Development of a shared feature representation space enabling the integration of data originating

The gap

Evidence profile

Sourced from the future-work section and future work of the source papers, classified as general, spanning 2 journals. Those papers have been cited 1 times in total.

Supporting evidence — 3 representative gaps

• Detection and Analysis of Conveyor Belt Damage: A Review of Sensing Technologies and Signal-Based Approaches (2026) · Sensors · doi

  Development of a shared feature representation space enabling the integration of data originating from different sensing technologies and construction of more universal diagnostic systems [71,72]. Further development of diagnostic methods requires improvements in classification accuracy but also increased robustness to industrial operating conditions and enhanced interpretability of obtained results.

  generalfuture-work section
  Keywords: development shared feature representation space enabling integration data
• Fault Diagnosis of Rotating Machinery Using Supervised Machine Learning Algorithms with Integrated Data-Driven and Physics-Informed Feature Sets (2026) · Sensors · cited 1× · doi

  This study presented a supervised machine learning-based diagnostic framework for vibration-based fault diagnosis of rotating machinery using integrated data-driven and physics-informed feature sets. Multiple signal representations, including time-domain, frequency-domain, baseband, and envelope spectra, were processed in parallel to extract complementary information associated with different fault mechanisms. The extracted fea- tures were evaluated using Support Vector Machines, Random Forests, Gradient Boosting, and a weighted ensemble classifier within a two-layer diagnostic strategy. The first layer identifies the dominant fault type and the second layer evaluates the presence of imbalance as a coexisting fault. Comparative analysis between purely data- driven and hybrid feature sets showed that the inclusion of physics-informed features significantly enhanced diagnostic performance across all evaluated classifiers. Among the tested models, the hybrid ensemble classifier achieved the best overall cross-validation results, confirming the benefit of combining statistical descriptors with features linked to physical fault signatures. An additional analysis of measurement axis selection showed that although single-axis models already provide competitive performance, combining multiple axes consistently improves diagnostic accuracy. However, the results also indicated that high performance can be achieved using only tangential and radial measurements on a single bearing, which corresponds to a realistic sensor configuration in practical condition monitoring systems. The proposed diagnostic framework was validated using an independent experimen- tal dataset acquired on a different test rig with a different bearing type, sensor configu- ration, and operating speeds. The validation results demonstrated that the model can be transferred from the training dataset to previously unseen experimental conditions with moderate but practically meaningful accuracy. The observed performance reduction compared to the cross-validation results reflects the inherent difficulty of generalizing data- driven models across different machines and measurement environments. Nevertheless, the two-layer strategy proved advantageous, particularly in combined fault scenarios, as the model was able to detect imbalance even when the dominant bearing fault type was not perfectly identified. In addition, the vibration severity levels expressed in terms of RMS velocity were consistent with the predicted machine conditions and aligned with international vibration evaluation standards. This enables the direct interpretation of diagnostic outputs in terms of fault severity and maintenance urgency, supporting practical decision-making in condition monitoring applications. Although deep learning approaches have attracted significant attention in recent ro- tating machinery diagnosis research due to their automatic feature learning capabilities and high classification performance on

  generalfuture workevidence 5/5
  Keywords: fault diagnostic performance using different layer learning vibration driven feature type models validation bearing machine
• A fault diagnosis method for aero-engine inter-shaft bearings based on 1DCNN-Transformer-BiGRU (2026) · Scientific Reports · doi

  This study develops a CT-BiGRU architecture for fault diagnosis of aero-engine inter-shaft bearings under noisy and varying operating conditions. A two-layer large-kernel 1DCNN is employed to enlarge the receptive field and improve hierarchical feature extraction from long vibration sequences. The Transformer encoder is used to model dependencies among the extracted features, while the BiGRU network captures bidirectional temporal variation patterns. The three components are arranged in a progressive manner to integrate local feature extraction, feature dependency modeling, and temporal information extraction. Experimental results on the HIT aero-engine inter-shaft bearing dataset show that the proposed architecture achieves favorable diagnostic performance under different operating conditions and maintains stable classification accuracy in the presence of noise interference. The feasibility of the proposed CT-BiGRU architecture is evaluated using a real aero-engine inter-shaft bearing dataset containing multiple rotational speeds. The current study focuses on normal conditions, inner ring faults, and outer ring faults. Additional fault types, including rolling element faults and cage faults, should be considered in future work to provide a more comprehensive evaluation. The Transformer encoder also increases computational cost, which may limit real-time deployment on resource-constrained devices. Model compression methods, such as attention pruning and knowledge distillation, will be investigated to reduce computational complexity while preserving diagnostic performance.

  generalfuture workevidence 5/5
  Keywords: faults bigru architecture aero engine inter shaft conditions feature extraction fault operating transformer encoder model

Questions about this gap