sign in sign up
<<
Home , Exponentiation

arXiv:1504.02332v1 [.data-an] 9 Mar 2015 .. . Wad4300 and kW 4.3 e.g., ~ ilcto fpyia ausa etradto n their and vector as values physical an of as tiplication Ω regard We values. value. physical same vi ofso ihtecrepnigrglraopera- regular corresponding the tions. with confusion avoid “ than together, more 3]. introduced little [2, with usually times are at in techniques them these omitting these Since of i.e., all unity, or to some setting units and the 2]; to [1, relevant constants field natural respective usually units, different base a of employing techniques: two following the of tion fntrluisa ela oefnaetlcnet of concepts fundamental some concept as algebra. the well linear introducing as pos- for units natural the formalism of discuss our we of Finally, merit level. sible vector axiomatic this an math- introduce on formalism’s space formally the and explore background transforma- we ematical transfor- basis addition, space), to this In dimension correspond In mations. systems the values. unit physical in between of tions (as thereby space and space vector [4] the vector Ref. to in it introduced extend space dimension algebra. the linear on builds only math- which transformations a and them introducing systems between by unit issue for them. this formalism apply address ematical to to how try and here validity We their about confused be hsclvle m 9 e,4 MeV, for 197 Examples fm, 1 [5]. are values quantity physical physical scalar mensionless, value eri rmuigi o lrt’ ae ednt vec- denote ( We symbols sake. encircled clarity’s here with for we units operations it formalism, tor base using our all complement from or easily refrain can some unity setting to of technique mentioned hsclvlecnb ersne ndffrn ways, different in represented be can value physical A . e eoetesto l oiieysge physical signed positively all of set the denote Ω Let h term The oti ups,w noprt ueia ausinto values numerical incorporate we purpose, this To oegnrlrmrso oain huhteafore- the Though notation: on remarks general Some I H ETRSAEO PHYSICAL OF SPACE VECTOR THE II. eest osbevleo iesoflo di- or dimensionful a of value possible a to refers ~ eoe h ulvector. null the denotes 0 aua units natural ASnmes 62.a 15.b 02.10.Ud 01.55.+b, 06.20.Fa, : PACS natural of concept systems. to the unit correspond introducing between for systems use unit find may different formalism As values. physical of in eepoetemteaia onain ftevco space vector the of foundations mathematical the explore We .Mkyoiz m .Py.4 3,1976, (3), 44 Phys. . Am. Maksymowicz, A. .INTRODUCTION I. J aua nt n h etrsaeo hsclvalues physical of space vector the and units Natural / VALUES s r w ersnain fthe of representations two are sal eest h combina- the to refers usually R vco pc,wt h mul- the with space, -vector ~ = hsclvalue Physical . 2, ” tdnsmay students 1”, = π ⊕ rtevleof value the or , ⊖ nvriyo on Germany Bonn, of University and rjust or ertAnsmann Gerrit ⊙ to ) n xedti omls otevco space vector the to formalism this extend and uhas such o xml,4 example, For xoetaina clrmlilcto,i.e.: , scalar as exponentiation 4 o with Ω for accordingly. of system avoided consequence are) unit a usually the not (and choosing be and applica- by can system in they unit up and formalism, the come our to these tied whether is A, tion III Sec. in discuss rprisi h process: the in properties . 2 omk lsdunder closed Ω make To ewl o esnta h etrsaeaim hold axioms vector-space the that reason now will We ⊕ • • • • • • (m α ic 1 Since Since rmtemlilcto fpyia values: physical of multiplication the from Distributivities: optblt fmultiplications: of Compatibility h elnme steiett fsaa multi- scalar of the is 1 1 plication: real The nes of inverse lmn of element omttvt n soitvt of associativity and Commutativity ( α x ⊙ π ffrn ae fti etrsae our space, vector this of bases ifferent ( nt n rnfrigpyia values physical transforming and units + ⊕ ⊕ ⊙ αβ ( α x x ⊕ β (( ( g=kg = kg x y ⊙ ⊕ fpyia iesosintroduced dimensions physical of ) ) ⊕ − and ⊕ ⊕ ⊙ ⊕ ⊙ x y 2) y . x ms 2 x x z ( = ) x x : : ⊕ x = = ⊙ = ) ⊙ ⊕ ⊙ under = = − 1 = π s1 is ) n N 2 and s)) x xy 1 x y n xlr oeo t fundamental its of some explore and − x x  α or xy x = αβ 2 = α = x ( + yz ) ∈ rte ihteeoeain is operations these with written 1 2 x xy α β ⊕ = = xx ⊙ 1 ,i.e., Ω, ( = ) = = ∀ is = = m = − x x ∀ ⊙ x x x 1 ∀ ,y x, x x β ∈ yx ehv oacp values accept to have we , o all for α α ,y z y, x, = 1 = xy − ·  x y ∀ s(2 is m 3 Ω; α 1 √ ∀ α β x ~ = 1. = 0 ) ∈ x ∈ = z .Hwvr sw will we as However, m. = = ,β α, ∈ y ∈ Ω; ( = ,i.e., Ω, α ∈ Ω. α α ∀ x ⊕ ~ o all for 0 Ω ,β α, ,y x, ⊙ Ω ⊙ ⊙ ∈ x α , ∈ x ( ⊕ ⊕ x x R β ⊕ ,teneutral the Ω, ∈ ∈ ∈ ⊕ N) ⊕ y ⊙ ⊖ r inherited are ) R Ω R x α β x ⊕ x ⊕ ) = ⊙ ⊙ ∈ (3 z x x y ,the Ω, ⊕ − 1 m). . 2

III. DIFFERENT UNIT SYSTEMS AND render the unit of length to be √qm. Using our formal- TRANSFORMING BETWEEN THEM ism and assuming that at least one basis element of a unit system is a positive number (usually 10), we can classify As the SI base units together with the scalar 10 are those unit systems for which the above requirement is ful- independent by means of multiplication and exponenti- filled: Suppose we have an established unit system corre- R ation, they are linearly independent and thus form in- sponding to the basis := a0,a1,...,an with a0 + A { } ∈ deed an algebraic basis of their span. For simplicity, and n N and this system fulfils the above requirement. ∈ we restrict the discussion to the subspace spanned by Suppose further that we want to introduce a new unit := 10, m, s, kg in the following examples. Using as system corresponding to the basis := b0,b1,...,bn S { } S R B { } an ordered basis, we can write for example and later use: with b0 + and span( ) = span( ). We can then ∈ B B A write the transformation TA from to and its 1 B A 10 = 10 =1 10 = (1, 0, 0, 0)S inverse as follows: ⊙ − 8 m log(3)+8 m 8.5 1 c 3 10 /s = 10 /s 10 ms ⋆ ⋆ ⋆ ≈ · ≈   =8.5 10 m ( 1) s=(8.5, 1, 1, 0)S ··· ⊙ ⊕ ⊕ − ⊙ − B 0 ~ −34.0 log(1.05)−34.0 TA =  . ; 1.05 10 Js = 10 Js  . ˆB  ≈ ·  . TA  ( 34.0) 10 2 m ( 2) s kg   ≈ − ⊙ ⊕ ⊙ ⊕ − ⊙ ⊕  0  = ( 34.0, 2, 1, 1)S − − ⋆ ⋆ ⋆ 1 MeV 1.6 10−13 J=10log(1.6)−13 J  ···  0 ≈ · A B −1 ( 12.8) 10 2 m ( 2) s kg TB = TA =  . −1 , ≈ − ⊙ ⊕ ⊙ ⊕ − ⊙ ⊕   . ˆB  = ( 12.8, 2, 2, 1) ,  . TA  S   − − (1)  0  where the subscript indicates the basis of the component representation. where ⋆ represents some . The leftmost In our formalism, different unit systems do correspond columns only contain non-zero entries in the first row to different bases. For example, one common system of because a0 = (log(a0)/ log(b0)) b0. ⊙ (partially) natural units used by high-energy physics cor- As fulfils the requirement, each bi for i 1 can be A ≥ responds to the basis = 10, c, ~, 1 MeV of span( ). expressed in units from without raising any ai for i 1 N { } S A ˆB ≥ Using Eq. 1, we obtain basis transformation matrices to a non- power and thus TA has only integer ele- from to and vice versa: ments. Analogously, fulfils the requirement if and only N S −1 B ˆB ˆB if TA has integer elements only. If TA having only −1 1 8.5 34.0 12.8 B − − integer elements is given, TˆA having only integer el- N  01 2 2    T ; S 0 1 1 2 ˆB ˆB ≈   ements is equivalent to detTA  = 1 and to TA being  00− − 1− 1  a unimodal matrix. −1 1 12.7 21.2 29.7 B B TˆA and TˆA correspond to transformation ma- S N −1  0 1 0 2    T = T . N S 0 1 1− 0 trices in the (or more precisely, the  ≈   Z  0 1 1 1  -module [6]) of physical dimensions introduced in − − Ref. [4]. With this, we can for example transform 1 fm to the above natural units, i.e., the basis (using T N ( 15, 1, 0, 0) ( 2.3, 1, 1, 1)): N IV. CONCLUSIONS S − ≈ − − We introduced the vector space of physical values and 1 fm = ( 15, 1, 0, 0) ( 2.3, 1, 1, 1) − S ≈ − − N thereby provided a mathematical foundation for unit sys- ~c ~c tems and transformations between them. From a practi- = 10−2.3 MeV ≈ 197 MeV cal perspective, our formalism allows transforming a rep- resentation of a physical value to a different unit system with one matrix–vector multiplication, using the same A. Avoiding non-integer powers of units matrix for a given pair of unit systems. In particular, we hope that our formalism has didactic A usual requirement for a unit system is that physi- values for introducing natural units as it not only ad- cally meaningful values can be represented without any dresses how to apply this technique but substantiates its unit being raised to a non-integer power. For example one mathematical foundations and validity. Though our for- would avoid using qm := m2 as a base unit, as this would malism only requires basic concepts from , 3 it may be challenging for some students due its unusual ioms was all we had to do to be able to safely apply vector addition and scalar multiplication. On the other well-known concepts from linear algebra to a new struc- hand, the latter may be beneficial if our vector space is ture. used as an example when teaching linear algebra, as it may help to illustrate how abstract, general and diverse the presented concepts are and act prophylactic against students conflating vector operations and regular opera- ACKNOWLEDGEMENTS tions. Furthermore, using transformations between unit systems as example for basis transformations may help overcoming certain difficulties as there is no straightfor- I am grateful to N. Ansmann, S. Bialonski and J. Wilt- ward canonical basis. Finally, our approach exemplifies ing for critical comments on earlier versions of the the value of axiomatic systems, as validating a few ax- manuscript.

[1] M. Planck, Annalen der Physik 306, 69 (1900). [5] Joint Committee for Guides in Metrology, rd [2] D. F. Bartlett, Am. J. Phys. 42, 148 (1974). International Vocabulary of Metrology, 3 ed. (2008). [3] E. A. Desloge, Am. J. Phys. 62, 216 (1994). [6] A Z-module is an that complies with [4] A. Maksymowicz, Am. J. Phys. 44, 295 (1976). the same axioms as a vector space, but with the scalars coming only from the of Z instead of a field.