For the option: | Do the following: |
|---|---|
Multi-objective Method | Select the method for combining multiple objectives into a single-objective optimization problem. ■Total Cost - Optimizes the total deviation from targets (for example, the sum of the deviations for all objectives). ■Total Squared Cost - Optimizes the total squared deviations from targets (for example, the sum of the squared deviations for all objectives). ■Worst Cost - Optimizes the worst case deviation from the target (for example, the worst deviation among all objectives). |
Optimization Solver | Select the method used to solve the optimization problem. The available methods are loaded when you start Adams Insight and may vary depending on your particular installation of Adams Insight. ■MSCADS MMFD (Modified Method of Feasible Directions) ■MSCADS SQP (Sequential Quadratic Programming) ■MSCADS SLP (Sequential Linear Programming) ■MSCADS SUMT (Sequential Unconstrained Minimization Technique) ■SDI (Stochastic Design Improvement) The MSCADS solvers are conventional gradient-based optimizers. Modified Method of feasible Directions (MMFD) is generally more robust, while Sequential Quadratic Programming (SQP) is generally faster. SDI is a stochastic solver. At each step (iteration), SDI randomly generates a set of trials around the current design point. It then selects the best trial, generates a new set of trials around that point, and continues. |
Solver Settings | Select to open the Solver Settings dialog box where you can change the settings for the selected optimization solver (each solver has its own set of settings). |