针对CDC减振器电磁阀研究中,只针对单一参数对电磁特性进行影响分析,且获得电磁阀最优结构参数组合速度较慢的不足,本文提出了基于 GA-BP 神经网络的电磁阀特性优化方法.首先,利用 ANSYS Max-well仿真分析,得到不同参数组合下的电磁力,然后训练样本集建立 BP 神经网络电磁力预测模型,再采用遗传算法优化,寻得最优的参数组合,建立GA-BP神经网络模型.结果表明,将GA-BP神经网络算法应用于电磁力的预测,能有效提高电磁阀的设计效率.
Optimization of CDC Shock Absorber Solenoid Valve Based on GA-BP Neural Network
In the study of CDC shock absorber solenoid valve,only a single parameter was analyzed for the in-fluence of electromagnetic characteristics,and the optimal structural parameter combination speed of solenoid valve was slow.In this paper,an optimal design method of solenoid valve characteristic based on GA-BP neural network is proposed.Firstly,using ANSYS Maxwell to simulate and analyze the CDC shock absorber solenoid valve,the electromagnetic force under different parameter combinations is obtained.Then,the sample set is trained to establish the BP neural network electromagnetic force prediction model,and then Genetic Algorithm(GA)is used to optimize the neural network,and the optimal structural parameter combination is found,and the GA-BP neural network model is established.The results show that GA-BP neural network algorithm is applied to the prediction of electromagnetic force,and the efficiency of solenoid valve design can be effectively improved.
Solenoid valveElectromagnetic force predictionStructural parametersNeural networkGe-netic algorithm
NETL
NSTL
万方数据
