Tao Fan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Tao Fan has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
- Puqi Ning
School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing, China
10 shared publications - Qi Li
State Key Laboratory of High Density Electromagnetic Power and Systems, Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, China
7 shared publications Xuhui Wen
State Key Laboratory of High Density Electromagnetic Power and Systems, Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, China
7 shared publications- Xuhui Wen
Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, 100080, P. R. China
6 shared publications - Dan Zheng
Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, China
5 shared publications - Xiaoshuang Hui
Institute of Electrical Engineering Chinese Academy of Sciences, Beijing, China
5 shared publications
Latest Publications
- Open Access
- 28 October 2026
- Published by MDPI AG in World Electric Vehicle Journal
- Vol. 17  (11) , 564
Remaining Useful Life Prediction of Electric-Vehicle Traction-Motor Winding Insulation Using a Dominant-Frequency Impedance Feature and Gaussian Process Regression
The reliability of traction-motor winding insulation directly affects the safe operation of electric vehicles. This study proposes a remaining useful life (RUL) prediction method for traction-motor main insulation using bipolar-pulse response measurements and Gaussian process regression (GPR). Simulated winding modules were subjected to full-lifecycle accelerated thermal aging at 300 °C. During the diagnostic stage of each aging subcycle, synchronized insulation-voltage and insulation-current waveforms were acquired at 50 MSPS. A 2 μs segment following the positive-polarity rising edge was processed using a Hann window, a 10th-order Butterworth band-pass filter from 2 to 12 MHz, and a 100-point fast Fourier transform. The frequency bin with the maximum current-spectrum magnitude was identified, and a frequency-weighted impedance feature was obtained from the voltage-to-current spectral-magnitude ratio at the same frequency. One complete lifetime trajectory containing 119 diagnostic measurements was evaluated retrospectively. The experimentally observed EOL at 428.405 h was independently established when the 10 kHz winding-to-ground capacitance fell to 44.6% of its initial value. A trend-informed GPR model was updated after each new measurement, and RUL was calculated from the first forecast crossing of the fixed feature threshold. Across ten rolling origins after 80% of the observed lifetime, the mean bounded relative accuracy was 80.08%, with an MAE of 6.455 h and an RMSE of 8.923 h. However, at the final origin, where the actual RUL was 4.185 h, the bounded relative accuracy was 0% because the terminal feature increase occurred abruptly between two measurements. The results constitute a single-specimen retrospective proof of concept; independent multi-specimen validation is required before predictive performance can be generalized.
- 28 October 2026
- Published by MDPI AG in World Electric Vehicle Journal
- Vol. 17  (11) , 564
Open AccessRemaining Useful Life Prediction of Electric-Vehicle Traction-Motor Winding Insulation Using a Dominant-Frequency Impedance Feature and Gaussian Process Regression
The reliability of traction-motor winding insulation directly affects the safe operation of electric vehicles. This study proposes a remaining useful life (RUL) prediction method for traction-motor main insulation using bipolar-pulse response measurements and Gaussian process regression (GPR). Simulated winding modules were subjected to full-lifecycle accelerated thermal aging at 300 °C. During the diagnostic stage of each aging subcycle, synchronized insulation-voltage and insulation-current waveforms were acquired at 50 MSPS. A 2 μs segment following the positive-polarity rising edge was processed using a Hann window, a 10th-order Butterworth band-pass filter from 2 to 12 MHz, and a 100-point fast Fourier transform. The frequency bin with the maximum current-spectrum magnitude was identified, and a frequency-weighted impedance feature was obtained from the voltage-to-current spectral-magnitude ratio at the same frequency. One complete lifetime trajectory containing 119 diagnostic measurements was evaluated retrospectively. The experimentally observed EOL at 428.405 h was independently established when the 10 kHz winding-to-ground capacitance fell to 44.6% of its initial value. A trend-informed GPR model was updated after each new measurement, and RUL was calculated from the first forecast crossing of the fixed feature threshold. Across ten rolling origins after 80% of the observed lifetime, the mean bounded relative accuracy was 80.08%, with an MAE of 6.455 h and an RMSE of 8.923 h. However, at the final origin, where the actual RUL was 4.185 h, the bounded relative accuracy was 0% because the terminal feature increase occurred abruptly between two measurements. The results constitute a single-specimen retrospective proof of concept; independent multi-specimen validation is required before predictive performance can be generalized.
- Open Access
- 15 September 2026
- Published by MDPI AG in Preprints.org
Remaining Useful Life Prediction of Electric-Vehicle Traction-Motor Winding Insulation Using a Dominant-Frequency Impedance Feature and Gaussian Process Regression
The reliability of traction-motor winding insulation directly affects the safe operation of electric vehicles. This study proposes a remaining-useful-life (RUL) prediction method for traction-motor main insulation using bipolar-pulse response measurements and Gaussian process regression (GPR). Simulated winding modules were subjected to full-lifecycle accelerated thermal aging at 300 °C. During the diagnostic stage of each aging subcycle, bipolar high-voltage pulses were applied, and synchronized insulation-voltage and insulation-current waveforms were acquired at 50 MSPS. A 2 μs oscillatory segment associated with the positive-polarity rising edge was processed using a Hann window, a 10th-order Butterworth band-pass filter from 2 to 12 MHz, and a 100-point fast Fourier transform. The frequency bin with the maximum current-spectrum magnitude was identified, and the impedance magnitude was calculated from the voltage-to-current spectral-magnitude ratio at the same frequency. The impedance magnitude was then multiplied by the corresponding frequency to obtain a frequency-weighted impedance feature. The feature was corrected to a reference temperature of 85 °C. One complete lifetime trajectory containing 119 diagnostic measurements was used for proof-of-concept evaluation. The initial feature was 13.7076 Ω·MHz, and the end-of-life threshold was defined as twice this value, 27.4152 Ω·MHz. The first threshold crossing occurred at 428.405 h. A trend-informed GPR model combining causal averaging, monotonic trend extraction, a quadratic mean function, and a short-memory Matérn-3/2 residual process was updated after each new measurement. Across ten rolling prediction origins after 80% of the observed lifetime, all ten origins produced finite threshold-crossing-based RUL estimates. The mean bounded relative accuracy was 80.08%, with an MAE of 6.455 h and an RMSE of 8.923 h. The results demonstrate the feasibility of waveform-feature-based chronological RUL prediction, while independent multi-specimen validation remains necessary.
- 15 September 2026
- Published by MDPI AG in Preprints.org
Open AccessRemaining Useful Life Prediction of Electric-Vehicle Traction-Motor Winding Insulation Using a Dominant-Frequency Impedance Feature and Gaussian Process Regression
The reliability of traction-motor winding insulation directly affects the safe operation of electric vehicles. This study proposes a remaining-useful-life (RUL) prediction method for traction-motor main insulation using bipolar-pulse response measurements and Gaussian process regression (GPR). Simulated winding modules were subjected to full-lifecycle accelerated thermal aging at 300 °C. During the diagnostic stage of each aging subcycle, bipolar high-voltage pulses were applied, and synchronized insulation-voltage and insulation-current waveforms were acquired at 50 MSPS. A 2 μs oscillatory segment associated with the positive-polarity rising edge was processed using a Hann window, a 10th-order Butterworth band-pass filter from 2 to 12 MHz, and a 100-point fast Fourier transform. The frequency bin with the maximum current-spectrum magnitude was identified, and the impedance magnitude was calculated from the voltage-to-current spectral-magnitude ratio at the same frequency. The impedance magnitude was then multiplied by the corresponding frequency to obtain a frequency-weighted impedance feature. The feature was corrected to a reference temperature of 85 °C. One complete lifetime trajectory containing 119 diagnostic measurements was used for proof-of-concept evaluation. The initial feature was 13.7076 Ω·MHz, and the end-of-life threshold was defined as twice this value, 27.4152 Ω·MHz. The first threshold crossing occurred at 428.405 h. A trend-informed GPR model combining causal averaging, monotonic trend extraction, a quadratic mean function, and a short-memory Matérn-3/2 residual process was updated after each new measurement. Across ten rolling prediction origins after 80% of the observed lifetime, all ten origins produced finite threshold-crossing-based RUL estimates. The mean bounded relative accuracy was 80.08%, with an MAE of 6.455 h and an RMSE of 8.923 h. The results demonstrate the feasibility of waveform-feature-based chronological RUL prediction, while independent multi-specimen validation remains necessary.
- 4 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. PP  (08858993) , 1-5
Decoupling Method for Simultaneous Online Tj and Bond-Wire Aging Monitoring of SiC MOSFETs
Junction temperature (Tj) monitoring using thermal sensitive electrical parameters (TSEPs) has been widely investigated. However, most TSEPs are susceptible to aging effects, making the decoupling of Tj and aging a critical challenge for the online condition monitoring of power modules. Based on the chip structural characteristics of MOSFETs, this letter proposes a novel TSEP to achieve decoupling of Tj and bond-wire aging. To this end, a dual-sampling and modulation strategy synchronized with the zero-vector is presented to achieve non-intrusive online measurement. The proposed decoupling method features a clear physical interpretation and an analytical model. Experimental results demonstrate that, for Tj monitoring, the proposed method is insensitive to bond-wire aging and achieves higher accuracy over the full operating range. For aging assessment, it enables converter-level online identification of bond-wire aging states.
- 4 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. PP  (08858993) , 1-5
Decoupling Method for Simultaneous Online Tj and Bond-Wire Aging Monitoring of SiC MOSFETs
Junction temperature (Tj) monitoring using thermal sensitive electrical parameters (TSEPs) has been widely investigated. However, most TSEPs are susceptible to aging effects, making the decoupling of Tj and aging a critical challenge for the online condition monitoring of power modules. Based on the chip structural characteristics of MOSFETs, this letter proposes a novel TSEP to achieve decoupling of Tj and bond-wire aging. To this end, a dual-sampling and modulation strategy synchronized with the zero-vector is presented to achieve non-intrusive online measurement. The proposed decoupling method features a clear physical interpretation and an analytical model. Experimental results demonstrate that, for Tj monitoring, the proposed method is insensitive to bond-wire aging and achieves higher accuracy over the full operating range. For aging assessment, it enables converter-level online identification of bond-wire aging states.
- 2 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Power Electronics
- Vol. PP  (21686777) , 1
A 400-kW Integrated Modular Machine Drive for Electrified Transportation: Architecture, Thermal Management, and EMI Filtering
The rapid electrification of transportation demands motor drive systems with high power density, high efficiency, and exceptional reliability. In this paper, the design and experimental validation of an 18-phase, 400-kW Integrated Modular Machine Drive (IMMD) system is presented. The architecture features deep integration of an open-end winding motor with single-phase full-bridge Silicon Carbide (SiC) inverters, forming 18 structurally and electrically decoupled modules that enable independent control, fault isolation, and rapid maintenance. Key innovations include a custom 1200-V/160-A stacked direct-bonded-copper (DBC) SiC power module with reduced parasitic inductance, and an ultra-conducting copper winding conductor incorporating an in-situ 3D graphene network to simultaneously enhance electrical conductivity, thermal conductivity, and mechanical strength. Both inverter and motor modules employ PINFIN-based direct liquid cooling, with thermal simulations confirming that hotspot temperatures remain below material limits under rated operation. A hybrid feedforward–feedback active EMI filter is proposed to suppress low-frequency common-mode (CM) interference, achieving significant attenuation in the 20 kHz–500 kHz band. Experimental results from single-module full-load and three-phase coordinated tests validate the design feasibility and performance predictions. The developed IMMD system demonstrates a viable pathway toward highly reliable, compact, and serviceable electric drive systems for next-generation electrified transportation.
- 2 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Power Electronics
- Vol. PP  (21686777) , 1
A 400-kW Integrated Modular Machine Drive for Electrified Transportation: Architecture, Thermal Management, and EMI Filtering
The rapid electrification of transportation demands motor drive systems with high power density, high efficiency, and exceptional reliability. In this paper, the design and experimental validation of an 18-phase, 400-kW Integrated Modular Machine Drive (IMMD) system is presented. The architecture features deep integration of an open-end winding motor with single-phase full-bridge Silicon Carbide (SiC) inverters, forming 18 structurally and electrically decoupled modules that enable independent control, fault isolation, and rapid maintenance. Key innovations include a custom 1200-V/160-A stacked direct-bonded-copper (DBC) SiC power module with reduced parasitic inductance, and an ultra-conducting copper winding conductor incorporating an in-situ 3D graphene network to simultaneously enhance electrical conductivity, thermal conductivity, and mechanical strength. Both inverter and motor modules employ PINFIN-based direct liquid cooling, with thermal simulations confirming that hotspot temperatures remain below material limits under rated operation. A hybrid feedforward–feedback active EMI filter is proposed to suppress low-frequency common-mode (CM) interference, achieving significant attenuation in the 20 kHz–500 kHz band. Experimental results from single-module full-load and three-phase coordinated tests validate the design feasibility and performance predictions. The developed IMMD system demonstrates a viable pathway toward highly reliable, compact, and serviceable electric drive systems for next-generation electrified transportation.
- 20 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 41  (4) , 4728-4739
Switching Loss Optimization for Multichip SiC Stacked DBC Power Modules Through Distributed Gate Damping Networks
To address the escalating risks of parasitic oscillations and high switching losses in multichip paralleled SiC mosfet modules with high device counts, this article proposes a distributed gate damping network featuring integrated high-value embedded resistors (30–40 Ω) in a stacked direct-bonded copper (DBC) package. Unlike conventional single-resistor solutions (<10 Ω), the proposed architecture embeds high-damping resistors into stacked DBC substrates, constructing localized RC networks to suppress inter-chip oscillation modes. A theoretical derivation and analytical model show that such per-die embedding reduces the total critical gate-damping requirement by 25%, thereby lowering switching loss and explaining why distributed damping outperforms a single small external resistor. The concept is implemented by integrating vertical chip-type resistors directly in each gate loop; no dedicated DBC routing is required and the parts are placed on the top DBC, avoiding high-temperature stress. A field–circuit co-design workflow is used to size the embedded values and verify ringing-free operation. Prototype modules, including a 16-chip, 1200 V/1000 A design, exhibit a 19.0–23.7% reduction in total switching loss while maintaining safe dv/dt and di/dt, confirming the effectiveness of the proposed approach for high-power-density applications.
- 20 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 41  (4) , 4728-4739
Switching Loss Optimization for Multichip SiC Stacked DBC Power Modules Through Distributed Gate Damping Networks
To address the escalating risks of parasitic oscillations and high switching losses in multichip paralleled SiC mosfet modules with high device counts, this article proposes a distributed gate damping network featuring integrated high-value embedded resistors (30–40 Ω) in a stacked direct-bonded copper (DBC) package. Unlike conventional single-resistor solutions (<10 Ω), the proposed architecture embeds high-damping resistors into stacked DBC substrates, constructing localized RC networks to suppress inter-chip oscillation modes. A theoretical derivation and analytical model show that such per-die embedding reduces the total critical gate-damping requirement by 25%, thereby lowering switching loss and explaining why distributed damping outperforms a single small external resistor. The concept is implemented by integrating vertical chip-type resistors directly in each gate loop; no dedicated DBC routing is required and the parts are placed on the top DBC, avoiding high-temperature stress. A field–circuit co-design workflow is used to size the embedded values and verify ringing-free operation. Prototype modules, including a 16-chip, 1200 V/1000 A design, exhibit a 19.0–23.7% reduction in total switching loss while maintaining safe dv/dt and di/dt, confirming the effectiveness of the proposed approach for high-power-density applications.
- 24 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-12
An Improved Method for Inverter-Fed Motor Insulation Online Condition Monitoring by High-Frequency Leakage Current Analysis
Insulation degradation is an important cause of inverter-fed motor failures. In order to improve the safety of electric motor systems and to avoid sudden failures, timely monitoring of motor insulation degradation is necessary. The high-frequency common-mode leakage current flowing through the groundwall insulation reflects the degradation of the groundwall insulation, especially that of the first few turns. In this paper, an insulation condition monitoring method based on leakage current is proposed. An improved common-mode model of the winding suitable for leakage current analysis is proposed, based on the characteristics of the common-mode impedance curve of the motor. By analyzing the mathematical relationship between high-frequency leakage current and insulation degradation in the first few turns, the insulation condition indicators are proposed, and it is validated by the experimental results. Based on this method, a suite of hardware for pre-processing leakage current waveforms is designed, thus providing a cost-effective insulation monitoring method that minimises both hardware and software requirements.
- 24 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-12
An Improved Method for Inverter-Fed Motor Insulation Online Condition Monitoring by High-Frequency Leakage Current Analysis
Insulation degradation is an important cause of inverter-fed motor failures. In order to improve the safety of electric motor systems and to avoid sudden failures, timely monitoring of motor insulation degradation is necessary. The high-frequency common-mode leakage current flowing through the groundwall insulation reflects the degradation of the groundwall insulation, especially that of the first few turns. In this paper, an insulation condition monitoring method based on leakage current is proposed. An improved common-mode model of the winding suitable for leakage current analysis is proposed, based on the characteristics of the common-mode impedance curve of the motor. By analyzing the mathematical relationship between high-frequency leakage current and insulation degradation in the first few turns, the insulation condition indicators are proposed, and it is validated by the experimental results. Based on this method, a suite of hardware for pre-processing leakage current waveforms is designed, thus providing a cost-effective insulation monitoring method that minimises both hardware and software requirements.
- 2 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 11  (6) , 13463-13475
Interturn Short-Circuit Fault Diagnosis of Permanent Magnet Synchronous Motor Based on No-Load Back Electromotive Force Analysis
Interturn short-circuit faults (ISCFs) in the stator winding are considered one of the most serious faults in permanent magnet synchronous machines (PMSMs). To address the need for timely and accurate ISCF diagnosis, the characteristics of the ISCF are studied, and corresponding fault diagnosis methods are proposed on this basis. Initially, an ISCF model that incorporates mutual inductance within the coil is developed, and its characteristics are thoroughly analyzed. A crucial mathematical relationship between fault severity and fault characterization is established, with a specific focus on how the variation of back electromotive force (EMF) correlates with the number of faulty coils. Based on the no-load back EMF characteristics of ISCF, a novel method is proposed to obtain the no-load back EMF during motor operation. Subsequently, ISCF diagnosis becomes feasible by analyzing the no-load back EMF. Through a combination of simulations and experiments, the effectiveness and reliability of the proposed principle and method are validated. Importantly, it is demonstrated that this diagnostic method remains robust and consistent, irrespective of the motor’s operating conditions.
- 2 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 11  (6) , 13463-13475
Interturn Short-Circuit Fault Diagnosis of Permanent Magnet Synchronous Motor Based on No-Load Back Electromotive Force Analysis
Interturn short-circuit faults (ISCFs) in the stator winding are considered one of the most serious faults in permanent magnet synchronous machines (PMSMs). To address the need for timely and accurate ISCF diagnosis, the characteristics of the ISCF are studied, and corresponding fault diagnosis methods are proposed on this basis. Initially, an ISCF model that incorporates mutual inductance within the coil is developed, and its characteristics are thoroughly analyzed. A crucial mathematical relationship between fault severity and fault characterization is established, with a specific focus on how the variation of back electromotive force (EMF) correlates with the number of faulty coils. Based on the no-load back EMF characteristics of ISCF, a novel method is proposed to obtain the no-load back EMF during motor operation. Subsequently, ISCF diagnosis becomes feasible by analyzing the no-load back EMF. Through a combination of simulations and experiments, the effectiveness and reliability of the proposed principle and method are validated. Importantly, it is demonstrated that this diagnostic method remains robust and consistent, irrespective of the motor’s operating conditions.
- 28 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 41  (1) , 150-155
Online Gate Leakage Current Monitoring for SiC MOSFET Based on Average Gate Drive Current Extraction
Monitoring the gate leakage current is crucial for silicon carbide mosfets, as their gate structure is more fragile than that of Si-based devices. However, the existing gate leakage current monitoring methods still fail to simultaneously achieve high monitoring resolution and compatibility with a flexible gate drive structure. This letter proposes an online gate leakage current monitoring method for SiC mosfets based on the extraction of the average gate drive current. Compared with existing methods, the proposed method achieves a monitoring resolution at the submicroamperes level. For verification, the monitoring circuit is implemented in a high-frequency boost converter. The experimental results demonstrate that an acceptable relative error is achieved across the entire range of estimated gate leakage current, and that the proposed method does not interfere with the normal operation of the power converter.
- 28 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 41  (1) , 150-155
Online Gate Leakage Current Monitoring for SiC MOSFET Based on Average Gate Drive Current Extraction
Monitoring the gate leakage current is crucial for silicon carbide mosfets, as their gate structure is more fragile than that of Si-based devices. However, the existing gate leakage current monitoring methods still fail to simultaneously achieve high monitoring resolution and compatibility with a flexible gate drive structure. This letter proposes an online gate leakage current monitoring method for SiC mosfets based on the extraction of the average gate drive current. Compared with existing methods, the proposed method achieves a monitoring resolution at the submicroamperes level. For verification, the monitoring circuit is implemented in a high-frequency boost converter. The experimental results demonstrate that an acceptable relative error is achieved across the entire range of estimated gate leakage current, and that the proposed method does not interfere with the normal operation of the power converter.