Temperature Measurement Correction for Power Cabinet Thermocouple Based on RBF Neural Network Parameter Optimization
In order to improve the accuracy of power cabinet temperature measurement,a power cabinet temperature correction method based on multi-channel high-accuracy measurement is proposed.Firstly,K-means clustering algorithm and artificial fish swarms algorithm(AFSA)were adopted to optimize the optimal basis function center and width of the radial basis function network(RBF),as well as the connection weights of the hidden layer and the output layer,thus the optimization of the RBF network was realized.Then,the improved RBF network was applied to the multi-channel high-accuracy power cabinet temperature measurement system,thus the temperature of the measuring power cabi-net was corrected.The simulation results show that the proposed method can realize the correction of the temperature nonlinear error of the pow-er cabinet based on multi-channel high-accuracy measurement,which makes the temperature error measured less than 0.4 ℃.Compared with the standard RBF network and BP neural network,the proposed method has faster training speed and smaller root mean square error,and can be used in temperature measurement environments that require higher temperature measurement accuracy.The innovation of this research lies in the realization of adaptive compensation of nonlinear temperature change based on the principle of thermocouple measurement and the im-provement of measurement accuracy.
temperature measurement correctionpower cabinet temperatureK-means clustering algorithmAFSA algorithmRBF networkthermocouple
万方数据
维普
