Research on ultra-high power laser curing based on RBF neural network

Aiming at the problem of large temperature error in the melting and clarifying process of borosilicate glass in the process of vitrification of high-level waste liquid with 100 kW fiber laser, a temperature control algorithm based on RBF neural network PID was proposed. The vitrification of borosilicate glass is a nonlinear, time-lag and large inertia process. Conventional manual control and PID control cannot meet the temperature control requirements during the vitrification of borosilicate glass. In this paper, a three-layer RBF neural network model is established, and the PID parameters are adjusted online by using the adaptability of the RBF neural network. The simulation results show that compared with manual control and PID control, the RBF neural network PID control has faster adjustment speed, smaller overshoot and stronger self-adaptation, and effectively reduces the temperature error during glass curing., improving the robustness of the system, resulting in higher waste containment and thermal efficiency.