Multimodal Bearing Fault Classification Under Variable Conditions: A 1D CNN with Transfer Learning

2025-02-23

AI Summary

To address poor generalization of bearing fault classification under varying operating conditions, this paper proposes a multimodal 1D CNN-based fault diagnosis framework integrating vibration and motor phase current signals. Methodologically, it employs late-feature fusion and introduces a novel hierarchical transfer learning strategy—freezing parameters up to the first pooling layer—combined with L2 regularization to enhance model robustness. The key contributions are: (i) the first application of dual-modal time-series signals in a 1D CNN architecture with late fusion; and (ii) the proposed hierarchical transfer learning significantly improves cross-speed and cross-load adaptability. Experimental results demonstrate a 96% classification accuracy under baseline conditions—2 percentage points higher than baseline methods—and consistently high accuracy across three distinct operating conditions, validating both strong generalization capability and engineering practicality.

Technology Category

• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Bearings play an integral role in ensuring the reliability and efficiency of rotating machinery - reducing friction and handling critical loads. Bearing failures that constitute up to 90% of mechanical faults highlight the imperative need for reliable condition monitoring and fault detection. This study proposes a multimodal bearing fault classification approach that relies on vibration and motor phase current signals within a one-dimensional convolutional neural network (1D CNN) framework. The method fuses features from multiple signals to enhance the accuracy of fault detection. Under the baseline condition (1,500 rpm, 0.7 Nm load torque, and 1,000 N radial force), the model reaches an accuracy of 96% with addition of L2 regularization. This represents a notable improvement of 2% compared to the non-regularized model. In addition, the model demonstrates robust performance across three distinct operating conditions by employing transfer learning (TL) strategies. Among the tested TL variants, the approach that preserves parameters up to the first max-pool layer and then adjusts subsequent layers achieves the highest performance. While this approach attains excellent accuracy across varied conditions, it requires more computational time due to its greater number of trainable parameters. To address resource constraints, less computationally intensive models offer feasible trade-offs, albeit at a slight accuracy cost. Overall, this multimodal 1D CNN framework with late fusion and TL strategies lays a foundation for more accurate, adaptable, and efficient bearing fault classification in industrial environments with variable operating conditions.

Problem

• Classify bearing faults using multimodal signals
• Enhance fault detection accuracy with 1D CNN
• Adapt to variable conditions via transfer learning

Innovation

• Uses 1D CNN for fault classification
• Employs transfer learning for robustness
• Integrates multimodal signal feature fusion