fmincon returns a message that no feasible solution found

I have a convex problem, linear objective, linear and nonlinear constraints. When I run a feasibility probleb (objective function is 0), the solver returns a feasible solution. However, with the same problem setup with a linear objective function, the solver returns a no feasible solution found. Any idea what the problem could be

Answer 1

[Vidip]

When you encounter a discrepancy in the feasibility of solutions between running a problem as a pure feasibility problem (objective function = 0) and introducing a linear objective function, several factors could be at play like different solvers may handle the balance between objective optimization and constraint satisfaction differently, also tolerances for feasibility and optimality of optimization solvers. When you introduce an objective function, the solver's behaviour regarding these tolerances might change, affecting the feasibility check. You can refer to the below documentation links to know about how to set solver tolerance and obtaining solution using feasibility mode–

• https://in.mathworks.com/help/gads/setting-tolerances-for-the-solver.html
• https://in.mathworks.com/help/optim/ug/optim.problemdef.optimizationproblem.optimoptions.html
• https://in.mathworks.com/help/optim/ug/feasibility-mode-optimization.html