Final Thoughts — Question 4

PDF ↗

Question 4

For the initial value problem y′=y2,y(0)=1,y'=y^2,\qquad y(0)=1, a candidate is f(x)=11−x.f(x)=\frac 1{1-x}. A report claims: “The right-hand side y2y^2 is smooth everywhere, so this formula gives a solution through x=1x=1 and for all real xx.”

Tasks

  1. Verify the candidate wherever it is defined. Find the largest open interval containing 00 on which it solves the IVP.

  2. Calculate its one-sided limits at x=1x=1 and explain why assigning any finite value at 11 cannot extend this solution across that point.

  3. Does the same formula solve the differential equation on (1,∞)(1,\infty)? Explain why that branch is not a continuation of the given IVP solution.

  4. Sketch both branches in your solution and correct the report’s inference about smooth equations and global solutions.

Original worksheet page 1: question and worked solution for 1-3-004
Show solutionHide solution

Question 4 – Solution

Strategy. Check the function’s domain as well as the equation’s right-hand side. A solution can become unbounded in finite time even for a smooth equation.

Step 1: Verification and interval. For x≠1x\ne 1, f′(x)=1(1−x)2=f(x)2,f(0)=1.f'(x)=\frac 1{(1-x)^2}=f(x)^2,\qquad f(0)=1. The largest open interval containing 00 that avoids 11 is (−∞,1)\boxed{(-\infty,1)}.

See the diagram in the original worksheet below.

Step 2: No finite extension. The limits are limx→1−f(x)=+∞,limx→1+f(x)=−∞.\boxed{\lim_{x\to 1^-}f(x)=+\infty,\qquad \lim_{x\to 1^+}f(x)=-\infty.} A differentiable extension across 11 would be continuous there and would have a finite value. The left-hand divergence rules this out, independently of what formula is proposed after 11.

Step 3: The separate branch. On (1,∞)(1,\infty), the same derivative calculation still proves f′=f2f'=f^2. This branch is a solution of the differential equation, but its interval does not contain the initial point 00 and cannot be joined through x=1x=1 to the IVP branch.

Step 4: Corrected inference. Smoothness of a right-hand side is not a guarantee of existence for all time. Here the equation is defined at every finite pair (x,y)(x,y), but the IVP solution escapes to unbounded yy as x→1−x\to 1^-. The failure belongs to global continuation of the solution, not to a singularity of y2y^2 at a finite point.

Original worksheet page 2: question and worked solution for 1-3-004

Original worksheet layout. Use Enlarge or open the PDF for a closer view.