摘要： 现有的比例积分微分(PID)优化设计算法难以兼顾系统对快速性、稳定性和鲁棒性的要求。为此，提出一种改进的Pareto遗传算法。该算法采用新的拥挤距离计算算法，改进非支配性的比较算法，引入双重精英机制，提高进化效率和解的质量，并且解的多样性好。将该算法应用于PID多目标优化设计，仿真结果表明，决策者可根据当前工作需求在所得的Pareto解集中选择最优的满意解。

关键词: 最优解, 传算法, 重精英机制, 例积分微分控制器, 目标优化

Abstract: The current Proportion Integration Differentiation(PID) optimization Design methods are often difficult to consider the system requirements for quickness, reliability and robustness. So this paper proposes an Improved Pareto Genetic Algorithm(IPGA), which uses a new method to calculate crowding distance, improves the comparative method of non-domination, introduces double elitism mechanism to improve efficiency of evolution and quality of solution, and increases diversity of the solution. The algorithm is applied to multi-objective optimization design of PID. Simulation results indicate that a satisfactory solution is selected in Pareto optimum set according to the requirements of the present system.

Key words: optimal solution, genetic algorithm, double elitism mechanism, Proportion Integration Differentiation(PID) controller, multi- objective optimization

中图分类号: 

• TP273