Coke Oven Flue Temperature Control Based on Particle Swarm Optimization and Fuzzy-PID with Variable Universe

Abstract

Coke oven flue temperature is an important factor to ensure coke output, coke quality and energy consumption in coking production process. The heating process of coke oven is an object with multiple control variables, nonlinearity and large delay. The traditional PID control algorithm cannot further improve the control performance of the coke oven system. This paper proposes a controller based on particle swarm optimization and variable universe Fuzzy-PID, and applies it to the control of coke oven flue gas temperature. In order to solve the problem of local self-adjustment optimization of controller parameters, the variable universe is adopted to ensure the global optimal control parameters of the system, which solves the problem that the fuzzy controller cannot guarantee high accuracy under given rules. At the same time, according to the performance index of the control system, the scale factor is adjusted online through particle swarm optimization (PSO) algorithm. The proposed controller combines the advantages of three different methods. Finally, according to the actual operation, the simulation under four working conditions, namely, normal working condition, given temperature change, random disturbance and coke oven parameter change. The simulation results show that the controller is feasible.

DOI: 10.61416/ceai.v25i4.8598