Aluminum Cell Voltage Modeling And Optimization Control Based On Multi-ELM Genetic Algorithm

In the aluminum electrolysis process, the pre-baked anode cell as the main equipment in the aluminum electrolysis, the power consumption is huge. The cell voltage is one of the most important economic indicators of the aluminum cell, reducing the cell voltage can significantly reduce power consumption. Since aluminum electrolysis is a nonlinear, strong coupling and large time delay system of industrial process, it is difficult to determine the cell voltage of the pre-baked aluminum reduction cell in the short term. Thus, it plays an extremely important role for the stable operation of the aluminum electrolysis production process while real-time monitoring the change of cell voltage and provide the best operating conditions.Through optimizing and control cell voltage to achieve the purpose to reduce the production costs, in this paper, we propose a optimization method of cell voltage based on multi extreme learning machine genetic algorithm to find the optimal production cell voltage and the corresponding production conditions. First, using kernel principal component analysis method to determine the key parameters affecting of aluminum electrolysis production, establish the model of BP neural network, a single-layer extreme learning machine and multi extreme learning machine to predict the cell voltage. Then, the genetic algorithm to find the global optimal value of cell voltage and corresponding production conditions of electrolyte temperature, alumina concentration, molecular ratio, the level of aluminum and the electrolyte levels. By actual production data to simulation, the results show that the model based on multi extreme learning machine and genetic algorithm to modeling and optimization can accurately predict cell voltage, at the same time to find the optimal cell voltage and the corresponding optimized production conditions.Finally, Simulation experiments of real-time control on voltage for aluminum electrolysis production process tank on the process operation optimization control platform. The platform can monitor real-time changes of the various parameters in the aluminum electrolysis process. The cell voltage and the corresponding production conditions can be controlled within the target range. The running results show that the control system can effectively control the cell voltage and the corresponding parameters of production conditions to ensure the normal operation of the aluminum electrolysis process and to achieve energy saving purposes.