Inter-turn short-circuit (ITSC) faults in permanent magnet synchronous machines (PMSMs) present a significant reliability challenge in electric vehicle (EV) drivetrains, particularly under non-stationary operating conditions characterized by inverter-driven transients, variable loads, and magnetic saturation. Existing diagnostic approaches, including motor current signature analysis (MCSA) and wavelet-based methods, are primarily designed for steady-state conditions and rely on manual feature selection, limiting their applicability in real-time embedded systems. Furthermore, the lack of publicly available, high-fidelity datasets capturing the transient dynamics and nonlinear flux-linkage behaviors of PMSMs under fault conditions poses an additional barrier to developing data-driven diagnostic solutions. To address these challenges, this study introduces a simulation framework that generates a comprehensive dataset using finite element method (FEM) models, incorporating magnetic saturation effects and inverter-driven transients across diverse EV operating scenarios. Time-frequency features extracted via Discrete Wavelet Transform (DWT) from stator current signals are used to train a Transformer model for automated ITSC fault detection. The Transformer model, leveraging self-attention mechanisms, captures both local transient patterns and long-range dependencies within the time-frequency feature space. This architecture operates without sequential processing, in contrast to recurrent models such as LSTM or RNN models, enabling efficient inference with a relatively low parameter count, which is advantageous for embedded applications. The proposed model achieves 97% validation accuracy on simulated data, demonstrating its potential for real-time PMSM fault detection. Additionally, the provided dataset and methodology contribute to the facilitation of reproducible research in ITSC diagnostics under realistic EV operating conditions.

Space Vector Theory, also known as Park's Vector Method, is a frequently utilized technique for analyzing, modeling, and controlling electrical machines and drives. In motor diagnostics applications based on Park's Vector trajectory patterns, the assistance of domain experts is indispensable for Fault Detection and Diagnosis (FDD). Yet, simultaneously occurring faults may lead to feature overlap or accumulation, posing additional challenges for identification. In this paper a comparative analysis of automating Park's Vector Approach is presented, with combining the method with advanced ML techniques. Utilizing a magnetics-based Permanent Magnet Synchronous Motor (PMSM) model, a custom simulated fault dataset is created for training purposes. The dataset is used to train three distinct models: a Siamese Network, EfficientNet and the Vision Transformer model. Throughout the evaluation, key classification metrics including the F1-score, precision, and recall are analyzed, alongside an examination of the training curves. The results indicate that the Park's Vector Approach combined with State-of-the-Art ML models efficiently detects anomalies and multiple faults. It highlights the models' ability to identify local features, accurately classify, and detect simultaneous faults, which is often challenging for domain experts due to the difficulty in distinguishing fault features of multiple faults even through visual inspection.

Various fault detection methods, particularly focused on onboard Condition-Based Monitoring (CBM) in Electrical Machines and Drives (EMDs), face limitations such as sensitivity to load variations, slow fault detection, and the absence of fully automated solutions. AI and Data-Driven methods offer flexible alternatives, utilizing historical data for pattern and anomaly identification. Among Electrical Signature Analysis techniques for electrical motor diagnostics, the Space Vector Theory (SVT) is extensively used, while Park's Vector based diagnostic solutions lack real-time Inter-Turn Short Circuit (ITSC) fault severity assessment, with available techniques often limited to binary classifiers. Implementing AI with SVT for real-time Electric Vehicle (EV) use is underdeveloped, hindered by data scarcity and diverse dataset collection challenges. Real-time simulation, accurate fault modeling, and hardware limitations pose challenges, especially for embedding AI models into processors. To achieve intelligent onboard diagnosis for ITSC fault severity in this paper, a multi-modal approach model is proposed, employing MobileNetV2 to classify Park's Vector trajectories based on the fault features related to the number of shorted turns. Performance assessments encompass both the standard MobileNetV2 and the proposed multi-modal approach model across various fault severity levels. Furthermore, to address the challenge of limited data availability, an accelerated real-time AI development environment is designed using an FPGA to generate synthetic fault pattern datasets, aligning with the standards of the Electric Vehicle industry. For modeling PMSM with ITSC faults, a fault circuit model is employed. The dataset of 900 Park's Vector trajectory images is automatically generated by varying the torque request from 10 to 100 Nm with a 10 Nm resolution. At each torque operating point, the motor currents are recorded by adjusting the number of shorted turns. Simulation results confirm the outstanding performance of MobileNetV2 in binary classification, achieving an accuracy of 99.26 %. In case of 5-class ITSC fault severity classification, the prediction accuracy reaches only 72.55 %. The here proposed multi-modal MobileNetV2 model excels, achieving a remarkable accuracy of 99.163 % in the 3-class fault severity classification and 84.907 % in the 5-class classification. These results support the superiority of the proposed multi-modal MobileNetV2 model, which is trained on the generated rich dataset. It outperforms existing Park's Vector Analysis based ITSC fault detection methods, particularly in early ITSC fault detection as it can detect faults from 6 shorted turns. Additionally, it allows for online fault severity assessment during transient operation and meets stringent requirements for onboard applications. Altogether, the results of investigations prove the presence and extractability of fine detail information in Park's Vector trajectories, for assessing ITSC fault severity. This contributes to a deeper understanding and analysis of faults in electrical motors through the use of Park's Vector trajectories.

Intelligent automatic transmission shift schedule design has been well established in the last decade. However, due to the paradigm change is currently taking place in mobility sector, which resulted in a rapid progress of Electric Vehicles and Autonomous Vehicles, intelligent automatic gear shift strategies are still in the focus of much research. In addition, the proper transmission shift schedule generation is especially important from the viewpoint of energy efficiency optimizing algorithms, which is affected by the driving style, power losses, etc. Fundamentally, conventional shift schedule design relies on lookup tables obtained from test-bench measurements and real-world driving measurements. During real time test data collection, the measurement of some variables may be impractical and/or patterns of important driving conditions may be unavailable during short-distance routes neglecting the comprehensive effects of the transient operation. Machine Learning methods in combination with model-based data generation is a promising alternative, which allows a significant reduction in development time and a more precise calibration by using rich historical data rich. Such models can be easily fitted to alternative drive systems also, which may raise more specific requirements regarding gear shift scheduling issues coupled with efficiency. In this paper the performances of Machine Learning models are investigated in automatic gear shift schedule generation based on simulated driving cycle test data. Results of simulation investigations validate the applicability and efficiency of the proposed approach.

AI-based data-driven methods are an emerging research direction in the field of power electronics. However, because of the absence of large datasets, the development of these solutions have some barriers to overcome. To properly train machine learning algorithms and neural networks a large amount of training data is necessary. This dataset can be a union of simulation and measured data. Generating simulation data with computer simulations can be slow process and gathering real data is not cost-effective. Real-Time simulators based on FPGAs can be powerful tools to accelerate simulation, and create datasets for AI applications in a cost-effective and accurate way. In this paper the possible FPGA-based solutions, which can be applicable for the problems, have been reviewed. Their applicability have been discussed, moreover a simplified FPGA-based concept have been designed and embedded into two possible AI-based application area.

Incipient inter-turn short-circuit (ITSC) faults in permanent magnet synchronous machines (PMSMs) pose substantial diagnostic challenges, particularly under transient operating conditions common in electric vehicle (EV) applications. Traditional frequency-domain techniques, such as the Fast Fourier Transform (FFT), exhibit poor time-frequency resolution, making them ineffective for capturing non-stationary fault signatures. This study employs FEM-based numerical simulations combined with Continuous Wavelet Transform (CWT) analysis to accurately detect and localize ITSC fault characteristics under dynamic load conditions. While magnetic saturation is known to influence machine behavior, its quantitative impact on electrical fault signal amplitudes has not been explicitly addressed in previous diagnostic approaches. To fill this gap, a high-fidelity FEM simulation framework was developed, encompassing the full operational envelope of PMSMs. The results demonstrate that magnetic saturation leads to a notable attenuation-approximately 20-30% - of fault-induced current components, significantly complicating onboard detection. To the best of the authors' knowledge, such an integrated, EV-specific onboard diagnostic approach for incipient ITSC faults has not yet been reported in the literature. Although onboard thermal management systems exist, incipient ITSC faults may rapidly escalate into severe winding damage within 10 to 60 minutes under continuous load. This highlights the critical need for early detection methods robust to transient dynamics and magnetic nonlinearities.