论文部分内容阅读
为降低电解铝的生产成本,提出了一种基于神经网络遗传算法的电解铝生产过程槽电压优化方法,以寻找最优生产槽电压和对应的生产条件。采用核主元分析法确定影响电解铝生产的关键参数,建立槽电压的神经网络模型。利用遗传算法寻找电解铝槽电压的全局最优值及对应的生产条件。通过实际生产数据进行仿真实验,结果表明,基于神经网络遗传算法全局寻优的能力,该优化方法能准确预测电解铝槽电压,同时能够找到电解铝生产过程中的最优槽电压及其对应的优化生产条件。
In order to reduce the production cost of electrolytic aluminum, a method based on neural network genetic algorithm is proposed to optimize the cell voltage in electrolytic aluminum production process in order to find the optimal production cell voltage and the corresponding production conditions. Kernel principal component analysis was used to determine the key parameters that affect electrolytic aluminum production and to establish a neural network model of cell voltage. Using Genetic Algorithm to Find the Global Optimal Value of Electrolytic Aluminum Cell Voltage and Corresponding Manufacturing Conditions. The simulation results show that the optimization method can accurately predict the voltage of electrolytic aluminum cell and find the optimal cell voltage in the electrolytic aluminum production process and its corresponding Optimize production conditions.