Identification of solenoid parameters based on fiber squeezer and neural network for Smart control

Solenoids are low-cost high-speed nonlinear actuators commonly used in switching mode in many applications. However, fluctuations on the performances of solenoid are a major problem, particularly in industrial applications. These fluctuations are essentially due to changes in the spring constant, in the coefficient of friction, in the inductance and the resistance of the coil. This paper proposes a new methodology for controlling the effect of solenoid parameters variation on the PID corrector coefficients. First at all, the effect of solenoid parameters variation on the PID corrector coefficients is analyzed, Then, the algorithm based on artificial neural networks (ANN) coupled with optical fiber polarization squeezer based on solenoid for polarization scrambling is used to monitor the solenoid parameters from the coefficients of the transfer function, these coefficient are determined from the step response of the squeezer fiber. the parameters identified are used to automatically adjust the coefficients of the PID correctors. The results of the simulation show the validity for monitoring the solenoid parameters for keeping an optimized dynamic response.