Open-Circuit Fault Diagnosis of NPC Inverter Based on Improved 1-D CNN Network

Yuan, Wanmai; Li, Zhigang; He, Yigang; Cheng, Ruiqi; Lü, Li; Ruan, Yi

doi:10.1109/tim.2022.3166166

Cited by 29 publications

(7 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to ensure the complexity of Z s , E, and D are fit by the similar n-layer 1-dimension convolution neural network (1-DCNN). The details of 1-DCNN can be found in Yuan et al (2022).…”

Section: Model Structure and Training Methodsmentioning

confidence: 99%

“…Compared with other methods, it has better diagnostic results under various conditions. Yuan et al (2022) used 1-DCNN to extract features from the original data, and 100% accuracy of fault diagnosis was realized perfectly in the experiment of the IGBT open-circuit fault diagnosis for NPC inverters. However, in order to train a reliable classification model, a large amount of fault data are required that can fully reflect the real operating conditions of the target system, which is particularly important for deep learning models.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Disturbance auto-encoder generation model: Few-shot learning method for IGBT open-circuit fault diagnosis in three-phase converters

Qiu²,

Zhang³

et al. 2023

Front. Energy Res.

View full text Add to dashboard Cite

With the rapid development of converters in a variety of industrial fields, the fault diagnosis of power switching devices has become an important factor in ensuring the safe and reliable operation of related systems. In recent years, machine learning has performed well in many fault diagnosis tasks. The success of these advanced methods depends on sufficient marked samples for each fault type. However, in most industrial applications, it is expensive and difficult to collect fault samples, and the fault diagnosis model trained under the limited samples cannot meet the requirements of fault diagnosis accuracy. In order to solve this problem, this study proposes a few-shot learning method based on fault sample generation to realize the open-circuit fault diagnosis of IGBT in a three-phase PWM converter. This method is the deformation of the auto-encoder called the disturbance auto-encoder generation model. By designing the model structure and training algorithm constraints, the encoder learns the nonlinear transferable disturbance from the normal operating sample pairs. Then, the disturbance is applied to the decoder to synthesize new fault samples to realize the training of the fault diagnosis model with limited samples. The biggest advantage of this method is that it can achieve 95.90% fault diagnosis accuracy by only collecting the samples in the normal operating conditions of the target system. Finally, the feasibility and advantages of the proposed method are verified by comparative experiments.

show abstract

Section: Model Structure and Training Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Disturbance auto-encoder generation model: Few-shot learning method for IGBT open-circuit fault diagnosis in three-phase converters

Qiu²,

Zhang³

et al. 2023

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…However, the diagnostic speed is influenced significantly due to the network’s extensive parallel structure. In [ 18 ], a one-dimensional CNN with an improved stochastic gradient optimization method is introduced for extracting and classifying inverter fault features. In [ 19 ], a multimodal deep residual filter network (DRFN) is proposed, achieving a 99.18% accuracy rate in identifying open-circuit faults in T-type three-level inverters.…”

Section: Introductionmentioning

confidence: 99%

An Open-Circuit Fault Diagnosis Method for Three-Level Neutral Point Clamped Inverters Based on Multi-Scale Shuffled Convolutional Neural Network

Yan,

Wu,

Cao

et al. 2024

Sensors

View full text Add to dashboard Cite

This study constructs a power switching device open-circuit fault diagnosis model for a three-level neutral point clamped inverter based on the multi-scale shuffled convolutional neural network (MSSCNN) and extracts and classifies the fault information contained in the output current of inverters. The model employs depthwise separable convolution and channel shuffle techniques to improve diagnostic accuracy and reduce model complexity. The experimental results show that the new model has lower model complexity, better noise resistance and higher average diagnostic accuracy compared with fault diagnosis models based on CNN, ResNet, ShuffleNet V2 and Mobilenet V3 networks.

show abstract

“…However, the shutdown is not safe for mitigating the effect of open switch and turn-on switch failures in power converters. Predictive modeling control methods have been used for the diagnosis of open and turn-on switch failures in the 3-L NPC inverter, using the sparse matrix representation and the support vector machine approach [16]- [20]. The diagnosis implements fault-tolerant operation by identifying space vectors, that can operate the converter safely.…”

Section: Introductionmentioning

confidence: 99%

Overvoltage Protection of Series-Connected 10kV SiC MOSFETs Following Switch Failures in MV 3L-NPC Converter for Safe Fault Isolation and Shutdown

Parashar,

Isik,

Kolli

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This paper presents a design methodology for overvoltage protection across 10kV SiC MOSFETs during turn-off after switch failure in a MV SST Power Conditioning System (PCS) enabled by a cascaded Three-Phase (3P) Three-level (3L) Neutral Point Clamped (NPC) Active Front-End Converter (AFEC) and Dual Active Bridge (DAB) using series-connected 10kV SiC MOSFETs and 10kV SiC JBS diodes. The methodology uses an active voltage clamp at the gate terminal and desat detection technique to identify abrupt open and turn-on switch failures across series-connected 10kV SiC MOSFETs. The analytical model estimates over-current time and turn-off voltage transition by considering bus bar inductance, device base plate capacitance and common mode (CM) choke tied between the heat sink and midpoint of the DC link capacitor. The transition model is used to evaluate the turn-off timing for series-connected MOSFETs, snubber resistors, snubber capacitors, and gate resistors to avoid MOSFET overvoltage during converter shutdown, without affecting the voltage balancing and efficiency during normal operation. The MOSFET turn-off transition during the shutdown has been verified in the Saber RD simulation using the validated Saber RD MAST model of 10kV SiC MOSFETs and 10kV SiC JBS diodes at 13.8kV AC/24kV DC level. The fault isolation and MV SST PCS shutdown have been verified in a real-time environment using HIL setup with Xilinx FPGAs and RTDS, at 13.8kV AC/24kV DC link under PCS operating conditions. The normal operation of 3L-NPC pole hardware with modified snubber resistors, snubber capacitors, and gate resistors is verified by experiments conducted at 7kV DC, 10A load current.

show abstract

Open-Circuit Fault Diagnosis of NPC Inverter Based on Improved 1-D CNN Network

Cited by 29 publications

References 38 publications

Disturbance auto-encoder generation model: Few-shot learning method for IGBT open-circuit fault diagnosis in three-phase converters

Disturbance auto-encoder generation model: Few-shot learning method for IGBT open-circuit fault diagnosis in three-phase converters

An Open-Circuit Fault Diagnosis Method for Three-Level Neutral Point Clamped Inverters Based on Multi-Scale Shuffled Convolutional Neural Network

Overvoltage Protection of Series-Connected 10kV SiC MOSFETs Following Switch Failures in MV 3L-NPC Converter for Safe Fault Isolation and Shutdown

Contact Info

Product

Resources

About