# Any methods to solve this system of ODE when the RHS is unknown?

Solve $$\eta_k\frac{d^2C_k}{dz}(z)=-e_k, k = 1,2,3$$

$$C_1(0)=0, C_2(0)=A, C_3(0)=0$$ $$C_1(L)=B, \frac{dC_2}{dz}(L)=0, \frac{dC_3}{dz}(L)=0$$ with $$e_1 = -\beta_1-\beta_3$$ $$e_2=-\beta_1-\beta_2$$ $$e_3=\beta_2-\beta_3$$

satisfy the constraint $$-e_k\le \frac{\eta_kC_k}{K_k+C_k}$$ where $$\eta_k, K_k$$ are some constant, choose $$\beta_1,\beta_2,\beta_3$$ such that $$19\beta_1+0.5\beta_2+16\beta_3$$ take the maximum.

Any ideas to solve this system of ODEs? I guess solving analytically is impossible, so how to solve it numerically? I have been working on this for about two weeks and I still have no clue. I found it difficult since the right hand side of the ODEs are known from optimization, so I cannot directly apply well-known methods like finite difference. Also the boundary condition contains first derivative, this makes it more difficult.

Any help is appreciated!

Thanks so much!

• Your question is about a very specific equation. It might help to view it a a special case of a more general problem. Try to write down a possible general problem as well as some general problems which are clearly simpler than your general problem. I am trying to get you to remove all distracting details and create a situation where the options are clear. – Carl Christian Mar 17 at 9:12
• @CarlChristian HI Carl! Thanks for your suggestion! I have posted another question with all the distracting details removed. math.stackexchange.com/questions/3154141/… – Tony 2 days ago