Model NOx, SO2 Emissions Concentration and Thermal Efficiency of CFBB Based on a Hyper-Parameter Self-Optimized Broad Learning System

At present, establishing a multidimensional characteristic model of boiler combustion system plays an important role in realizing its dynamic optimization and real-time control, so as to achieve the purpose reducing environmental pollution saving coal resources. However, complexity process makes it difficult using traditional mathematical methods. In this paper, kind hyper-parameter self-optimized broad learning by sparrow search algorithm is proposed NOx, SO2 emissions concentration thermal efficiency circulation fluidized bed (CFBB). A (BLS) novel neural network algorithm, which shows good performance feature learning. BLS has several hyper-parameters be set wide range, that optimal combination between determine. This paper uses (SSA) select system, namely SSA-BLS. To verify effectiveness SSA-BLS, ten benchmark regression datasets are applied. Experimental results show SSA-BLS obtains accuracy stability. Additionally, applied parameters 330 MW circulating boiler. reveal can establish accurate prediction models for efficiency, NOx emission concentration, separately. Altogether, effective modelling method.