Uday Deshpande
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Uday Deshpande has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Narayan C. Kar
Department of Electrical and Computer Engineering, University of Windsor, Windsor, ON, Canada
4 shared publications - Hadi Mohajerani
CHARGE Lab, University of Windsor, Windsor, ON, Canada
3 shared publications - Ehsan Majma
Mechanical Engineering Department, McMaster University, Hamilton, ON, Canada
2 shared publications - Ryan Ahmed
Department of Mechanical Engineering, Centre for Mechatronics and Hybrid Technologies (CMHT), McMaster University, Hamilton, ON, Canada
2 shared publications Saeid Habibi
Mechanical Engineering Department, McMaster University, Hamilton, ON, Canada
2 shared publications- Hongfu Cheng
Department of Electrical and Computer Engineering, University of Windsor, Windsor, ON, Canada
1 shared publication
Latest Publications
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 12 (2) , 3975-3990
Power-Hardware-in-the-Loop Electric Motor Emulation for Transportation Electrification Considering Mass Production Variations
This article presents a compact and scalable modeling framework for real-time emulation of permanent magnet synchronous machines (PMSMs) using artificial neural networks (ANNs). The proposed method replaces conventional lookup-table (LUT)-based maps with an ANN to represent the nonlinear flux–current and torque–current relationships of PMSMs. Trained on finite element analysis (FEA)-derived data and fine-tuned using parameter distributions, the proposed framework captures manufacturing variability without requiring extensive retraining or LUT regeneration. In this research, parameter deviation denotes statistical dispersion within production limits and temperature-driven drift; it is not intended to identify or reproduce a specific localized motor defect. The ANN architecture is designed for deployment on FPGA-based real-time simulators (RTSs), consuming less than 1 kB of memory and 173 floating-point operations (FLOPs) per iteration. Experimental validation using a power-hardware-in-the-loop (PHIL) testbed demonstrates excellent agreement between the emulator and measured responses, achieving average mean-squared errors below 0.03% across dynamic conditions. This approach offers significant improvements in adaptability, resource efficiency, and scalability over traditional LUT-based models, making it especially well-suited for production-line testing of intelligent drive systems in industrial and automotive applications.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 12 (2) , 3975-3990
Power-Hardware-in-the-Loop Electric Motor Emulation for Transportation Electrification Considering Mass Production Variations
This article presents a compact and scalable modeling framework for real-time emulation of permanent magnet synchronous machines (PMSMs) using artificial neural networks (ANNs). The proposed method replaces conventional lookup-table (LUT)-based maps with an ANN to represent the nonlinear flux–current and torque–current relationships of PMSMs. Trained on finite element analysis (FEA)-derived data and fine-tuned using parameter distributions, the proposed framework captures manufacturing variability without requiring extensive retraining or LUT regeneration. In this research, parameter deviation denotes statistical dispersion within production limits and temperature-driven drift; it is not intended to identify or reproduce a specific localized motor defect. The ANN architecture is designed for deployment on FPGA-based real-time simulators (RTSs), consuming less than 1 kB of memory and 173 floating-point operations (FLOPs) per iteration. Experimental validation using a power-hardware-in-the-loop (PHIL) testbed demonstrates excellent agreement between the emulator and measured responses, achieving average mean-squared errors below 0.03% across dynamic conditions. This approach offers significant improvements in adaptability, resource efficiency, and scalability over traditional LUT-based models, making it especially well-suited for production-line testing of intelligent drive systems in industrial and automotive applications.
- 9 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 12 (2) , 2962-2977
Interturn Short-Circuit Fault Detection and Diagnosis in Permanent Magnet Synchronous Motors Using Interactive Multiple Model Strategy
This article introduces a novel approach for diagnosing interturn short-circuit (ITSC) faults in permanent magnet synchronous motors (PMSMs) using a two-layer interactive multiple model (IMM) strategy tailored for real-time applications. The first layer employs an IMM-constrained extended Kalman filter (IMM-CEKF) framework for fault detection, while the second layer uses an extended Kalman filter (EKF) framework for fault diagnosis. In the first layer, four models are employed: the healthy motor model and three models with ITSC faults in each of the PMSM phases. The short-circuit resistance is estimated as an augmented state alongside system states. Upon detecting a fault in one of the phases, the second layer estimates the number of shorted turns and refines the estimated short-circuit resistance value. This study addresses three key challenges in applying the IMM framework to this problem. First, distinguishing between the healthy motor and faulty models with large short-circuit resistances is difficult due to their similar dynamic responses. This issue is resolved by imposing constraints on the estimated resistance, enabling better model separation. Second, the diversity of state variables across models is managed by zero-padding the state vector and estimation error covariance matrix, with matrix-based mode probabilities mitigating errors from augmented zeros. Third, uncertainty in the number of shorted turns is addressed in the second layer by using a secondary model set to determine the exact number of shorted turns and refine the short-circuit resistance estimate. Experimental validation, conducted with a custom relay box for controlled fault injection, demonstrates the method’s effectiveness in detecting and diagnosing ITSC faults using the first and second model banks, respectively.
- 9 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 12 (2) , 2962-2977
Interturn Short-Circuit Fault Detection and Diagnosis in Permanent Magnet Synchronous Motors Using Interactive Multiple Model Strategy
This article introduces a novel approach for diagnosing interturn short-circuit (ITSC) faults in permanent magnet synchronous motors (PMSMs) using a two-layer interactive multiple model (IMM) strategy tailored for real-time applications. The first layer employs an IMM-constrained extended Kalman filter (IMM-CEKF) framework for fault detection, while the second layer uses an extended Kalman filter (EKF) framework for fault diagnosis. In the first layer, four models are employed: the healthy motor model and three models with ITSC faults in each of the PMSM phases. The short-circuit resistance is estimated as an augmented state alongside system states. Upon detecting a fault in one of the phases, the second layer estimates the number of shorted turns and refines the estimated short-circuit resistance value. This study addresses three key challenges in applying the IMM framework to this problem. First, distinguishing between the healthy motor and faulty models with large short-circuit resistances is difficult due to their similar dynamic responses. This issue is resolved by imposing constraints on the estimated resistance, enabling better model separation. Second, the diversity of state variables across models is managed by zero-padding the state vector and estimation error covariance matrix, with matrix-based mode probabilities mitigating errors from augmented zeros. Third, uncertainty in the number of shorted turns is addressed in the second layer by using a secondary model set to determine the exact number of shorted turns and refine the short-circuit resistance estimate. Experimental validation, conducted with a custom relay box for controlled fault injection, demonstrates the method’s effectiveness in detecting and diagnosing ITSC faults using the first and second model banks, respectively.
- Open Access
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 96635-96645
The Interactive Multiple Model Strategy With Matrix Form Mode Probability for Models With Different State Vectors
The Interactive Multiple Model (IMM) strategy is widely used for systems with multiple operating modes characterized by varying dynamics. This approach is particularly effective when a system undergoes sudden transitions between different modes, such as trajectory tracking problems or fault detection in systems. In some instances, as the system transitions from one mode to another, the describing model may undergo structural changes resulting in the IMM requiring different dimensionality state vectors. To accommodate these cases, model augmentation with rows and columns of zeros is used to ensure continuity of transition in the context of IMM. However, this form of augmentation with zeros can introduce inaccuracies in the final estimation produced by the IMM. To overcome this limitation, this paper proposes a modification to the combination step of the IMM strategy by introducing two distinct mode probability matrices: a diagonal matrix form and a symmetric matrix form. These matrices help mitigate the negative effects of augmented zeros during the mixing stage when applying the IMM. A classic trajectory-tracking scenario is considered in this paper, demonstrating the effectiveness of this method in estimating states that are not dynamically present across all models. This results in higher accuracy in estimating certain states compared to the standard IMM approach, outperforming state-of-the-art techniques.
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 96635-96645
Open AccessThe Interactive Multiple Model Strategy With Matrix Form Mode Probability for Models With Different State Vectors
The Interactive Multiple Model (IMM) strategy is widely used for systems with multiple operating modes characterized by varying dynamics. This approach is particularly effective when a system undergoes sudden transitions between different modes, such as trajectory tracking problems or fault detection in systems. In some instances, as the system transitions from one mode to another, the describing model may undergo structural changes resulting in the IMM requiring different dimensionality state vectors. To accommodate these cases, model augmentation with rows and columns of zeros is used to ensure continuity of transition in the context of IMM. However, this form of augmentation with zeros can introduce inaccuracies in the final estimation produced by the IMM. To overcome this limitation, this paper proposes a modification to the combination step of the IMM strategy by introducing two distinct mode probability matrices: a diagonal matrix form and a symmetric matrix form. These matrices help mitigate the negative effects of augmented zeros during the mixing stage when applying the IMM. A classic trajectory-tracking scenario is considered in this paper, demonstrating the effectiveness of this method in estimating states that are not dynamically present across all models. This results in higher accuracy in estimating certain states compared to the standard IMM approach, outperforming state-of-the-art techniques.
- 23 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Energy Conversion
- Vol. 40 (3) , 2242-2255
A Neural-Network-Based Electric Machine Emulator Using Neuro-Fuzzy Controller for Power-Hardware-in-the-Loop Testing
The emulation of permanent magnet synchronous machines (PMSMs) is critical for the advancement of power electronics and drive converter testing, particularly within power-hardware-in-the-loop (PHIL) platform. Despite significant progress, and developing accurate machine models, the amount of resources and memory used by these accurate models are not ideal for real-time applications due to added latency. Hence, a research gap exists in developing models that while accurately and efficiently replicate the dynamic behaviors of the machine model under various operating conditions, are light in resource usage. This paper addresses this gap by introducing an artificial neural network (ANN)-based machine modeling approach and combines it with a neuro-fuzzy-based control strategy to ensure robust and precise performance of the system, that is to minimize the error between the electric machine emulators (EME) and physical PMSM test results. The ANN model requires only 0.68 KB of memory compared to the 4 MB needed for traditional 1,000 × 1,000 LUT-based models, which incur greater latency due to cache limitations and interpolation demands despite lower floating-point operation (FLOP) requirements. By using this optimized ANN model with an adaptive ANFIS controller, the proposed system So, the main objective is to enhance the performance and accuracy of EMEs in PHIL testing environments. The ANN model provides a resource-efficient yet precise representation of the PMSM, while the adaptive neuro-fuzzy inference system (ANFIS)-based controller dynamically adjusts its membership functions to adapt to changing system dynamics and loading conditions and provide proper control command.
- 23 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Energy Conversion
- Vol. 40 (3) , 2242-2255
A Neural-Network-Based Electric Machine Emulator Using Neuro-Fuzzy Controller for Power-Hardware-in-the-Loop Testing
The emulation of permanent magnet synchronous machines (PMSMs) is critical for the advancement of power electronics and drive converter testing, particularly within power-hardware-in-the-loop (PHIL) platform. Despite significant progress, and developing accurate machine models, the amount of resources and memory used by these accurate models are not ideal for real-time applications due to added latency. Hence, a research gap exists in developing models that while accurately and efficiently replicate the dynamic behaviors of the machine model under various operating conditions, are light in resource usage. This paper addresses this gap by introducing an artificial neural network (ANN)-based machine modeling approach and combines it with a neuro-fuzzy-based control strategy to ensure robust and precise performance of the system, that is to minimize the error between the electric machine emulators (EME) and physical PMSM test results. The ANN model requires only 0.68 KB of memory compared to the 4 MB needed for traditional 1,000 × 1,000 LUT-based models, which incur greater latency due to cache limitations and interpolation demands despite lower floating-point operation (FLOP) requirements. By using this optimized ANN model with an adaptive ANFIS controller, the proposed system So, the main objective is to enhance the performance and accuracy of EMEs in PHIL testing environments. The ANN model provides a resource-efficient yet precise representation of the PMSM, while the adaptive neuro-fuzzy inference system (ANFIS)-based controller dynamically adjusts its membership functions to adapt to changing system dynamics and loading conditions and provide proper control command.
- 12 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 60 (9) , 1-6
Differential Model-Based Parameter Estimation of IPMSMs From Multi-State Measurements
Accurate parameter estimations are essential for efficient operation and high performance of interior permanent magnet synchronous machines (IPMSMs). Voltage source inverter (VSI) nonlinearity can adversely affect parameter estimation in IPMSM drive systems. Cross influence can compromise the accuracy of parameter estimation. This article proposes a differential model-based decoupling scheme to eliminate VSI nonlinearity effects and cross influence for accurately estimating key IPMSM parameters, including permanent magnet (PM) flux linkage, winding resistance, and machine inductances. The adverse effect of measurement noise and observational error on parameter estimation can be reduced in the proposed differential model. Utilizing the decoupling scheme, each parameter is estimated individually with high efficiency and accuracy leveraging the least square algorithm. The proposed differential model-based decoupling scheme is particularly well-suited for accurately estimating parameters over a wide speed range and diverse load conditions. The estimated parameters can improve the accuracy of predicting electromagnetic torque. Furthermore, the proposed method is noninvasive, robust, and does not require extra signal injection.
- 12 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 60 (9) , 1-6
Differential Model-Based Parameter Estimation of IPMSMs From Multi-State Measurements
Accurate parameter estimations are essential for efficient operation and high performance of interior permanent magnet synchronous machines (IPMSMs). Voltage source inverter (VSI) nonlinearity can adversely affect parameter estimation in IPMSM drive systems. Cross influence can compromise the accuracy of parameter estimation. This article proposes a differential model-based decoupling scheme to eliminate VSI nonlinearity effects and cross influence for accurately estimating key IPMSM parameters, including permanent magnet (PM) flux linkage, winding resistance, and machine inductances. The adverse effect of measurement noise and observational error on parameter estimation can be reduced in the proposed differential model. Utilizing the decoupling scheme, each parameter is estimated individually with high efficiency and accuracy leveraging the least square algorithm. The proposed differential model-based decoupling scheme is particularly well-suited for accurately estimating parameters over a wide speed range and diverse load conditions. The estimated parameters can improve the accuracy of predicting electromagnetic torque. Furthermore, the proposed method is noninvasive, robust, and does not require extra signal injection.
- 23 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Industrial Electronics
- Vol. 5 (4) , 1420-1434
RNN-Based High Fidelity Permanent Magnet Synchronous Motor Emulator Considering Driving Inverter Switching Faults
This paper presents a novel interior permanent magnet synchronous motor (IPMSM) model with a new PI-PR controller for emulation applications with the goal to accurately mimic the behavior of the lower order harmonics in the motor caused by drive inverter faults. The proposed method utilizes recurrent neural networks (RNNs) to model the machine under study with the goal to alleviate the computational burden of the emulator and reduce the overall latency in the system. To validate the accuracy of the proposed model, a comparative analysis with a Look-Up-Table (LUT)-based model under varying loading conditions has been conducted. The empirical findings validate the efficacy of the IPMSM emulator in facilitating drive inverter fault testing and demonstrate its utility in mitigating the risk of inverter impairment that may result from the emulation of machine behavior under such faulty circumstances. The proposed emulator is a significant advancement in the field of drive inverter fault testing, allowing for more accurate and efficient simulation of machine currents under defective conditions. This research provides a viable resolution to emulate the behavior of the machine at the presence of drive inverter switching failure.
- 23 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Industrial Electronics
- Vol. 5 (4) , 1420-1434
RNN-Based High Fidelity Permanent Magnet Synchronous Motor Emulator Considering Driving Inverter Switching Faults
This paper presents a novel interior permanent magnet synchronous motor (IPMSM) model with a new PI-PR controller for emulation applications with the goal to accurately mimic the behavior of the lower order harmonics in the motor caused by drive inverter faults. The proposed method utilizes recurrent neural networks (RNNs) to model the machine under study with the goal to alleviate the computational burden of the emulator and reduce the overall latency in the system. To validate the accuracy of the proposed model, a comparative analysis with a Look-Up-Table (LUT)-based model under varying loading conditions has been conducted. The empirical findings validate the efficacy of the IPMSM emulator in facilitating drive inverter fault testing and demonstrate its utility in mitigating the risk of inverter impairment that may result from the emulation of machine behavior under such faulty circumstances. The proposed emulator is a significant advancement in the field of drive inverter fault testing, allowing for more accurate and efficient simulation of machine currents under defective conditions. This research provides a viable resolution to emulate the behavior of the machine at the presence of drive inverter switching failure.