基于改进粒子群优化算法的气体源定位研究

Research on gas source localization based on improved PSO algorithm

为了提高粒子群优化(PSO)算法在气体泄漏源中的定位精度,针对标准PSO算法中存在的收敛早熟等问题,提出了一种惯性权重非线性递减和异步变化的学习因子相结合的改进PSO(IPSO)算法。该方法能够提高算法的性能,并加快粒子的收敛速度,引入二阶振荡环节来增加种群的多样性。通过函数优化实验与其他PSO算法对比,进行有效性分析和误差分析,由气体扩散模型仿真实验得出:定位结果误差值在1%范围内,表明IPSO算法不仅能够优化粒子学习能力,还能够有效提高算法的收敛精度和稳定性。

In order to improve positioning precision of the particle swarm optimization(PSO)algorithm in gas leakage source,aiming at the problems of premature convergence in the standard PSO algorithm,an improved PSO(IPSO)algorithm that combines the non-linear decrease of inertia weight and the asynchronously changing learning factor is proposed.This method can improve the performance of the algorithm,speed up the convergence rate of particles,and a second-order oscillation link is introduced to increase the diversity of the population.The function optimization experiment is compared with other PSO algorithms for effectiveness analysis and error analysis.The error value of the positioning result obtained from the gas diffusion model simulation experiment is within 1%,indicating that the IPSO algorithm can not only optimize the particle learning ability,but also effectively improve the convergence precision and stability of the algorithm.