arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.12222cs.LGcs.SYeess.SY
收藏

可解释性:基于相关性的化工过程动态深度学习故障检测与诊断算法

Explainability: Relevance based Dynamic Deep Learning Algorithm for Fault Detection and Diagnosis in Chemical Processes

University of Waterloo(滑铁卢大学);Sanofi Pasteur(赛诺菲巴斯德)

Piyush Agarwal, Melih Tamer, Hector Budman

摘要:

这项工作的重点是统计过程控制(SPC)的制造过程的基础上可用的测量。SPC在工业环境中的两个重要应用是故障检测和诊断(FDD)。在这项工作中,提出了一种基于深度学习(DL)的FDD方法。我们研究了可解释性概念的应用,以提高用相对少量样本的数据集训练的深度神经网络模型的FDD准确性。的可解释性量化的输入变量是从分层相关传播(LRP)算法计算的一种新的相关性措施。结果表明,相关性可以用来丢弃冗余的输入特征向量/变量迭代,从而减少过拟合的噪声数据,增加输出类和优越的FDD测试精度之间的可扩展性。该方法的有效性证明了对基准田纳西伊士曼过程。

英文摘要:

The focus of this work is on Statistical Process Control (SPC) of a manufacturing process based on available measurements. Two important applications of SPC in industrial settings are fault detection and diagnosis (FDD). In this work a deep learning (DL) based methodology is proposed for FDD. We investigate the application of an explainability concept to enhance the FDD accuracy of a deep neural network model trained with a data set of relatively small number of samples. The explainability is quantified by a novel relevance measure of input variables that is calculated from a Layerwise Relevance Propagation (LRP) algorithm. It is shown that the relevances can be used to discard redundant input feature vectors/ variables iteratively thus resulting in reduced over-fitting of noisy data, increasing distinguishability between output classes and superior FDD test accuracy. The efficacy of the proposed method is demonstrated on the benchmark Tennessee Eastman Process.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑