Neural-Network-Based Algorithm for Fault Detection and Identification in PMSM Drive for UAV Electric Propulsion

Fault detection and classification of electric vehicle motor drive using ensemble subspace k-nearest neighbour model

An ensemble subspace k-nearest neighbour model for fault detection and classification of switch open-circuit and short-circuit faults and indicates that the proposed model provides an effective and computationally efficient solution for reliable fault diagnosis in electric vehicle motor drive systems.

Fault Diagnosis Method Based on Temperature Rise Detection for Switched Reluctance Motor Drive Systems in Electrical Transportation

In this paper, a fault diagnosis method based on temperature rise detection is proposed for power converters in switched reluctance motor drive systems used in electrical transportation equipment. First, the total power losses of all power devices are calculated and recorded under different operating conditions in both healthy and faulty states. A finite-element electrothermal model is then established to characterize the relationship between fault-induced power-loss redistribution and variations in the temperature rise of the converter devices. Based on the power-loss analysis, temperature rise is used as a key characteristic, and a corresponding fault diagnosis method is proposed. To account for the influence of operating conditions on the diagnostic criterion, three independent backpropagation neural network (BPNN) models are developed to predict fault-specific temperature-rise thresholds using rotor speed, load torque, and ambient temperature as inputs. During diagnosis, the real-time temperature evolution of the power diodes is compared with the selected thresholds to detect power converter faults. Finally, experimental results demonstrate the validity of the proposed fault diagnosis method.

Robust Open-Circuit Fault Diagnosis of PMSMs Using Feature Fusion of Current Signatures With Deep Neural Networks

Permanent magnet synchronous motors (PMSMs) are broadly used in diverse applications due to their inherent advantages. Open-circuit faults (OCFs) are among the major fault classifications in PMSMs, posing significant concerns due to their contribution to torque ripples, vibrations, and efficiency degradation. Therefore, accurate and real-time OCF diagnosis is essential for reliable operation and predictive maintenance practices. This underscores the importance of a robust diagnostic framework that enables early fault detection and localization, supports embedded integration, and requires no additional dedicated sensors. However, existing studies rarely address these requirements together. To overcome these limitations, this article proposes a novel OCF diagnostic framework that fuses features derived from multiple strategies, including wavelet energy-based features, frequency-domain features extracted from current waveforms, and speed measurement data. The extracted feature vector is used as input to a lightweight deep neural network. The proposed approach enhances interpretability and enables seamless embedded integration compared to conventional raw-data-driven machine learning models. In addition, an extended refinement layer is incorporated to enable integrated fault detection and classification for OCF while enhancing diagnostic transparency. The effectiveness of the proposed method is demonstrated through MATLAB/Simulink simulations using the PLECS Blockset and further validated in real-time with an RTBox-based hardware-in-the-loop setup using a C2000 launchpad. Furthermore, experimental validation is conducted using a domain-adaptation strategy based on transfer learning. Performance evaluation confirms diagnostic accuracy exceeding 99% across varying operating conditions. The validation process achieves fault detection within 22% of a fundamental electrical cycle, with fault localization occurring within 40% of an average, demonstrating the robustness and adaptability of the proposed method. A sensitivity analysis of the proposed algorithm’s feature vector validates the effectiveness of high-frequency features. Furthermore, the risk distribution matrix provides insights supporting informed maintenance decisions.

Fault Detection and Isolation in Electric Vehicle Powertrain

An Improved Power Factor Correction Converter With AI-Powered Predictive Maintenance for PMSM Drives

In this work, an advanced PFC converter was designed and AI-based predictive maintenance (PM) module was suggested to be included in complex PMSM drives.In this paper, an advanced PFC converter was designed and the AI based predictive maintenance (PM) module was proposed to be added in a complex PMSM drives system. The power quality of the PFC stage is good and reduces the generation of harmonic distortion caused by PWM based soft switching and effectively controls the power factor close to unity by using the capacitor bank controls, and noise suppression. The converter, motor drive, communication and data processing is controlled by a microcontroller. The use of a multi-sensor approach, which uses a three-axis accelerometer for vibration monitoring, a current sensor for electrical fault detection and a temperature sensor for thermal condition assessment, is used in the predictive maintenance system. These are then inputted into machine-learning algorithms, that detect unabnormal vibration, temperature and current fluctuations, which happen at the occurrence of faults such as bearing wear, rotor misalignment, inverter switching problems, thermal stress and so on. The integrated solution provides greater efficiency and reliability of the PMSM drives, increased equipment life and maintenance effectiveness, reduced unexpected downtime and operating costs.

FPGA – EMBEDDED FUZZY LOGIC-PROCESSOR FOR ONLINE DETECTION OF BROKEN BARS ON INDUCTION MOTORS