Intelligent Optimal Control of Furnace Temperature Using Multi‐loop Controller and PSO Optimization

To address the challenge of effectively obtaining the optimal furnace temperature setpoint value, which is influenced by multiple manipulated variables (MVs) and directly affects pollutant emission concentrations in exhaust gas, a furnace temperature optimization control method was proposed. This method combines multi-loop control with particle swarm optimization (PSO) to minimize pollutant emission concentrations. First, the Tikhonov regularized linear regression decision tree (TR-LRDT) algorithm is employed to establish a model for the furnace temperature control object. Then, leveraging domain expert knowledge, a multi-loop temperature controller is designed using an improved single neuron adaptive PID (ISNA-PID) algorithm. Further, NO x and CO 2 indicator models are developed, and the PSO algorithm is applied to determine the furnace temperature setpoint value that minimizes pollutant emission concentrations. Finally, the intelligent optimization control framework is validated. Experimental results demonstrate that the optimal furnace temperature setpoint value can effectively reduce NO x and CO 2 emission concentrations.