【摘要】 针对现有约束粒子群优化(PSO)算法存在的算法复杂、应用范围受限、优化效果不佳等缺陷,提出一种新型约束粒子群算法。该算法采用目标函数替换的方法将约束优化问题转化为非约束优化问题,具有简便易用的优点。通过典型测试函数测试并和其他具有代表性的约束PSO算法进行对比,表明该算法在处理约束优化问题上的优越性。为了验证该算法应用于工程的可行性,以样例导弹纵向模型为对象,针对经典Raytheon控制结构,采用该算法设计了μ-PID控制器。仿真结果表明,样例导弹控制器可以在满足多种时域指标的同时具有良好的鲁棒性能,达到了设计指标要求,验证了所提出算法的有效性。更多还原

【Abstract】 The existing constrained particle swarm optimization(PSO) algorithms have the disadvantages of algorithm complexity,limited application and poor optimization. A kind of novel constrained particle swarm optimization algorithm is proposed. The proposed algorithm converts a constrained optimization problem into an unconstrained one using the method of objective function substitution method. It is tested with the typical test functions. The algorithm shows its superiority in handling the constrained conditions clearly compared with other constrained PSO algorithms. In order to verify the feasibility of the proposed constrained PSO algorithm applying in the engineering,the proposed algorithm is used to design a μ-PID fixed-structure robust controller by taking a sample missile longitudinal model for the classical Raytheon control structure. The simulated results indicate that the sample missile controller can not only satisfy a variety of time domain indexes but also have strong robustness,and the proposed algorithm is effective.更多还原