摘要
电流效率作为铝电解过程的重要参数,获得实时准确的测量结果对实现过程的有效控制至关重要。基于数据挖掘的思想,提出基于优化核极限学习机(KELM)的铝电解电流效率预测模型。通过分析铝电解机理,获得影响电流效率的过程参数,采用核主元分析法对试验数据进行降维,并用聚类算法剔除数据异常点,建立基于KELM的铝电解电流效率模型。使用鲸鱼优化算法与模拟退火的混合算法(WOASA)优化KELM模型的关键参数,从而提高模型的精度和泛化能力。通过实际生产数据进行仿真试验,将本文的方法与原始KELM、PSO-KELM、GWO-KELM、CGWO-KELM算法进行对比,结果证明了该预测模型的有效性,可以实现铝电解过程电流效率的准确预测。
Current efficiency is an important parameter in aluminum electrolysis process,and obtaining realtime and accurate measurement results is essential for effective control of the process. Based on the idea of data mining,a prediction model of aluminum electrolysis current efficiency based on improved kernal extreme learning machine( KELM) is proposed. The process parameters affecting the current efficiency were obtained by analyzing the mechanism of aluminum electrolysis. The kernal principal component analysis( KPCA) method was used to reduce the dimension of the test data,and the clustering algorithm was used to remove the abnormal data points. The current efficiency model of aluminum electrolysis based on KELM was established. The WOASA was used to optimize the key parameters of the KELM model,so as to improve the accuracy and generalization ability of the model. Through the simulation test of actual production data,the proposed method was compared with the original KELM,PSO-KELM,GWO-KELM and CGWO-KELM algorithms. The simulation results show that the model is effective and can be used to predict the current efficiency of aluminum electrolysis.
作者
徐辰华
张进智
XU Chen-hua;ZHANG Jin-zhi(School of Electrical Engineering,Guangxi University,Nanning 530004,China)
出处
《测控技术》
2020年第10期73-78,共6页
Measurement & Control Technology
基金
广西重点研发项目(2018AB67003)
广西自然科学基金项目(2017GXNSFAA198225)。
关键词
铝电解
电流效率
优化核极限学习机
剔除异常点
aluminum electrolysis
current efficiency
optimized K ELM
eliminate the abnormal points
作者简介
徐辰华(1976-),女,博士,副教授,硕士生导师,主要研究方向为智能系统;张进智(1995-),男,硕士研究生,主要研究方向为复杂工业系统智能控制、智能计算。
