基于AFSA-PSO-LSSVM的风电机组齿轮箱故障诊断  

Fault Diagnosis of Gearbox of Wind Turbine Based on AFSA-PSO-LSSVM

作　　者：王一宁 甄成刚[1] 韩瑶瑶 WANG Yining;ZHEN Chenggang;HAN Yaoyao(School of Control and Computer Engineering, North China Electric Power University, Baoding 071003, China)

机构地区：[1]华北电力大学控制与计算机工程学院,河北保定071003

出　　处：《郑州大学学报（理学版）》2022年第3期81-87,共7页Journal of Zhengzhou University:Natural Science Edition

基　　金：国家自然科学基金项目(51677072);北京市自然科学基金项目(4182061)。

摘　　要：针对LSSVM算法参数优化选取的问题,提出一种结合人工鱼群(AFSA)和粒子群优化(PSO)的混合智能算法,优化LSSVM的参数,利用AFSA算法进行全局寻优搜索参数初值,PSO算法局部更新最优解、加速跳出局部最优。最后通过对风电机组齿轮箱振动加速度数据进行模拟实验,建立了PSO-LSSVM、AFSA-LSSVM和AFSA-PSO-LSSVM算法模型。实验结果表明,AFSA-PSO-LSSVM相较于PSO-LSSVM和AFSA-LSSVM模型,收敛速度更快、精度更高,验证了方法的有效性。Aiming at the problem of optimal selection of LSSVM algorithm parameters,a hybrid intelligent algorithm,which combined artificial fish swarm algorithm(AFSA)and particle swarm optimization(PSO)was proposed to optimize LSSVM parameters.The AFSA algorithm was used to search the initial values of the global optimization search parameters,the PSO algorithm was used to locally update the optimal solution and accelerate the jump out of the local optimal.Finally,the PSO-LSSVM,AFSA-LSSVM and AFSA-PSO-LSSVM algorithm models were established through simulation experiments on the vibration acceleration data of the wind turbine gearbox.Experimental results showed that AFSA-PSO-LSSVM had faster convergence and higher accuracy than PSO-LSSVM and AFSA-LSSVM models,which verified the effectiveness of this method.

关 键 词：风电机组 齿轮箱 故障诊断 AFSA-PSO-LSSVM 

分 类 号：TM315[电气工程—电机]