APU控制参数多策略融合粒子群算法寻优
APU Control Parameter Optimization by Multiple-Strategies Particle Swarm Optimization Algorithm
应世泽 1魏民祥 1丁玉章 1邢致毓1
作者信息
- 1. 南京航空航天大学能源与动力学院 江苏 南京 210016
- 折叠
摘要
为使辅助动力单元(APU)在工作点切换过程中的动态性能更优,提出了一种多策略融合粒子群算法优化模糊PID的控制参数.该算法对粒子群算法的常规迭代过程引入概率可调的干扰策略、混沌搜索策略、惯性权重自适应和学习因子线性调节策略,使其更适合优化维度多达 187 个的模糊PID控制参数.针对该算法中关键的增强全局搜索能力p和增强局部搜索能力pp两个参数的取值进行研究,搭建了APU仿真模型.仿真结果表明:多策略融合粒子群算法优化后的模糊PID控制参数相较于经典PID在转速变化、转矩变化和节气门开度变化的动态性能上均有较大的提升.
Abstract
In order to improve the dynamic performance of the auxiliary power unit(APU)in the switching process of the work point,a multiple-strategies particle swarm optimization algorithm was proposed to op-timize the control parameters of the fuzzy PID.This algorithm introduced probability adjustable interfer-ence strategy,chaos search strategy,inertia weight adaptive and learning factor linear adjustment strategy to the conventional iterative process of particle swarm optimization algorithm,making it more suitable for optimizing the fuzzy PID control parameters with 187 dimensions.The APU simulation model was built based on the research on the value of two key parameters of the algorithm,namely,enhancing the global search capability p and enhancing the local search capability pp.The simulation results showed that the dynamic performance of the fuzzy PID control parameters optimized by multiple-strategies particle swarm optimization algorithm was greatly improved compared with the classical PID control parameters in terms of speed change,torque change and throttle opening change.
关键词
辅助动力单元/模糊PID/参数优化/粒子群算法/APU仿真Key words
Auxiliary power unit/Fuzzy PID/Parameter optimization/Particle swarm optimization/APU simulation引用本文复制引用
出版年
2024
NETL
NSTL
万方数据