Research On Optimal Control Of Aluminum Electrolysis Process

Aluminum electrolysis is an industry of high energy consumption and high pollution, so it is important to have an optimal control and management about aluminum electrolysis production process. Due to the existence of high temperature,large current, strong corrosion and strong magnetic field, alumina concentration and electrolytic temperature, the two key parameters in aluminum electrolysis process cannot be measured on line and continuously. However, they both have to be controlled within a certain range; otherwise, it will affect pre-roasting aluminum electrolytic tank and even the whole electrolysis system adversely.Firstly, the fundamental principle and flow path of production techniques about modern aluminum electrolysis are analyzed, and relevant technological parameters in production process are also introduced. The material balance and energy balance of aluminum electrolysis production process are studied specially.Secondly, because of the characteristic of the difficulty of constructing exact mathematical model of aluminum electrolysis process, an intelligent control algorithm based on fuzzy neural network is designed. The algorithm adopting BP learning algorithm has tracked and controlled alumina concentration in aluminum electrolysis process effectively.Finally, the mutual correlation of main factors is analyzed, which influences the material balance and energy balance of pre-roasting aluminum electrolytic tank in allusion to such characteristics as opening, nonlinearity, large time-delay and correlation of multivariable in aluminum electrolysis process. Then a multistage-distributed correlation model about aluminum electrolysis process is established based on behavioral approach and chain system control method. The system model consists of two secondary-correlated systems and six relevant third-level subsystems which are related to the concentration and electrolytic temperature of alumina. Furthermore, the multistage-distributed control scheme aboutthe concentration and electrolytic temperature of alumina in aluminum electrolysis process is proposed, multistage-distributed predictive algorithms and control algorithms are used to coordinate the operation of different subsystems. The global control performance is achieved through division and cooperation of the subsystems.The simulation results show that alumina conservation and electrolytic temperature in the dynamic aluminum electrolysis process can be controlled effectively using the control algorithm.The research on optimal control of alumina concentration and electrolytic temperature in aluminum electrolysis process may improve production efficiency of aluminum electrolysis, and further promote energy conservation and emission reduction.