变论域双模糊算法在铝电解槽电压滤波中的应用研究

Research on application of variable universe double fuzzy algorithm in voltage filtering of aluminum electrolysis cell

电解槽电压滤波处理是电解铝系统智能控制中的关键环节,电解槽电压滤波算法为一阶滤波算法,该算法在电压幅值变化较大时迟滞性严重,滤波效果差,为了解决这些问题,设计一种基于变论域双模糊算法的自适应滤波器。用相邻电解槽电压的变化量以及电解槽电压与校正后电压的变化量作为2个模糊控制器的输入变量,并对这2个输入变量论域伸缩,采用模糊算法分别计算2个模糊控制器的滤波系数和校正系数,使用MATLAB软件进行仿真实验。研究结果表明:该滤波算法在电压幅值变化较大时不仅能保持良好的平稳性,而且能保持较高灵敏度;与一阶滤波算法相比,该算法的跟踪时间缩短93%,滤波后电解槽电压的均方根误差减少93%;在某铝厂500 kA电解槽的应用中,阳极效应预报准确率达98%。

Electrolytic cell voltage filtering is the key link in the intelligent control of electrolytic aluminum system. At this stage, the electrolytic cell voltage filtering algorithm is a first-order filtering algorithm. This algorithm has serious hysteresis and poor filtering effect when the voltage data changes a lot. Aiming at the above problems, an adaptive filter was designed based on the variable universe dual fuzzy algorithm. The amount of change in the voltage of the adjacent electrolytic cell and the amount of change in the voltage between the electrolytic cell and the corrected voltage as the two input variables of the two fuzzy controllers were used. Filter coefficients and correction coefficients were calculated by the fuzzy algorithm. Simulation tests were made using MATLAB software. The results show that the filtering algorithm not only maintains strong stability when the voltage data changes are small, but also maintains high sensitivity when the voltage data change is large. The tracking time of the algorithm is shortened by 93% compared with the first-order filtering algorithm, and the root mean square error is reduced by 93% compared with the first-order filtering algorithm. In the practical application of a 500 kA electrolytic cell in an aluminum plant, the anode effect prediction accuracy rate reaches 98%.