基于改进PSO算法的Volterra模型在磨损预测中的应用

Wear Prediction by Volterra Model Improved by Particle Swarm Optimization Algorithm

为准确预测机械设备的磨损问题,提出基于改进PSO算法的Volterra级数预测模型。该预测方法首先根据Volterra级数的性质,建立Volterra级数模型;然后利用改进的PSO算法对模型参数进行优化,得到Volterra级数的预测模型。利用轴承钢试件的磨损实验数据,采用建立的预测模型对数据进行建模和磨损预测。仿真结果表明,与基于PSO算法的Volterra模型、多项式模型、AR模型、RBF神经网络模型及BP神经网络模型相比,基于改进PSO算法的Volterra预测模型结构简单、预测精度高,具有一定的实用性。

In order to predict the wear of mechanical equipment accurately, Volterra series prediction model based on improved particle swarm optimization（PSO） algorithm was put forward.This prediction method firstly establishes the Volterra series model according to the nature of the Volterra series, then establishes Voherra series prediction model by optimizing the parameters of the model with the improved PSO algorithm.The wear experimental data of the bearing steel gear specimen was used to establish the prediction model ,and the wear of the bearing steel gear was predicted.The prediction results show that,in comparison with Volterra model based on PSO algorithm, polynomial model, AR model, RBF neural network model and BP neural network model, the proposed model has simpler structure and higher prediction precision, which has certain practicability.