论文部分内容阅读
目的为了提高生产效率、降低成本、安全生产,通过对铝电解故障进行有效的检测和预报,减少铝电解过程中阳极效应、热槽、冷槽故障的发生.方法通过对铝电解故障发生机理和故障发生时相关特征量变化趋势的分析,基于模糊逻辑理论,建立了多级模糊故障检测模型,采用BP神经网络建立了故障分类模型,实现对铝电解故障的检测和预报.结果降低了模糊系统的维度,减少了规则数量,采用多级模糊与神经网络相结合的故障诊断预报的方法,提前了预报时间,提高了预报准确率.结论铝电解模糊神经网络故障诊断方法,有效地降低铝电解的故障发生率,降低了能耗,提高了铝的产量和质量,具有良好的应用前景.
Aim To improve the production efficiency, reduce the cost and improve the safety of production, through the effective detection and prediction of the aluminum electrolysis failure, it can reduce the anode effect, hot tank and cold tank failure in the process of aluminum electrolysis.Methods Through the mechanism of aluminum electrolysis failure and Based on the fuzzy logic theory, a multilevel fuzzy fault detection model was established, and a fault classification model was established by using BP neural network to detect and predict aluminum electrolysis failure.Results Reduced fuzzy system The number of rules is reduced and the method of fault diagnosis and prediction combined with multi-level fuzzy and neural network is adopted to advance the forecasting time and improve the forecasting accuracy.Conclusion The fault diagnosis method of aluminum electrolysis fuzzy neural network can effectively reduce the output of aluminum electrolysis Of the failure rate, reduce energy consumption and improve the output and quality of aluminum, has a good prospect.