrence of plugging in open-ended piles during driving can be represented by the The occurrence of plugging in open-ended piles during driving can be represented by the Incremental Filling Ratio (IFR), which affects base resistance in the definition of SRD. In fully coring Incremental Filling Ratio (IFR), which affects base resistance in the definition of SRD. In fully coring conditions (IFR = 1), end bearing is mobilized on the pile annulus (qann) only, and both internal (τ f ,in) conditions (IFR = 1), end bearing is mobilized on the pile annulus (qann) only, and both internal (,) and external (τ f ) shaft friction are mobilized along the shaft surface area. Alm and Hamre [8] and and external ( ) shaft friction are mobilized along the shaft surface area. Alm and Hamre [8] and Schneider and Harmon [2] suggest reducing unit shaft friction to 50% and applying on both the internal Schneider and Harmon [2] suggest reducing unit shaft friction to 50% and applying on both the and external pile wall, which is approximately the same as applying the full external shaft friction internal and external pile wall, which is approximately the same as applying the full external shaft

Energies 2020, 13, 3128 4 of 19 without any internal shaft friction. Methods to derive estimates of shaft friction are presented in Section 2.3. The shaft stress distribution associated with each pile penetration increment is altered due to friction fatigue, whereby the shear resistance at a given point below ground level reduces with advancement of the pile tip. Some existing SRD models incorporate this by inclusion of a degradation term typically of the form (h/R)n, where h is the vertical distance from the pile tip to the point in the ground where the stress is being calculated and R is the pile radius. This term changes the shape of the soil shear stress distribution as the pile penetrates, therefore an averaging technique to obtain one global SRD proﬁle for input to the wave equation solver is required. Schneider and Harmon [2] suggest that the shape of the shaft stress distribution has a negligible eﬀect on the bearing graph and suggest calculating the pseudo-average shaft friction between successive depth increments (∆τ f ,avg) as follows: P P QS,L QS,L 1 ∆τ = − − (2) f ,avg πD∆L P P where QS,L is the cumulative shaft resistance at the pile tip depth, QS,L-1 is the cumulative shaft resistance at the previous depth increment, ∆L is the depth increment, and D is the pile diameter. Friction fatigue is incorporated to provide a more realistic estimate of the shaft friction experienced by a pile during installation.

2.3. Application of CPT-Based Axial Capacity Approaches to Predicting Pile Installation Cone Penetration Tests (CPT) are a common and useful site investigation tool used in geotechnical characterization. Site investigations must be undertaken to determine strength parameters of soil to inform on design and construction-related matters. The correlation between cone tip resistance (qc), which is the stress experienced by a cone tip as it penetrates into soil, and pile shaft friction and base resistance has been developed over several years. Various traditional SRD approaches have been proposed by Toolan and Fox [6], Stevens et al. [5] and Alm and Hamre [8] among others. For piles installed in sand, Toolan and Fox [6] proposed that unit base resistance and unit skin friction be determined as a weighted average and a fraction of CPT qc respectively. The qc value used in the base resistance should be averaged over a number of pile diameters above and below the pile tip. Stevens et al. [5] proposed determining both the unit base and skin resistance by limiting these for plugged and coring conditions. Alm and Hamre [8] developed a model-based CPT approach using back-calculated driveability studies, which incorporated friction fatigue eﬀects. While these approaches have been used with some success to date, there is a question over their general applicability to piles with diameters beyond those used in the respective datasets from which these were each derived. Methods that use the CPT qc value as a primary input parameter in estimating the axial capacity of piles in sand include Imperial College (IC-05) [9], University of Western Australia (UWA-05) [10], and Fugro-05 [19]. Recent CPT-based approaches consider friction fatigue and plugging eﬀects on piles. The ultimate capacity of a pile is the summation of the shaft and base resistance as written in the following equation: Z Qt = Qs + Qb = P τ f dz + Ab qb (3) where Qs is the total shaft capacity, Qb is the total base capacity, P is the pile perimeter (P = πD for a circular pile and P = 4B for a square pile), τ f is the local shaft friction at failure along the pile shaft, z is πD2 2 the embedded shaft length, Ab is the base area (Ab = 4 for a circular pile, and Ab = B for a square pile), qb is the base resistance, which assumes a displacement of 0.1D as the failure criterion (at the pile head for IC-05, and at the pile tip for UWA-05 and Fugro-05), and D is the pile outer diameter. Energies 2020, 13, 3128 5 of 19

2.3.1. Shaft Capacity The shaft capacity is formulated diﬀerently for the various CPT-based axial capacity approaches. The IC-05 approach for estimating ultimate shaft resistance is shown in the following equation:

 !0.13" !# 0.38   ∆σ h −  = a b q 0v0 max + tan τ f 0.029 c , 8 ∆σ0rd δ f (4) pre f R∗ where a = 0.9 for open-ended piles in tension and 1 in all other cases, b = 0.8 for piles in tension and 1 for piles in compression and R* = (R2 R 2)0.5 where R and R are the external and internal pile radius − i i respectively. Using this approach it is assumed that no plugging has occurred during installation. ∆σ0rd is the change in radial stress experienced at the interface due to dilation. The UWA-05 approach for estimating shaft resistance is shown as follows:

 " !# 0.5  ft  0.3 h −  τ f = 0.03 b qc Ar,e f f max , 2 + ∆σ0rdtan δ f (5) fc  D 

f where t (ratio of tension to compression capacity) = 1 for compression and 0.75 for tension, fc A (effective area ratio) = 1 IFR (D /D)2 where D and D are the external and internal pile diameter, r,e f f − i i IFR = the incremental increase in soil plug length over the pile penetration depth, ∆hplug/∆Lpile. Given that IFR data is not available prior to pile installation, it is recommended to assume IFR = Plug Length Ratio, (PLR = final plug length/penetration depth) based on experience of similar pile geometries and sand types or to assume IFR = 1.0 if no experience is available. The Fugro-05 approach for estimating shaft friction is defined as follows:

!0.05 ! 0.9 ! σ0v0 h − h τ f = 0.08 qc , compression loading f or 4 (6) pre f R∗ R∗ ≥

!0.05 ! ! σ0v0 0.9 h h τ f = 0.08 qc (4)− , compression loading f or < 4 (7) pre f 4R∗ R∗ !0.15" !# 0.85 σ0v0 h − τ f = 0.045 qc max , 4 , tension loading (8) pre f R∗ which varies depending on what portion of the pile shaft is being analyzed, i.e., dependent on (h/R*).

2.3.2. Base Resistance Similarly to the shaft capacity, the base resistance is formulated diﬀerently for the various axial capacity approaches. The IC-05 approach postulates two expressions depending on whether the pile is open or closed-ended. Furthermore, for the open-ended pile, the expression varies depending on the plugging condition, as follows: " ! # qb D 0.1 = max 1 0.5 log , 0.3 , for closed ended piles (9) qc,avg − DCPT

qc,avg If Di 2 (Dr 0.3) or Di 0.083 DCPT, pile is unplugged (Dr is the relative density of the sand). ≥ − ≥ Pre f qb0.1 2 = Ar,Ar (area ratio) = 1 (Di/D) qc,avg − (10) " ! # qb0.1 D = max 0.5 0.25 log , 0.15 , Ar , pile is plugged. (11) qc,avg − DCPT Energies 2020, 13, 3128 6 of 19

where qb0.1 is the base resistance corresponding to a displacement of 0.1D, qc,avg is the average CPT stress (depending on which averaging technique is adopted), DCPT is the diameter of the penetrating cone, and Ar is the area ratio. For the UWA-05 approach, the base resistance is formulated as follows:

qb0.1 = 0.15 + 0.45 Ar,e f f (12) qc,avg

For the Fugro-05 approach it is formulated as shown here:

P !0.5 qb0.1 re f 0.25 = 8.5 Ar (13) qc,avg qc,avg

2.4. Modiﬁcations to CPT-Based Axial Capacity Approaches to Predict Pile Installation In order to account for the conditions experienced at installation, some modiﬁcation to the axial capacity approaches can be undertaken, as discussed in the following subsections.

2.4.1. Pile Ageing The methods described previously to calculate the axial capacity of piles (IC-05, UWA-05, and Fugro-05) were each derived empirically from pile load tests conducted between 10 and 30 days after pile installation. Various studies suggest that shaft capacity of piles driven in sand increases with time, a phenomenon termed ageing. Jardine et al. [11] investigated the ageing eﬀect in dense sand by determining Intact Ageing Curves (IAC), which indicate that the shaft capacity experienced one day after driving is approximately 70% that experienced 10–30 days after driving. Gavin et al. [15] found that ageing does not depend on whether the piles are driven in dry or partially saturated sand with salty or fresh water, however, driving reduced the resistance during installation, causing higher gains in capacity with time. Lehane et al. [20] collected recent pile ageing information and collated into a database, including ﬁeld test data at Dunkirk [11], Blessington [12] and Larvik [13]. Using this database, an equation was derived to estimate the time-variation in capacity after installation, determined as:

1 Ftime = + do f f set (14) exp( 0.1 t0.68) + 0.45 − where t is the pile age in days and doffset is the vertical offset that best fits the data points. This time factor can be applied to determine the shaft friction expected during installation by setting the time to zero, which reveals a time factor of 0.69 can be applied to determine the shaft friction for SRD calculations. Gavin and Igoe [21] found that changes in shaft capacity caused by ageing effects on piles installed in Blessington sand were concentrated in a zone within five pile diameters of the tip. Therefore, this reduction factor should be used with caution on long, slender piles as it may be unconservative.

2.4.2. Base Mobilization during Driving The unit base resistance expressions postulated in the CPT-based axial capacity methods such as UWA-05, IC-05, and Fugro-05 calculate the resistance mobilized at a pile displacement of 0.1D (outer pile diameter). During pile installation, however, the displacement per blow is expected to be signiﬁcantly less than 0.1D. The UWA-05 method considers partially plugging conditions, which are represented by the Final Filling Ratio (FFR) value. This suggests a potential means for considering actual expected displacements during installation within a modiﬁed UWA-05 approach [1,16]. Byrne et al. [1] propose the implementation of a three-stage base resistance-displacement model [16] to estimate the expected base resistance mobilized during driving. Figure2 shows an idealized base resistance-displacement model consisting of the unit base resistance (qb) versus the pile tip displacement Energies 2020, 13, 3128 7 of 19 Energies 2020, 13, x FOR PEER REVIEW 7 of 19

b displacement(wb) normalized (w ) by normalized pile diameter by pile (D ).diameter The model (D). consistsThe model of threeconsists portions: of three the portions: initial settlementthe initial by settlementresponse is response assumed is linear assumed until linear a yield until strain a (yieldwby/D strain= 1.5%) (w is/D exceeded, = 1.5%) andis exceeded, this is followed and this by is a followednon-linear by parabolic a non-linear stage parabolic until astrain stage ofuntil 0.1D a strain is achieved. of 0.1D An is achieved. initial residual An initial stress residualqb,res must stress be qovercomeb,res must be prior overcome to any prior movement to any occurring movement (see occurring Section 2.4.3(see ).Section 2.4.3).

FigureFigure 2. 2. BaseBase resistance-settlement resistance-settlement model model [16]. [16].

The linear portion of the curve is governed by the small-strain Young’s modulus of the soil (E0), The linear portion of the curve is governed by the small-strain Young’s modulus of the soil (E0), whichwhich can be obtainedobtained fromfrom Multi-Channel Multi-Channel Analysis Analysis of of Surface Surface Waves Waves [22 ][22] or fromor from correlations correlations to CPT to CPTtip resistance tip resistance [23– 25[23–25].]. The linearThe linea portionr portion of the of curve the curve follows: follows: wb q = k + q for w /D < 0.015 (15) = b +1 forb, reswb/D < 0.015b (15) D, 4 E k = 0 (16) 4 1 2 (16) = π 1 v 1− − where ν is the Poisson’s ratio, wb is the pile tip displacement (at a given blow), D is the outer pile wherediameter, ν isand the qPoisson’sb,res is the ratio, residual wb baseis the resistance. pile tip displacement (at a given blow), D is the outer pile diameter,The parabolicand qb,res is portion the residual of the base curve resistance.wby/D < w b/D < 0.1 is modiﬁed from the original source [1,16] andThe approximated parabolic portion as a line of as the follows: curve wby/D < wb/D < 0.1 is modified from the original source [1,16] and approximated as a line as follows: wb qb = k2 0.015 + 0.015k1 + qb,res (17) D = − 0.015− + 0.015 + (17) , qb01,UWA 0.015 k1 k = − (18) 2 0.085 , − 0.015 (18) = where qb01,UWA is the unit base resistance from0.085 the UWA-05 method. The reason for approximating the parabolic portion of the curve as a line is for ease of the analysis in this paper. To modify the where qb01,UWA is the unit base resistance from the UWA-05 method. The reason for approximating the base resistance using the base resistance-settlement model from [16], it is anticipated that the errors parabolic portion of the curve as a line is for ease of the analysis in this paper. To modify the base introduced in this respect will be minimal. Residual base stresses associated with the action of previous resistance using the base resistance-settlement model from [16], it is anticipated that the errors hammer blows can be incorporated as a proportion of the CPT end resistance [1], as discussed in the introduced in this respect will be minimal. Residual base stresses associated with the action of next section. previous hammer blows can be incorporated as a proportion of the CPT end resistance [1], as discussed2.4.3. Residual in the Base next Stressessection.

2.4.3. DuringResidual driving, Base Stresses the pile experiences compression under the action of a hammer blow, and subsequently tension under zero loading (rebounding). The tension force tends to cause some During driving, the pile experiences compression under the action of a hammer blow, and rebound to occur between hammer blows. The development of this condition can be considered subsequently tension under zero loading (rebounding). The tension force tends to cause some as a residual base stress (qb,res) developing on the pile base area. The development of this residual rebound to occur between hammer blows. The development of this condition can be considered as a residual base stress (qb,res) developing on the pile base area. The development of this residual base

Energies 2020, 13, x FOR PEER REVIEW 8 of 19 stress implies that negative skin friction (,) must also develop along the pile shaft to ensure equilibrium between blows. Figure 3 illustrate the development of the stress condition. Paik et al. [26] suggest that the presence of residual base stresses does not affect the ultimate bearing capacity of piles under axial static loading since the summation of residual shaft and base resistance will equal to zero. However, qb,res must be acknowledged when considering pile driveability, since its presence alters the proportion of resistance acting at the shaft and the base, which influences the response under wave equation analyses [1]. When a hammer impacts the pile, Energiesthe residual2020, 13 ,negative 3128 skin friction must first go to zero then tend towards its limiting positive value.8 of 19 Meanwhile, the effective stress generated at the pile base due to a hammer impact is added to the residual base resistance already present. base stress implies that negative skin friction (τ f ,neg) must also develop along the pile shaft to ensure equilibrium between blows. Figure3 illustrate the development of the stress condition.

(a) (b)

Figure 3. The development of residual base stress during pile driving. (a) Pile in compression. Figure 3. The development of residual base stress during pile driving. (a) Pile in compression. (b) (b) Residual stresses at zero pile loading. Residual stresses at zero pile loading. Paik et al. [26] suggest that the presence of residual base stresses does not affect the ultimate Estimating the magnitude of residual base stress is challenging. Alawneh and Malwaki [27] bearing capacity of piles under axial static loading since the summation of residual shaft and base propose a method to estimate the post-driving residual base stresses as a function of the pile resistance will equal to zero. However, q must be acknowledged when considering pile driveability, penetration length, diameter, area, shearb,res and Young’s modulus. This method suggests residual since its presence alters the proportion of resistance acting at the shaft and the base, which influences stresses that range between 0 and 4000 kPa. Paik et al. [26] measured residual base stresses on 0.356 the response under wave equation analyses [1]. When a hammer impacts the pile, the residual negative m diameter closed and open-ended piles which suggests these stresses are of the order of 11–14% of skin friction must first go to zero then tend towards its limiting positive value. Meanwhile, the effective CPT qc. Similarly, Byrne et al. [1] suggest residual base stresses could be of the order of 1–10% of qc. stress generated at the pile base due to a hammer impact is added to the residual base resistance Gavin and Lehane [16] found that the residual load developed on an open-ended pile was related to already present. the IFR values, with fully-coring piles (IFR = 100%) developing low residual loads and fully plugged Estimating the magnitude of residual base stress is challenging. Alawneh and Malwaki [27] piles developing higher residual loads than closed-ended piles with the same external diameter. propose a method to estimate the post-driving residual base stresses as a function of the pile penetration Gavin and Igoe [21] measured the residual load on Pile S6 considered in this paper and found that length, diameter, area, shear and Young’s modulus. This method suggests residual stresses that the residual base stress at the end of installation was 4 MPa, ≈ 20% of the qc value at this pile tip level. range between 0 and 4000 kPa. Paik et al. [26] measured residual base stresses on 0.356 m diameter This was for an IFR value of 40% which is unlikely to be developed by an open-ended pile in practice closed and open-ended piles which suggests these stresses are of the order of 11–14% of CPT q . as at such a value it would be extremely difficult to continue pile driving due to high base resistances.c Similarly, Byrne et al. [1] suggest residual base stresses could be of the order of 1–10% of q . Gavin In reality, the estimation of residual stress magnitude is quite an uncertain process, and dependsc on and Lehane [16] found that the residual load developed on an open-ended pile was related to the IFR the relative relationship between the developing negative skin friction and the internal stresses in the values, with fully-coring piles (IFR = 100%) developing low residual loads and fully plugged piles pile, among other factors. developing higher residual loads than closed-ended piles with the same external diameter. Gavin and

Igoe [21] measured the residual load on Pile S6 considered in this paper and found that the residual base stress at the end of installation was 4 MPa, 20% of the qc value at this pile tip level. This was for ≈ an IFR value of 40% which is unlikely to be developed by an open-ended pile in practice as at such a value it would be extremely diﬃcult to continue pile driving due to high base resistances. In reality, the estimation of residual stress magnitude is quite an uncertain process, and depends on the relative relationship between the developing negative skin friction and the internal stresses in the pile, among other factors. EnergiesEnergies2020 2020, 13, 13, 3128, x FOR PEER REVIEW 99 of of 19 19

3. Analysis and Results 3. Analysis and Results 3.1. Driven Pile Database 3.1. Driven Pile Database In this paper, the driveability results of seven steel open-ended piles at Blessington, Ireland are In this paper, the driveability results of seven steel open-ended piles at Blessington, Ireland are used. Soil properties and ground conditions at the Blessington site are reported in [12,14,16,23] and used. Soil properties and ground conditions at the Blessington site are reported in [12,14,16,23] and the significant parameters are reported herein. The soil profile at this site consists of very dense over- the significant parameters are reported herein. The soil profile at this site consists of very dense consolidated sand. The groundwater table is approximately 13 m below ground level (bgl). The in- over-consolidated sand. The groundwater table is approximately 13 m below ground level (bgl). situ water content is relatively uniform at 10–12% above the water table. Piles are installed above the The in-situ water content is relatively uniform at 10–12% above the water table. Piles are installed water table where pore pressure dissipates almost immediately. The sand relative density ranges above the water table where pore pressure dissipates almost immediately. The sand relative density between 90% and 100%. The particle size (D50) varies between 0.1 mm and 0.15 mm based on particle ranges between 90% and 100%. The particle size (D50) varies between 0.1 mm and 0.15 mm based on size distribution analyses from samples located between 0.7–2 m bgl. The soil is well-graded angular particle size distribution analyses from samples located between 0.7–2 m bgl. The soil is well-graded sand with 5–10% fines content. The unit weight is approximately 20 kN/m3, and the constant volume angular sand with 5–10% fines content. The unit weight is approximately 20 kN/m3, and the constant friction angle is approximately 37°. volume friction angle is approximately 37◦. A total of 10 CPTs were conducted at the site, the average, maximum and minimum profiles A total of 10 CPTs were conducted at the site, the average, maximum and minimum profiles shown in Figure 4a. The CPTs are quite uniform across the site. The shear wave velocity (Vs) profiles shown in Figure4a. The CPTs are quite uniform across the site. The shear wave velocity ( Vs) profiles measured using MASW are shown in Figure 4b, and the subsequently derived shear modulus profiles measured using MASW are shown in Figure4b, and the subsequently derived shear modulus profiles (G0) from both the CPT and MASW measurements are shown in Figure 4c. The G0 profiles were (G0) from both the CPT and MASW measurements are shown in Figure4c. The G0 profiles were derived from the CPT measurements using a correlation known as a rigidity index, G0/qc. For a given derived from the CPT measurements using a correlation known as a rigidity index, G0/qc. For a given deposit G0/qc increases with sand age and cementation [25], and this parameter has been studied in deposit G0/qc increases with sand age and cementation [25], and this parameter has been studied in detail by several authors [25,28]. For an aged, over-consolidated material such as that at Blessington, detail by several authors [25,28]. For an aged, over-consolidated material such as that at Blessington, a rigidity index in the range of 5–8 is expected. In the present analysis a value of 6 was adopted. a rigidity index in the range of 5–8 is expected. In the present analysis a value of 6 was adopted.

(a) (b) (c)

FigureFigure 4. 4.Blessington Blessington site site properties. properties. ( a(a)) Cone Cone tip tip resistance, resistance,q cq;(c; b(b)) Shear Shear wave wave velocity, velocity,V Vs;(s; c(c)) Shear Shear modulus,modulus,G G0.0.

SevenSeven open-endedopen-ended steelsteel piles named named S1–S7 S1–S7 were were each each driven driven 7 m 7 minto into the the ground, ground, except except for forpile pileS6, S6,which which was was driven driven 6.5 6.5m. Each m. Each pile pile has hasan external an external diameter diameter (D) (ofD )0.34 of 0.34 m, internal m, internal diameter diameter (Di) (Dofi )0.312 of 0.312 m, and m, andwall wall thickness thickness (tw) of (tw 0.014) of 0.014 m. Blow m. Blowcount count records records for each for pile each are pile plotted are plotted in Figure in Figure5a. All5 a.piles All exhibit piles exhibit an increase an increase in blow in blowcounts counts with withpenetration penetration depth. depth. Blow Blowcounts counts for pile for S4 pile are S4low are compared low compared to S2, S3 to and S2, S3S5. and Pile S5.S6 encounters Pile S6 encounters the highest the blow highest counts blow compared counts compared to the remaining to the remainingpiles. Some piles. of the Some differences of the di inﬀerences the blow in counts the blow can counts be accounted can be accounted for by variations for by variationsin the properties in the of the driving equipment (hammer type and stroke heights). Table 1 provides information on the

Energies 2020, 13, x FOR PEER REVIEW 10 of 19 hammer properties adopted to drive each pile. Piles S1–S5 and S6, S7 were driven using 4000 kg Junttan PM16 and 5000 kg Junttan PM20 hammers, respectively. Piles S1 and S2–S5 were driven with the same constant stroke height of 0.4 m and 0.3 m accordingly. Piles S6 and S7 used a combination of stroke heights with higher energies (e.g., drop-heights) being used for Pile S7 in order to limit the blow-counts during installation and thus protect sensitive radial stress sensors used in the ageing study reported by Gavin and Igoe [21].

Table 1. Hammer properties used to install piles at Blessington.

Pile Penetration length Hammer Cushion Stroke height (m) Name (m) 4000 kg Junttan S1 7 None 0.4 PM16 4000 kg Junttan S2 7 None 0.3 PM16 4000 kg Junttan S3 7 None 0.3 PM16 4000 kg Junttan S4 7 None 0.3 PM16 4000 kg Junttan S5 7 None 0.3 PM16 5000 kg Junttan 50 mm ash 0.2 (0–4 m) & EnergiesS62020 , 13, 3128 7 10 of 19 PM20 timber 0.35 (4–7 m) 5000 kg Junttan 50 mm ash 0.2–0.3 m (increment 0.025 S7 6.5 properties of the driving equipment (hammerPM20 type and stroketimber heights). Table1 providesm) information on theFigure hammer 5b shows properties the adoptedIFR measured to drive during each pile. pile Pilesinstallation. S1–S5 and All S6, piles S7 weredeveloped driven usinga similar 4000 IFR kg profile.Junttan PM16All piles and were 5000 kgnearly Junttan fully PM20 coring hammers, or unplugged respectively. (IFR Piles = 1) S1 over and S2–S5the first were meter driven of withpile penetrationthe same constant and became stroke partially height of plugged 0.4 m and (IFR 0.3 = m 0.4) accordingly. at the end Piles of driving S6 and with S7 used 2.45 a m combination final plug length.of stroke Pile heights S4 experienced with higher less energies plugging (e.g., over drop-heights) the final metre being of penetration used for Pile and S7 had in order a final to IFR limit = 0.75 the atblow-counts the end of driving during and installation the longest and final thus plug protect length. sensitive radial stress sensors used in the ageing study reported by Gavin and Igoe [21].

(a) (b)

Figure 5. Recorded data during pile installation. ( (aa)) Blow Blow counts; counts; ( (b)) IFR. IFR.

Table 1. Hammer properties used to install piles at Blessington.

Pile Name Penetration Length (m) Hammer Cushion Stroke Height (m) S1 7 4000 kg Junttan PM16 None 0.4 S2 7 4000 kg Junttan PM16 None 0.3 S3 7 4000 kg Junttan PM16 None 0.3 S4 7 4000 kg Junttan PM16 None 0.3 S5 7 4000 kg Junttan PM16 None 0.3 S6 7 5000 kg Junttan PM20 50 mm ash timber 0.2 (0–4 m) & 0.35 (4–7 m) S7 6.5 5000 kg Junttan PM20 50 mm ash timber 0.2–0.3 m (increment 0.025 m)

Figure5b shows the IFR measured during pile installation. All piles developed a similar IFR profile. All piles were nearly fully coring or unplugged (IFR = 1) over the first meter of pile penetration and became partially plugged (IFR = 0.4) at the end of driving with 2.45 m final plug length. Pile S4 experienced less plugging over the final metre of penetration and had a final IFR = 0.75 at the end of driving and the longest final plug length.

3.2. Comparison of Unmodified CPT-Based Approaches Piles S1 and S2 In this section, the driveability data for piles S1 and S2 are compared to predictions using unmodified CPT-based axial capacity approaches. The approaches are detailed in Section 2.3. The IC-05, UWA-05 and Fugro-05 methods are applied in their unmodified forms to derive SRD profiles for shaft and toe resistance using the average CPT profile from the site, which are subsequently input into the wave equation analysis software. The friction fatigue effect is incorporated using the pseudo-average Energies 2020, 13, x FOR PEER REVIEW 11 of 19

3.2. Comparison of Unmodified CPT-Based Approaches Piles S1 and S2 In this section, the driveability data for piles S1 and S2 are compared to predictions using unmodified CPT-based axial capacity approaches. The approaches are detailed in Section 2.3. The IC- 05, UWA-05 and Fugro-05 methods are applied in their unmodified forms to derive SRD profiles for Energiesshaft 2020and, 13 toe, 3128 resistance using the average CPT profile from the site, which are subsequently11 ofinput 19 into the wave equation analysis software. The friction fatigue effect is incorporated using the pseudo- average shaft friction as detailed in Section 2.2. For the IC-05 and Fugro-05 approaches, both shaftunplugged friction and as detailed plugged in models Section are2.2 .used For to the derive IC-05 the and base Fugro-05 resistance approaches, since it bothis not unplugged possible to andaccount plugged for modelspartial plugging are used using to derive these the approaches base resistance (as experienced since it is notby the possible real piles). to account Analysing for partialfully pluggingplugged usingor unplugged these approaches represents (as experiencedthe extreme bycases the realthat piles). can be Analysing considered. fully The plugged hammer or unpluggedproperties represents and stroke the heights extreme used cases on thatthe real can bepiles considered. are modelled The as hammer closely properties as possible and within stroke the heightswave equation used on the analysis real piles to ensure are modelled a fair prediction as closely in as each possible case. within The results the wave are equationshown in analysis Figure 6a,b to ensurefor pile a fair S1 and prediction S2, respectively. in each case. These The results aresuggest shown that in the Figure IC-056a,b and for Fugro-05 pile S1 and unplugged S2, respectively. models Theseprovide results a reasonable suggest that prediction the IC-05 (though and Fugro-05 slight under-prediction) unplugged models whereas provide the a reasonable IC-05 plugged prediction model (thoughprovides slight a slight under-prediction) over-prediction whereas in estimated the IC-05 blow plugged counts, model for both provides S1 and a slight S2. The over-prediction UWA-05 and inFugro-05 estimated plugged blow counts, models for appear both S1 to and over-predict S2. The UWA-05 the blow and counts Fugro-05 required plugged to modelsinstall the appear piles. to It over-predictshould be noted the blow that counts an over-prediction required to install is desireable the piles. as Itit shouldis considered be noted conservative that an over-prediction in respect of pile is desireabledriving. Even as it isthough considered the CPT conservative profiles are in relatively respect of uniform pile driving. across Even the thoughsite, since the the CPT average proﬁles profile are relativelyis used in uniform the driveability across the predictions, site, since the this average is a potential proﬁle is source used inof thesome driveability of the disparity predictions, evident. this is a potential source of some of the disparity evident.

(a) (b)

FigureFigure 6. 6.Predicted Predicted blow blow counts counts against against recorded recorded data data for unmodifiedfor unmodified approaches. approaches. (a) Pile (a) S1,Pile (b S1,) Pile (b) S2. Pile S2. 3.3. Comparison of Modified CPT-Based Approaches Piles S1 and S2 3.3.In Comparison this section, of theModified static capacityCPT-Based approaches Approaches from Piles Section S1 and 2.3 S2 are modified to account for the actual base resistanceIn this section, experienced the static during capacity driving approaches and the influencefrom Section of pile 2.3 ageingare modified on the to shaft account resistance. for the Sectionactual 2.4 base details resistance the modifications experienced adopted. during Thedrivin baseg and resistance the influence estimate of for pile the UWA-05ageing on method the shaft is updatedresistance. by applying Section 2.4 a base details resistance-settlement the modifications modeladopted. [16 ]The as discussedbase resistance in Section estimate 2.4.2. for During the UWA- pile driving,05 method the pile is updated tip displacement by applying experienced a base resistan during eachce-settlement blow will model be lower [16] than as thediscussed failure criterionin Section typically2.4.2. During adopted pile by driving, the UWA-05 the pile approach, tip displacement 0.1D. It is experienced therefore necessary during toeach reduce blow the will estimate be lower of thethan actingthe failure resistance criterion mobilized typically under adopted each blow. by the In UWA- the present05 approach, study, the 0.1D. actual It is displacements therefore necessary for each to hammerreduce blowthe estimate are back-calculated of the acting from resistance the blow mobi countslized recordedunder each at Blessington.blow. In theThe present residual study, base the stressesactual aredisplacements initially assumed for each to behammer zero. The blow results are ba ofck-calculated applying the base-settlementfrom the blow counts model arerecorded shown at in Figure7 for Pile S1 and Figure8 for Pile S2. Figures7 and8 show the base resistance-settlement curves at various depths of the penetration of piles S1 and S2, respectively. Both piles exhibit similar behavior. The initial linear portion of each curve is governed by Equation (16), which is modelled using the small-strain stiffness properties (E0) of the deposit. The base resistance value from the UWA-05 approach (qb01,UWA) is used as the limiting resistance when wb exceeds the failure criterion of 0.1D. The pile tip displacement per blow (averaged Energies 2020, 13, x FOR PEER REVIEW 12 of 19 Energies 2020, 13, x FOR PEER REVIEW 12 of 19

Blessington. The residual base stresses are initially assumed to be zero. The results of applying the Blessington. The residual base stresses are initially assumed to be zero. The results of applying the base-settlement model are shown in Figure 7 for Pile S1 and Figure 8 for Pile S2. base-settlement model are shown in Figure 7 for Pile S1 and Figure 8 for Pile S2. Figures 7 and 8 show the base resistance-settlement curves at various depths of the penetration Figures 7 and 8 show the base resistance-settlement curves at various depths of the penetration of piles S1 and S2, respectively. Both piles exhibit similar behavior. The initial linear portion of each of piles S1 and S2, respectively. Both piles exhibit similar behavior. The initial linear portion of each curve is governed by Equation (16), which is modelled using the small-strain stiffness properties (E0) Energiescurve is2020 governed, 13, 3128 by Equation (16), which is modelled using the small-strain stiffness properties12 of(E 190) of the deposit. The base resistance value from the UWA-05 approach (,) is used as the limiting of the deposit. The base resistance value from the UWA-05 approach (,) is used as the limiting resistance when wb exceeds the failure criterion of 0.1D. The pile tip displacement per blow (averaged resistance when wb exceeds the failure criterion of 0.1D. The pile tip displacement per blow (averaged into 0.5 m layers) and normalized by the pile diameter (wb/D) during pile driving is shown as the red intointo 0.50.5 mm layers)layers) andand normalizednormalized byby thethe pilepile diameterdiameter ((wbb/D/D)) duringduring pilepile drivingdriving isis shownshown asas thethe redred dot for each layer of penetration. Almost all normalized displacements per blow in each layer (wb/D) dot for each layer of penetration.penetration. Almost all normalized displacements per blow in each layer ( wb/D/D) occur at the second linearized stage of the base settlement-resistance model for both piles. wb/D values occur at the second linearized stage of the base settlement-resistance modelmodel forfor bothboth piles.piles. wbb/D/D values decreased as the blow counts increased at deeper pile penetrations for both piles. The displacement decreased asas thethe blowblow countscounts increasedincreased atat deeperdeeper pilepile penetrationspenetrations forfor bothboth piles.piles. TheThe displacementdisplacement experienced during driving is less than the failure criterion of 0.1D except over the first 1.5 m for pile experienced duringduring drivingdriving isis lessless than the failure criterioncriterion ofof 0.1D0.1D exceptexcept overover thethe ﬁrstfirst 1.51.5 mm for pile S1, and 1 m for pile S2, when the piles are fully coring (IFR = 1). The resistance mobilized exceeds the S1, and 11 m for pile S2, whenwhen thethe pilespiles areare fullyfully coringcoring (IFR(IFR == 1). The resistance mobilized exceeds the estimate due to the initial soft response (low blow counts) experienced upon initial installation. estimate duedue toto thethe initialinitial softsoft responseresponse (low(low blowblow counts)counts) experiencedexperienced uponupon initialinitial installation.installation.

Figure 7. Application of base-settlement model to Pile S1. Figure 7.7. Application of base-settlement model to Pile S1.

Figure 8. Application of base-settlement model to Pile S2. Figure 8. Application of base-settlement model to Pile S2.

In addition to to modifying modifying the the base base resistance resistance to toaccount account for for the the actual actual mobilized mobilized resistance, resistance, the In addition to modifying the base resistance to account for the actual mobilized resistance, the theinfluence influence of pile of pile ageing ageing onon the the shaft shaft resistance resistance is is also also incorporated. incorporated. Based Based on on normalized ageingageing influence of pile ageing on the shaft resistance is also incorporated. Based on normalized ageing curves derivedderived fromfrom ageingageing studies, studies, the the shaft shaft capacity capacity as as derived derived by by each each method method is reducedis reduced to 69%to 69% of curves derived from ageing studies, the shaft capacity as derived by each method is reduced to 69% itsof its value value to accountto account for for the the fact fact that that ageing ageing would would have have occurred occurred and and each each of the of designthe design approaches approaches was of its value to account for the fact that ageing would have occurred and each of the design approaches developedwas developed based based on load on tests load conducted tests conducted 10–30 days 10–30 after days installation, after installation, see Section see 2.4.1 Section. The driveability 2.4.1. The was developed based on load tests conducted 10–30 days after installation, see Section 2.4.1. The resultsdriveability using results the modified using the axial modified static capacityaxial static approaches capacity approaches are shown inare Figure shown9a,b in Figures for pile S19a,b and for driveability results using the modified axial static capacity approaches are shown in Figures 9a,b for S2, respectively. The IC-05 and Fugro-05 unplugged models both under-predict the blow-counts required to install the piles. In their unmodified form, they provided a closer estimate, as observed in Figure6 (though still an under-prediction). The IC-05 plugged models provide a better estimate in their modified form (Figure9) than in their unmodified form (Figure6). In Figure6, the UWA-05 and Fugro-05 plugged models significantly over-predicted the blow-counts, however in their modified form Energies 2020, 13, x FOR PEER REVIEW 13 of 19

pile S1 and S2, respectively. The IC-05 and Fugro-05 unplugged models both under-predict the blow- counts required to install the piles. In their unmodified form, they provided a closer estimate, as observed in Figure 6 (though still an under-prediction). The IC-05 plugged models provide a better Energiesestimate2020, in13 ,their 3128 modified form (Figure 9) than in their unmodified form (Figure 6). In Figure13 of6, 19the UWA-05 and Fugro-05 plugged models significantly over-predicted the blow-counts, however in intheir Figure modified9, they provide form in a Figure closer prediction.9, they provide The UWA-05a closer prediction. model in particular The UWA-05 provides model a much in particular closer estimate,provides as a a much result closer of altering estimate, the base as a resistanceresult of alte to accountring the for base the resistance actual mobilized to account resistance. for the actual It is noteworthymobilized thatresistance. the modiﬁed It is noteworthy UWA-05 approachthat the mo slightlydified over-predictsUWA-05 approach the blow slightly counts over-predicts in this paper, the whereasblow counts when appliedin this paper, to monopiles whereas with when larger applied diameters to monopiles in Byrne with et al. larger [1], an diameters under-prediction in Byrne was et al. observed[1], an under-prediction for the cases considered. was observed As mentioned for the previously,cases considered. an over-prediction As mentioned is considered previously, desirable. an over- Onceprediction more, since is considered the SRD were desirable. derived Once from more, the average since the CPT SRD proﬁle were across derived the site,from this the may average account CPT forprofile some across of the variabilitythe site, this in may the predictions. account for some of the variability in the predictions.

(a) (b)

FigureFigure 9. 9.Predicted Predicted blow blow counts counts against against recorded record dataed data for modifiedfor modified approaches. approaches. (a) Pile (a) S1,Pile (b S1,) Pile (b) S2. Pile S2. 3.4. Incorporation of Residual Base Stresses 3.4.The Incorporation presence of of residualResidual baseBase stressesStresses could influence the predicted blow counts required to install piles byThe changing presence the proportionof residual ofbase the resistancestresses could acting in atfluence the base the relative predicted to the blow shaft incounts the subsequent required to driveabilityinstall piles analyses. by changing The the base proportion resistance-settlement of the resistance model acting in Figure at the2 baseallows relative the incorporation to the shaft in of the thesesubsequent potential driveability residual stresses, analyses which. The are presentbase resistance-settlement as a result of the unloading model caused in Figure between 2 allows hammer the blows.incorporation These residual of these stresses potential must residual be overcome stresses, under which the are action present of subsequentas a result hammerof the unloading blows priorcaused to inducing between any hammer net downward blows. These movement. residual In thisstresses paper, must the basebe overcome resistance-settlement under the action model of wassubsequent applied to hammer the UWA-05 blows approach prior to toinducing estimate any the net likely downward mobilised moveme base resistancent. In this under paper, hammer the base driving.resistance-settlement In this section, model this estimate was applied is further to the modified UWA-05 to account approach for to potential estimate residual the likely stresses. mobilised baseEstimating resistance the under amount hammer of residual driving. stress In this potentially section, this contributing estimate is to further the resistance modified is complicated.to account for Inpotential this study, residual the procedure stresses. used in Byrne et al. [1] is adopted, whereby residual stresses were specified as percentagesEstimating of the the CPT amountqc value of residual at various stress depths. potentially Residual contributing base stresses to the varying resistance from is 1% complicated. qc to 10% qcInthat this are study, in keeping the procedure with values used measured in Byrne during et al. the [1] installation is adopted, of whereby Pile S6 were residual added stresses to the toewere resistancespecified estimated as percentages by the of modified the CPT UWA-05qc value at approach. various depths. The results Residual are shown base stresses in Figure varying 10a,b from for piles1% S1qc to and 10% S2, q respectively.c that are in keeping with values measured during the installation of Pile S6 were added to theFor toe increasing resistance amounts estimated of residualby the modified base stress UW addedA-05 approach. to the piles, The the results estimated are shown blow in counts Figure required10a,b for to piles install S1 the and piles S2, respectively. increases. Unlike in the study of Byrne et al. [1], which considered larger diameter monopiles mostly with diameters of 4.2 m, the addition of residual base stresses causes the UWA-05 prediction to deviate away from the measured blow count response. This highlights the uncertainty in the estimation of the influence of residual stresses on the response to driving. It is possible that residual stresses have a greater impact on larger diameter piles, such as those studied Energies 2020, 13, x FOR PEER REVIEW 14 of 19

For increasing amounts of residual base stress added to the piles, the estimated blow counts required to install the piles increases. Unlike in the study of Byrne et al. [1], which considered larger diameter monopiles mostly with diameters of 4.2 m, the addition of residual base stresses causes the UWA-05 prediction to deviate away from the measured blow count response. This highlights the Energiesuncertainty2020, 13 in, 3128 the estimation of the influence of residual stresses on the response to driving.14 ofIt 19is possible that residual stresses have a greater impact on larger diameter piles, such as those studied in Byrne et al. [1], than on the pile geometries in the present study. This requires further research to ininvestigate Byrne et al.the [ 1underlying], than on the mechanisms. pile geometries in the present study. This requires further research to investigate the underlying mechanisms.

(a) (b)

Figure 10. Predicted blow counts against recorded data for UWA-05 modified approach with varying Figure 10. Predicted blow counts against recorded data for UWA-05 modified approach with varying residual base stresses. (a) Pile S1, (b) Pile S2. residual base stresses. (a) Pile S1, (b) Pile S2. 4. Parameter Study 4. Parameter Study In this section, a parameter study is undertaken to highlight the relative importance of the parametersIn this adoptedsection, ina theparameter wave equation study is analyses undertaken on the to resultinghighlight predictions. the relative The importance purpose ofof thisthe sectionparameters is to highlightadopted in the the importance wave equation of accurate analyses information on the resulting to ensure predictions. that predictions The purpose are as fairof this as possiblesection is for to proposedhighlight installationsthe importance ofpiles. of accurate information to ensure that predictions are as fair as possible for proposed installations of piles. 4.1. Influence of Damping 4.1. Influence of Damping In a driveability analysis, dynamic forces and viscous rate effects are represented by damping values.In Thesea driveability values vary analysis, depending dynamic on soil forces type and and viscous in general rate standard effects valuesare represented are adopted. by However,damping thisvalues. parameter These values can have vary a relatively depending large on influence soil type on and the in predicted general blowstandard counts, values as demonstrated are adopted. herein.However, For this the UWA-05parameter method, can have the a prescribed relatively dampinglarge influence values on for the sand predicted are 0.25 sblow/m and counts, 0.5 s/ m,as respectivelydemonstrated for herein. skin and For toethe damping.UWA-05 method, Figure 11 the shows prescribed the results damping of varying values thefor sand damping are 0.25 values s/m adoptedand 0.5 fors/m, the respectively prediction offor blow skin counts and fortoe piledampin S2. Figureg. Figure 11a shows11 shows the sensitivitythe results of of the varying prediction the todamping variations values in skinadopted damping for the and prediction Figure 11of bblow shows counts the for sensitivity pile S2. Figure for variations 11a shows in the toe sensitivity damping. Inof boththe prediction plots, the redto variations lines represent in skin the damping standard and values Figure adopted 11b shows and is thethe samesensitivity data as for that variations presented in intoe Figure damping.6b. Changes In both plots, in toe the damping red lines result represent in a prediction the standard of greatervalues adopted blows than and changes is the same in skin data damping,as that presented highlighting in Figure that 6b. the Changes dissipation in toe of energydamping at theresult base in resultsa prediction in a significant of greater increaseblows than in thechanges required in skin blows damping, to install highlighting a pile. This that is notthe surprising,dissipation of as energy the base at resistancethe base results tends in to a govern significant the pileincrease driveability in the required as observed blows in theto install previous a pile. analyses. This isThe not esurprising,ffect of changing as the thebase damping resistance parameter tends to isgovern not constant the pile asdriveability the pile is as driven observed but leadsin the toprev increasedious analyses. predicted The blow effect counts of changing for deeper the damping depths. Atparameter the final is penetration not constant (7 as m), the anpile increase is driven of bu 50%t leads in theto increased skin and predicted toe damping blow values counts relativefor deeper to their nominal specified values (marked as UWA on the plot legends) increases the predicted blow counts by 16% and 22%, respectively. Generally, the blow counts and damping value are positively correlated. Soil damping in the wave equation analysis represents the energy loss within the soil at the pile-soil interface during pile driving, hence it is sensible that higher damping values lead to an Energies 2020, 13, x FOR PEER REVIEW 15 of 19

depths. At the final penetration (7 m), an increase of 50% in the skin and toe damping values relative to their nominal specified values (marked as UWA on the plot legends) increases the predicted blow counts by 16% and 22%, respectively. Generally, the blow counts and damping value are positively Energiescorrelated.2020, 13Soil, 3128 damping in the wave equation analysis represents the energy loss within the 15soil of 19at the pile-soil interface during pile driving, hence it is sensible that higher damping values lead to an increased number of blows to install a pile. This brief study highlights the sensitivity of this parameter increased number of blows to install a pile. This brief study highlights the sensitivity of this parameter to obtaining accurate predictions. to obtaining accurate predictions.

(a) (b)

FigureFigure 11.11. Predicted blow counts for pile S2 usingusing thethe UWA-05UWA-05 approachapproach withwith modiﬁedmodified dampingdamping parameters.parameters. ((aa)) modifyingmodifying skinskin damping,damping, ((bb)) modifyingmodifying toetoe damping.damping.

4.2.4.2. InfluenceInfluence of Quake QuakeQuake isis thethe termterm usedused toto describedescribe thethe displacementdisplacement requiredrequired toto achieveachieve yield,yield, andand standardstandard valuesvalues forfor thisthis parameterparameter havehave beenbeen putput forwardforward forfor didifferentfferent soilsoil typestypes [2[2,5,8,29].,5,8,29]. IncreasingIncreasing thethe quakequake valuevalue isis equivalentequivalent toto extendingextending thethe soilsoil elasticelastic displacementdisplacement rangerange priorprior toto yield.yield. TheThe quakequake valuevalue suggestedsuggested byby thethe UWA-05UWA-05 methodmethod isis 2.52.5 mmmm bothboth forfor thethe skinskin andand toe.toe. FigureFigure 1212 investigatesinvestigates howhow quakequake valuesvalues influenceinfluence thethe predictedpredicted blowblow countscounts fromfrom thethe wavewave equationequation analysis.analysis. TheThe datadata correspondingcorresponding toto unmodifiedunmodified UWA-05UWA-05 approachapproach appliedapplied toto pilepile S2S2 isis usedused inin thisthis analysis,analysis, similarsimilar toto SectionSection 4.14.1.. ChangesChanges inin toetoe quake, quake, Figure Figure 12 12b,b, have have a a lesser lesser influence influence on on the the response response than than changes changes in thein the skin skin quake, quake, Figure Figure 12a. 12a. Increasing Increasing theskin the andskin toeand quake toe quake by 50% by relative 50% relative to their to standard their standard values asvalues prescribed as prescribed by UWA-05 by UWA-05 results in results an increase in an increase in predicted in predicted blow counts blow for counts 7 m penetration for 7 m penetration of 41.5% andof 41.5% 30.5% and respectively. 30.5% respectively. TheThe resultsresults ofof thisthis analysisanalysis andand thatthat ofof changing thethe damping parametersparameters inin thethe previousprevious sectionsection suggestsuggest thatthat thesethese parametersparameters shouldshould reallyreally bebe obtainableobtainable basedbased onon measurablemeasurable soilsoil data,data, andand thethe useuse ofof standardstandard valuesvalues isis potentiallypotentially misleading.misleading. GivenGiven thethe large change in prediction caused byby changing thethe dampingdamping andand quake,quake, thisthis isis aa potentialpotential sourcesource ofof significantsignificant disparitydisparity betweenbetween actualactual pilepile drivingdriving recordsrecords andand predictionspredictions mademade usingusing standardstandard values.values.

EnergiesEnergies 20202020,, 1313,, 3128x FOR PEER REVIEW 1616 ofof 1919

(a) (b)

FigureFigure 12. Predicted blow counts for pile S2 S2 using using the the UWA-05 UWA-05 approach with modified modiﬁed quake values. ((aa)) modifyingmodifying skinskin quake,quake, ((bb)) modifyingmodifying toetoe quake.quake.

4.3.4.3. InfluenceInfluence of Stroke Height and HammerHammer EEfficiencyfficiency TheThe heightheight ofof hammerhammer stroke influencesinfluences thethe energyenergy appliedapplied toto drivedrive aa pile,pile, andand thereforetherefore hashas anan impactimpact onon predictedpredicted blowblow countscounts requiredrequired toto drivedrive aa pile.pile. Figure 13 showsshows thethe blowblow countscounts/0.25/0.25 mm resultingresulting fromfrom varyingvarying thethe hammerhammer strokestroke heightheight forfor thethe analysisanalysis usingusing thethe unmodifiedunmodified UWA-05UWA-05 approachapproach appliedapplied toto pilepile S2.S2. Increasing the stroke heightheight decreasesdecreases thethe predictedpredicted blowsblows requiredrequired toto installinstall thethe pile.pile. The estimated blow counts at the finalfinal pilepile penetrationpenetration decreasesdecreases byby 61%61% whenwhen thethe hammerhammer strokestroke heightheight isis increasedincreased fromfrom thethe valuevalue usedused (0.3(0.3 m)m) byby 50%50% toto 0.450.45 m.m. ThisThis suggestssuggests aa significantsignificant sensitivitysensitivity to to the the stroke stroke height height used andused highlights and highlights the importance the importance of accurately of measuringaccurately thismeasuring parameter this onparameter site to ensure on site the to predictionsensure the predictions made are sensible made are and sensible reasonable. and reasonable. Other changes Other in strokechanges height in stroke and theirheight influence and their on influence the predicted on the results predicted can be results seen in can Figure be seen 13a. in It Figure is important 13a. It tois noteimportant that installing to note that a pile installing with excessive a pile with energy excess canive lead energy to pile can damage, lead to pile whereas damage, driving whereas a pile driving using aa lowpile strokeusing heighta low canstroke lead height to premature can lead refusal. to premature Pile stresses refusal. should Pile stresses be explicitly should analyzed be explicitly when proposinganalyzed when changes proposing to hammer changes stroke to height.hammer stroke height. HammerHammer eefficiencyfficiency accountsaccounts for thethe energyenergy losseslosses thatthat cannotcannot bebe calculatedcalculated directlydirectly duringduring thethe pilepile drivingdriving process.process. The standard value of thethe hammerhammer eefficiencyfficiency depends on thethe typetype ofof hammerhammer adopted.adopted. TheThe installationinstallation of of pile pile S2 S2 at at the the Blessington Blessington site site used used a hydraulic a hydraulic impact impact hammer, hammer, which which has anhas e ffianciency efficiency value value of 0.8 of according 0.8 according to the recommendationsto the recommendations of the pile of driveabilitythe pile driveability analysis softwareanalysis manualsoftware [ 18manual]. Figure [18]. 13 Figureb shows 13b that shows increasing that increasing hammer hammer efficiency efficiency results results in lower in lower predicted predicted blow countsblow counts required required to install to install a pile. a This pile. mainly This mainly influences influences the predicted the predicted blow counts blow counts at depth at wheredepth diwherefferences differences in efficiency in efficiency have a largerhave influencea larger in onfluence the predicted on the predicted results overall. results A overall. 10% reduction A 10% inreduction efficiency in (fromefficiency 80% to(from 70%) 80% results to 70%) in a 76%results increase in a 76% in the increase predicted in the blow predicted counts requiredblow counts at a penetrationrequired at a depth penetration of 7 m. depth Changes of 7 inm. hammerChanges einffi cientlyhammer have efficiently the largest have influencethe largest on influence the results on outthe ofresults the various out of parametersthe variousinvestigated parameters ininvestigat this section,ed in highlighting this section, the highlighting paramount the importance paramount of accuratelyimportance estimating of accurately this es parameter.timating this parameter.

EnergiesEnergies2020 2020,,13 13,, 3128x FOR PEER REVIEW 1717 ofof 1919

(a) (b)

FigureFigure 13.13. Predicted blow counts for pile S2 using the UWA-05 approach with modifiedmodified strokestroke heightsheights andand eefficiencies.fficiencies. ((aa)) modifyingmodifying strokestroke height,height, ((bb)) modifyingmodifying hammerhammer eefficiency.fficiency.

5.5. Conclusions InIn thisthis paper,paper, anan investigationinvestigation ofof thethe applicabilityapplicability ofof CPT-basedCPT-based axialaxial staticstatic capacitycapacity approachesapproaches forfor pilespiles to to estimating estimating pile pile driveability driveability is conducted. is conducted. Specifically, Specifically, the study the study investigates investigates various various factors thatfactors can that influence can influence the prediction the ofprediction pile driveability, of pile anddriveability, how these and factors how a fftheseect the factors prediction affect of thethe blowprediction counts of requiredthe blow to counts install required a pile. Theto install results a pile. of field The tests results on openof field ended tests steelon open piles ended driven steel at Blessington,piles driven Ireland)at Blessington, are used Ireland) to appraise are used the methodsto appraise postulated. the methods postulated. ThreeThree CPT-basedCPT-based axial axial capacity capacity methods methods are investigated,are investigated, namely namely the IC-05, the UWA-05 IC-05, UWA-05 and Fugro-05 and approaches.Fugro-05 approaches. The methods The are methods applied are in theirapplied unmodified in their formunmodified and with form appropriate and with modifications appropriate tomodifications match deviations to match in deviations the expected in the behavior expected during behavior driving during as driving opposed as to opposed under staticto under loading. static Modificationsloading. Modifications include changing include changing the base resistancethe base resistance estimate proposedestimate inproposed the UWA-05 in the approach, UWA-05 incorporatingapproach, incorporating the influence the of influence ageing (capacity of ageing increases (capacity over increases time), whichover time), require which a reduction require in a shaftreduction capacity in shaft to account capacity for to theaccount lower for shaft the lower resistance shaft mobilized resistance duringmobilized installation, during installation, and assessing and theassessing influence the ofinfluence residual of base residual stresses. base Installation-related stresses. Installation-related data (blow data counts) (blow measured counts) frommeasured two fieldfrompiles two field installed piles at installed Blessington at Blessington are used. Forare theused. unmodified For the unmodified approaches, approaches, the IC-05 andthe IC-05 Fugro-05 and unpluggedFugro-05 unplugged models provide models a provide reasonable a reasonable prediction prediction of the blow of counts the blow with counts some with under-prediction, some under- whereasprediction, the whereas IC-05 plugged the IC-05 model plugged slightly model over-predicts slightly over-predicts the blow counts. the blow For the counts. modified For the approaches, modified theapproaches, IC-05 and the Fugro-05 IC-05 unpluggedand Fugro-05 models unplugged further under-predictmodels further the under-predict blow counts thanthe blow their unmodifiedcounts than counterparts.their unmodified The IC-05counterparts. plugged model,The IC-05 on theplugg othered hand,model, provides on the aother better hand, estimate provides in its modified a better formestimate than in its its unmodifiedmodified form form. than The its unmodifi modifieded UWA-05 form. The approach modified provides UWA-05 a approach substantially provides closer a estimatesubstantially in its closer modified estimate form thanin its unmodified, modified form as a than result unmodified, of altering the as basea result resistance of altering to be the more base in keepingresistance with to thebe more actual in expected keeping mobilization with the ac undertual expected hammer mobilization impacts. However, under ithammer leads to impacts. a slight over-predictionHowever, it leads in theto a blowslight counts over-prediction for both piles, in the which blow contradictscounts for both the findings piles, which of a previouscontradicts study the (albeitfindings this of previous a previous study study was (albeit on larger this previous diameter study piles). was Furthermore, on larger diameter the incorporation piles). Furthermore, of residual basethe incorporation stresses withthe of modifiedresidual base UWA-05 stresses approach with causethe modified a further UWA-05 deviation approach in the predicted cause response,a further suggestingdeviation in these the predicted might not response, be so critical suggesting for smaller thes diametere might pilesnot be than so critical evident for in studiessmaller ondiameter larger diameterpiles than piles evident conducted in studies previously. on larger diameter piles conducted previously.

Energies 2020, 13, 3128 18 of 19

Finally, a parametric study is conducted where the parameters used in the wave equation analysis are varied to assess their relative influence on predicted blow counts. The data from the unmodified UWA-05 prediction for pile S2 are used and the quake, damping, stroke height and hammer efficiency parameters are varied. Results suggest that hammer efficiency has the largest influence on the predicted results highlighting the importance of accurate characterization of this information for use in driveability analyses. The present study has investigated the various acting components contributing to pile driveability analyses, and has highlighted the complex and inter-related nature of the various parameters governing the predicted response. The eventual goal could be to derive a pile driveability approach based directly on an empirical correction to CPT data, without the requirement for estimating residual stresses, damping and quake data, or even without the need to undertake a wave equation analysis. It is anticipated that this would be a significant challenge, however it would solve the issue that, at present, there are potentially too many variables contributing to the response. The ability to obtain a fair estimate of pile driveability remains a challenge.

Author Contributions: Conceptualization, K.G.; methodology, L.J.P.; software, P.G.; validation, P.G., K.G. and L.J.P.; formal analysis, P.G.; investigation, P.G., K.G. and L.J.P.; resources, K.G.; data curation, K.G.; writing—original draft preparation, L.J.P. and P.G.; writing—review and editing, K.G.; visualization, P.G.; supervision, L.J.P. and K.G.; project administration, K.G.; funding acquisition, K.G. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Acknowledgments: Gandina’s work is a development of the earlier work by Tiernan Byrne whose contribution is acknowledged together with members of the Geotechnical Research Group formerly at University College Dublin for testing, data acquisition, software and organization. The first author wishes to acknowledge the Faculty of Engineering, University of Nottingham, for travel funding to enable this collaboration. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Byrne, T.; Gavin, K.; Prendergast, L.J.; Cachim, P.; Doherty, P.; Pulukul, S.C. Performance of CPT-based methods to assess monopile driveability in North Sea sands. Ocean Eng. 2018, 166, 76–91. [CrossRef] 2. Schneider, J.; Harmon, I. Analyzing drivability of open ended piles in very dense sands. J. Deep. Found. Inst. 2010, 4, 32–44. [CrossRef] 3. Wu, W.H.; Prendergast, L.J.; Gavin, K. An iterative method to infer distributed mass and stiffness profiles for use in reference dynamic beam-Winkler models of foundation piles from frequency response functions. J. Sound Vib. 2018, 431, 1–19. [CrossRef] 4. Prendergast, L.J.; Wu, W.H.; Gavin, K. Experimental application of FRF-based model updating approach to estimate soil mass and stiffness mobilised under pile impact tests. Soil Dyn. Earthq. Eng. 2019, 123, 1–15. [CrossRef] 5. Stevens, R.; Wiltsie, E.; Turton, T. Evaluating drivability for hard clay, very dense sand, and rock. In Proceedings of the Offshore Technology Conference, Houston, TX, USA, 3–6 May 1982. 6. Toolan, F.E.; Fox, D.A. Geotechnical planning of piled foundations for offshore platforms. Proc. Inst. Civ. Eng. 1977, 62, 221–244. [CrossRef] 7. Semple, R.; Gemeinhardt, J. Stress history approach to analysis of soil resistance to pile driving. In Proceedings of the Offshore Technology Conference, Houston, TX, USA, 4–7 May 1981. 8. Alm, T.; Hamre, L. Soil model for pile driveability predictions based on CPT interpretations. In Proceedings of the International Conference on Soil Mechanics and Geotechnical Engineering, Istanbul, Turkey, 27–31 August 2001; pp. 1297–1302. 9. Jardine, R.J.; Chow, F.C.; Overy, R.F.; Standing, J. ICP Design Methods for Driven Piles in Sands and Clays; Institution of Civil Engineering (ICE): London, UK, 2005. 10. Lehane, B.; Schneider, J.; Xu, X. The UWA-05 method for prediction of axial capacity of driven piles in sand. In Proceedings of the International Symposium on Frontiers in Offshore Geotechnics, Perth, Australia, 19 September 2005. Energies 2020, 13, 3128 19 of 19

11. Jardine, R.; Standing, J.; Chow, F. Some observations of the effects of time on the capacity of piles driven in sand. Geotechnique 2006, 55, 227–244. [CrossRef] 12. Gavin, K.; Kirwan, L.; Igoe, D. The effect of ageing on the axial capacity of piles in sand. Proc. Inst. Civ. Eng. Geotech. Eng. 2013, 166, 122–130. [CrossRef] 13. Karlsrud, K.; Jensen, T.G.; Lied, E.K.W.; Nowacki, F.; Simonsen, A.S. Significant ageing effects for axially loaded piles in sand and clay verified by new field load tests. In Proceedings of the Offshore Technology Conference, Houston, Texas, USA, 5–8 May 2014. 14. Kirwan, L. Investigation into Ageing Mechanisms for Axially Loaded Piles in Sand. Ph.D. Thesis, University College Dublin, Dublin, Ireland, 2015. 15. Gavin, K.; Jardine, R.J.; Karlsrud, K.; Lehane, B.M. The effects of pile ageing on the shaft capacity of offshore piles in sand. In Proceedings of the International Symposium Frontiers in Offshore Geotechnics (ISFOG), Oslo, Norway, 10–12 June 2015. 16. Gavin, K.; Lehane, B. Base load–displacement response of piles in sand. Can. Geotech. J. 2007, 44, 1053–1063. [CrossRef] 17. Xu, X. Investigation of the End Bearing Performance of Displacement Piles in Sand. Ph.D. Thesis, The University of Western Australia, Perth, Western Australia, Australia, 2007. 18. Pile Dynamics GRLWEAP Software Brochure. 2010. Available online: http://www.pile.com/wp-content/ uploads/2020/05/PDI-GRLWEAP-Brochure.pdf (accessed on 1 June 2020). 19. Kolk, H.J.; Baaijens, A.E.; Senders, M. Design criteria for pipe piles in silica sands. In Proceedings of the First International Symposium on Frontiers in Offshore Geotechnics, Perth, Australia, 19 September 2005. 20. Lehane, B.M.; Lim, J.K.; Carotenuto, P.; Nadim, F.; Lacasse, S.; Jardine, R.; van Dijk, B.F.J. Characteristics of Unified Databases for Driven piles. In Proceedings of the 8th International Conference of Offshore Site Investigation and Geotechnics OSIG, London, UK, 12–14 September 2017. 21. Gavin, K.; Igoe, D. A field investigation into the mechanisms of pile ageing in sand. Geotechnique 2019. [CrossRef] 22. Donohue, S.; Long, M.; Gavin, K.; O’Connor, P. Shear wave stiffness of Irish glacial till. In Proceedings of the International Conference of Site Characterisation ISC-2, Porto, Portugal, 19–22 September 2004; pp. 459–466. 23. Prendergast, L.J.; Hester, D.; Gavin, K.; O’Sullivan, J.J. An investigation of the changes in the natural frequency of a pile affected by scour. J. Sound Vib. 2013, 332, 6685–6702. [CrossRef] 24. Robertson, P. Soil classification using the cone penetration test. Can. Geotech. J. 1990, 27, 151–158. [CrossRef] 25. Schnaid, F.; Lehane, B.M.; Fahey, M. In situ test characterisation of unusual geomaterials. In Proceedings of the International Conference of Site Characterisation, Porto, Portugal, 19–22 September 2004; pp. 49–73. 26. Paik, K.; Salgado, R.; Lee, J.; Kim, B. Behavior of open- and closed-ended piles driven into sands. J. Geotech. Geoenviron. Eng. 2003, 296–306. [CrossRef] 27. Alawneh, A.S.; Malkawi, A.I.H. Estimation of post-driving residual stresses along driven piles in sand. Geotech. Test. J. 2000, 23, 313–326. 28. Lunne, T.; Robertson, P.K.; Powell, J.J.M. Cone Penetration Testing in Geotechnical Practice; Blackie Academic and Professional: London, UK, 1997. 29. Hirsch, T.J.; Carr, L.; Lowery, L.L. Pile driving analysis-wave equation user manual. In TTI Programme -Final Report; Federal Highway Administration (FHWA): Washington, DC, USA, 1976; Volume I–IV.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).