LAUSR

Abstract The eccentric rotor extruder is a new kind of extrusion equipment with novel structure and outstanding engineering performance. As the structure of the eccentric rotor extruder is different from that of the traditional screw extruder, the control of the barrel temperature becomes important, including avoiding the influence of heating coupling and achieving high control accuracy. A neuron proportional-integral-derivative (neuron-PID) control algorithm of barrel temperature for the eccentric rotor extruder is introduced. The neural self-learning algorithm is able to tune PID parameters online, and the particle swarm optimization (PSO) algorithm is adopted to optimize the initial weight coefficients of the neuron. The experimental results show that the PSO-neuron-PID controller has the advantages of low overshoot and high control accuracy, and the influence of heat coupling can be counteracted effectively.