Graph-Based Model Compression for HSR Bogies Fault Diagnosis at IoT Edge via Adversarial Knowledge Distillation
- 27 September 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (2) , 1787-1796
Abstract
Deep graph neural networks (GNNs) have demonstrated their exceptional expressive capability in detecting fault features from multisensor signals for fault diagnosis. However, the excessive depth of these models often hinders their deployment on Internet of Things (IoT) systems. In order to facilitate the implementation of graph-based models on IoT devices for fault diagnosis of HSR bogie, this paper proposes a novel compression technique called GraKD. GraKD distills the latent knowledge from the teacher model to a lightweight student model in an adversarial manner. The discriminator of GraKD comprises a representation identifier and a logit identifier. The former effectively distinguishes between the node representations of the student and the teacher by discerning the local affinity of connected node patch-patch pairs and the global affinity represented by the patch-global pairs. The latter employs a residual multi-layer perceptron block to differentiate between the logits of the student and the teacher. The effectiveness of GraKD is validated using data collected from test rigs of bogies. A series of experiments substantiate the broad applicability of GraKD in compressing various GNN-based models.Keywords
Funding Information
- National Natural Science Foundation of China (52275129)
- Guangxi science and technology major project (Guike AA22068001)
- Key Research and Development Projects of Shaanxi Province (2022LL-JB-19)
- Fundamental Research Funds for the Central Universities (XZY022022026, XZY022021006)
This publication has 33 references indexed in Scilit:
- Compacting Deep Neural Networks for Internet of Things: Methods and ApplicationsIEEE Internet of Things Journal, 2021
- Cooperative Intelligent Transport Systems: Choreography-Based Urban Traffic CoordinationIEEE Transactions on Intelligent Transportation Systems, 2021
- Graph Neural Networks with Convolutional ARMA FiltersIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
- A Review of Intelligent Fault Diagnosis for High-Speed Trains: Qualitative ApproachesEntropy, 2020
- Data-Driven Fault Diagnosis for Traction Systems in High-Speed Trains: A Survey, Challenges, and PerspectivesIEEE Transactions on Intelligent Transportation Systems, 2020
- Data-Free Network Quantization With Adversarial Knowledge Distillation2020
- Graph Representation Learning via Graphical Mutual Information Maximization2020
- Short-Frequency Fourier Transform for Fault Diagnosis of Induction Machines Working in Transient RegimeIEEE Transactions on Instrumentation and Measurement, 2017
- Maintenance of bogie components through vibration inspection with intelligent wireless sensors: A case study on axle-boxes and wheel-sets using the empirical mode decomposition techniqueProceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 2016
- Fault Diagnosis of Rotating Electrical Machines in Transient Regime Using a Single Stator Current’s FFTIEEE Transactions on Instrumentation and Measurement, 2015