Construction of beam quality factor prediction model based on support vector machine

The effectiveness of the laser can be evaluated by the beam qua lity factor, the beam quality factor is related to the initial parameters of the outgoing laser and various atmospheric parameters on the transmission path. Therefore, the prediction of the beam quality factor can be regarded as a statistical identification problem. Support vector machine has unique advantages in dealing with small sample, nonlinear and high dimensional problems, and can be used to deal with such statistical identification problems. This paper simulates the transmission process of Gaussian laser in the marine atmosphere based on the numerical simulation model of multi-layer phase screen, obtaining simulation data of initial laser radius, initial power, transmission distance, refractive index structure constant, atmospheric visibility, and spot radius, beam drift, and energy circle rate at the receiving surface; then constructing the prediction models of these beam quality factors by support vector machine, the optimal parameter model is obtained by adjusting the kernel function of the model, the insensitive loss coefficient and the penalty factor; finally, the prediction error and accuracy of the model are analyzed. The research results show that the support vector machine can fit the multiple regression relationship between input and output well, and the prediction accuracy of the model is high. The research results can provide a feasible basis for the application of support vector machine in the evaluation of the effectiveness of the laser in the marine atmosphere.