Abstract:Improving the fitting accuracy of the calibration data of the infrared target simulator is of great significance for the measurement of the irradiance and other radiation characteristics of the infrared target. In view of the strong nonlinearity of the calibration data of the infrared standard simulator and the poor accuracy of the traditional fitting algorithm, a particle swarm optimization extreme learning machine (PSO-ELM) was introduced in this paper. Taking the standard black body radiation temperature as the input factor and the irradiance actually measured by the MCT detector as the output factor, the PSO-ELM-based method was established. In the PSO-ELM-based method, the connection weight matrix from the input layer to the hidden layer and the bias vector of the hidden layer were optimized by the PSO, and A nonlinear relationship between input parameters and output parameters was fitted. The optimization of these two parameters has greatly improved the predictive ability of original ELM. Main advantages of this method are that it has strong fault tolerance, better processing performance for complex nonlinear data, and the optimization mechanism in a kernel parameter setting of ELM. Comparing with genetic algorithm extreme learning machine (GA-ELM), extreme learning machine (ELM), it is verified the superior performance of the PSO-ELM-based method compared to the conventional data fitting method, which provided a new method for infrared target simulator calibration data fitting.