Study On Full Load NOx Emission Reduction Model For Coal Fired Power Plant

As the most important power generation method in China,thermal power generation generates a large amount of polluting gases such as NOx and dust in the combustion process,which not only causes serious pollution to the environment,but also endangers the health of the human body.People's awareness of environmental protection is increasing,and it is imperative to reduce NOx emissions.In this paper,a 660 MW ultra-supercritical boiler of a coal-fired power plant is taken as the research object,and NOx emission reduction is carried out for the unit under the operating conditions of medium and high load and low load respectively,so as to achieve the low NOx emission under the operating conditions of full load.For the normal operation of SCR equipment under medium and high load operation conditions,Python programming language and scikit-learn extension library are used as tools,and the operation data under 500 MW load is selected as samples,and use the GBDT,random forest and support vector machine to establish boiler NOx emission prediction model,the maximum relative error and the average relative error of the model are 3.16% and 1.93% respectively.The maximum relative error in the test set of the random forest prediction model is 3.40%,and the average relative error is 2.28%.The maximum relative error of SVM prediction model test set is 2.98%,and the average relative error is 2.17%,the maximum relative error of GBDT under 660 MW load is 2.62%,and the average relative error is 1.67%.The maximum relative error of the random forest prediction model test set is 2.23%,and the average relative error is 2.03%.The maximum relative error of SVM prediction model test set is 2.66%,and the average relative error is 1.94%.The results show that the three models have higher prediction accuracy and stronger generalization,comprehensive prediction accuracy and operational efficiency,and finally select the SVM as the final prediction model.The SVM is selected as the boiler combustion prediction model,and the particle swarm optimization algorithm is used to optimize the parameters of the support vector machine model.The optimized parameters take the value C=14.3,gamma = 0.0065,and the optimized support vector machine is used.The maximum relative error in the model test set is 2.50%,and the average relative error is 1.78%.Compared with the pre-optimization error,the error is significantly reduced,which indicates that using the particle swarm optimization algorithm to optimize the parameters of the model can improve the accuracy of the prediction model.Based on the vector machine,the particle swarm optimization algorithm is used to optimize the NOx emission.The results show that the optimized NOx emission is significantly reduced,but the boiler combustion thermal efficiency is also reduced.In order to ensure the high efficiency combustion of the boiler under the NOx emission reduction,the utilization is high.The target particle swarm optimization algorithm performs multi-objective optimization of boiler combustion.Taking the NOx emission concentration and boiler thermal efficiency as the objective functions,a pareto solution set composed of multiple feasible solutions is obtained,which can ensure the thermal efficiency of the boiler under the premise of reducing NOx emissions.When the actual operating load of the boiler is too low,the SCR denitration equipment is shut down.At this time,if only the above optimization algorithm is used to adjust the boiler operating parameters,the final emission concentration of NOx has little effect.For this practical engineering problem,The tail flue of the coal-fired boiler was engineered to make the SCR inlet flue temperature reach the normal reaction temperature,and the SCR denitration equipment was operated normally.The thermal calculations were carried out for the three reform schemes.The calculation results showed that the post-economizer was under 50% BMCR conditions.When the flue gas fraction is adjusted to 0.5,the flue gas temperature is raised to 323 ° C;when the feed water bypass water supply fraction is about 0.8,the flue gas temperature is raised to 322 ° C;after the economizer area is 0.3 of the total area,the flue gas temperature is raised to 325 °C,the minimum flue temperature of 320 °C required for the normal operation of the SCR denitration equipment is achieved.