R-SCN based Model Predictive Control for NOx Emissions of MSWI Process

Model predictive control (MPC) provides supports for controlling the NOx emission in municipal solid waste incineration (MSWI) process. The MSWI process suffers from dynamic characteristic, it is a challenge to precisely construct a prediction model online. In terms of this problems, an R-SCN based MPC method is proposed in this work. First, the nonlinear prediction model of NOx emissions is established offline based on stochastic configuration network (SCN). Then, the output weight of SCN hidden layer is updated online by recursive least square method. Second, the particle swarm optimization (PSO) method is designed for calculating the optimal control law of each control sequence in the constrained receding-horizon optimization. Finally, the historical data of a waste treatment plant is used to validate the proposed method. And the experimental results show that the dynamic nonlinear prediction model of NOx emissions can accurately predict the variation trend of NOx value and the proposed control method can realize the stable control of NOx concentration in MSWI process.