Control Strategy Of Aluminium Electrolysis Energy Saving And Automation Management System Research

Countries proposed resource saving society,the environment-friendly society,reduce the concentration of PM2.5,saving energy consumption,the overall requirements of energy efficiency.Aluminum electrolytic industry as energy-intensive,polluting industries is the key of the optimized adjustment.Saving energy and reducing consumption for aluminium electrolytic control,optimal control is an important way for energy saving and emission reduction,aluminum electrolytic industry aluminium electrolytic control system is a complicated multi-variable,nonlinear control system,the effect of aluminium electrolysis energy saving and current efficiency of variable has a lot of,how can through multi-objective optimization control strategy,in order to achieve energy conservation and environmental protection effect is the key of the research.This thesis first analyzed the process features of aluminium electrolysis,according to the aluminum electrolysis process is complex,variable characteristics,aluminum electrolytic multi-objective optimization control strategy is proposed.Summarizes the main influence factors of aluminium electrolytic process and on the basis of mutual relations,set up the mechanism of aluminium electrolytic process of alumina concentration and electric current efficiency model,and finally chooses the Elman neural network as the corresponding network modeling and the preliminary optimization.By alumina concentration influence on current efficiency in aluminum electrolysis production,and then put forward to control the cell resistance control of alumina concentration control ideas,design including prediction,optimization and output feedback of the overall design scheme of aluminum electrolytic multi-objective control system,also caught some can make exceptions to ascend to save energy consumption and the specific operation details such as control.But as a result of BP neural network inherent shortcomings,also need to be further improved.So for multi-objective optimization problems,and puts forward the quantum optimization method.The algorithm using qubits to real number coding chromosome of quantum bit probability amplitude as feasible solution of multi-objective problem.With quantum revolving door update quantum bit phase,design the quantum selection operator,crossover operator,mutation operator and population update operators,and use phase incremental space according to the index level decline and dynamically adjust strategy,using selected non dominated sorting operator,and based on the gravity of the Pareto individual selection strategy,the improvement measures to improve the algorithm efficiency obviously.Test results show that the proposed algorithm can effectively solve the problem of multi-objective optimization,relative to other multi-objective optimization algorithm has stronger global searching ability.Using the proposed fusion algorithm and the BP neural network controller,this paper expounds the quantum multi-objective genetic algorithm(ga)and BP neural network fusion of the specific implementation steps,and give full play to the quantum multi-objective the global search ability of genetic algorithm,improves the convergence speed and generalization ability of neural network.And introducing quantum optimization is easy to understand,easy to implement,practice is more effective than genetic multi-objective particle swarm optimization(pso)algorithm.The energy consumption of aluminum electrolysis process to optimize the overall scheme design,and expounds the general steps and optimization strategy.Multi-objective particle swarm optimization(pso)algorithm used for the optimization,the computed solution,and compared with the NSGA II algorithm optimization.Achieved a certain amount of energy consumption optimization,to provide operating instructions for personnel engaged in the production and scheduling.Effect of aluminium electrolysis automation managment system has carried on the simple introduction,and the system from the top management module,the function modules of the management to the subroutine module is divided into three levels,management is composed of seven function modules,then down decomposition of a subroutine module layer,each function constitute a module to module "coupling" small as far as possible,"cohesive degree" as far as possible.Finally,the main module design,reflect the value of the computer automation management system.