#14 in 2.4 > with(DEtools): > dfieldplot(diff(y(x),x)=y*(3-x*y),y(x),x=-5..5,y=-5..5,arrows=LINE);This is hard to look at. What does it realy mean? Notice that y' = 0 on the x-axis (y=0) and alone the hyperbola y=3/x. This may help you to understand the behavior as x gets large. If we start at a point on the negative y-axis it should be clear that y goes to -infinity as x goes to infinity. If we start at y(0)=0 then the limit is 0 since y(x)=0 is a solution. (Is this solution unique? It is hard to tell from the plot alone.) Now for y(0) positive it is less clear what happens as x goes to infinity. The curves start out going up but then turn sharply down. Do they go to 0 in the limit or to some small number? Let's "zoom in". > dfieldplot(diff(y(x),x)=y*(3-x*y),y(x),x=120..130,y=0..0.2,arrows=LINE);
I have zoomed in several times now and it seems to be going to zero. This makes sense since 3/x goes to zero. To be sure I solved the equation with dsolve: > dsolve(diff(y(x),x)=y*(3-x*y),y(x)); 1 1 1 ---- = --- x - --- + exp(-3 x) _C1 y(x) 3 9 For the life of me I don't know why Maple gives 1/y instead of y itself. But as x goes to infinity it is clear that y(x) goes to 0 for any C. But that does not make any sense!! From the graph it seemed so clear that the limit was -infinity for y(0) negative. What is going on? We will go over this in class, but see if you can figure out what is happening. Hint: plot several solution curves. BTW: We can rewrite the equation as y'-3y=xy². This is known as a Bernoulli Equation and can be solved using the method given in exercise #37 of Section 2.2.