摘要
针对辊道窑进行异常检测时存在的检出率低、误报、发现滞后等问题,引入移动窗主元分析算法,并提出一种自适应步长移动窗主元分析算法。在该算法中,利用主元分析法中的T2统计量为依据,以积累多个新样本进行数据块更新的方式代替移动窗主元分析逐个新样本更新模型的方式,有效提升了算法性能。通过实验证明,该算法具有较好的准确性和较高的性能,能有效运用于辊道窑的异常检测。
Aiming at the problems of inaccuracy,false alarms and lag detection in the abnormal detection of roller kiln,the moving window principal component analysis algorithm was introduced,and an adaptive step size moving window principal component analysis algorithm was proposed.In this algorithm,the T2 statistic in the principal component analysis method was used as the basis,and the method of accumulating multiple new samples to update the data block was used instead of the moving window principal component analysis to update the model one by one,which effectively improved the performance of the algorithm.The experiment proves that the algorithm has better accuracy and higher performance,and can be effectively used in abnormal detection of roller kiln.
作者
邹振弘
印四华
Zou Zhenhong;Yin Sihua(School of Computers,Guangdong University of Technology,Guangzhou 510006,China;School of Electromechanical Engineering,Guangdong University of Technology,Guangzhou 510006,China)
出处
《机电工程技术》
2020年第10期110-114,共5页
Mechanical & Electrical Engineering Technology
基金
国家自然科学基金项目(编号:U1501248)。
