To solve the poor real-time performance of the existing fault diagnosis algorithms on transmission system rotating components, this paper proposes a novel high-dimensional OT-Caps (Optimal Transport–Capsule Network) model. Based on the traditional capsule network algorithm, an auxiliary loss is introduced during the offline training process to improve the network architecture. Simultaneously, an optimal transport theory and a generative adversarial network are introduced into the auxiliary loss, which accurately depicts the error distribution of the fault characteristic. The proposed model solves the low real-time performance of the capsule network algorithm due to complex architecture, long calculation time, and oversized hardware resource consumption. Meanwhile, it ensures the high precision, early prediction, and transfer aptitude of fault diagnosis. Finally, the model’s effectiveness is verified by the public data sets and the actual faults data of the transmission system, which provide technical support for the application.

With the development of the information age, the importance of edge computing has been highlighted in industrial site monitoring, health management, and fault diagnosis. Among them, the processing and computing of signals in edge scenarios is the cornerstone of realizing these scenarios. While the performance of edge devices has been dramatically improved, the demand for signal processing in the edge side has also ushered in explosive growth. However, the deployment of traditional serial or parallel signal processing architectures on edge devices has problems such as poor flexibility, low efficiency, and low resource utilization, making edge devices unable to exert their maximum performance. Therefore, this paper proposes a resource-saving customizable pipeline network architecture with a space-optimized resource allocation method and a coordinate addressing method for irregular topology. This architecture significantly improves the flexibility of multi-signal processing in edge devices, improves resource utilization, and further increases the performance potential of edge devices. Finally, we designed a comparative experiment to prove that the resource-saving and customizable pipeline network architecture can significantly reduce resource consumption under the premise of meeting real-time processing requirements.

Strong broadband noise can mask the weak impact responses produced by bearing faults, while wavelet reconstructions selected only by signal-domain scores may not provide useful inputs for diagnosis. To address this problem, this paper presents WPMSR-1D-ResNet, a training-aware wavelet-domain controlled-noise augmentation framework. PKPM-guided particle swarm optimization first generates scale-specific perturbation candidates in the DWT detail coefficients. Signal-fidelity constraints remove distorted reconstructions, and a proxy CNN with identity fallback selects a global candidate according to validation Macro-F1. The final 1D-ResNet is trained jointly with the measured waveform and the selected augmented view, whereas inference uses only the raw signal. Under one fixed data construction and a common fixed 20-epoch budget, WPMSR-1D-ResNet achieved $0.8311\pm 0.0607$ Macro-F1 at the predefined CWRU low-SNR endpoint and ranked third among eleven methods, placing it within the leading statistical group. Its paired mean exceeded raw-signal 1D-ResNet and per-slice PKPM replacement by 0.05582 and 0.05227, respectively, with gains in nine of ten paired computational seeds. Mixed-SNR training increased mean Macro-F1 across eight mismatched-noise conditions from $0.5463\pm 0.1320$ to $0.6105\pm 0.1292$. The method ranked second on PU and third on acquisition-held-out AT data; on AT, it reduced the normal-state false-alarm rate from 0.2854 to 0.1646 while maintaining 0.9708 fault sensitivity. Raw-only inference required 0.266 ms per slice. WPMSR-1D-ResNet, therefore, aligns wavelet augmentation with diagnostic performance, preserves useful waveform information, and removes wavelet reconstruction and PSO from the deployment path.