基于IPSO-BP神经网络的DSC温度修正方法研究OA

In allusion to the temperature measurement error problem caused by the changes in the heating rate β and sample mass m of the differential scanning calorimeter(DSC),a method of DSC temperature correction based on the BP neural network optimized by improved particle swarm optimization algorithm(IPSO-BP)is proposed.Based on the non-equilibrium thermodynamic characteristics of the core structure of DSC,the influence of the changes in β and m on the temperature measurement of DSC was systematically analyzed.Experimental schemes covering different combinations of β and m were designed,and a training dataset was constructed for the neural network to learn the mapping relationship between temperature measurement errors and β and m.The improved particle swarm optimization(IPSO)algorithm was used to optimize the weights and biases of the hidden layers in the BP neural network,and a DSC temperature correction model was established.The experimental results show that after temperature correction using this method,the absolute value of the maximum absolute error of DSC temperature measurement is decreased from 2.509℃before correction to 0.322℃,and the average absolute error is decreased from 2.238℃to 0.203℃.This method can effectively suppress the DSC temperature measurement error introduced by the changes of β and m,which significantly reduces the influence of systematic errors caused by its core structure on temperature measurement,providing an intelligent correction solution for high-precision DSC measurement.