基于SVR与BP神经网络的水电机组瓦温预测

Prediction of the Bearing Bush Temperature in Hydropower Units based on SVR and BP Neural Network

基于支持向量回归和反向传播神经网络算法,结合实际生产经验,对水电机组运行过程中轴瓦温度以及影响其变化的主要因素进行关联映射,建立相关映射模型,通过对比模型预测精度,优选模型对实时轴瓦温度进行预测。有效实现水电机组瓦温的智能实时预测,一定程度上解决了传统瓦温监测中阈值预警判断信息单一的弊端。

Based on the support vector regression(SVR)and the back-propagation(BP)neural network algorithms,and the practical production experience,correlation mapping is carried out between the bearing bush temperature and the major factors affecting the temperature variations during the operation of hydropower units.A correlation mapping model is then established.By comparing the prediction accuracy of the models,the optimal model is selected to predict the real-time bearing bush temperature.It realizes the intelligent and real-time prediction of the bearing bush temperature in hydropower units,and solves the shortcomings of the traditional monitoring scheme that the judgement information in threshold warning is limited.