Average gradient value and improved multi-objective particle swarm optimization based robust optimization system

Abstract

The invention discloses an average gradient value and improved multi-objective particle swarm optimization based robust optimization system. The average gradient value and improved multi-objective particle swarm optimization based robust optimization system comprises a parameter receiving module, an objective function obtaining module, a robustness calculation module, a function integration module and a function solving module; the parameter receiving module is used for receiving process parameters to be optimized according to the engineering actual situation and an optimization objective; the objective function obtaining module is used for obtaining an objective function model of the optimization objective and the process parameters to be optimized; the robustness calculation module is used for setting an uncertain domain and calculating the robustness through an average gradient value of uncertain domain samples; the function integration module is used for enabling an objective function and the robustness to form a double-objective optimization objective function; the function solving module is used for performing double-objective optimum value solution with the objective function and the robustness as the optimization objective. According to the average gradient value and improved multi-objective particle swarm optimization based robust optimization system, the applicable engineering problem range is broad and the evaluation of the robustness and the calculation of the original objective function serve as two optimization sub-objectives simultaneously to enable a designer to obtain a reasonable compromise solution between the performance index and the variation amplitude generated by the uncertain domain according to actual needs.

Description

technical field [0001] The invention relates to the field of computer data processing, in particular to a robust optimization system based on improved multi-objective particle swarm optimization. Background technique [0002] Robust optimization is a new optimization method to solve the uncertain environment of internal structure (such as parameters) and external environment (such as disturbance). It considers the uncertainty of the optimization model at the beginning of optimization. Through the optimization method, the The optimized result is insensitive to uncertain factors and the optimal unity of performance index. The classic robust optimization method is mainly used in operations research, and studies convex problems such as linear programming problems, quadratic programming problems and semidefinite programming problems with different forms of data uncertainty. However, engineering problems in the real world are often non-convex, and even have no mathematical expres...

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved in the present invention is to provide a robust optimization system based on improved multi-objective particle swarm optimization for the difficulties encountered in solving engineering problems in the prior art, which can combine the evaluation of robustness with the original objective function The calculation of is used as two optimized sub-objectives at the same time, so that the designer can choose a reasonable compromise solution between the performance index and the range of change in the uncertainty domain according to actual needs

Images