Machinery Probabilistic Few-Shot Prognostics Considering Prediction Uncertainty

被引:13
|
作者
Ding, Peng [1 ]
Jia, Minping [2 ]
Ding, Yifei [2 ]
Cao, Yudong [2 ]
Zhuang, Jichao [2 ]
Zhao, Xiaoli [3 ]
机构
[1] Yangzhou Univ, Coll Mech Engn, Yangzhou 225127, Peoples R China
[2] Southeast Univ, Sch Mech Engn, Nanjing 211189, Peoples R China
[3] Nanjing Univ Sci & Technol, Sch Mech Engn, Nanjing 210094, Peoples R China
基金
中国国家自然科学基金;
关键词
Bayesian approximation; machinery degr-adation prognostics; meta-learning; prediction uncertainty; probabilistic few-shot prognostics; variational inference; FAULT-DIAGNOSIS; LIFE PREDICTION; NETWORKS;
D O I
10.1109/TMECH.2023.3270901
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Few-shot learning based machinery prognostics are feasible for intelligent operation and maintenance with scarce monitoring data. In fact, single-point estimations of existing few-shot prognostics (FSP) algorithms suppress the high-reliability predictive maintenance. To alleviate this dilemma, the probabilistic few-shot prognostics problem is formulated to conquer the challenges of uncertainty estimation in FSP. We propose a novel Bayesian approximation enhanced probabilistic meta-learning (BA-PML) algorithm to convert learnable parameter uncertainty into final prediction uncertainty. It consists of two main components: the designed base probabilistic predictor and its corresponding episodic training strategy. The former reshapes Seq2Sep models with Bayesian theories and accomplishes variable-length degradation prediction. The latter follows the few-shot learning paradigm and extends variational inference driven posterior approximations to meta-level training that assists in mining general degradation knowledge from probabilistic aspects. Finally, run-to-failed vibration data proves our proposed BA-PML holds well-calibrated uncertainty prognostics under cross-domain decision-making tasks.
引用
收藏
页码:106 / 118
页数:13
相关论文
共 50 条
  • [41] Fractal Few-Shot Learning
    Zhou, Fobao
    Huang, Wenkai
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 35 (11) : 1 - 15
  • [42] Effect of Probabilistic Similarity Measure on Metric-Based Few-Shot Classification
    Lee, Youngjae
    Park, Hyeyoung
    APPLIED SCIENCES-BASEL, 2021, 11 (22):
  • [43] Few-shot Classifier GAN
    Ali-Gombe, Adamu
    Elyan, Eyad
    Savoye, Yann
    Jayne, Chrisina
    2018 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2018,
  • [44] Few-shot Multimodal Sentiment Analysis Based on Multimodal Probabilistic Fusion Prompts
    Yang, Xiaocui
    Feng, Shi
    Wang, Daling
    Zhang, Yifei
    Poria, Soujanya
    PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2023, 2023, : 6045 - 6053
  • [45] A review of few-shot classification
    Lim, Jia Min
    Lim, Kian Ming
    Lee, Chin Poo
    Lim, Jit Yan
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 275
  • [46] Interventional Few-Shot Learning
    Yue, Zhongqi
    Zhang, Hanwang
    Sun, Qianru
    Hua, Xian-Sheng
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
  • [47] Few-Shot Lifelong Learning
    Mazumder, Pratik
    Singh, Pravendra
    Rai, Piyush
    THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 2337 - 2345
  • [48] Few-shot fault diagnosis of rotating machinery with two-branch prototypical networks
    Cuixia Jiang
    Hao Chen
    Qifa Xu
    Xiangxiang Wang
    Journal of Intelligent Manufacturing, 2023, 34 : 1667 - 1681
  • [49] Few-shot fault diagnosis of rotating machinery with two-branch prototypical networks
    Jiang, Cuixia
    Chen, Hao
    Xu, Qifa
    Wang, Xiangxiang
    JOURNAL OF INTELLIGENT MANUFACTURING, 2023, 34 (04) : 1667 - 1681
  • [50] Uncertainty Guided Progressive Few-Shot Learning Perception for Aerial View Synthesis
    Gao, Zihan
    Li, Lingling
    Liu, Xu
    Jiao, Licheng
    Liu, Fang
    Yang, Shuyuan
    IEEE TRANSACTIONS ON MULTIMEDIA, 2025, 27 : 1177 - 1192