Modeling And Optimization Of Power Plant Boilers Based On Dual-Population Grey Wolf Optimization Algorithm

For power plant boilers,their combustion needs a lot of fossil fuel,a large number of fossil fuels will emit a lot of acid oxides,thus polluting the atmosphere.Aiming at the comprehensive goal of improving boiler thermal efficiency and reducing acid oxide emissions,this paper designs an optimization modeling method for power plant boilers based on the new grey Wolf algorithm optimization,which has very superior performance compared with the original grey Wolf algorithm and the model optimized by the unused algorithm,and has high stability for power plant boilers.Effective operation and reduction of acid oxides have important guiding significance.In order to achieve the goal of improving boiler thermal efficiency while reducingSO₂ and NO_xemissions,a dual grey wolf optimization algorithm?DGWO?optimized kernel extreme learning machine?KELM?boiler modeling method was proposed.The main work of this article is as follows:First,the structure and composition of the circulating fluidized bed boiler are introduced,and the combustion system of the boiler is introduced.According to the emission concentration of all kinds of air pollutants and the calculation method,the calculation method aboutSO₂?NO_xis expounded.Secondly,the detailed process of solving the boiler thermal efficiency is analyzed.Finally,the objective of combustion optimization is put forward.Secondly,aiming at the problem that the grey wolf optimization algorithm has low convergence precision and may fall into the local optimum of single population,this paper uses the dual population method and the elite strategy to improve the grey wolf optimization algorithm,and completes the optimization of the key parameters C and^?about KELM.The optimization of these two parameters will make the prediction of the boiler combustion system model more accurate.In addition,According to the sample data of different time periods,the prediction ability of the optimal kernel limit learning machine model?DGWO-KELM?proposed by the dual-population gray wolf algorithm is tested,and the prediction of the kernel extreme learning machine model?GWO-KELM?is optimized with the original gray wolf algorithm.The simulation results show that the dual-population gray wolf algorithm has strong parameter optimization ability,which can effectively improve the accuracy of the nuclear limit learning machine,and improve the optimization accuracy of the nuclear limit learning machine.Compared with the extreme learning machine model,the new dual-species gray wolf algorithm is optimized to optimize the nuclear extreme learning machine model with higher precision and faster convergence,which can predict boiler thermal efficiency and emissions more accurately.Finally,based on the built combustion model,the DGWO algorithm is used to optimize the single target and multi-objective.The optimization results show that the proposed combustion optimization scheme can effectively improve boiler efficiency and reduceSO₂andNO_x emissions.