Chiller faults detection and diagnosis with sensor network and adaptive 1D CNN
作者机构：Department of the Built EnvironmentNational University of Singapore4 Architecture Drive117566Singapore Faculty of Data ScienceShiga UniversityHikone5228522Japan RIKEN Center for Advanced Intelligence ProjectRIKENTokyo1030027Japan
出 版 物：《Digital Communications and Networks》 (数字通信与网络（英文版）)
年 卷 期：2022年第8卷第4期
页 面：531-539页
核心收录：
学科分类：0810[工学-信息与通信工程] 080202[工学-机械电子工程] 08[工学] 0802[工学-机械工程]
主 题：Chiller Fault detection and diagnosis Deep learning neural network Long short term memory Recurrent neural network Gated recurrent unit
摘 要：Computer-empowered detection of possible faults for Heating,Ventilation and Air-Conditioning(HVAC)subsystems,e.g.,chillers,is one of the most important applications in Artificial Intelligence(AI)integrated Internet of Things(IoT).The cyber-physical system greatly enhances the safety and security of the working facilities,reducing time,saving energy and protecting humans’*** the current trends of smart building design and energy management optimization,Automated Fault Detection and Diagnosis(AFDD)of chillers integrated with IoT is highly *** studies show that standard machine learning techniques,such as Principal Component Analysis(PCA),Support Vector Machine(SVM)and tree-structure-based algorithms,are useful in capturing various chiller faults with high accuracy *** the fast development of deep learning technology,Convolutional Neural Networks(CNNs)have been widely and successfully applied to various ***,for chiller AFDD,few existing works are adopting CNN and its extensions in the feature extraction and classification *** this study,we propose to perform chiller FDD using a CNN-based *** proposed approach has two distinct advantages over existing machine learning-based chiller AFDD ***,the CNN-based approach does not require the feature selection/extraction *** CNN is reputable with its feature extraction capability,the feature extraction and classification processes are merged,leading to a more neat AFDD framework compared to traditional ***,the classification accuracy is significantly improved compared to traditional methods using the CNN-based approach.
维普期刊数据库