Research On Multi-objective Collaborative Optimization Of Large Coal-fired Boilers For Pollutant Reduction And Efficiency Improvement

With the rapid development of China and the improvement of national living standards,the country’s demand for energy is more and more intense.Coal is still the main energy consumption due to the restriction of the national energy structure of"rich in coal,poor in oil and little in gas".Domestic thermal power plants will consume a large amount of coal resources while providing a large amount of electric power resources.In the process of coal combustion,a variety of pollutants will be generated,which will cause adverse effects on the environment and people.The pollutants generated in coal combustion and heat efficiency of furnace type,burning method and so on the many kinds of parameters involved,so how to predict the pollutant and the thermal efficiency,and further combined with the multi-objective optimization algorithm,the concentration of pollutants generated,and to establish predictive control strategy to be a source of large boiler thermal efficiency and efficiency of the important research direction.Based on the field operation data of 1000MW hedging boiler in a power plant,this paper carries out research on the strategy of source emission reduction and energy efficiency improvement of large boilers.The main research results are as follows:Aiming at the difficult problem of how to achieve the prediction of boiler pollutants and thermal efficiency,a prediction model of NO_xgeneration concentration and thermal efficiency was established based on Spearman correlation analysis and random forest algorithm based on the data collected from PI database..The results show that the random forest model has a root-mean-square error（RMSE）of 16.31mg·m^-3for the NO_xconcentration prediction set,and a RMSE of 0.08%for the thermal efficiency prediction set,which shows that the random forest model based on feature selection can achieve fast and accurate prediction of the outlet NO_xand thermal efficiency of the opposed firing boiler.In order to find a balance point between economy and environmental protection in the operation of hedging boilers,the NO_xgeneration concentration and thermal efficiency were established as optimization objectives,and the multi-objective optimization process of hedging boilers NSGA-Ⅱcombined with random forest prediction model was established.The results show that the NO_xconcentration was reduced from the original 430 mg·m^-3to 362.57 mg·m^-3,with an emission reduction ratio of nearly 15.7%;the thermal efficiency was increased from the original 94.20%to 94.41%,the multi-objective optimization of hedging boiler under full load is realized.Aiming at the problem of how to realize the coordinated control of pollution reduction and efficiency improvement of hedging boilers under various influencing factors,the control strategy of the two-layer control structure combined with MPC and optimization process was established to realize the coordinated control of boiler load,thermal efficiency and NO_xgeneration concentration.The simulation results show that the thermal efficiency can be effectively increased by 0.3%and NO_xemission concentration by about 60mg·m^-3by double-layer control..