New Operation Optimization Method Based On Data-driven For Desulfurization Slurry Recycle Pump In Thermal Power Plant

As the flue gas SO₂ emission standards become increasingly stringent,the flue gas desulfurization equipment has been installed in most thermal power plant in China.Limestone-gypsum wet flue gas desulfurization?WFGD?process technology is currently the main method used in large coal-fired unit.This desulfurization process can effectively reduce the SO₂ content of the flue gas,but also bring great material energy consumption.The slurry circulation pump is the main power consumption equipment of the limestone-gypsum wet flue gas desulfurization system,and its power consumption accounts for more than 50%of the total power consumption of wet flue gas desulfurization system.When the flue gas emission meets environmental protection requirements,it is an important way to realize the optimal control of the slurry recycle pump operation to save energy and reduce consumption in power plant.Operation combination of the slurry recycle pump is researched in this paper,which is in Huadian International Shandong power plant.Based on the desulfurization principle and the historical operation data of the wet flue gas desulfurization system,the operation optimization method of the slurry recycle pump has been researched in this paper.The key influencing factors of the power consumption for the slurry recycle pump have been obtained by combining the qualitative and quantitative methods.Based on the qualitative analysis of the reaction mechanism and correlation computation for the historical data of the wet flue gas desulfurization system,the seven main factors have been determined.The combination of clustering and comprehensive evaluation methods are adopted to realize the optimal working condition database of the slurry recycle pump.The record of the historical operating data is clustered by fuzzy clustering algorithm based on the unit load and the inlet sulfur dioxide concentration,and the comprehensive evaluation index is used to determine the optimal operating record of the slurry recycle pump under the current operating conditions.The optimal working condition database is created by gathering the optimal operation records in different working condition clusters.Finally,the validity of the target working condition database is tested by the statistical inspection method.The support vector machine model based on genetic algorithm optimization parameters is applied to the optimization of the slurry recycle pump operation.Based on the key influencing factors of the slurry recycle pump power consumption and the target working condition database,a support vector machine classifier model is established,which can be used to predict the slurry recycle pump operation scheme.The genetic algorithm is used to optimize the parameters of the support vector machine model.The simulation results based on actual operation data show that:The proposed optimization method of the desulphurization slurry recycle pump in this paper is feasible,and the maximum power consumption can be reduced by 21.55%.The study has made a new exploration of the operation optimization method of the desulfurization slurry recycle pumps in coal-fired power plants,and further engineering application research can be carried out on this basis.