Research on Online Monitoring Method of Axial Flow Fan Performance based on Improved RBF Neural Network
To solve the problem of long-term serious deviation from design operating conditions of auxilia-ry equipment in power plants under the flexible operation of thermal power units,an online monitoring method for axial flow fan performance in thermal power plants was proposed.The static performance curves for axial flow fan under design conditions were analyzed,the static performance model of fan was established by a method of radial basis function(RBF)neural network,and an improved particle swarm optimization(IPSO)algorithm was used to optimize the center and width of the hidden layer basis func-tion of the RBF neural network,as well as the connection weights between the hidden layer and the out-put layer.The simulation results show that the goodness of fit of the model on the training set is 0.999 4,and the root mean square error(RMSE)is 0.006 3.Compared with models established by BPSO-RBF algorithm,traditional RBF algorithm and BP algorithm,its goodness of fit is closer to 1,and the RMSE is smaller,proving its effectiveness.Combined with the above static performance model,measured parame-ters and similarity laws of fan,the dynamic performance model is built,then the surge warning model of the fan is formulated.The visual online monitoring platform of fan is developed.The experiment proves that the method can real-timely monitor the working point status and performance parameters such as flow of fan under actual working conditions.
axial flow fanonline monitoringimproved PSO algorithmRBF neural networkdynamic performance model