Research On The Model And System Of Dynamic Coal Blending For Gasification Process

Coal gasification technology is one of the important production technology widely used by large coal chemical enterprises,its long-term operation practice shows that the instability of coal quality will affect the long-term and effective running of gasifier device and using coal blending to resolve the problem is one of the convenient and economical and feasible methods.Existing coal blending technology can not forecast the coal quality which is nolinear changing accurately,and can not take the coal quality requirement of gasifier into consideration when executing proportion calculation,as a result of which,the coal quality after coal blending is still with fluctuation,the gasifier device can not take long-term and stable running.In addition,manual coal blending calculation can not satisfy the development need of coal enterprises in the new period.In this paper,first of all,according to understanding of coal quality changing rules after coal blending,the coal blending forecasting model based on ash flowing temperature is built using multiple linear regression,BP neural network and optimal BP neural network with genetic algorithm respectively.Through comparing the imitative effect,error analysis and forecast result of the three model,it finds that the ash flowing temperature forecast model built with GA-BP neural network has some feasibility and superiority.Furthermore,through adaptability research to current topic of existing classical ash viscosity forecast model,a coal blending forecast model based on ash viscosity which is easily realized and can forecast ash viscosity features of gasification coal blending is implemented in that the coal quality can be forecasted accurately.Next,according to the actural coal quality key restrictive factors of gasifier in Ningdong coal chemical industrial base,a multi-objective optimization model of coal blending which uses moisture,ash content,volatile component,sulfur content,calorific value,ash flowing temperature,ash viscosity as constraint condition and uses coal price,sulfur content,flowing temperature as objective function is built.Through model solution using MATLAB optimization function and genetic algorithm respectively,analysis and comparison from theoretical and practical outcomes,it finds that it gets better result when genetic algorithm is used to solve multi-objective optimization model,and ratio optimization calculation can be realized by building a optimization coal blending model based on genetic algorithm.At last,through analysis of coal chemical industrial base gasification coal blending production process,a system of dynamic coal blending apply to gasification coal using C/S system architecture,Microsoft Visual Studio 2008 development platform and MFC is designed and implemented,what’s more,a coal blending forecast model based on GA-BP neural network and a optimization model of coal blending based on genetic algorithm is applied into this system.In this paper,the research on the model and system of dynamic coal Blending for gasification process can take the constrains and objective of coal quality ofter blended into consideration fully.According to the existing raw coalin warehouse,compute the most reasonable proportion of coal blending to meet the demand for different gasifier,and the coal quality can be forecasted accurately lower the gasifier failure rate caused by coal quality fluctuation,and update the information of stock row coal and gasifier.All of this shows,the research work of this paper can liberate the workforce,increase enterprises economic returns,and lowers the air pollution rate caused by high surfur content in coal quality.