Monitoring Model of Coal Mill in Power Plant Based on Big Data and BP Neural Network

Abstract

In order to monitor the wear condition of grinding roller of coal mill in power plant and improve the reliability of production equipment, it is necessary to establish a state monitoring model with high accuracy and good prediction effect. It has been shown that the power of coal mill can reflect the wear degree of grinding roller. If the voltage and power factor of coal mill are constant, grinding current can be used to replace the power of coal mill. In this paper, through collecting field historical operation data and data preprocessing, the current model of coal mill is established by using double hidden layers BP (Back Propagation) neural network to predict the wear state of grinding roller. The simulation results show that compared with single hidden layer, double hidden layers BP neural network can improve the performance of the network, so as to improve the prediction accuracy of the model and provide basis for the follow-up maintenance of coal mill, which has certain practical engineering significance.