In some cases, an optimization may not converge. In this case, GoldSim will provide one of three warning messages:
Unable to create valid initial complex. Optimization failed. In order to start the optimization, GoldSim must find 2N valid solutions, where N is the number of optimization variables. This error indicates that GoldSim could not find 2N valid solutions. This probably indicates that your problem is over-constrained. That is, GoldSim could not find any combination of optimization variables (within the bounds specified) that met the specified constraint (the Required Condition).
Cannot improve on best solution, but failed to converge. Optimization is probably successful. This indicates that the optimization has found a number of valid solutions, but is “stuck” and cannot find any better solutions (while still failing to meet the convergence criterion). In this case, the solution is likely a good one, and it simply indicates that the convergence criterion was too strict. You should carefully examine the Top Results to determine if the optimal values are still significantly changing.
Solution still slowly improving, but failed to converge. Optimization is probably successful. This indicates that the optimization has found a number of valid solutions, and continues to slowly improve the solution, but has not converged after a large number of iterations (100 for low, 1000 for medium, 10,000 for high precision). In this case, the solution is likely a good one, and it simply indicates that the convergence criterion was too strict. You should carefully examine the Top Results to determine if the optimal values are still significantly changing.