Acoustic Temperature Field Reconstruction Interpolation Based on Selective Kernel CNN
The boiler temperature reflects the combustion in the furnace.Accurate positioning of the high temperature area in the furnace plays an important role in improving combustion efficiency and identifying fault conditions.Traditional acoustic temperature field reconstruction interpolation depends on the arrangement of acoustic transducer and the choice of radial basis function,so it is difficult to achieve high resolution temperature field reconstruction.In order to solve this problem,a temperature field reconstruction interpolation network(TRIN)based on the selective convolution kernel CNN is designed to optimize the interpolation problem in the acoustic temperature field reconstruction,and the high precision interpolation of temperature field is realized.In order to verify the validity of the model,experiments are carried out on the simulated temperature field data set and the measured data of the boiler plant,and good results are obtained.
convolutional neural networkinterpolationacoustic temperature field reconstructionselective kernel
万方数据
维普
