0% found this document useful (0 votes)
78 views131 pages

Acfrogdebcq7n6igzkwjcj9v4caujukzvi1yx7oqyfelzxc-Dkwr592icqbpexw6liox29kp4 Ol7bcq1lev9z2ejdffxwzu4hpuhh3umlcbvdetoavmph3nfn59 7ntpvkicluxbspzvambrudr

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
78 views131 pages

Acfrogdebcq7n6igzkwjcj9v4caujukzvi1yx7oqyfelzxc-Dkwr592icqbpexw6liox29kp4 Ol7bcq1lev9z2ejdffxwzu4hpuhh3umlcbvdetoavmph3nfn59 7ntpvkicluxbspzvambrudr

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 131
aor 2H om. eames Elementary Differential Equations tammernenee eer SEVENTH EDITION Earl D. Rainville Late Professor of Mathematics University of Michigan Phillip E. Bedient Professor of Mathematics - Franklin and Marshall College Macmillan Publishing Company New York Macmillan es ‘Copyright © 1989, Macmillan Publishing Company, a division of Macmillan, tnc Printed in the Philippines All rights reserved. No part of this book may be reproduced or ‘transmitted in any form or by any means, electronic or mechanical, including photocopying. recording, oF any information storage and retrieval system, without permission in writing from the publisher Earlier editions copyright © 1981, 1974, 1969, 1964, 1998, 1952, and 1949 by ‘Macmillan Publishing Company ‘Macmillan Publishing Company {866 Third Avenue, New York, New York 10022 Collier Macmillan Canada, Inc. Library of Congress Cataloging in Publication Data Rainville, Earl David (date) Elementary differential equations/Ear! D. Rainville, Phillip E Bedient—Tth ed. Pp. cm. Includes index. ISBN 971-1055-52-X 1. Differential equations. 1, Bedient, Phillip Edward (date) IL Title, QA37LR29 1989 811120 S1S3'S—de 19 cP Priming: 12345678 Year 90123456 Reprinted under Authority of P.D.#285 fas Amended by PD. No. 400 and P.D.No. 1203 by EDCA Publishing and Distributing Corporation 488 Quezon Avenuc, Quezon City Tel. Nos. 71-97-09; 732- 14-83 EXPORT OF THIS BOOK FROM THE PHILIPPINES PUNISHABLE BY LAW Preface tothe Seventh Edition The publication of a new edition of Professor Rainville’s book is an acknowledgment of the effectiveness of the simple and direct style that has U iade earlier editions most popular with students of differential equations The balance between developing techniques for solving equations and the theory necessary to support those techniques remains essentially unchanged, although a number of sections have been revised and some new ‘applications have been added. Sections 56 through $9 are new and present “A derivation of Kepler's laws from Newton's laws. Section 135 is new and Gives a presentation of a modified Euler method for solving differential uations numerically © The material is arranged to permit great fexibility in the choice of topics for a semester course, Except for Chapters 1, 2, 5, 16, 17, and 18, and either yand 7 or 11 and 12, any chapter on ordinary dif ‘omitted without interfering with the study of later chapters. Parts of chap~ omitted in many instances. 7 For a course that aims at reaching power series as rapidly as is consistent fh some treatment of more elementary methods, a reasonable syllabus ‘should include Chapters 1 and 2; Chapters $, 6, 7, ana 8; parts of Chapters 3 and 15; Chapters 17 and 18; and whatever applications the instructor Fes to insert. Chapters 1 through 16 of this book appear separately as A Short Course "Differential Equations, Seventh Edition, The shorter version is intended ‘courses that do not include discussion of infinite series methods. ‘ ‘The author wishes to thank those students and colleagues at Franklin, Marshall College whose $0 sey 5 upport of Professor George Rosenstein are gratefully acknowledged, er, Pennsylvania 1 Definitions, Families of Curves 1. Examples of differential equations 2. Definitions 3 3. The elimination o” arbitrary constants: 4, Families of curves 10, 2 "Equations of Crder One §. The isoclines of an equation WW 6. An existence theorem 19 7, Separation of variables 21 8. Homogencous functions’ 26 Gama iam 9, Equations with homogeneous coefficients 10. Exact equations 41. The linear equation of order one 37 12, The general olution of linear equation 41 Miscellaneous exercises 440 vl Contents Contents 4 eo Topics on Equations of Order One 63 . Integrating factors found by ins facto spection: 63 19. The determination of integrating factors 67 20. Substitution suggested by the equation 72 21. Bernoulli's equation © 74 22. Coefficients linear in the two variables 77 23. Solutions involving nonelementary integrals. 82 Miscellaneous exercises 85 8 Variation of Parameters 44, Reduction of order 139 45. Variation of parameters 143 46. Solution of y’+y=J(x) 148 Miscellaneous exercises 151 | 9 Inverse Differential Operators 47. The exponential shift 152 48. The operator 1/f(D) 156 49. Evaluation of [1/f(Dyle" 157 $0. Evaluation of (D? + a?)~! sin ax and (D? +a2)"!cosax 158 5 Linear Differential Equations 87 24. The general linear equation 87 25. An existence and uniqueness theorem 89 26. Linear in: 90 27. The Wronskian a1 28. General solution of a homogeneous equation 94 29. General solution of a nonhomogeneous equation 96 30. Differential operators 97 31. The fundamental laws of ope 32. Some properties of differenti 10 Applications 51. Vibration ofa spring 162 Undamped vibrations 164 j. Resonance 167 }, Damped vibrations 170 The simple pendulum — 174 . Newton's laws and planetary motion 176 Central force and Kepler's second law 178 Kepler's first law 178 Kepler's third law 181 ion 100 operators 102 6 Linear Equations with Constant Coefficients 105 33. Introduction 105 34, The auxiliary equation; distinct roots 105 35. The auxiliary equation; repeatea roots 108 36. A definition of exp zforimaginaryz — 112 37. The euxiliary equation; imaginary roots 114 38. A note on hyperbolic functions 116 yus-exercises 119 7 Nonhomogencous Equations: Undetermined Coefficients 121 39. Construction of a nomogeneous equation from a specified solution 121 * Contents Contents 12 Inverse Transforms 209 70. Definition of an inverse transform 209 71, Partial fractions 213 72. Initial value problems 216 73. Astep function 222 74, Aconvolution theorem 229 75. Special integral equations 234 76. Transform methods and the vibration of springs 239 77. The deflection of beams 243 } Nonlinear Equations 97. Preliminary remarks 324 98. Factoring the left member 325 99. Singular solutions 328 100, The c-discriminant equation 329 101. The p-discriminant equation 331 102. Eliminating the dependent variable 333 103. Clairaut’s equation 335 404, Dependent variable missing 339 105. Independent variable missing 340 106, Thecatenary 343 13 Linear Systems of Equations 250 Miscellaneous exercises 345 78. Introduction 250 79. Elementary elimination calculus 250 80. First-order systems with constant coefficients 254 81, Solution ofa first-order system 256 82. Some matrix algebra 257 83. First-order systems revisited 264 84. Complex eigenvalues 273 85. Repeated eigenvalues 279 86, Nonhomogeneous systems 287 87. Arms races 291 18 Solutions Near Regular Singular Points 364 88. The Laplace transform = 295 112. Regular singular points 364 | 113. The indicial equation 367 114, Form and validity of the solutions near a regular Power Series Solutions 107 Linear equations and power series 347 108. Convergence of power series 348 109. Ordinary points and singular points 351 110, Validity of the solutions near an ordinary point 352 111, Solutions near an ordinary point 353 4 Electric Circui k 301 singular point 369 ; 99, ae ie: 115, Indicial equation with difference of roots 90. Simple a Ks 304 nonintegral 369. 3 aan 116. Differentiation of a product of functions 375 117, Indicial equation with equal roots 377 ¥ : Indicial equation with equal roots, an alternative 383 15 The Existence and Uniqueness of Solutions 314 119. Indicial equation with difference of roots a positive Dicandeasy somal integer, nonlogarithmic case 387 a sa see 24 Indieiai equation with difference of roots a positive ‘An existence and uniqueness theorem = 314 ti Ss a A iets sondlton 7 integer, logarithmic case 391 94. A proof of the existence theorem = 318 Solution for large x 397 3 Many-term seucee relations 401. 95. A proof of the uniqueness theorem 321 96. Other existence theorems — 322 xii ‘Contents 19 Equations of Hypergeometric Type 124. Equations to be treated in this chapter 410 12S. The factorial function 410 126. The hypergeometric equation 411 127. Laguerre polynomials. 413 410 128. Bessel’s equation with index not an integer 415 129. Bessel's equation with index an integer 415 130. Hermite polynomials 417 131. Legendre polynomials 417 132. The confluent hypergeometric equation 418 20 Numerical Methods 133. General remarks 421 134, Euler's method 421 135. A modification of Euler’s method 425 136. A method of successive approximation 426 421 137. An improvement on the preceding method 427 138. The use of Taylor's theorem 429 139. The Runge-Kutta method — 431 140. Acontinuing method 435 21 «Partial Differential Equations 141. Remarks on partial differential equations — 438 142. Some partial differential equations of applied mathematics 438 143, Method of separation of variables 441 144. A problem on the conduction of heat in a slab 22 Oxsomsass tes of Pusctions Ble wa tas ol pctrnonas 453 Orthogonal polynomials 454 Se gage roe 436 ‘orthogonal sets 458 438 446 452 Contents | Fourier Series 151. Orthogonality of a set of sines and cosines 152. Fourier series: an expansion theorem 463 153, Numerical examples of Fourier series 407 154. Fourier sine series. 476 188. Fourier cosine series 479 156. Numerica! Fourier analysis 483 © 157. Improvement in rapidity of convergence 484 Boundary Value Problems 158. The one-dimensional heat equation 486 159. Experimental verification of the validity ofthe heat equation 493 160. Surface temperature varying with time 495 161. Heat conduction inasphere 497 162. The simple wave equation 499 163. Laple2e's equation in two dimensions $02 Additional Properties of the Laplace Transform 164. Power series and inverse transforms 507 © 365. Theerror function S11 166. Bessel functions 518 “167. Differential equations with variable coefficients "Partial Differential Equations; Transform ethods Boundary value problems The wave equation 527 ‘Canonical vari Diffusion in a slab of finite width | Diffusion in a quarter-infinite 1 Definitions, Families — of Curves Examples of Differential Equations pin equation (3), a derivative may ‘Of differentials. Our aim is to find methods for solving differential ons, that is, to find the unknown function or functions that satisfy the ial equation. = following are examples of differential equations: oe ft ena (+ Pde 2 ‘Definitions, Families of Curves Ich du (Ou, Ow aie & * a @ ai at Laat Rt Gin Bm conan, (5) @v ev oe tap ©) ‘aw\? Ce.-. ., @x a tegen ® ty oe ® #y ax ae ae * (19) Ge a Baas (i) ‘When an equation involves one or more derivatives with respect to a ‘particular variable, that variable is called an independent variable. A vari- able is called dependent if a derivative of that variable occurs. In the equation ai dit eta cg Shr et (5) 4 is the dependent variable, ¢ the independent variable, and L, R, C, E, and @ are called parameters. Tae equation ov oy, Ox? * dy? has one dependent variable V and two independent variables. Since the equation 0 (6) (2? + y?) dx — Ixy dy = 0 @) “may be written w+y-ry Zo 2 yy eee G+ NG, — BO consider either variable to be dependent, the other being the it one. ify the gations given as examples in this section, Definitions F order of a differential equation is the order of the Righest-ordered jive appearing in the equation. For instance, noe B+ mz) +9=0 m equation of “order two." It is also referred to as a “second-order ation.” ey iyi egbeton FU IV OD =O ‘ an “nth-order” ordinary differential equation. Under Bee (ie tancion Fea pof the ‘other n + I variables x, y, yi = Le HY purposes of this book we shall assume that this i fan equation of the form of equation (2) may ‘one equation of the form of equation (3) ple, the equation Definitions, Families of Curves Ich _ A function $, defined on an interval a < x 0, (2/,/3)x*? is a solution of the equation in Exercise 6, 3. The Elimination of Arbitrary Constants In practice, differential equations arise in many ways, some of which we shall encounter later. There is one way of arriving at a differential equation, however, that is useful in that it gives us a feeling for the kinds of solutions to be expected. In this section we start with a relation involving arbitrary constants and, by elimination of those arbitrary constants, come to a differ ‘ential equation consistent with the o-iginal relation. In a sense we start with the answer and find the prob!=m ‘Methods for the elimination of 2rburary constants vary ith the ws y in ‘which the constants enter the given relation. A mv'ho chat is eificient 0” ‘one problem may be poor for anothe:. Une fact persists throughout Because each differentiation yields a new relation, the number of derivatives that need be used is the same as the number of arbitrary constants to be liminated. We shall in each case determine the differential equation that is (a) Of order equal to the number of art tion. Fy Constents in the given ris (b) Consistent with that relation. (€) Free from arbitrary constants. EXAMPLE (a) Eliminate* the arbitrary constants c, and c, from the relation yace™+c,e%. a ‘Since two constants are to be eliminated “tain the two derivatives, ¥ = —deye"™* + Jeze, 2 ¥ = Acye"™* + 9e,e", @ ‘The elimination of ¢, from equations (2) and (3) yields of + Dy = Weze™; + By diflerentiations and pertinent leplimate mathematical procedures, Elimination 6) ‘ecasare fr instance, snot permitied. es 4 The EI ination of Arbitrary Constants - jon of c, from equation (1) and (2) yields yt Seze* y' + dy = 3 + Ih a yi —y' —6y =0, TAnother method for obtaining the differential equation in this example ceeds as follows: We know from a theorem in elementary algebra that f three equations (1), (2), and (3), considered as equations in the two ynowns c, and c, , can have solutions only if 4 eae «| te * Je @ -21 ge%| flows immediately this latter method has the advantage of making it easy to see that the ination of the constants ¢,.¢2...-+ &,{tom a relation of the form +een" ye cet eee iil always lead to a lincar differential equation fy thy ay ; a Ooo Yay ad, oat Nae dx yhich the coefficients do... a ate constants. The study of such — tial equations will receive much of our attention. cee Definitions, Families of Curves Ich Direct differentiation of the relation yields Ax — a) + 2yy' = 0, from which anxtyy. ‘Therefore, using the original equation, we find that Ww?'+ P= + w?, or vax + doy, which may be written in the form (x? = y?) dx + 2xy dy = 0. Another method will be used in this example as an illustration of a device that is often helpful. The method is based upon the isolation of an arbitrary constant. ‘The equation (oF +P =a ‘may be put in the form x? + y? — Jax =0, or Bey ‘Then differentiation of both members leads to (2x dx + Jy dy) — (02 + y*) dx =0, (x? = y?) dx + Day dy = 0. as desired. . It is interesting to speculate here about the significance of x = 0 upon the argument just used. The student should draw a few of the members of this family of circles and observe what is peculiar about their behavior at « = 0. ‘The Elimination of Arbitrary Constants: E (©) nate B and « from the relation <= B cos (wt +), hh w is a parameter (not to be eliminated). ‘we obtain two derivatives of x with respect tof: & 2 —wB sin (wt + 2 . exy $x +4=0. dy tay) th =O. em -y +4¥) ‘substitution into the original equation leads us to the result ely? + ey +450. . examples suggest that in a certain sense the totality of solutions of stder equation depends on n arbitrary constants, The sense in Which ‘true will be stated in Sections 5 and 12. w Definitions, Families of Curves {Chi he aaa ea oer os ee oe SF's -0 6 -acectgul ere fees [Lysates Am yen’ +0 +1 8 y= mx + Asha parameter, mto be eines Ane. 9. y= tax. WW ywar siete Amy Me yee tee 2 yds ge yaa tae M4 ya qe tee IS yaxteetee™ Me ye ce te Tye tte set Ame Sy + 6y = 6x07 10x +? IR ym eet + ese Am yf —2f +y=0 19. y= Ae + Bre™ Am y'—4y +4y~0 BR y= cet cos 3x + ce sin Ix. Am 7 —4/' + 13y -0. BL y= c,e cos bx + cp sin br; a and b are parameters. ‘Ams. — 2ay + (0? + b*)y = 0 BR yeent ae Am (e—y — xy +9 Dynex? + ee* Aas ate + 29" +(x? - Dy — x + Ny Moyer tertaet Ame (e+ Wy # xy y= + 2+? Boyne sget Ame afl — wy" + (2 — Ny = 22x — y= Am 7 —5y+6y-0 4. Families of Curves ‘An equation involving » parameter, as well as one or both of the coordi nates of 4 point in a plane, may represent a family of curves, one curv: corresponding to each value of the parameter. For instance, the equation (x - oP + (ye = 207, a) Families of Curves FIGURE 1 ‘the constant c in equation (1) or in equation (2) is treated as an arbi- ‘constant and eliminated as in the preceding section, the result is called ‘differential equation of the family represented by equation (1). In this the elimination of c is easily performed by isolating ¢, then differ- throughout the equation with respect to x. Thus, from BE x+y (x + yh2x dx + 2y dy) — (2 + dx + dy) +P rE (x2 + dey — y?) de — (0? = 2xy - yy differential equation of the family of circles represented by P. n Definitions, Families of Curves 1ch.1 Families of Curves y yeti -VDe Perievoy \ FIGURE 2 FIGURE 3 a vertical tangent. From ay = 0 jows that ‘we see that xy =2y30 o ya(-1+ Six 6) differential equation of the family, We note that (9) is a first-order ce differential equation. . ya(-1- JS. 6 E (b) In Figure 2, the straight lines (5) and (6) appear along with the family (1). differential equation of the family of circles (Figure 4) having tbeir It is seen that the lines (5) and (6) cut the members of the family of circles in ‘on the y-axis. ‘a member of the family of circles of this example may have its | precisely those points of vertical tangency. For a two-parameter family of curves, the differential equation will be of order two, and such a simple geometric interpretation is not available. ‘anywhere on the y-axis and its radius of any magnitude, we are, ‘with the two-parameter family +a EXAMPLE (a) ‘ Find a differential equation satisfied by the tamily of parabolas (Figure 3). hall eliminate both b and r ‘and obtain a second-order ‘having their vertices at the origin and their foci on the y-axis. jon for the family (10). “An equation of this family of parabolas ymax, ” x+0-dY =O so that ais. y = 2x. ®) Reeatiy “: i “4 “ Definitions, Families of Curves [cnt y FIGURE 4 Then Sw + OPI — Ve + WI _ o, oF 2 ‘50 the desired differential equation is xf - WP —¥ =0. . Exercises In cach exercise, obtain the differential equation of the family of plane curves described and sketch several representative members of the family. 4. Straight lines through the origin Amy dx -x dy =0. 2 Straight lines through the fixed point (h,k);h and k not to be eliminated. ‘Am (y ~ k) dx ~ (x ~ h) dy = 0. 2 Straight lines with slope and y-intercept equal. Ama y dx —(x + I) dy = 0. ‘Straight lines with slope and x-intercept equal. Ame (y= xy Straight lines with algebraic sum of the intercepts fixed as k ‘Am (xy — Wy ~ 1) + hy’ = 0. (6 Straight lines at » fixed distance p from the origi Ams. (xy — ym pith + 00) 7. Circles with center at the origin. ‘Am. x dx +y dy = 0 ‘& Circles with center on the x-axis. Am yy +(yP +1 =0 “9, Circles with fixed radius r and tangent to the x-axis. . Am (yt MY) + y? + Dry = 0. Famities of Curves tangent to the x-axis, Ase. (1+ PP = Dy +14 0FP with center on the line y = —x, and pasting through the origin. Ans. (x? — dey ~ yt} de +O? + xy = yd dy =O lf radius unity. Use the fact that the radius coe oi ‘Am (7)? = 01 +072 Am U1 + OF = YF olas with vertex on the x-axis, with axis parallel to the y-axis, and with from focus to vertex fixed asa Am ay? my. jas with vertex on the y-axis, with axis parallel to the x-axis, and with tance from focus to vertex fixed as a As is parallel to the y-axis and with distance from vertex to focts Am ay = 1. is parallel to the x-axis and with distance from vertex to focus. Am 2ay +P =O. AG a4 (Gyo ty \dy, dy di \dx, ow svers to Exercises 17 and 18 are equivalent ith vertex and focus on the x-axis. Am yy + (YP =O is parallel to the ax Am yy" HF =. ntral conics with center atthe origin and vertioes on the coordinate axes. Ane xy" + xy — yy =O. ‘onfocal central conics vee 1d b held fixed, Aas (xy! — yoy + x) = (a? = ate cubies cy’ = x'(e — 0) with a held fixed, Aum. 2x(x — aly = yi3x = 2a cubics of Exerrise 2¢ with ¢ held fixed and a to be eliminatsd, ‘Ans. 2eytxy'— y= x quate c#y* = x(x — 0)? witha held xed | Ame ‘aux — aly = Woe — ah uartics of Exercise 26 with c held fixed and a to be eliminated. © cM axy = yP = 2Ixty, ets 5 ae ax 16 Definitions, Families of Curves (cht ‘31. Circles through the intersections of the circle x* + y* = I and the line y = « ‘Une the “u + ko” form; that ia, the equation BeyPoitky—ya0 Am (3? — xy ~ y? + 1) dx + (x2 Bay — y?— 1) dy = 0 ‘32. Circles theough the fixed points (a, 0) and (—4, 0). Use the method of Exercise a Am 2xy dx + ()? +a? — x2) dy = 0, 3X The circles r = 2atain 8 ~ cos ‘Ams (con 6 — sin 0) dr + r(c0s 8 + sin 8) d0 ~ 0 34. The cardioids r = o(1 — sin 8). ‘Am. (I sin 6) di +r cos 0:d0— 0 ‘3K The cissoids r = a sin 0 tan 6, (See Exercise 29) ‘Ame sin O.cos 6 dr ~ x(t + cos? 6) 40 ~ 0 34 The strophoide r = atsec 4 + tan 8) pm Harsco ‘37, The trisectrices of Maclaurin r = a(4 cos @ ~ sec 8), (See Exercise 30.) ‘Ams. cos 4 cos? @— 1) dr +r sin M4 cos? 8+ 1) do ~ 0 The Isoclines of an Equation this chapter we study several elementary methods for solving first-order ial equations. Before we begin that study, we take a brief look at ‘ofthe basic geometrical ideas that are involved. jider the equation of order one dy F=f. wo ‘can think of equation (1) as a machine that assigns to each point (a, 6) ‘domain of f some direction with slope f(a, 6). We can thus speak of ction field of the differential equation, In @ real sense any solution of {(1) must have a graph, which at each point has the direction equa- this basic idea is to draw a short mark at ‘associated with that point. This can * Equations of Order One ICh2 ont ° ge. FIGURE 5 EXAMPLE (a) ‘Consider the equation 4 7 ‘The isoclines are the straight lines f(x, y) = y=. For each value of « < ‘obtain a line in which, at each point, the direction dictated by the differe ‘ial equation is that number c. For example, at each point along the line y= 1, equation (2) determines a direction of slope |. In Figure 5 we have drawn several of these isoclines, indicating the direction associated wh ‘each isocline by short markers. If one start: at any point in the plane an moves along a curve whose direction is always in the di of the direc tion marks, then a solution curve is obtained. Several solution curves hove “ween drawn in Figure 5. . EXAMPLE (b) ‘Use the method of isoclines (o sketch some of the solution curves for ' «em Here the isoclines will be the circles x? + y? = c, with ¢ > 0. When c= } the isocline has radius $; for ¢ = 1, adit. '; for c = 4, radius 2. In Figure © ‘we have drawn these isoclines, marking each of them with the appropria(e direction indicator, and finally sketching several curves that represent sol! tions of equation (3). . each ‘of the following differential equations, draw several isoctines with appro- FIGURE 6 tion markers and sketch several solution curves for the equation, ayy a b 20 Equations of Order One Ich. 2 the simplest of functions. Before discussing some of the analytic techniques for finding solutions, we shall state an important theorem concerning the existence,and uniqueness of solutions, a theorem which will be discussed in detail in Chapter 15 Consider the equation of order one Let T denote the rectangular region defined by Ix-xolSa and = yol Sb, a region with the point (xo, yo) at its center Suppose that f and (1/¢y ar continuous functions of x and y in T. Under the conditions imposed on f(x, y) above, an interval exists about Xo] — Xo] Sh, and a function y{x) which has the properties. (a) y = x(x) is a solution of equation (1) on the interval |x ~ xo| Sh {b) On the interval |x — xo| Sh, (x) satisfies the inequality Iyeo = yol $b: (6) Atx= xo. 9 = x0) = Yor (@) 4x) is unique on the interval |x — xo] = h im the sense that it is the ‘only function that has all of the properties (a), (b), and (c) The interval {x — xo] Sh may or may not need to be smaller than the interval |x — xo| S a over which conditions were imposed upon f(x, }) In rough language, the theorem states that if f(x, y) is sulliciently well behaved near the point (xo, yo). then the differential cyuaiion aj a hhas a solution that passes through the point (xo. Yo) and that solution » unique near (x9. yo). In Example (a) of Section 5 we can consider (xo. Yo) to be any point 10 the plane, since f(x, y) = y and its partial derivative df/dy = | are contin lupus in any rectangle. Therefore, our existence theorem assures us tha! through any point (xp, yo) there is exactly one solution, a situation that we ‘assumed when we sketched the solution curves in Figure 5 “Again in Example (b), Section 5, the function f(x, y) = x7 + y* and its ‘partial derivative dffay = 2y are continuous in any rectangle. It follows that through any point (xo, yo) in the plane there is exactly one solution curve. fact that is suggested by the solution curves in Figure 6. Separation of Variables au "Separation of Variables Ibegin our study of the methods for solving first-order, equations By ng an equation of the form M dx +N dy M and N may be functions of both x and y. Some equations of this s0 simple that they can be put in the form A(x) dx + Bly) dy = 05 a Fis, the variables can be separated. Then a solution oan be written at Bor itis only a matter of finding a function F whose total differential Heft member of (1). Then F = c, where c is an arbitrary constant, is the d result \MPLE (a) the equation ye ax x fe note that for the function in equation (2), the theorem of the preed- Tsection applies and assures the existence of a unique continuous solu= Sthrough any point in the first quadrant. By separating the variables we write for x > Oand y>0. @ ee ce we obtain a family of solutions In jy =2 In Lx] +6 @ use we are in the first quadrant, yas? @) 420 & 6 LE (b) Be quation of Example (a) for x i ewan of PM ee, Fit iy ed as before to equation (2). However equation (3) must typeerxt ty ee i ‘ Sa 2 Equations of Order One Ich. 2 Second, if y = 0, we see immediately that since x #0, y = 0 is a solution of the differential equation (2) ‘As a matter of convenience the solutions given by equation (6) are usually written yeex, ” where c, is taken to be an arbitrary real number, Indeed, this form for the solutions incorporates the special case y = 0. Thus we say that the family of curves given in Figure 3, page 13. represents a family of solution curves for the differential equation (2), ‘We must be cautious, however. The function defined by a) =<, x>0 =-43, x50, obtained by piecing together two different parabolic arcs could also be con sidered a solution of the differential equation, even though this function 1s, not included in the family of equation (7). The uniqueness statement in the theorem of Section 6 indicates that, as long as we restrict our attention to « point (xo. yo) with x #0 and consider a rectangle with center at (ss. 5 containing no points at which x = 0. then in that rectangle there is a un»quc solution that passes through (Xo, yo) and is continuous in the rectangle. m EXAMPLE (c) Solve the equation (+ dx Ul + Vay with the “initial condition” that when x = 0, y If we write this .quation in the form. dy _ 0+) dx te we observe that the right member and its partial derivative with respect (0 ) are continuous near (0, ~1). It follows that a unique solution exists for equation (8) that passes (urough the point (0, — 1). equation we get from which it follows at once that are tan x + arc tan y = ¢. OO} Separation of Variables tangent and is subject to the restriction — $n 1 In this ext ‘onto the left branch, equation (10), by the initial condition | x=0. c tions (10) and (11) can be Rion betwen acetate aaa =1), would be: ua Equations of Order One Ich.2 FIGURE 7 prevent the computer from learning of the existence of the other branch of the hyperbola (11). . EXAMPLE (d) Solve the initial value problem Bay + 1) de ~ y dy = 0, (2) where x = Oand y = ~2. Separating the variables in equation (12), we obtain 1 sa) ov we get a family of solutions given implicitly by Pr aee(1 Integra ay-iniy+ tite. (13) ‘Since we seek a member of this family that passes through the point (0, ~2), we must have O=-2-Inj-1] +6, or en? ‘Thus the solution to the problem is given implicitly by Pomyoiniy+ i+? ae Separation of Variables should note how the theorem of Section 6 applies to 1 through 6, obtain the particular solution satisfying the initial condi ‘each exercise interpret your answer in the light of the existence Ams p= ro exp (<2). met + y?p when x = 2, Am ior fee Bie wisn s = 27 ~ pe = ei ohen x= iy a1 ka = fie = 3x dy: when x= —2.y=1 A y= 729. idx = 3x dy; when x = 2,y =~. Ans. y= —(x/2P" condi- 7 through 10, obtain the particular solution satisfying the 2 An. = In oe An. y= effet = x= 1) y= 0. x exp (y — x2), when x dx + e* dy <0; when x-+ 05, 9+ =P) dr =P? sin 8 d0; when 0 = 0, =a fax) = gi when x = Xo.0= Y% 11 through 37, obtain the general solution, Ans. v7? — 05 = 2gtx — xa mt +2)dy =O te dy = ri ee tye ay Barts +? + OS Wola ae dx + tan y dy = 0. snetdy= 0. - arte tert One ena Bye ers «, 26 yy wet eee: 37. tan? y dy = sin? x de. ee ee stan ye 2B y= cos? x cosy. eae ee eta 1m Be + ain 2x a adh dat A lee Ans. y= ofeec x + tan x) 30. dx = 1 + 1°) sec? x de Aes, 2x4 sin xm e+ (140) 31 ("+ My =», Tm yh de) — 02 32 a dp + fda + 93 da + ap) = 0. Rie AEB = exp (—30.— 9, BB (1+ In dx +1 + In dy = 0. ee wine t yiny~< Moxa — few dy 0. i Ree enteral srthe circle? + (y — 0) = 0 Moxdx + =x dy = 0. TAR pe aati tae per hall of the crc x* + (y — 36 a? dx = xa dy, Ams. x =asec®*< 37. yin xn yd + dy = 0, Ans. xInx+In|Iny|=x +e 8. Homogeneous Functions Polynomials in which all terms are of the same degree, such as x3 — xy + 4y3, ey, 0) xty + IV. are called homogeneous polynomials. We wish now to extend the concept of ‘homogeneity so it will apply to functions other than polynomials. ‘If we assign a physical dimension, say length, to each variable x and y the polynomials in (1), then each polynomial itself also has a physical dimension, length to some power. This suggests the desired generalization Mf, when certain variables are thought of as lengths, a function has the physical dimension length to the kth power, then we shall call that function homogeneous of degree k in those variables. For example, the function x Sx, y= 2y? exp () ve lengthy’ when x and y are lengths, Therefore, that function | to be homogeneous of degree 3 in x and y. Homogeneous Functions mit the degree k to be any number. The function JE 47 sneous of degree zero in x and y. ft definition of homogeneity is: The function f(x, 9) is said £0 be aus of degree k in x and y if, and only if fax, dy) = 4% Ws nition is easily extended to functions of more than two vanables: Bf the function f(x, y) of equation (2), the formal definition of homo- y Ieads us to consider ° 7 ix + Uy His = orale ‘see at once that slax, ay) = BF Mi ix, y) is homogeneous of degree 3 in x and y, as stated previously. following theorems prove useful in the next section. ‘4, If M(x, y) and N(x, ») are both homogeneous and af the same The function Mix, »V/N(x, ) 's homogeneous of degree zero. fof Theorem I is left to the student. 2, IF f(x. y) ts homogeneous of degree zero in x and y. Sts 15-@ ‘of y/x alone. Let us put y = ox. Then Theorem 2 states that if/(x, 9) is homage: | Sof degree zero. f(x, ») is a function of v alone, Now A Fx, y) =Fo% 0x) = XFL, 0) = Fle Oh o Z ch the x is now playing the role taken by 2 in the definition (3) By” ¥)) depends on v alone, as stated in Theorem 2: 2 i sin each exercise whether or not the function is state the degree of the function. ca Equations of Order One {ch.2 se 6. tan x rep (2) 8 an y x 40, xsin = y sin. 9. (xt + yep a a a) a ee Bt dy n : im x= dy 13, (u? + oP 1S»? tan = acre +46 oe 1 xtnx—yiny, 1. 18 n~ ¥ 20, x In x—xIny, ‘Ams. All functions are homogeneous except those of Exercises 2, 5. 6, and 19 9. Equations with Homogeneous Coefficients ‘Suppose that the coefficients M and N in an equation of order one M(x, y) dx + N(x. y) dy = 0, a are both homogeneous functions and are of the same degree in x and y. By Theorems 1 and 2 of Section 8, the ratio M/N is a function of y/x alone Hence equation (1) may be put in the form BE 2. a? =o @ ‘This suggests the introduction of a new variable » by putting y = vx. Then (2) becomes eee tiien ql in which the variables are separable. We can obtain the solution cof (3) by ‘the method of Section 7, insert y/x for v. and thus arrive at the solution ! (1), We have shown that the substitution y = vx will transform equation a) “imto an equation in v and x in which the variables are separable, Peace Equations with Homogeneous Coefficients, a above would have been equally successful had we used to obtain from (1) an equation in y and v. See Example (6). (x? — xy + y?) dx — xy dy =O. @ the coefficients in (4) are both homogeneous and of degree two in = Tet us put y = vx. Then (4) becomes (x2 — x20 + x0) dx — x2x{v dx + x do) = 0 ‘which the factor x? should be removed at once. That done, we have 1 (1 — 0 4 0) dx — 0 dx + x do) = 0, (1-0) dx — xv dv = 0. of solutions is seen to De In |x| #o+Injo— t= intel, xe Nera e é ‘of the original variables, these solutions are given by eI Equations of Order One 1Ch. 2 EXAMPLE (b) Solve the equation xy dx + (x? + y3) dy =0. (3) Again the coefficients in the equation are homogeneous and of degree two. We could use y = ox, but the relative simplicity of the dx term in (5) suggests that we put x= vy. Then dx =o dy + y dv, and equation (5) 1s, replaced by wytu dy + y dv) + (y+ y*) dy = wo dy + y de) + (v* + 1) dy =0. Hence we need to solve vy dv + (20? + 1) dy =0, o which leads at once to In 20? + 1) +4 In |y| = In, « YiQu? + I) =e ‘Thus the desired solutions are given by that is, V2x? + yy) =e. a Since the left member of equation (7) cannot be negative, we may, for sym ‘metry’s sake, change the arbitrary constant to cf, writing PRxe + v?)— eh. It is worthwhile for the student to attack equation (5) using y = vx. That ‘method leads directly to the equation (@? + 20) dx + x{0? + 1) dv = 0. Frequently in equations with homogeneous coefficients, it | terial whether one uses y = vx oF x = vy. However, it is sometimes: -guibstitute for the variable whose differential has the simpler coefficient. Equations with Homogeneous Coefficients ses 1 through 21, obtain a family of solutions. xt +?) dx — 2xy dy =0 = 29) dx + (2x + y)dy = 0. B+ y3) dx — xy dy ide — (37 + 397) dy ~ 0. me + Tay + 29% dx +(x? +?) dy = 0. wa An. 2 = 2 + yh Ame inet +99 Aiea ‘Ane. xt =} A ‘Ans, x4 = 6y? In |y/el Am. ty +'20) = y+ ‘iw ae ae Ans. xy +2)? = ety — 2a ‘Ams (Go+ 1)? = cf — wh Ans, 32 + ay? mtx +h Ans. (x7 + y*P = 6y* In Je/¥|- Am. (x! + 2y)) mee ‘Ans. yx + y)= cet Idx + x(x + 0) dv = 0 ‘Ans. x0? = ofx + 20), tes (y/x) — y] dx + x dy =0. Ans. In.|x/c| = cos (y/a) Das + sin? (y/x)Ly dx —x dy] = 0. Amn 4x in |t/el = 2y + sin yl Gay iny ty inx) dx + xin y— In x) dy=0. ‘Ans. (x —3)lnx +yln y= ex Ey arc tan (y/x)) dx + x arc tan (y/x) dy = 9. ‘Ans. 2y are tan (y/x) = x tn Cet? + Ve. ‘Ans. y In Lylel = = ‘aos. 5? = —2t7 In fest. ‘Ans. arc sin (x/y) = In |y/el ‘Aas. (y ~xXy + 3x) =O. eA xy — 4y") de — (3? he + yy de — x dy) +9 dy den (e+ Jy" Gxt — xy 39°) de vipa) dx + Hx, yky dx = x dy) = OF Fe H(-x, y) is homogeneous in x and 5 23 through 35, find the particular solution indicated: ays? Ane. 22x +39) + (8 + In (x +9) = 0. EEE de ~ x dy = Ocwhen x= Oya t Ame adap UCP) dx — xy = Ovwben x= J =), A when x = 1 =n my) dx +x ‘Gay + V6x2) dx + x2 dy & Of when x= 1 yh f ss Am x—y= Sy +4 2) dy = 0; when x = y= I te 4 (x? + dey + 4’ be + 6x2 + 29°) dy = 0; when x= 0 y= egy + yas — x'0x — dra Oi whee 32 Equations of Order One Ic Bi. 9x — 2y) dx — 16x — y)dy = O; when x= I, y= I. Am. 3x? -x*y ~ 2)? = 0, 32. ye? + Pax + Sx? — Sy") dy = 0; when x = 2 y= 1 Am 2y? — 2x77? + 3x <0, BB. (16x + Sy) dx + (3x + y) dy = 0; the curve to pass through the point (1, — 3), ‘Am y+ 3x (y+ 43)In(y + 40, 34 x + 20) dx — x? dv = 0; when x = |. Am 2x? + 2x" — v= 0. 3S Gx? = 2y")y' = 2xy; when x = 0, Am x= 27+ 1 ‘36 From Theorems | and 2, page 27, it follows that if F ixhomogeneous of degree ‘ein x and y, F can be written in the form raed) " ‘Use (A) to prove Euler's theorem that, if F is a homogeneous function of degrce Kin xand y, ar oF Rtg 10. Exact Equations In Section 7 it was noted that when an equation-can be put in the form Alx) dx + Bly) dy = 0, ‘a set of solutions can be determined by integration, that is, by finding » function whose differential is A(x) dx + B(y) dy. ‘That idea can be extended to some equations of the form M(x, ») dx + N(x, y) dy = 0 0 in which separation of variables may not be possible. Suppose that a func tion F(x, y) can be found that has for its total differential the expression M dx +N dy; that is, dF = M dx +N dy. @) ‘Then certainly Fix ym ° defines implicitly a set of solutions of (1). For, from (3) it follows th: dF =O, or, in view of (2), M dx + Ndy=0, Exact Equations | : ‘things, then, are needed: first, to find out under what conditions on Na function F exists such that its total differential 1s exactly WN dy; second, if those conditions are satisfied, actually to deter- function F. Ifa function F exists such that M dx +N dy y the total differential of F, we call equation (1) an exact equation: ‘equation M dx+Ndy=0 a then by definition F exists such that dF = M dx +N dy. calculus, aF Sax +54, +5” derivatives are continuous. Therefore, if (1) 15 4m d that these partial ‘equation, then om oy Ip to be exact itis necessay ly ‘show that, if condition (4) is satisfied, y) be a function for which ei oo M Equations of Order One [Cha The function ¢ is the result of integrating M dx with respect to x white holding y constant. Now te _ aM dy ax” ay’ hence, if (4) is satisfied, then also 26 oN Gx dy” Ox” 6) Let us integrate both sides of this last equation with respect to x, holding y fixed, In the integration with respect to x, the “arbitrary constant” may be ‘any function of y. Let us call it B'(y), for ease in indicating its integral. Then integration of (5) with respect to x yields a = =N4 BU) ) ‘Now a function F can be exhibited, namely, F = (x, ») — BO), for which ote 22 ry) dy aF 5g ay oY — BU) dy = M dx +(N + B(y)) dy — BY) dy =Mdx+N dy. Hence equation (1) is exact. We have completed a proof of the theorem stated below Theorem 3. If M,N, 2M/dy, and ON/2x are continuous functions of x and y then a necessary and sufficient condition that M dx+Ndy=0 qa) be an exact equation is that am _ aN. “) dy Ox Furthermore, the proof contains the germ of a method for obtaining a s), where Q(x) will be determined from (12). The latter yields =xy? — 2y + Q(x) = 2x? — xy? — 2y + 3, Q(x) = 2x? +3. ‘Therefore, Q(x) = tx* + 3x, and the desired set of solutions of (11) is defined implicitly by a dxty! — dey + dxt + 3x = 4c, A et eam . Exercises “Test each of the following equations for exactness and solve the equation. The equa tions that are ot exact may be solved by methods discussed in the preceding I AG tx + pax + x — yNdy = 0. Ame tp dey yin M6x + y") dx + f2x — By) dy = 0. Am 3x? + xy? ye Qxy — 323) dx +(x? + yy =0. Am ty + by = 6 (f) ~ Bry + 6x) dx ~*~ 2xy + Nady =O. Am xy? —x*y + Ix? — Dy =o ay + yhde +(e? — x)dy =O. ed ene = 1" Be — det Ay — ah dy 0 Am x + dy? = ary te phere ‘The Linear Equation of Order‘Ove a” 6 by another method, 3y) dx + (2y ~ 32) dy = 0. 8 by another method B= 3) di + (3u?e? — 3u + 40) dv = 0. Am dite —3u + 2m © 3x2)?) dx + (008 2y — 2x sin 2y — 2x*y) dy =O hon} tin 2y + x 008 2y = 2998 =e ‘Ans. 2 arc tan x +In(1 + 9) a Am 2x + (1 +2P =e Ans xt + yhm day te idx 4 (xty + 9) dy = 0. 4 ny?) di + (xy + y + Day) dy =a) dw + + wie — w) de An (w+ 2) = awe +6 de + (42 — x sec! y) dy = 0 ‘Am. xty—x tan ye fpieee y — cot x) dx — sin x sin y dy =0. Ame sin xcos y=ln|e sin sh i tin @ — cos 6) + risin 0 + cos 0) db = 0 ‘Asa. 1? + 2nsin @ — con 8) = 6 Say? +6) dx + (x? — 6x" Am Oya 2ely + Isto yee IB — 2r cos? @ dr +r cos H2r sin 9 +1 do = 9. Am rain 0— 7? cost Ome. hu 2 sin Gy) =e Am yx? + Ppa Am xi + y= ey am. xy? + xy ee when x = 0,y= Am. xXx? 3) = 41 — er 2 4 x41 — xy?) dy = 0; when x= 2p = I a” me Am xy —y? + Sxy— 3a = 2 x) dx + (x + 2xy — y sin 2x) dy Y + 2y? sin® x) dx +(x + 2 Oa ee 24 ay) 4 exp (2) dy = An. xy +) + 2x? + yeap yee be + xy dy = 2004 9) dy when Ba meee tn? he Tay = 2 4 (et a) = B92 1S: Ub y cos (xy)] dx + x cos (xy) dy = 0 dx + (9? + x1) dy =0. idx + (y? — x7) dy = 0. ay — 2) dx + xixy + 1) dy = Si) dx + (x? + By - 3) dy 4+ y—yexp(— x79] dx +e The Linear Equation of Order One Qn ied first-order differential equations that were exact, If 0 i i ate ‘of an appropriate factor, which is then called an d, in Section 7 ws multiplied by an integrating factor ‘variables and thereby obtain an exact equation. — very little can be said about the theory of int ‘equations, In Chapter 4 we shall prove some ftasistance in a few isolated situations: J Equations of Order One 1Ch. 2 class of equations, however, where the existence of an inte, : " grating factor can be dcraountrated, Thir its caus oflieear equations of order one. An equation that is linear and of order one in the dependent variable y ‘must by definition (Section 2) be of the form dy Als) Fe + Bey = Cl). (1) By dividing each member of equation (1) by A(x), we obtain a a 2 2 + Peay =O @ which we choose as the standard form for the linear equation of order onc For the moment, suppose that there exists for equation (2) a positi« integrating factor (x) > 0, a function of x alone. Then ce 2, rey] = x) Cx) @) must be an exact equation. But (3) is easily put into the form M dx +Ndy=0 with M = vPy — 0 and Nao, in which v, P, and Q are functions of x alone. Therefore, ifequation (3) is to be exact, it follows from the requirement ae ek éy ~ ox that » must satisfy the equation (4) “From (4), v may be obtained readily, for Pdxa®, 7 Equation of Order One » inv [pax v= exp ({ Pax) © if equation (2) has a positive integrating factor independent of ¥, factor must be as given by equation (5). ins to be shown that the v given by equation (5) is actually am dy aa + Pely= O08), 2 ‘multiply (2) by the integrating factor, obtaining D wio({ rte) Z+ren({rep=cen (fra) member of (6) is the derivative of the product yero({? a) st member of (6) is a function of x only. Hence equation (6) is exact, is what we wanted to show. WF course, one integrating factor is sufficient. Hence we may use in the nt (f P dx) any function whose deriv P. of the great importance of the ideas just discussed and the fre- Seecurrence of linear equations of first order, we summarize the steps in solving such equations: Put the equation into standard form: eyes Z+ Pye d tain the integrating factor exp (f P dx) {ultiply both sides of the equ ting factor. Wve the resultant exact equation. n integrating the exact equation that the ys the product of the dependent “ Equations of Order One tena EXAMPLE (a) ‘Solve the equation My — 4x3) dx + x dy = 0. ‘The equation is linear in y. When put in standard form it becomes, 2 Bely-sx when x40 © exp (2 In |x) = exp (In x?) = x’. ‘Next apply the integrating factor to (7), thus obtaining the exact equation 4 eZ +ry a8, ® which may immediately be rewritten as 4 Ee=&. o By integrating (9) we find that aty = xt +0. (10) ‘This can be checked. From (10) we get (8) by differentiation. Then ‘he original differential equation follows from (8) by « simple adjustment ‘Hence (10) defines a set of solutions of the original equation. . EXAMPLE (b) Solve the equation y dx + (3x - xy + Ddy =O. Since the product y dy occurs Fe etn Derm the equation isnot linear in y. 1 ydx+(3- yx dy= —2dy ‘and pass to the standard form, S+Q-)eag meres JQ -1)a-simi-yren co) ‘The General Solution of a Linear Equation a integrating factor for equation (11) is. exp (3.n |y| — y) = exp In|yl)-€- = exp (in |yP)-e-? =ly Ihat for y > 0, ye~” is an integrating factor for equation (11) and =yPe"? serves as an integrating factor. In either case we are led equation yre-? dx + y'3— ye?x dy = ~2y8e? dye we get yreP'dy xe = 2y2e? + dye? + de amily of solutions is defined implicitly by xy? = 2y? + 4y + 4+ cet + The General Solution of a Linear Equation Beginning of this chapter we stated an existence and. walateases Te: tratorder differential equations. If the differential equation im happens to be a linear equation, we eam prove & ‘statement. the linear differential equation B + Pixdy = Qt) a that P and Q are continuous functions on the interval <2 -<05 aaa e nn any number in that interval I vy i Sal arbiiracy eas Mae Geists 2 unique solution y = 248) of differential tuations G9) ‘satisfies the initial condition this soluti of this theorem has essentially Deen ‘of equation (1) by the = Equations of Order One [ch integration gives pairs: Since v # 0, we can write It is a simple matter to show that since » #0 and v is continuous on a Mey +x Day A dm p= Xe +) WS Gx? — dey — Ph dx sy eo Am x? me ciy— x)exp(y Me He! + Phas + Se — Sy) dy = Ov when v= Dy =| Am. 2y? — 2x*y? + 3x = ‘97. ¥ + ay = b; 2 and b constants Solve by two methods, Am. y= bia + ce © ‘TR &e — idx — (x + yi dy = 0 Solve by two methods. «Am. x’ — 2xy — = 19, delat = con x con? ‘Am 4injsecx + tan x| = 20+ sin 2 + aoe Am xsiny+ycosx=c BA (1+ ay — Aap de + (x — 2) dy =O; when x = 2 y= Am ixtye + des 2inix—! ity = 247-9 Am yecty—dexpix* ER Gy con x + sin* x) dx = sin x dy; when x = bx y= 1, me? Ams ym 2sin? x sin? $< UA yids — ay) = 2 dy + 7 de ney inicry erate asp ecco aa ‘Am Dare tan (y/a) = In [efx + ab(x - «) Bo-w 2eserty ate eases sca = es mame ons Ate Am wis dny tyne ME Bey Be) dx + (2! + Dy) dy = Am ty- ty nercine 29 through 3 reall that the principal ofthe aes ano “hae seein x + are sn y = ‘Soma ep ada ta temee oo “The General Solution of 2 Linear Equation s x TS — {x or that arc of the ellipse: any sli nein Figure Son of Ese condition that when «=O ame wenn > Grrr that ar of the ligne Iv te xy + y= 2 that is indicated DY 2 light solid line in Figure 8 toa ee and. 3i have been deleted ihe that after the answers ig arcs of the ellipse erent pe not solutions of the differential equation: [iar ax + (Ta ay 8 thi purpose consider the sgn ofthe slope of the carne ‘the equation Aap ax- /T= Rayne xercises 29 through 32. fend solve four problems analogous 10 E Am tum cel — (e+ Uh eo = aye x= tye ol — saps ey t2en% aye me Bay = Pde — cP oy ant ae -hoel Stay ty? - Day = orwnen 1 “ Equations of Order One Ich. BD. Wy? — 3x9) dx + 2? dy = M0. y= cos x — y sec x; when x = 0, y = 1. ‘Abe, y(1 + tin x) = cos x(x + 2 ~ co 4 Fie that sation oy = 3x +> which passes cvough the point —t.0, Am. 2x8 = xt +0) i Am. y= 3x + en cnn dior Oo poi” ‘Ass. = exp (« + 1) — mone es: 4. (e? = 1 + 2) dx + (1 = 7) dy = Of when x = By =). Ame tee tie Hie 4 1 +210 0) ema ees. S07 ey Ca ae dew yox- y LP + yh de ebay + dye Or when xy — tA 2 + Steamy Am Bax + yim ex's ty) @ yew ra punk Am yotnztec eas ee Am. itesy| en xy mx y+ xy tan, Ame xy con cmc +08 x + x sin Re eee 0 ee Ans. erty +0) x sin y + tan x) dx — cos x cos y dy = 0. Ams, x win |c sec ‘ So ey Am. 2x%(x — 2)y = x? — 5. felocity of Escape from the Earth 1 probtems involve differential equations of erder one: if the problem of determining the velocity of a particle projected Bet direction outward from the earth and acted upon by nly one Pgravitational attraction of the earth. Sail assume an initial velocity in & radial direction so thar the the particle takes place entire Jy on a line through the center of to the Newtonian law of gravitation, the acceleration of His Ie inversely proportional to the square of the distanes fam the Bike canter of the earth. Let r be that variable Gistanes sams I Peeine earth, If represents time, ¢ the velocity, Othe! Bartels a gad k the constant of proportionality i. she Newianlian ion is ne Hs negative. When r=. “s Elementary Applications Icha gravity at the surface of the earth. Thus from which ‘We wish to express the acceleration in terms of the velocity and the dis tance. We have a = dv/de and v = dr/dt. Hence 5 do a-$-5 s0 the differential equation for the velocity is now seen to be dv _gR? Fig Reale af a ‘The method of separation of variables applies to equation (1) and leads at ‘once to the set of solutions 2gR? es . Suppose that the particle leaves the earth's surface with the velocity vo ‘Then v= vo when r = R, from which the constant {is easily determined to be C= 03 — 20R Thus, « particle projected in a radial direction outward from the earth's surface with an initial velocity og will travel with a velocity v given by the ‘equation 0 = WE 5 0 — 29R @ It is of considerable interest to determine whether the particle will escape rs velocity is ‘= #9. An examination of the rit member of equation (2) shows ‘that the velocity of the particle will remain positive if, and only if 03 — 29R ZO. eo Newton's Law of Cooling, - ity (3) is satisfied, the velocity given by equation (2) will remain use it cannot vanish, is continuous, and is positive at r= hand, if (3) is not satisfied, the vj ~ 2gR <0, and there will be fue of f for which the right member of equation (2) is zero, That ‘would stop, the velocity would change from positive to mens @ the particle would return to the earth. Projected from the earth with a velocity 9 such that Be from the earth. Hence, the minimum such velocity of projections Jak Oy Bi the velocity of escape. Bessie of the earth is approximately R = 3960 miles. The senties: previty at the surface of the earth is approximately g = 32.16 feet Biper second (ft/sec), or g — 60X10) > mile/see', For the earth, ty of escape is easily found to be vy soltne gravitational pull of other celestial bodies the moos. Sanus’ and so on, has been neglected in the idealized! proves Se 1k is not difficult to see that such approximations are j Be Yatereated in only the critical initial velocity ¥.= Whekies She Bercy recedes from the earth forcver or becomes for eReOass Be ioine heavenly body is of no consequence i i: Bears study we happen to be thinking of the particle 88 80 idealization Hate rocker then other elements most be considered A resist= ible. Methods for overcoming feulties are not suitable topics for discussion here He aoalized that the formula v, ~ /20R applies ealually well for Bat escape from the other members of the sales ESSA Jong, fare given their appropriate values Newton’s Law of Cooling Sit has shown that under certain conditions, a good |to the temperature of an object can be obtained by ‘The temperature of & the difference in temperatu 0 Elementary Apptications Ich. 3 the body itself. We shall assume here that the constant of proportionality is the same whether the temperature is increasing or decreasing. ‘Suppose, for instance, that a thermometer, which has been at the reading 70°F inside a house, is placed outside where the air temperature is 10°! ‘Three minutes tater it is found that the thermometer reading is 25°F. Ww. wish to predict the thermometer reading at various later times. ‘Let w (°F) represent the temperature of the thermometer at time ¢ (rn the time being measured from the instant the thermometer is place: outside. We are given that when ¢ = 0, u = 70 and when t = 3,u = 25 According to Newton’s law, the time rate of change of temperature, dis is proportional to the temperature difference (u — 10). Since the thermo ‘meter temperature is decreasing, it is convenient to choose (—K) as the con stant of proportionality. Thus the wis to be determined from the differential ‘equation ~~ Mu 10, () and the conditions that when t= 0,u= 70 2 and when t= 3,u= 25. QB) We need to know the thermometer reading at two different times because there are two constants to be determined, k in ¢quation (1) and the “arbitrary” constant that occurs in the solution of differential equation (1) From equation (1) it follows at once that w= 10+ Ce™ ‘Then condition (2) yields 70 = 10 + C from which C = 60, so we have u= 10 + 60e 4 ‘The value of k will be determined now by using condition (3). Putting f= 3 and u = 25 into equation (4), we get 25 = 10 + 60e-™, from which e~** = 4,20 k = $n 4, ‘Thus the temperature is given by the equation w= 10 + 60 exn /—4¢ In 4), ” Since In 4 = 1.39, equation (5) may be replaced by w= 10 + 60 exp (0.461), © Simple Chemical Conversion st ple Chemical Conversion mount of unconverted substance be known at some specified fet x = Xo at ¢=0. Then the amount x at any time t >is by the differential equation i di o dition that x = xo when t = 0, Since the amount x is decreasi pases, the constant of proportionality in equation (1) is taken to ion (1) it follows that Biwhen 1 = 0. Hence C = xq. Thus we have the result seer Q fadd another condition, which will enable us to determine k. Ge known that at the end of half a minute, at ¢ = 30 (sec), two fofiginal amount xo has already been converted. Let us deter- jmuch unconv 60 (sec), Gsthirds of the substance has been converted, one-third remains d. Hence x 30. Equation (2) now yields the rela- .d substance remains at € 4x0 when fk is easily found to be ys In 3, Then with ¢ measured in seconds, Bt of unconverted substance is given by the equation Ast In 3). a) x = Xo exp ( 21n 3) = xo"? = Bor Xp &XP ( Bs of the moon is roughly 1080 miles, The acceleration oo oe ic 0163p, where g i the acceleration of grav oe i apomine the velocity of eveape (Or the RANA is BN 2 Elementary Applications res 2 Determine, to two significant figures, the velocity of escape for exch *he ce tial bodies listed below. The data given are rough and g may be Wuke,. (a 6.1110)" * milejne?. ‘ahd ‘Acceleration of Gravity Radius Answer a Supece Crlles) _(milesiic) Venus, 0859 3,800, 63 Mars 038g 2,100 31 Jupiver 26g 43,000 ” ‘Sun 28g 432,000 380 Ganymede O12 1.780 16 3. A thermometer reading 18°F is brough om where the temperatur. 70°F; 1 min later the thermometer reading is 31°F. Determine the temperature reading as a function of time and, in particular, find the temperature reading § ‘min alter the thermometer is first brought in’o the "sor Am wm 70— 5: exp (—1 291); whens = 5,u~ 58 4 A thermometer reading 75°F is taken out where the temperature is 20 F The reading is 30°F 4 min later. Find (a) the thermometer reacling 7 min after 1 thermometer was brought outside and (b) the time taxen for the reading to drop from 75°F to within a half degree of the air temperature * Am (a) 23°F; (b) 11.5 min & At 1:00 pa, a thermometer reading 70°F is taken outside where the air tem perature is —10°F (ten below zero). At 102 F.m., the reading is 26°F At 1:05 Ps, the thermometer is taken back indoors where the air is at 70°F What is the thermometer reading at 1:09 p.? Am. 56°F 6 At 9 4m, a thermometer reading 70°F is taken outuvors where the temper is 15°F Ar 9:05 am, the thermometer reading is 45°F. At 9:10 Aq. the ther mometer is taken back indoors where the temperature is fixed at 70°F. Find (a) the reading at 9:20 am. and (b) when the reading, 10 the nearest degree, ll show ‘ne correct (70°F) indoor temperature. ‘Ams. (a) 58°F; (b) 9:46 40 7. At 2:00 rm, a thermometer reading 80°F is taken outside where the air term perature is 20°F. At 2:03 ra. the temperature reading yielded by the therm ‘meter is 42°F. Later, the thermometer is brought inside where the air is at 80 ‘At 2:10 9. the reading is 71°F. When was the thermometer brought indoors” im. At 2:05 7 | Suppose that « chemical reaction proceeds according (o the law given in Section 15 above. If half the substance 4 has been converted at the end of 10 sec, find when nine-tenths of the substance will have been converted Awe. 33 s0c '%. The conversion of a substance B follows ine law o°ed in Section 15. If only 8 fourth of the substance has been converted at the end of 10 sec, find when ‘nine-tenths of the substance wil, |..ve been converted, ‘Ama. 80 sec ‘I. Foor « substance C, the time rate of conversion is proportional to the square of the amount x of ure mveried substance Let k be the numerical value of (he Simple Chemica! Conversion ce ‘of proportionality and let the amount of unconverted substance Be BE Determine x for ale z0 fon <~ oll + xpht) tance D, the time rate of conversion is proportional tothe square Amount x of unconverted substance, Lat & be the numerical value of tof proportionality. Show thatthe substance ill appear i finite ibtarmine that time As ta Dale lances, A and B, nrc being converted into u single compound C tn the Wt has been shown that, for these substances, the following law of holds: the time rate of change of the amount x of compound € fs onal to the product of the «mounts of unconverted substances A and B. ithe units of measure so chosen that one unit of compound C is formed. feombination of one unit of A with one unit of B. If at time ¢ = O there Bits of substance A, b units of substance B, and none of compound © ‘show that the law of conversion may be expressed by the equation dx So ka — xyb — x) a Jequation with the given initial condition abtexp (b- okt =). py og xo tHE Pe bts ob omer ake +1 lution of Exercise 12, assume that k > 0 and investigate the behavior of Ame Wb2a,xa:ifb Sax—b decomposes at a rate proportional (o the quantity of radiam present, that it is found that in 25 years approximately 1.1% of a certain quan- rad how long it will take thaif the original amount of radium to decompose. Ama. 1600 years itadioactive substance has a half-life of 38 ht. Find how long it takes M% of the radinactivity to be dissipated Acs. 120 hv. {population B is known (o have a rate'of growth proportional to & ween noon and 2 p.m. the population triples, at what time, no con- ing exerted, should B become 100 times what it was at noon? ‘Ama. About “ ion of an object through a certain medium (air at certain pressures is the medium furnishes a resisting force proportional to the square locity of the moving object. Suppose a body falls, due to the action of pthrough such a medium. Let ¢ represent time, and v represent velocity, ;nward, Let g be the usual constant acceleration of gravity and Tet w: of the body. Use Newton's law, force equals mass times aoceler- that the differential equation of the motion is wa @ == ho? “ Elementary Applications Icha I Solve the differential equation of Exercise 17, with the initial condition ip, ‘=r when ¢ = 0. Introduce the constant o* = w/k to simplify the formu. 19, There are mediuims that resist motion through them with a force propoctioy to the first power of the velocity. For such a medium, state and solve problems analogous to Exercises 17 and 18, except that for convenience 1 convton, ~ w/k may be introduced to replace the a* of Exercise 18. Show that b huss dimensions of a velocity Ae = b+ (0 ~d) exp ( 20, Figure 9 shows a weight, w pounds (Ib), sliding down an inclined plane that makes an angle a with the horizontal. Assume that no force other than gravis is acting on the weight; that is, there is no friction, no air resistance, and so At time ¢ = 0, let x = xg and let the initial velocity be vo. Determine x for» 0 Ams x= dat sina + 191 + 5, FIGURE 9 Along. very smooth board is inchned at an angle of 10° with the horizontal § ‘weight starts from rest 10 ft from the bottom of the board and slides down tinder the action of gravity alone. Find how ong it will take the wight (© r the bottom of the board and determine the terminal speed Ans. 1.9 sec and 10.5 tse 22. Add to the conditions of Exercise 1) a retarding force of magnitude kr. he ie the velocity. Determine » and x under the assumption that the weight sc (som rest with x = x9. Use the notation a = ha/w Ams. v =a 'gsinall —¢ “);x=xo +a *gsinaj—1 te “ +o") 23, A man, standing at 0 in Figure 10, holds a rope of length » 19 which a wen! sttached, initially at Wy. The man walks to ‘he’. sagging the weigh! 3! him. When the man is at M, the weight is at W Fiad the differential equation the path (called the Wactis) of the weight and solve the equation, gay ete fat 24. A tank contains 80 gallons (gal) of pure water. A brine solution with 2 Ib/ea! ‘salt enters at 2 gal/min, and the well-stirred mixtur- leaves at the same ‘2 Loglatic Growth and the Price of Commodities s FIGURE 10 lhe amour of salt in the wank at any time and () the time at which the fang wi contain 1 b/g ofa i . ‘Ans. (a). 160{1 — exp (—1/40)); (b) ¢ = 40 In 2 min. Bank in Exercise 26 determine the limiting vale forthe amount of elt ink after a long time. How much time must pass before the amount of Ma youches 80% of this lmiting value? yy ‘Ans. (a) = 160 Ib; (b) ¢ = 64 min, am of money P draws inveret compounded continuously, tf ats Bc dilare in the acsount, determine the time when the re reac vaiac 2Ps dollars the anal interest cates (a) 2% oF () ‘km. (a) 30m 2 yeas (e) 25 2 yea afters 5% interent compounded continuously In savings assent ri fa) the amount of interest earned in | year on deposit of $100 and fate if the compounding were done annually, valent rate if the compounding. ae istic Growth and the Price of Commodities re been jevelop models to study the growth: ict ar een mae 0 se te he average birthrate per individual is a positive constant anid gage death rate pet individual is proportional to the population, ~ St) represent the population at time f, then the above: assump the differential equation sj ; - Elementary Applications Ich where b and a are positive constants. This equation is commonly called the Togistic equation and the growth of population determined by it is calieg logistic growth. "The variables in the logistic equation may be separated to obtain dx o-a~* or ate. Grrta)e-o* Integrating both sides gives us ee ea or lpaal-ee a To expedite the study of equation (2), let us further assume that at (= 0 the population is the positive number x. Then equation (2) may be »ri't ee Bax" b— ax, and upon solving for x, we have bxoe™ Ma, range It is interesting to not (3) has a limiting value iat the population function obtained in equation bxoe then ih Shea a ee Bate Horee ~ im BES e , ee where we have used I"Hospital’s rule to evaluate the limit. ‘We should also note that the logistic equation (1) will dictate a growth or ‘s decline in the population depending upon whether the initial population is Jess than or greater than b/a. Logistic Growth and the Price of Commodities ” ‘example of an application in which a first-order differential is, we consider an economic model of a certain commodity ‘assume that the price P. the supply 5, and the demand D of ity are functions of time and that the rate of change of the tional to the difference between the demand and the supply. @ sitive so that the price will ithe demand exceeds the supply Bat models of the commodity marke ‘of the demand and netions swe assume (hat Dec-dP an s will result depending upon hat are indicated. If, for a+ bP, o Ie. and d are positive constants, we obtain a differential equation aP = k{lc - a) — (d+ OPI ©) in P. The assumptions (5) reflect the tendency for the demand fas the price increases and the’ tendency for the supply to the price increases, both reasonable assumptions for many com We should also assume that 0 < P-

You might also like