Next Article in Journal
Stacking Machine Learning Model for the Assessment of R&D Product’s Readiness and Method for Its Cost Estimation
Previous Article in Journal
Minimizing Dependency Ratio in Spain through Demographic Variables
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Natural Approximation to the Complete Elliptic Integral of the First Kind

Department of Mathematics, Zhejiang Gongshang University, Hangzhou 310018, China
Mathematics 2022, 10(9), 1472; https://doi.org/10.3390/math10091472
Submission received: 8 March 2022 / Revised: 22 April 2022 / Accepted: 22 April 2022 / Published: 27 April 2022
(This article belongs to the Section Mathematics and Computer Science)

Abstract

:
Let K ( r ) be the complete elliptic integral of the first kind. Then, the inequality 2 K ( r ) / π > tanh 1 ( r ) / sin 1 r holds for all r ( 0 , 1 ) . This conclusion does not match those in the existing literature.

1. Introduction

Legendre’s incomplete elliptic integral of the first kind is defined by
F ( λ , k ) = 0 λ d t 1 t 2 1 k 2 t 2 .
The research on various series expansions and asymptotic approximations for the function F ( λ , k ) is in the ascendant, and interested readers can refer to literature [1,2,3,4,5,6,7,8,9,10,11,12,13,14].
This article focuses on Legendre’s complete elliptic integral of the first kind. For r ( 0 , 1 ) , Legendre’s complete elliptic integrals of the first kind (see [15,16]), denoted K ( r ) , is defined by,
K ( r ) = 0 π / 2 d t 1 r 2 sin 2 ( t ) ,
which is the particular case of the Gaussian hypergeometric function and can be a special power series ([17,18,19,20,21,22,23]):
K ( r ) = π 2 F 1 2 , 1 2 ; 1 ; r 2 = π 2 n = 0 1 2 n 2 n ! 2 r 2 n ,
where 1 2 n = 1 2 1 2 + 1 1 2 + 2 1 2 + n 1 .
Many researchers have obtained the upper and lower bounds of this special function ([24,25,26,27,28,29,30,31,32,33]). Since 1991, some scholars have considered using the logarithmic function related to r to approximate 2 K ( r ) / π . Carlson and Vuorinen [24], Vamanamurhty and Vuorinen [25], Qiu and Vamanamurthy [26] and Alzer [27] proved that the double inequalities
A r = : log r r 1 < K ( r ) < π 2 log r r 1 ,
B r = : 1 + π 4 log 2 1 r 2 log 4 r < K ( r ) < 1 + 1 4 r 2 log 4 r
hold for all r > 0 , where r = 1 r 2 .
Anderson, Vamanamurhty and Vuorinen [28] found and proved that the double inequality
C r = : π 2 tanh 1 ( r ) r 1 / 2 < K ( r ) < π 2 tanh 1 ( r ) r
holds for all r ( 0 , 1 ) , where tanh 1 ( r ) = log [ ( 1 + r ) / ( 1 r ) ] / 2 is the inverse hyperbolic tangent function.
The lower bound in (4) was improved by Alzer and Qiu [29], and Yang, Song and Chu [30] independently as follows:
K ( r ) > π 2 tanh 1 ( r ) r 3 / 4 : = D r
holds for all r ( 0 , 1 ) .
In order to find a new bound function, we first analyze the special function 2 K ( r ) / π in the case of two breakpoints:
2 π K ( 0 + ) = 2 π 0 π / 2 d t = 1 , 2 π K ( 1 ) = .
The last fact is based on the following reasoning: when doing transformation sin t = x , we have
2 π K ( 1 ) = 2 π 0 π / 2 d t 1 sin 2 ( t ) = 2 π 0 π / 2 cos t d t cos 2 t = 2 π 0 π / 2 d sin t 1 sin 2 t = 2 π 0 π / 2 1 2 1 1 + sin t + 1 1 sin t d sin t = 2 π lim x 1 1 2 ln 1 + x 1 x = 2 π lim x 1 tanh 1 ( x ) = 2 π lim r 1 tanh 1 ( r ) .
We then consider
lim r 1 1 sin 1 r = 2 π ,
try to use the function tanh 1 ( r ) / sin 1 ( r ) as the bound for the function 2 K ( r ) / π , and draw the following conclusion.
Theorem 1.
The inequality
2 π K ( r ) > tanh 1 r sin 1 r
or
K ( r ) > π 2 tanh 1 r sin 1 r : = G ( r )
holds for all r ( 0 , 1 ) .
It must be pointed out that the lower bound for the function K ( r ) given in this paper is concise and accurate. In order to express this advantage, we specially list the following two absolute error tables, one is the error table between the function K ( r ) and the lower bound G ( r ) given in this paper (see Table 1), and the other is the error tables between the lower bound G ( r ) given in this paper and the ones A ( r ) , B ( r ) , D ( r ) (see Table 2), which given in [24,25,26,27,29,30], respectively.
The latter table shows the fact that the new lower bound G ( r ) for K ( r ) is better than the one A ( r ) , and is more advantageous than the one B ( r ) near r = 0 while G ( r ) is more advantageous than the one D ( r ) near r = 1 . These facts illustrate the uniqueness of this new lower bound G ( r ) .

2. Lemmas

This article needs the following lemma. Let us give a simple proof of Lemma 1.
Lemma 1
([34,35,36,37,38]). For x < 1 ,
2 x sin 1 x 1 x 2 = n = 1 2 x 2 n n 2 n n .
Proof. 
Let
h ( x ) = sin 1 x 1 x 2 .
Then, h ( 0 ) = 0 , and
1 x 2 h ( x ) = sin 1 x .
Differentiating (9) gives
1 x 2 h ( x ) x h ( x ) = 1 .
Since h ( x ) is an odd function, we can express it as a power series
h ( x ) = n = 0 a n x 2 n + 1 .
Differentiation of (11) yields
h ( x ) = n = 0 2 n + 1 a n x 2 n .
Substituting the series of h ( x ) and h ( x ) into (10) yields
a 0 + n = 0 2 n + 3 a n + 1 2 n + 1 a n x 2 n + 2 = 1 .
Equating coefficients of x 2 n + 2 , we can obtain
a 0 = 1 , a n + 1 = 2 n + 1 2 n + 3 a n , n 0 .
Then,
a n = 2 n n 2 n + 1 n 1 2 n 1 2 5 1 3 1 = 2 n n ! 2 n + 1 ! ! , n 0 ,
and
h ( x ) = sin 1 x 1 x 2 = n = 0 2 n n ! 2 n + 1 ! ! x 2 n + 1 = n = 1 2 n 1 n 1 ! 2 n 1 ! ! x 2 n 1 = n = 1 n ! 2 n 1 n 2 n 1 ! ! x 2 n 1 = n = 1 2 x 2 n 1 n 2 n n ,
the last equality is founded because of the identity
2 n n = 2 n 2 n 1 ! ! n ! .

3. Proof of Main Result

The main conclusion of this paper is equivalent to the following theorem.
Theorem 2.
The inequality
2 π K ( r ) sin 1 r > tanh 1 r
holds for all r ( 0 , 1 ) .
Proof. 
Let
F ( r ) = 2 π K ( r ) sin 1 r tanh 1 r , 0 < r < 1 .
Then
F ( r ) = 2 π K ( r ) sin 1 r + 2 π K ( r ) 1 1 r 2 1 1 r 2 ,
and
F ( r ) 1 r 2 = 2 π K ( r ) sin 1 r 1 r 2 + 2 π K ( r ) 1 1 r 2 1 1 r 2 3 / 2 .
From (1), (7) and Cauchy products we have
F ( r ) 1 r 2 = n = 1 2 n 1 2 n 2 n ! 2 r 2 n 1 n = 1 2 2 n 1 n 2 n n r 2 n 1 + n = 0 1 2 n 2 n ! 2 r 2 n n = 0 r 2 n n = 0 3 2 n n ! r 2 n = 1 r 2 n = 1 2 n 1 2 n 2 n ! 2 r 2 n n = 1 2 2 n 1 n 2 n n r 2 n + n = 0 1 2 n 2 n ! 2 r 2 n n = 0 r 2 n n = 0 3 2 n n ! r 2 n
= 1 r 2 n = 1 k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 n k ! 2 r 2 n + n = 0 k = 0 n 1 2 k 2 k ! 2 r 2 n n = 0 3 2 n n ! r 2 n = n = 1 k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 ( n k ! ) 2 r 2 n 2 + n = 1 k = 0 n 1 1 2 k 2 ( k ! ) 2 r 2 n 2 n = 1 3 2 n 1 n 1 ! r 2 n 2 = : n = 1 f n r 2 n 2 = n = 2 f n r 2 n 2 ,
where
f n = k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 n k ! 2 + k = 0 n 1 1 2 k 2 k ! 2 3 2 n 1 n 1 ! , n 2 .
Below, we will prove f n > 0 for n 2 , which is equivalent to
k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 n k ! 2 > 3 2 n 1 n 1 ! k = 0 n 1 1 2 k 2 k ! 2 , n 2 .
We prove (14) by mathematical induction. It is not difficult to verify that (14) is true for n = 2 . Assuming that (14) holds for n 3 , let’s prove (14) holds for n + 1 . Since
k = 1 n + 1 2 2 k 1 k 2 k k 2 n + 1 k 1 2 n + 1 k 2 n + 1 k ! 2 k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 n k ! 2 = k = 1 n 2 2 k 1 k 2 k k 2 n + 1 k 1 2 n + 1 k 2 n + 1 k ! 2 k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 n k ! 2 = k = 1 n 2 2 k 1 k 2 k k 1 2 n k + 1 1 2 n k 2 n k ! 2 > 2 2 n 1 n 2 n n 1 2 = 2 2 n 2 n 2 n ! n ! 2
holds for all n 1 , we have that for n 3 ,
k = 1 n + 1 2 2 k 1 k 2 k k 2 n + 1 k 1 2 n + 1 k 2 n + 1 k ! 2 = k = 1 n 2 2 k 1 k 2 k k 2 n + 1 k 1 2 n + 1 k 2 n + 1 k ! 2 > k = 1 n 2 2 k 1 k 2 k k 2 n k 1 2 n k 2 n k ! 2 + 2 2 n 2 n ! 2 n 2 n ! > 3 2 n 1 n 1 ! k = 0 n 1 1 2 k 2 k ! 2 + 2 2 n 2 n ! 2 n 2 n ! .
The proof of (14) is complete when proving
3 2 n 1 n 1 ! k = 0 n 1 1 2 k 2 k ! 2 + 2 2 n 2 n ! 2 n 2 n ! > 3 2 n n ! k = 0 n 1 2 k 2 k ! 2 2 n 1 2 n n ! + 2 2 n 2 n ! 2 n 2 n ! > 2 n + 1 1 2 n n ! 1 2 n 2 n ! 2
1 2 n 2 ( n ! ) 2 > 1 2 n n ! 2 2 n 2 n ! 2 n 2 n !
We prove (15) by mathematical induction too. It is not difficult to verify that (15) is true for n = 3 . Assuming that (15) holds for n 4 , let us prove (15) holds for n + 1 . Since
1 2 n + 1 2 n + 1 ! 2 = 2 n + 1 2 4 n + 1 2 1 2 n 2 n ! 2 > 2 n + 1 2 4 n + 1 2 1 2 n n ! 2 n + 1 2 4 n + 1 2 2 2 n 2 n ! 2 n 2 n ! ,
we can complete the proof of (15) when proving
2 n + 1 2 4 n + 1 2 1 2 n n ! 2 n + 1 2 4 n + 1 2 2 2 n 2 n ! 2 n 2 n ! > 1 2 n + 1 n + 1 ! 2 2 n n + 1 ! 2 n + 1 2 n + 2 ! ( 2 n + 1 ) 2 4 ( n + 1 ) 2 1 2 n n ! ( 2 n + 1 ) 2 4 ( n + 1 ) 2 2 2 n 2 n ! 2 n 2 n ! > ( 2 n + 1 ) 2 n + 1 1 2 n n ! 2 2 n n + 1 ! 2 n + 1 2 n + 2 ! 1 4 2 n + 4 n 2 1 n 2 n + 1 n + 1 2 n ! 2 2 2 n 2 2 n ! > 1 4 2 n + 1 n + 1 2 1 2 n n !
1 2 n n ! < 2 2 n 2 n ! 2 2 n + 4 n 2 1 n 2 n ! 2 n + 1 2 .
By mathematical induction, we can prove (16). It is not difficult to verify that (16) is true for n = 4 . Assuming that (16) holds for n 5 , let us prove (16) holds for n + 1 . Since
1 2 n + 1 n + 1 ! = 2 n + 1 2 n + 1 1 2 n n ! < 2 n + 1 2 n + 1 2 2 n 2 n ! 2 2 n + 4 n 2 1 n 2 n ! 2 n + 1 2 ,
we can complete the proof of (16) when proving
2 n + 1 2 n + 1 2 2 n 2 n ! 2 2 n + 4 n 2 1 n 2 n ! 2 n + 1 2 < 2 2 n n + 1 ! 2 2 n + 1 + 4 n + 1 2 1 n + 1 2 n + 2 ! 2 n + 3 2 = 2 2 n n + 1 ! 2 10 n + 4 n 2 + 5 2 n + 2 ! 2 n + 3 2 n + 1 4 n n + 1 10 n + 4 n 2 + 5 2 n + 3 2 2 n + 4 n 2 1 > 1 .
The last inequality obviously holds due to
4 n n + 1 10 n + 4 n 2 + 5 2 n + 3 2 2 n + 4 n 2 1 = 4 n 2 + 14 n + 9 > 0
holds for n 1 . Based on the above fact f n > 0 for n 2 , we have F ( r ) > 0 for 0 < r < 1 . Since F ( 0 + ) = 0 , we obtain that F ( r ) > F ( 0 + ) = 0 for 0 < r < 1 . This completes the proof of Theorem 2.

4. Conclusions

Scholars have been exploring the precise and concise bounds of the complete elliptic integral of the first kind K ( r ) . This paper gives a concise and novel lower bound that conforms to the behavior of the function at two endpoints. I believe that the exploration of this paper provides the possibility to explore more concise and better upper and lower bounds for the complete elliptic integral of the first kind K ( r ) .

Funding

This research received no external funding.

Acknowledgments

The author is thankful to reviewers for reviewers’ careful corrections and valuable comments on the original version of this paper.

Conflicts of Interest

The author declares that he has no conflict of interest.

References

  1. Carlson, B.C. Some series and bounds for incomplete elliptic integrals. J. Math. Phys. 1961, 40, 125–134. [Google Scholar] [CrossRef]
  2. Carlson, B.C. Special Functions of Applied Mathematics; Academic Press: New York, NY, USA, 1977. [Google Scholar]
  3. Carlson, B.C.; Gustafson, J.L. Asymptotic expansion of the first elliptic integral. SIAM J. Math. Anal. 1985, 16, 1072–1092. [Google Scholar] [CrossRef]
  4. Carlson, B.C.; Gustafson, J.L. Asymptotic approximations for symmetric elliptic integrals. SIAM J. Math. Anal. 1994, 25, 288–303. [Google Scholar] [CrossRef] [Green Version]
  5. Gustafson, J.L. Asymptotic Formulas for Elliptic Integrals. Ph.D. Thesis, Iowa State University, Ames, IA, USA, 1982. [Google Scholar]
  6. Karp, D.; Sitnik, S.M. Asymptotic approximations for the first incomplete elliptic integral near logarithmic singularity. J. Comput. Appl. Math. 2007, 205, 186–206. [Google Scholar] [CrossRef] [Green Version]
  7. Karp, D.; Savenkova, A.; Sitnik, S.M. Series expansions and asymptotics for the third incomplete elliptic integral via partial fraction decompositions. J.Comput. Appl. Math. 2007, 207, 331–337. [Google Scholar] [CrossRef] [Green Version]
  8. Kaplan, E.L. Auxiliary table for the incomplete elliptic integrals. J. Math. Phys. 1948, 27, 11–36. [Google Scholar] [CrossRef]
  9. Kelisky, R.P. Inverse elliptic functions and Legendre polynomials. Am. Math. Mon. 1959, 66, 480–483. [Google Scholar] [CrossRef]
  10. López, J.L. Asymptotic expansions of symmetric standard elliptic integrals. SIAM J. Math. Anal. 2000, 31, 754–775. [Google Scholar] [CrossRef]
  11. López, J.L. Uniform asymptotic expansions of symmetric elliptic integrals. Constr. Approx. 2001, 17, 535–559. [Google Scholar] [CrossRef]
  12. López, J.L. Asymptotic expansions of Mellin convolutions by means of analytic continuation. J. Comp. Appl. Math. 2007, 200, 628–636. [Google Scholar] [CrossRef] [Green Version]
  13. Nellis, W.J.; Carlson, B.C. Reduction and evaluation of elliptic integrals. Math. Comput. 1966, 20, 223–231. [Google Scholar] [CrossRef]
  14. Radon, B. Sviluppi in serie degli integrali ellipttici. Atti. Accad. Naz. Lincei Mem. Cl. Sci. Fis. Mat. Nat. Ser. 1950, 2, 69–109. [Google Scholar]
  15. Bowman, F. Introduction to Elliptic Function with Applications; Dover Publications: New York, NY, USA, 1961. [Google Scholar]
  16. Byrd, P.F.; Friedman, M.D. Handbook of Elliptic Integrals for Engineers and Scientists; Springer: New York, NY, USA, 1971. [Google Scholar]
  17. Anderson, G.D.; Vamanamurthy, M.K.; Vuorinen, M. Conformal Invariants, Inequalities, and Quasiconformal Maps; John Wiley & Sons: New York, NY, USA, 1997. [Google Scholar]
  18. Anderson, G.D.; Qiu, S.-L.; Vuorinen, M. Precise estimates for differences of the Gaussian hypergeometric function. J. Math. Anal. Appl. 1997, 215, 212–234. [Google Scholar] [CrossRef] [Green Version]
  19. Ponnusamy, S.; Vuorinen, M. Univalence and convexity properties for Gaussian hypergeometric functions. Rocky Mt. J. Math. 2001, 31, 327–353. [Google Scholar] [CrossRef]
  20. Song, Y.-Q.; Zhou, P.-G.; Chu, Y.-M. Inequalities for the Gaussian hypergeometric function. Sci. China Math. 2014, 57, 2369–2380. [Google Scholar] [CrossRef]
  21. Wang, M.-K.; Chu, Y.-M.; Jiang, Y.-P. Ramanujan’s cubic transformation inequalities for zero balanced hypergeometric functions. Rocky Mt. J. Math. 2016, 46, 679–691. [Google Scholar] [CrossRef] [Green Version]
  22. Wang, M.K.; Chu, Y.M.; Song, Y.Q. Asymptotical formulas for Gaussian and generalized zero balanced hypergeometric functions. Appl. Math. Comput. 2016, 276, 44–60. [Google Scholar]
  23. Wang, M.-K.; Chu, Y.-M. Refinenemts of transformation inequalities for zero-balanced hypergeometric functions. Acta Math. Sci. 2017, 37B, 607–622. [Google Scholar] [CrossRef]
  24. Carlson, B.C.; Vuorinen, M. Inequality of the AGM and the logarithmic mean. SIAM. Rev. 1991, 33, 653–654. [Google Scholar] [CrossRef]
  25. Vamanamurthy, M.K.; Vuorinen, M. Inequalities for means. J. Math. Anal. Appl. 1994, 183, 155–166. [Google Scholar] [CrossRef] [Green Version]
  26. Qiu, S.-L.; Vamanamurthy, M.K. Sharp estimates for complete elliptic integrals. SIAM. J. Math. Anal. 1996, 27, 823–834. [Google Scholar] [CrossRef]
  27. Alzer, H. Sharp inequalities for the complete elliptic integral of the first kind. Math. Proc. Camb. Philos. Soc. 1998, 124, 309–314. [Google Scholar] [CrossRef]
  28. Anderson, G.D.; Vamanamurthy, M.K.; Vuorinen, M. Functional inequalities for hypergeometric functions and complete elliptic integrals. SIAM. J. Math. Anal. 1992, 23, 512–524. [Google Scholar] [CrossRef]
  29. Alzer, H.; Qiu, S.-L. Monotonicity theorem and inequalities for the complete elliptic integrals. J. Comput. Appl. Math. 2004, 172, 289–312. [Google Scholar] [CrossRef] [Green Version]
  30. Yang, Z.-H.; Song, Y.-Q.; Chu, Y.-M. Sharp bounds for the arithmetic-geometric mean. J. Inequal. Appl. 2014, 2014, 192. [Google Scholar] [CrossRef] [Green Version]
  31. Yang, Z.-H.; Qian, W.-M.; Chu, Y.-M.; Zhang, W. On approximating the arithmetic-geometric mean and complete elliptic integral of the first kind. J. Math. Anal. Appl. 2018, 462, 1714–1726. [Google Scholar] [CrossRef]
  32. Yang, Z.-H.; Tian, J.-F.; Zhu, Y.-R. A rational approximation for the complete elliptic integral of the first kind. Mathematics 2020, 8, 635. [Google Scholar] [CrossRef]
  33. Zhu, L. Concise high precision approximation for the complete elliptic integral of the first kind. AIMS Math. 2021, 6, 10881–10889. [Google Scholar] [CrossRef]
  34. Lehmer, D.H. Interesting series involving the central binomial coefficient. Amer. Math. Mon. 1985, 92, 449–457. [Google Scholar] [CrossRef]
  35. Sury, B.; Wang, T.-M.; Zhao, F.-Z. Identities involving reciprocals of binomial coefficients. J. Integer Seq. 2004, 7, 3. [Google Scholar]
  36. Chen, H.W. A power series expansion and its applications. Int. J. Math. Educ. Sci. Technol. 2006, 37, 362–368. [Google Scholar] [CrossRef]
  37. Sofo, A. Integral Identities Involving Reciprocals of Single and Double Binomial Coefficients; RGMIA Seminar Series; Austral Internet Publishing: Victoria, Australia, 2007. [Google Scholar]
  38. Borwein, J.M.; Chamberland, M. Integer powers of Arcsin. Int. J. Math. Math. Sci. 2007, 2007, 19381. [Google Scholar] [CrossRef]
Table 1. The error table between K ( r ) and G ( r ) .
Table 1. The error table between K ( r ) and G ( r ) .
r0.20.40.80.9
K ( r ) G ( r ) 5.3480 × 10 3 2.2894 × 10 2 0.13430 0.21407
Table 2. The error tables between G ( r ) and A ( r ) , B ( r ) , D ( r ) .
Table 2. The error tables between G ( r ) and A ( r ) , B ( r ) , D ( r ) .
r00.40.90.99999
G ( r ) A ( r ) 0.570800.572880.593220.68627
G ( r ) B ( r ) 2.5244 × 10 29 2.1093 × 10 2 0.2075 0.67575
G ( r ) D ( r ) 0 0.02284 0.20684 2.1124 × 10 2
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Zhu, L. A Natural Approximation to the Complete Elliptic Integral of the First Kind. Mathematics 2022, 10, 1472. https://doi.org/10.3390/math10091472

AMA Style

Zhu L. A Natural Approximation to the Complete Elliptic Integral of the First Kind. Mathematics. 2022; 10(9):1472. https://doi.org/10.3390/math10091472

Chicago/Turabian Style

Zhu, Ling. 2022. "A Natural Approximation to the Complete Elliptic Integral of the First Kind" Mathematics 10, no. 9: 1472. https://doi.org/10.3390/math10091472

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop