基于径向基神经网络的半导体激光器温度自抗扰控制OA

In order to solve the problem that semiconductor lasers are easily affected by the operating temperature during gas detection,resulting in wavelength shift,and then reducing the concentration detec-tion accuracy,an auto disturbance rejection control(ADRC)method based on radial basis function neural network(RBFNN)was proposed.An active disturbance rejection controller combined with RBF neural network was designed,and the gradient descent method was used to adjust the nonlinear parameters in the dilated state observer in real time to optimize the dynamic response and anti-disturbance performance of the temperature control system.The experimental simulation results show that the proposed method can significantly improve the temperature control accuracy,the steady-state error is only 0.004 ℃,the over-shoot is 0.93％,and the adjustment time is 13.8 s,which has better control performance and higher practi-cal value than the traditional PID and ADRC methods.