Main Parameters Prediction of the Hot Water Boiler Based on the LSTM Neural Networks

This paper presents a data-driven method for the main parameter's prediction of a hot water boiler. Principal component analysis is used to compress the input dimensions of the model and reserves the main information of the monitored parameters. The validity of the model is demonstrated by a case study of a coal water slurry circulating fluidized bed hot water oiler belong to a heating company. The historical data of the boiler is employed to establish a deep long short memory cell neural network as the prediction. The prediction results of the main parameters could fulfil the demand of the actual engineering. (paper)