Design And Implementation Of The Energy Management System In Ansteel Cold Rolrd Works

Industry energy consumption is a large aspect of energy consumption of the country,and it accounts the total energy consumption of the country for about 70%.And the steel industry is One of the high energy consumption industries.The industrial enterprises of different types of processes,devices,product type,and energy management level of energy consumption will have.different effects of energy consumption.Construction of a whole plant-level centralized energy management system call complete all online energy data collection,calculation,analysis and play an importantrole Oil processing the energy materials balance,scheduling and optimization,energym anagement,equipment operation.Based on the analysis of Ansteel cold rolrd works's energy management status and energy management needs,from the advancement of energy management,usability,operability perspective,energy management systems and energy consumption prediction for a more in-depth study is designed thoroughly,main work shows as follows:In the firsr part,Overall design of the energy management system.According to demand analysis,the overall system architecture design is complete.With advanced,practical principles for software and hardware selection.Data acquisition subsystem real-time monitor the energy consumption of medium classification.Information management subsystem the realization of the objective information based on the energy planning,energy equipment,energy cost,energy consumption prediction,energy report and other functions.This thesis detail the process based on energy planning and energy consumption prediction,and design its HMI.In the second part,genetic algorithms artifical neural network model of the energy consumption based on the manager system.As genetic algorithms have been widely used to optimize feedforward neural networks foryears,especially in the field of predicting.This thesis will do some research on the topic:making full use of the global searching trait of genetic algorithms to optimize feedforward neural networks,for the sake of predicting.In this thesis,three-layered feedforward neural network which has wide applicability being the object of optimizing,a real-coded genetic algorithm with synthetic control strategy is presented,synthetically considering the effects of coded pattern,fitness function,distributing of initial population and genetic operators,which forms high-efficient algorithm in optimizing neural networks.