Zhen Cheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhen Cheng has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Baiqiang Yin
Hefei University of Technology
6 shared publications - Lei Zuo
Hefei University of Technology
5 shared publications - Lifen Yuan
Hefei University of Technology
5 shared publications - Chengwei Ding
Hefei University of Technology
4 shared publications - Yigang He
The State Key Laboratory of Power Grid Environmental Protection, School of Electrical Engineering and Automation, Wuhan University, Wuhan, China
4 shared publications - Bing Li
Hefei University of Technology
3 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75, 3521513
High-Impedance Grounding Fault Detection Method for Distribution Networks Based on a Knowledge Distillation Optimized Lightweight Learning Model
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75, 3521513
High-Impedance Grounding Fault Detection Method for Distribution Networks Based on a Knowledge Distillation Optimized Lightweight Learning Model
- 7 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-11
Improved Singular Value Decomposition for Faulty Feeder Detection in Resonant Grounding Distribution System
In a resonant grounding distribution system (RGDS), the transient zero-sequence current component generated by single-phase-to-ground fault (SPGF) has unique modal characteristics, which is affected by fault initial angles. However, the complex working environment of distribution system complicates the extraction of transient components, especially under high noise levels. Singular value decomposition (SVD) is an effective technique to separate signal components from noise, while the effective singular values are the primary determinant of the separation accuracy. To improve the effective signal separation accuracy, a faulty feeder detection method based on improved SVD is proposed. By analyzing the transient features of zero-sequence current under different fault initial angles, an efficient singular value feature extraction method focusing on proper singular value order selection is developed. Furthermore, the similarity feature of the exponential decay sinusoidal components of zero-sequence current is integrated to achieve faulty feeder detection. Simulation and field tests validate the effectiveness of the proposed method under different fault initial angles. When the signal-to-noise ratio (SNR) exceeds 5 dB, the feeder detection accuracy exceeds 92.5%, which is 47% higher than that of traditional method. Moreover, the proposed method shows excellent robustness under high noise levels.
- 7 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-11
Improved Singular Value Decomposition for Faulty Feeder Detection in Resonant Grounding Distribution System
In a resonant grounding distribution system (RGDS), the transient zero-sequence current component generated by single-phase-to-ground fault (SPGF) has unique modal characteristics, which is affected by fault initial angles. However, the complex working environment of distribution system complicates the extraction of transient components, especially under high noise levels. Singular value decomposition (SVD) is an effective technique to separate signal components from noise, while the effective singular values are the primary determinant of the separation accuracy. To improve the effective signal separation accuracy, a faulty feeder detection method based on improved SVD is proposed. By analyzing the transient features of zero-sequence current under different fault initial angles, an efficient singular value feature extraction method focusing on proper singular value order selection is developed. Furthermore, the similarity feature of the exponential decay sinusoidal components of zero-sequence current is integrated to achieve faulty feeder detection. Simulation and field tests validate the effectiveness of the proposed method under different fault initial angles. When the signal-to-noise ratio (SNR) exceeds 5 dB, the feeder detection accuracy exceeds 92.5%, which is 47% higher than that of traditional method. Moreover, the proposed method shows excellent robustness under high noise levels.
- 6 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dielectrics and Electrical Insulation
- Vol. PP (10709878) , 1
Small-Sample GIS Partial Discharge Type Identification Method Based on Fusion of One-Dimensional AT-DRSN and IDRN Models
Different types of partial discharges (PD) in gas insulated switchgear (GIS) cause different degrees of GIS insulation damage, and correct PD identification is critical to GIS insulation status. This paper proposes a multi-model fusion PD pattern recognition method based on one-dimensional adaptive transfer deep residual shrinkage network (AT-DRSN) and improved deep residual network (IDRN), which fully utilizes PD time-domain waveform images generated from field inspection. Firstly, the finite-difference time-domain (FDTD) method and mathematical models are used to simulate GIS partial discharges, and the time-domain waveform image datasets of four typical partial discharge defects are established. Secondly, the network model is built based on time-domain feature sequences and time-order feature parameters respectively, and effectively combined with transfer learning method, and finally, the AT-DRSN and IDRN models are weighted and fused for PD pattern recognition. The results show that the model proposed in this paper can effectively achieve high-precision diagnosis of small-sample GIS PDs, with a recognition accuracy of 98.3%, and the accuracy under strong noise is still greater than 95%, which has higher accuracy and anti-interference performance compared with other methods, and is of reference value for small-sample PD recognition in the field.
- 6 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dielectrics and Electrical Insulation
- Vol. PP (10709878) , 1
Small-Sample GIS Partial Discharge Type Identification Method Based on Fusion of One-Dimensional AT-DRSN and IDRN Models
Different types of partial discharges (PD) in gas insulated switchgear (GIS) cause different degrees of GIS insulation damage, and correct PD identification is critical to GIS insulation status. This paper proposes a multi-model fusion PD pattern recognition method based on one-dimensional adaptive transfer deep residual shrinkage network (AT-DRSN) and improved deep residual network (IDRN), which fully utilizes PD time-domain waveform images generated from field inspection. Firstly, the finite-difference time-domain (FDTD) method and mathematical models are used to simulate GIS partial discharges, and the time-domain waveform image datasets of four typical partial discharge defects are established. Secondly, the network model is built based on time-domain feature sequences and time-order feature parameters respectively, and effectively combined with transfer learning method, and finally, the AT-DRSN and IDRN models are weighted and fused for PD pattern recognition. The results show that the model proposed in this paper can effectively achieve high-precision diagnosis of small-sample GIS PDs, with a recognition accuracy of 98.3%, and the accuracy under strong noise is still greater than 95%, which has higher accuracy and anti-interference performance compared with other methods, and is of reference value for small-sample PD recognition in the field.
- 18 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-10
An Improved Bayesian Learning Method for Distribution Grid Fault Location and Meter Deployment Optimization
With the development of both distribution automation and management system in distribution grid, the fault location methods based on wide-area measurement show great potential. Timely fault location based on wide-area measurements depends largely on the computing time of the algorithm. In addition, the meters layout method also affects the cost and accuracy of fault location. In this article, the efficiency of sparse Bayesian learning (SBL) algorithm and the meter configuration strategy for distributed grid fault location have been improved. Different from the previous calculation for all parameters in the multidimensional vector of Bayesian learning, the proposed method uses the $\boldsymbol {K}$ -means algorithm to extract partial parameters in the vector for update, which improves the computational efficiency without reducing the reconstruction accuracy. Then, for the sparse meter configuration using the principle of sparse recovery, which can be optimized by improving the coherence principle of the sensing matrix, the simulation results show that the improved Bayesian learning of the proposed $\boldsymbol {K}$ -means algorithm significantly improves the computational efficiency, regardless of low- or high-dimensional signals. At the same time, use the low coherence principle of sparse recovery to guide the meter configuration in distribution grid, which could increase the overall fault location accuracy of about 4% compared to recent work. To verify the effectiveness, the performance of the method is tested on IEEE standard 69 bus distribution grids. Simulation results indicate that the suggested strategy performs well for various scenarios.
- 18 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-10
An Improved Bayesian Learning Method for Distribution Grid Fault Location and Meter Deployment Optimization
With the development of both distribution automation and management system in distribution grid, the fault location methods based on wide-area measurement show great potential. Timely fault location based on wide-area measurements depends largely on the computing time of the algorithm. In addition, the meters layout method also affects the cost and accuracy of fault location. In this article, the efficiency of sparse Bayesian learning (SBL) algorithm and the meter configuration strategy for distributed grid fault location have been improved. Different from the previous calculation for all parameters in the multidimensional vector of Bayesian learning, the proposed method uses the $\boldsymbol {K}$ -means algorithm to extract partial parameters in the vector for update, which improves the computational efficiency without reducing the reconstruction accuracy. Then, for the sparse meter configuration using the principle of sparse recovery, which can be optimized by improving the coherence principle of the sensing matrix, the simulation results show that the improved Bayesian learning of the proposed $\boldsymbol {K}$ -means algorithm significantly improves the computational efficiency, regardless of low- or high-dimensional signals. At the same time, use the low coherence principle of sparse recovery to guide the meter configuration in distribution grid, which could increase the overall fault location accuracy of about 4% compared to recent work. To verify the effectiveness, the performance of the method is tested on IEEE standard 69 bus distribution grids. Simulation results indicate that the suggested strategy performs well for various scenarios.
- 16 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-9
Sparse Bayesian Learning-Based Topology Reconstruction Under Measurement Perturbation for Fault Location
Topology reconstruction is of great importance for fault location in smart grids. However, with the increasing development of grid infrastructure, the measurement perturbation increases and the topology reconstruction accuracy decreases. By analyzing the topology sparsity variation of the power grid before and after faults, a new method for fault location based on topology recognition is proposed. Traditional Bayesian inference with synthesized noise term is introduced to solve the measurement perturbation problem, and a fixed-point iteration method is adopted for inference to solve intractable posterior in Bayesian learning. Extensive simulations are conducted to verify the feasibility of the method, which show that the root mean square error (RMSE) of the proposed method is lower than 0.03 when the noise level varies from $5\mathrm {e}^{-6}$ to $5\mathrm {e}^{-3}$ . The fault location results are more accurate compared with traditional methods when facing ground fault. The proposed method is of great significance for fault location under unknown topology information.
- 16 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-9
Sparse Bayesian Learning-Based Topology Reconstruction Under Measurement Perturbation for Fault Location
Topology reconstruction is of great importance for fault location in smart grids. However, with the increasing development of grid infrastructure, the measurement perturbation increases and the topology reconstruction accuracy decreases. By analyzing the topology sparsity variation of the power grid before and after faults, a new method for fault location based on topology recognition is proposed. Traditional Bayesian inference with synthesized noise term is introduced to solve the measurement perturbation problem, and a fixed-point iteration method is adopted for inference to solve intractable posterior in Bayesian learning. Extensive simulations are conducted to verify the feasibility of the method, which show that the root mean square error (RMSE) of the proposed method is lower than 0.03 when the noise level varies from $5\mathrm {e}^{-6}$ to $5\mathrm {e}^{-3}$ . The fault location results are more accurate compared with traditional methods when facing ground fault. The proposed method is of great significance for fault location under unknown topology information.