Aiming at the problem that because of the disturbance and time delay in the extrusion control system of cable insulation thickness

which cause difficulty in establishing mathematical model

low control precision and poor adaptability

a global optimization strategy for control system based on NARX neural network and particle swarm optimization（ PSO） algorithm is proposed. Firstly

NARX neural network is used to approximate the original transfer function and the disturbance in the field

and to construct a transfer function which is more conform to the field data. Then PSO algorithm is used to optimize proportional-integral-differential（ PID） parameters

according to the constructed fitness value function. Finally

the constructed transfer function and the optimized PID controller are substituted into the control system. Experimental results show that this algorithm has the advantages of small overshoot

fast response

strong adaptability and small steady-state error when controlling disturbance and timedelay systems

and has significant performance advantages over other algorithms.