Machine Learning-Based Fault Diagnosis for a PWR Nuclear Power Plant
Machine Learning-Based Fault Diagnosis for a PWR Nuclear Power Plant
复制标题
DOI：
10.1109/access.2022.3225966
复制
发表时间：
2022
期刊：
影响因子：
3.9
通讯作者：
Amine Naimi;Jiamei Deng;Paul Doney;Akbar Sheikh-Akbari;S. Shimjith;A. Arul
中科院分区：
文献类型：
--
作者：
Amine Naimi;Jiamei Deng;Paul Doney;Akbar Sheikh-Akbari;S. Shimjith;A. Arul
在核电行业，安全性和可靠性至关重要。传感器和执行器是此类系统中不可或缺的组件，潜在的故障可能会对系统性能产生不利影响。因此，设计一种达到最高安全标准的故障检测和诊断（FDD）系统势在必行。本文提出了一种基于机器学习的压水堆（PWR）执行器和传感器故障检测和诊断（FDD）技术。在提出的 FDD 框架中，首先使用浅层神经网络检测故障。其次，使用 MATLAB Classification Learner 工具箱中提供的 15 种不同分类器执行故障诊断，包括支持向量机 (SVM)、K 最近邻 (KNN) 和集成。研究发现多种分类器可提供卓越的分类性能，包括中型 KNN、三次 KNN、余弦 KNN、加权 KNN、精细高斯 SVM、二次 SVM、中高斯 SVM、粗高斯、袋装树和子空间 KNN。使用一组仿真结果证明了 FDD 方法的准确性。
In the nuclear power industry, safety and reliability are of the utmost importance. Sensors and actuators are integral components in such systems, and potential faults may adversely impact system performance. It is therefore imperative to design a fault detection and diagnosis (FDD) system that achieves the highest standards of safety. This paper presents a machine learning-based fault detection and diagnosis (FDD) technique for actuators and sensors in a pressurized water reactor (PWR). In the proposed FDD framework, faults are first detected using a shallow neural network. Second, fault diagnosis is performed using 15 different classifiers provided in the MATLAB Classification Learner toolbox, including support vector machine (SVM), K-nearest neighbor (KNN), and ensemble. Several classifiers were found to provide superior classification performance, including medium KNN, cubic KNN, cosine KNN, weighted KNN, fine Gaussian SVM, quadratic SVM, medium Gaussian SVM, coarse Gaussian, bagged trees, and subspace KNN. The accuracy of the FDD approach was demonstrated using a set of simulation results.