Multiobjective Operation Optimization for Municipal Solid Waste Incineration Process

This paper proposes a multi-objective operation optimization method for the municipal solid waste incineration (MSWI) process to reduce NOx emissions and enhance combustion efficiency. In the optimization modeling stage, a self-organizing radial basis function (SORBF) neural network is employed to establish index models that capture the relationship between operational parameters and performance index of MSWI. Real data obtained from a MSWI plant in Beijing is utilized to train and validate models. By employing the established index models for fitness evaluation, the multi-objective particle swarm optimization (MOPSO) algorithm is developed to obtain optimal Pareto solutions of the operational parameters. Ultimately, the operation settings are determined by entropy weight method and utility function. The modeling results demonstrate that the SORBF neural network outperforms the kriging model in terms of fitting performance index. The MOPSO algorithm is effective in reducing NOx emissions and enhancing combustion efficiency. Furthermore, experimental findings indicate that the optimization effect using the kriging model is inferior to that using the SORBF neural network. These findings suggest that the choice of performance index modeling method significantly influences the optimization outcome.