Interpretation and explanation of convolutional neural network-based fault diagnosis model at the feature-level for building energy systems

被引：10
|
作者：
Li, Guannan [1 ,2 ,6 ,8 ,9 ]
Chen, Liang [1 ]
Fan, Cheng [3 ,4 ,5 ]
Li, Tao [1 ,6 ,9 ]
Xu, Chengliang [1 ,9 ]
Fang, Xi [7 ]
机构：
[1] Wuhan Univ Sci & Technol, Sch Urban Construct, Wuhan 430065, Peoples R China
[2] Anhui Jianzhu Univ, Anhui Prov Key Lab Intelligent Bldg & Bldg Energy, Hefei 230601, Peoples R China
[3] Shenzhen Univ, Minist Educ, Key Lab Resilient Infrastruct Coastal Cities, Shenzhen, Peoples R China
[4] Shenzhen Univ, Sino Australia Joint Res Ctr BIM & Smart Construct, Shenzhen, Peoples R China
[5] Shenzhen Univ, Coll Civil & Transportat Engn, Shenzhen, Peoples R China
[6] Xian Univ Architecture & Technol, State Key Lab Green Bldg Western China, Xian 710055, Peoples R China
[7] Hunan Univ, Coll Civil Engn, Changsha 410082, Peoples R China
[8] Chongqing Univ, Minist Educ China, Key Lab Low grade Energy Utilizat Technol & Syst, Chongqing 400044, Peoples R China
[9] Wuhan Univ Sci & Technol, Hubei Prov Engn Res Ctr Urban Regenerat, Wuhan 430065, Peoples R China
来源：
ENERGY AND BUILDINGS | 2023年 / 295卷
基金：
中国国家自然科学基金;
关键词：
Building energy system; Deep learning; Fault diagnosis; Model interpretation and explanation; Shapley additive explanations (SHAP); Layer-wise relevance propagation (LRP); STRATEGY; METHODOLOGY;
D O I：
10.1016/j.enbuild.2023.113326
中图分类号：
TU [建筑科学];
学科分类号：
0813 ;
摘要：
electrolyte membrane electrolysis. to one based on clean energy requires implementing policies, stakeholders, technological adchallenges. While the UAE is clean energy, the contribution energy mix remains relatively Barakah plant) and solar (2000 constitute steps toward clean contribution to the 37 GW More support is critical to a at 8% share in 2021) under an more ambitious approach is needed to eliminate the reliance on coal, which has a 12% target by 2050. Hydrogen energy can significantly promote renewable energy adoption in hard-to-decarbonize sectors like energy-intensive industries, and utilizing solar energy for green hydrogen production can decarbonize industries such as aluminum, steel, and cement. A national hydrogen strategy is necessary to drive hydrogen energy applications and accelerate the deployment of clean hydrogen projects domestically and internationally. Credit author statement
引用
收藏
页数：19
相关论文
共 50 条
  • [41] Multi-level features fusion network-based feature learning for machinery fault diagnosis
    Ye, Zhuang
    Yu, Jianbo
    APPLIED SOFT COMPUTING, 2022, 122
  • [42] Convolutional neural network-based models for diagnosis of breast cancer
    Mehedi Masud
    Amr E. Eldin Rashed
    M. Shamim Hossain
    Neural Computing and Applications, 2022, 34 : 11383 - 11394
  • [43] A Feature Difference Convolutional Neural Network-Based Change Detection Method
    Zhang, Min
    Shi, Wenzhong
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2020, 58 (10): : 7232 - 7246
  • [44] A lightweight convolutional neural network-based feature extractor for visible images
    He, Xujie
    Jin, Jing
    Jiang, Yu
    Li, Dandan
    COMPUTER VISION AND IMAGE UNDERSTANDING, 2024, 249
  • [45] AGFL: A Graph Convolutional Neural Network-Based Method for Fault Localization
    Qian, Jie
    Ju, Xiaolin
    Chen, Xiang
    Shen, Hao
    Shen, Yiheng
    2021 IEEE 21ST INTERNATIONAL CONFERENCE ON SOFTWARE QUALITY, RELIABILITY AND SECURITY (QRS 2021), 2021, : 672 - 680
  • [46] A Deep Double-Convolutional Neural Network-Based Fault Detection
    Wang, Xiuli
    Li, Zhongxin
    Liang, Jing
    Li, Yang
    2023 IEEE 2ND INDUSTRIAL ELECTRONICS SOCIETY ANNUAL ON-LINE CONFERENCE, ONCON, 2023,
  • [47] Intelligent fault diagnosis of bearings based on feature model and Alexnet neural network
    Shi, Xiaoyu
    Cheng, Yuhua
    Zhang, Bo
    Zhang, Haonan
    2020 IEEE INTERNATIONAL CONFERENCE ON PROGNOSTICS AND HEALTH MANAGEMENT (ICPHM), 2020,
  • [48] Multiscale holospectrum convolutional neural network-based fault diagnosis of rolling bearings with variable operating conditions
    Zhang, Xining
    Liu, Shuyu
    Li, Lin
    Lei, Jiangeng
    Chang, Ge
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2021, 32 (10)
  • [49] A Convolutional Neural Network-Based Method of Inverter Fault Diagnosis in a Ship's DC Electrical System
    Yan, Guohua
    Hu, Yihuai
    Shi, Qingguo
    POLISH MARITIME RESEARCH, 2022, 29 (04) : 105 - 114
  • [50] Neural network-based sensor online fault diagnosis and reconfiguration for flight control systems
    Liu, Xiaoxiong
    Zhang, Weiguo
    Huang, Yijun
    Wu, Yan
    WCICA 2006: SIXTH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-12, CONFERENCE PROCEEDINGS, 2006, : 5577 - +
12345