Hello all,

I am trying to solve a simple PDE with one unknow theta and three boundary conditions. Unfortunately, I receive error messages. In this case, the integer s seems to cause troubles. The error message is:

Error, (in pdsolve/numeric/process_PDEs) variable(s) {s} are in the PDE system but are not dependent or independent variables

When I assign s a value (such as 1 or 2), no error messages, but still no response. I get:

pds:=module() export plot,plot3d,animate,value,settings; ... end module

Would anyone have an idea how to resolve this issue? Thank you for your suggestions. Below is the small portion of code

restart;
with(PDEtools);

PDE := [(1-y^(s+1))*(diff(theta(y, z), z)) = diff(theta(y, z), `$`(y, 2))+y^(s+1)*(s+1)^(1/s+1)];
IBC := {theta(1, z) = 0, theta(y, 0) = 0, (D[1](theta))(0, z) = 0};

pds := pdsolve(PDE, IBC, numeric);
Error, (in pdsolve/numeric/process_PDEs) variable(s) {s} are in the PDE system but are not dependent or independent variables

Hello,

I have 20 by 20 matrix (with only two symbolic components, say a and b), other entries of the matrix are either populated with zeros or real numbers. I am trying to obtain the inverse.

All the command I have tried returned an error message. In particular M^(-1) yields:

Error, (in radnormal/ifactors) too many levels of recursion.

I would appreciate any suggestion that could solve this issue. Thank you

ps: I am essentially trying to solve Ax=b. The commands within LinearAlgebra all yielded similar error messages.