Research on reconstruction algorithm of boiler combustion temperature field based on AT

Abstract:

In order to monitor the combustion state and measure the temperature distribution in the furnace, a threelayer progressive grid temperature field reconstruction algorithm based on acoustic tomography (AT) is proposed. The algorithm first computes the sparse grid temperature information with the least squares method (LSM), and then uses the least squares support vector machine (LSSVM) to establish a fine grid model for the temperature field. In order to further improve the reconstruction accuracy, the parameters of the LSSVM reconstruction model are tuned by the differential evolution algorithm (DE) (Referred to as the LLD algorithm) to establish a detailed temperature description of the entire measured temperature field. Through the reconstruction of three different types of temperature fields provided by Dongfang Boiler Industry Group, the results and error analysis show that the LLD algorithm can describe the temperature distribution information of the measured combustion zone globally and achieve high-precision reconstruction of the temperature field.