1 January 2026 in IEEE Transactions on Industrial Electronics
9 June 2026 in IEEE Internet of Things Journal
Lightweight fault diagnosis (FD) models are essential for efficient inference on resource-constrained IoT edge devices and practical industrial deployment. While existing lightweight FD research has achieved notable progress in network architecture compression, the substantial benefits brought by quantization techniques remain largely overlooked. However, directly applying mainstream quantization methods to FD tasks is problematic, as they suffer from two main drawbacks: 1) Uniform quantization struggles to simultaneously maintain local resolution in dense regions and dynamic range coverage in sparse regions. 2) Mainstream quantization-aware training (QAT) methods rely on the straight-through estimator (STE) to approximate discrete gradients, introducing gradient bias that further limits model accuracy improvement. To address these issues, this paper proposes an Adaptive Quantization Error Minimization Method (AQEMM) for fault diagnosis, with the following main contributions: 1) An adaptive quantizer is introduced to replace the traditional fixed quantization step size, expanding the search space of quantization points from discrete integer grids to a continuous subspace, enabling quantization levels to adaptively concentrate in data-dense regions. 2) A block coordinate descent strategy is incorporated to alternately optimize quantization parameters, with quantization reconstruction error minimization as the direct optimization objective, fundamentally circumventing the gradient bias introduced by STE. Validation on FD tasks for two typical industrial devices, induction motors (IM) and permanent magnet synchronous motors (PMSM), demonstrates that the proposed 2-bit quantization model saves up to approximately 82% of hardware resources compared to full-precision (FP) deployment on an FPGA platform while maintaining competitive diagnostic accuracy. Furthermore, by optimizing the hardware design to leverage the released hardware resource margin, a significant inference speedup of approximately 20.6× is achieved over pure software inference.
6 May 2026 in IEEE Transactions on Energy Conversion
This paper presents a hybrid analytical–optimization framework for identifying the equivalent orthotropic material parameters of stator core and winding assemblies in electrical machines. An analytical cylindrical model is developed based on first-order shear deformation theory and Love's shell theory to establish the relationship between material properties and modal characteristics. The stator core and its components are modeled using equivalent layered cylindrical shells, incorporating structural features such as longitudinal ribs. The Firefly Algorithm is employed to solve the inverse identification problem by minimizing the discrepancy between predicted and measured natural frequencies. The material parameters of epoxy resin and winding insulation are identified, and the corresponding equivalent orthotropic material parameters are derived using composite material theory. The proposed method is validated through both analytical and finite element analyses, showing good agreement with experimental results, with relative errors within 3%. Compared with conventional optimization and data-driven approaches, the proposed framework demonstrates improved accuracy, convergence stability, and computational efficiency. The developed method provides an effective tool for vibration analysis and parameter identification in electromagnetic devices and offers potential for application in structural optimization and noise reduction design.
1 October 2026 in IEEE Sensors Journal