1, Use the method of Lagrange multipliers to maximize 2%y° subject to the constraint 2 + Answer: Let j(e, y) — 2%y5 and ge, y) — x+y —8. The points (e,y) which will maximize f(e, y) subject to the constraint 9(.y) =0 will be the set of points (2,y) that satisfy the equations V4(7,y) = AVg(2, y) and g(e, y) = 0. Writing out the system of three equations and three unknowns explicitly: wry = A wo sty! = A @ sty = 8 @) Suppose A=0. Then eqn.(1) would be 3224° =0 which wonkd imply that cither x =0 or y = 0. (They cannot both be equal to 0, because that would contradict eqn.(3).) If 2 = 0, then eqn.(3) says y =8. And if y —0, wo would instoad have from ogn.(3) that # —8. Eithor way, wo got f(8,0) — f(0,8) ~ 0. Now suppose that #0. We can set equations (1) and (2) equal to each other and simplify: aot = sty! 3e’y® -5r%y! = 0 2y'(By—52) = 0 @) Since A#0, eqn.(I) implies that 2 #0, y #0. Therefore, for eqn.(4) to be true, it must be that 3) —! and solving this equation for y: y= 3. Plugging this into eqn.(3) and solving for 2 yields: ot fr <8 =o tr = 8 r= Band thereore y ‘And so we have /(3,5) is maximized at the point (3, = 27) (3,125) = 84, 395. Since clearly /(3, 5) > f(8,0) and (0,8), f(e,0)2. Find the distance from the origin to the plane x + 2y +22 = (i) using a geometric argument (no calculus) (ii) by reducing the problem to an unconstrained problem in two variables, and (ii) using the method of Lagrange nuultipliers, Answer: @ equation is / we know that the distance s from the point (ro, yo, 29) to the plane whose + By + €z =D ws given by the formul ro + By +O — DI VAT + B+ CE And so, plugging in the appropriate numbers:Gi): The distance from any point on the plane to the origin is f(r,y,2) = V77 FP Fe. This is the function I wish to minimize, To make Life (and taking derivatives) easier, I will instead minimize its square: Pleyy,2) = 07 +y? +22. The point (#, y, 2) that minimizes F (2, y, 2) will also certainly minimize f(x, y, 2). Le solve the equation of the plane for 2: = ea—2 =2y), and then phig this 2 into F(x, y, Foy) = L = Paes (s L = tepato = P47 +50 I find the critical imum. distance. Let that last equation equal g(x, y). Inow need to find the critical points of g(r, y). Once points, I plug them into g(@, y) and the smallest number I get will be the square of the mi (Remember, I'm minimizing the distance-squared function.) To find the critical points, I have tofind the (x,y) such that Vg(z, y) = 0i +0). So the equations I need to solve are 3,5 g(ey) = -S4zr+y=0 g(ty) = —d+r+4y=0. Solving the above for 7 and y: 2 == and y= and then we get g ( ) = 1, and so the distance is 1(iii): As in part (ii), Il instead minimize the distance-squared: F(z, y,2) =2? +y?+ 27. The constraint I have is g(x,y, 2) =2 + 2y+22—3=0, and this means the system of equations I need to solve is w= 6) 2y = 2d 6) Qe = 20 (7) rtmytoe = 3 8) Suppose X= 0. Then eqns.(5-7) would imply 2 = 04040 =3. Therefore, we must have A #0, Since A 4 0, setting eqn.(6) equal to eqn.(7) yields: 2y = 22 > y = 2. (Note that to get this, Ihave to divide by A. Since I know A 4 0, I can legally do so.) And using equs.(6) and (5), I see that y = 2x. Substituting all this into eqn.(8) and solving: =0. But then that would contradict eqn (8): © +2(2r) +2(22 L B—or=3—r=5, = rng 122 PUI sot so web a And s0 (a big surprise here): F( )) = snd that mons the minis distance is 13, Use the Lagrange multiplier method to find the greatest and least distances fram the point (2, 1,2) to the sphere with the equation 22 + y? +2? = 1. Answer: The distance from (x, y, 2) to the point (2,1, =! i f(e.y.2) = (@- 2? + Wr + +2) 24 yet and maximize sub ject to the constraint that g(z,y, 2) Using the same reasoning given in the previous problem, I will instead minin F(a,y,2) = (@ —2)? + (y- LP + (2 +2)”. Taking the appropriate derivatives, the system of equations I need to solve is 2(e—2)=22\ which simplifies to: 2(1—d) =2 (9) 2y—1)=2yA which simplifies to: y(l—d) =1 (ao) 2A24+2)=22A which simplifies to: 2(1—A) = — ay x2 4y?+22=1 which stays the same: 2244? +22 = 2) Now, eqns.(9-L1) all imply that 2, y,2 #0. (Otherwise, we would have 0 = 2) Since I know that 2, y,2 40, I can legally divide both sides of eqn.(9) by 2, both and both sides of eqn.(11) by 2. (Before di idles of eqn. (10) by ing, I need to first make sure that P'm not dividing by 0.) And (a3) (4) (15)Setting eqns.(13) and (15) equal to each other gets: and from eqns.(18) and (14): Lot's phug all this into eqn.(12): 2+(2) +caPai te =1s:? The two points we need to check are ( Therefore, the least distance from the point (2,1,—2) to the unit sphere centered at: the origin is V7= 2 and the greatest distance is WIG@ Two Constraints Suppose now that we want to find the maximum and minimum values of a function f(x, ¥, 2) subject to two constraints (side conditions) of the form g(x, y, 2) =k and h(t, y, 2) = c. Geometrically, this means that we are looking for the extreme values of f when (x, y, 2) is restricted to lie on the curve of intersection C of the level surfaces g(x, y, 2) = k and A(x, y, 2) = c. (See Figure 5.) Suppose f has such an extreme value at 4 point P(x, yo, Zo). We know from the beginning of this section that Vf is orthogonal to C at P. But we also know that Vg is orthogonal to g(.x, y, z) = k and Vi is orthogonal to h(x, y, 2) = ¢, so Vg and Vi are both orthogonal to C. This means that the gradient vector Vf(xo. Yo, Zo) is in the plane determined by Vg(xa, yo, 20) and Vi(x0, yo, Z0). (We assume that these gradient vectors are not zero and not parallel.) So there are numbers A and p. (called Lagrange multipliers) such that Vfl, AVgl%0. Yos Zo) + we Vio. Yos Zo) FIGURE 5In this case Lagrange’s method is to look for extreme values by solving five equations in the five unknowns x, y,2, A, and pL. In terms of its components and using the constraint equations: Sfi=Agi + ph: f= Agy + why L=Ag. + ph ax, y,2) =k AG y, 2) =eEXAMPLE 5 Find the maximum value of the function f(x,y, 2) =x + 2y + 3z onthe curve of intersection of the plane x — y + z = I and the cylinder x? + SOLUTION We maximize the function f(x, y,z) =x + 2y + 32 subject to the con- straints g(x, y.z) x —y +2=Tand h(x, y, 2) = 1x7 + y? = 1. The Lagrange condi- tion is Vf= A Vg + 11 Vh, so we solve the equations (17) (i) 3) (29) a] 1=A+ 2p 2=—-A + yp B=A Putting A = 3 [from (19) ] in (17), we get 2x. = —2, sox = —1/p. Similarly, (18) gives y = 5/(2u). Substitution in(21) then gives 1 Be | waand sop? = 2 y = +./29/2. Then x = ¥2/V29, y = +5/,/29, and, from (20), z= 1 —x+y=1 + 7/¥29. The corresponding values of f are “4 ‘Therefore the maximum value of f on the given curve is 3 + /29. Theceylinder * + 9? ~ Limtersects the plane x ~ y == Tin an ellipse (Figure 6). Example 5 asks forthe ‘maxim value of f when (x.y, 2) is restricted toe on the elise FIGURES.4, Find the maximum and mini the plane z= 2 -+y and the ell um values of the function f(y, 2) =2 over the curve of intersection of oid 2? 4+ 2y? + 222 = 8, Answer: So we have to minimize and maximize the function f(z,y,2) = 2 subject to the two constraints g@,y,2) = +y—2 =O and A(z, y,2) =2? + 2y2+22?—8 =0. Therefore, the equations we have to solve are VF(e, y,2) = AV Ge, y, 2) + eWH(2,y,2), gla. y,2) =O, and h(z,y,z)=0. That means, we solve LoS A+2ep ey) 0 = A+4yp (22) 0 = ~A+4ye (23) wty = 2 (24) +o 4922 = 8 (25) Note that it must be the case that pr 0. (If a =0, then equ.(21) would say A= 1 and eqn.(22) would say A=0, and we would get a contradiction.) Since I know I can legally divide by p1, from equs.{22) and (23) we have yy = — 429 > y =-2 and using that in eqn.(24): et (e)ae—perareLet's plug all this into eqn.(25): (22)? +2(-z)? +227 Ae? 4227 422? st For = =1, we have r= 2and y J(-2,1,-1) = 2. Therefore, the mi 2and y = 1, and