Research On Optimization Of Coal Blending System For Thermal Power Plant

At the present stage,the combustion of a single coal is not the type of combustion used in thermal power plants.How to meet the requirements of coal-fired boiler combustion conditions,through a variety of coal blending with combustion in order to achieve the purpose of reducing fuel costs of thermal power plant is an urgent need to solve the problem of thermal power enterprises.Coal blending technology makes use of the differences in properties of various types of coal.Through optimization of different proportioning in each coal,the coal that is allocated out finally achieves the optimal performance in comprehensive performance to meet the combustion index of thermal power plant.The incoming coal data of a 350 MW coal-fired unit and boiler coal combustion coal quality indicators is took as the research object.A coal blending optimization mathematical model is established by analyzing the relationship of quality indicators between mixed coals with the single coal,which is in line with the actual power plant coal blending optimization requirements.The predictive model of coal blending cost with four kinds of coal and five kinds of coal is established by the Support Vector Machine algorithm respectively,as the main objective function of the optimization model.The analysis of prediction results shows that the model can accurately predict the blending cost of coal blending with two methods and meet the actual needs of the project.The two predictive models and the Chaotic Particle Swarm Optimization algorithm are respectively used to optimize the mathematical model of dynamic coal blending.After optimization and calculation,eight groups of four kinds of coal blending optimization combinations and 12 groups of five kinds of coal blending are obtained.In the optimal combination,the prices of all the optimal combinations of blended coal have been raised relative to the prices of inferior coal,which are relatively lower than the prices of high-quality coal.Finally,the optimization algorithm is compiled into COM components,and the software of coal blending optimization system is developed by using hybrid programming method.The software can adapt to different computer operating environments,and can add and update real-time information on coal quality of incoming coal.Set the constraints in the existing coal type data based on the current conditions of the optimal coal blending program,which provide convenient digital management of thermal coal yard.