Support Vector Machine Predictive Control for Superheated Steam Temperature Based on Particle Swarm Optimization

The processes with characters of nonlinear and time-varying is common in power plant. In order to achieve high control performance and effectiveness, a predictive control strategy combining support vector machine model and particle swarm optimization algorithm is proposed. Simulation of the superheated steam temperature system was done in a 600 MW supercritical once-through boiler. Experimental results show good control performance in terms of reference command tracking ability and steady state errors. This method is better than PID control in robustness and dynamic performance.