基于改进遗传算法的飞行控制律参数设计

在进行大包线飞行控制律设计中，需将飞行包线分为几个区域，针对各个区域设计线性控制律，采用多个线性控制律来近似替代所需的非线性控制律，本文中线性控制律采用PID控制结构，若用传统PID控制律参数整定方法，工作烦琐、设计时间长，而且控制效果不是很好，提出运用遗传算法优化控制律参数，并针对遗传算法局部寻优能力差、收敛速度差、易早熟等缺点，对其进行了几点改进，为提高鲁棒性，对其适应函数进行了设计，针对某机型某个区域的俯仰控制进行了PID控制律设计仿真，利用改进的遗传算法对其参数进行优化，结果表明，该方法在23代就能得到最优控制参数，而且使系统动态性能有了较明显的提高。

Design of Parameters of Flight Control Law Based on Improved Genetic Algorithm

In the design of large envelope flight control law, flight envelope can be divided into several regions, and linear control law is designed for each region. Multiple linear control laws are employed to approximate alternative nonlinear control law that shall be present. In this article, linear control law was designed using PID control structure, the traditional method of PID control law parameters definition takes tedious effort and long time, and the control is not good enough, as a result, a method using genetic algorithm to optimize the parameters of the control law is proposed. Aiming at traditional genetic algorithm＇s poor local optimization ability, poor convergence rate, prematurity and other shortcomings, several improvements are proposed. In order to improve robustness, its fitness function is designed. One region＇s PID controller of pitch is designed and simulated, then the improved genetic algorithm is used to optimize controller parameters . The results show that the method is able to get the optimal control parameters in the 23th generation, and the system dynamic performance is significantly improved.