Jun Wang

Jun Wang

Auto-generated profile, based on publicly available information

Jun Wang

College of Electrical Engineering, Zhejiang University, Hangzhou China 310027

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 March 2017 in IEEE Transactions on Power Electronics
Verified
Journal Article
Closed access
JWJun Wang
Jun Wang
HMHao Ma
Hao Ma
ZBZhihong Bai
Zhihong Bai
The modular multilevel converter (MMC) with nearest level modulation (NLM) is widely used in the high voltage applications for low switching frequency and easy implementation. Existing literature has not provided a complete submodule (SM) fault ride-through scheme for MMC with NLM. In this paper, a strategy including fault detection, localization, redundancy, and recovery is proposed to ensure continuous operation of MMC under IGBT open-circuit faults conditions. It only requires a few hardware and software resources. The features of MMC and SMs with three types of failures are studied, respectively. Based on these, the fault detection method is proposed by using a simple hardware circuit, thus high computation complexity is avoided. Since current fault localization schemes are limited to MMC with carrier phase shifted pulse width modulation, this paper further proposes a strategy for MMC with NLM to locate the faulty SM and identify the fault type. After this, the fault redundancy and the proposed fault recovery method are applied to eliminate the fault and then exit the failure state. Therefore, the ability of SM fault ride-through can be realized. Analysis of failure characteristics are verified in simulation. Experimental results based on a single-phase MMC prototype with 11 SMs per arm are presented to demonstrate the validity of the proposed fault ride-through strategy.

23 February 2018 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
JWJun Wang
Jun Wang
XHXu Han
Xu Han
HMHao Ma
Hao Ma
Zhihong BaiZhihong Bai
Zhihong Bai
The circulating current of modular multilevel converter (MMC) has a great impact on the power losses, the submodule capacitor voltage ripples and the rating of power devices. Existing analysis of circulating current has considered the capacitor voltage. In this paper, the effect of semiconductor devices on the circulating current is further presented. In order to realize the different circulating current injection targets under different conditions, a unified on-line calculation scheme by using the instantaneous information of MMC is proposed to generate the circulating current injection references. It avoids detecting the amplitude and phase of the output current or using an extensive lookup table. Besides, a concise analysis model is built and it reveal the impact on the capacitor voltage ripple and the RMS value of arm current with different injection references, thus guide the determination of the injection ratio. Experimental results based on a three-phase MMC prototype with 6 submodules per arm are presented to demonstrate the analysis of circulating current and the proposed circulating current injection references generating method is verified by different circulating current control targets with different loads.

7 January 2019 in IEEE Transactions on Power Electronics
Verified
Journal Article
Closed access
Zhan LiZhan Li
Zhan Li
BWBorong Wang
Borong Wang
YRYini Ren
Yini Ren
JWJun Wang
Jun Wang
Zhihong BaiZhihong Bai
Zhihong Bai
HMHao Ma
Hao Ma
This paper presents a transistor open-circuit fault diagnosis method based on average model and post-fault reconfiguration algorithms for grid-tied single-phase inverter that transfers power bi-directionally. The transistor open-circuit fault occurrence is detected by comparing average bridge arm pole-to-pole voltage deviation and threshold first. Then, the exact faulty transistor is identified by developed algorithms based on post-fault reconfiguration of control (PFRC) and of utilizing redundant leg (PFRUR). These two fault identification algorithms are suitable for different situations. Compared with the PFRUR, the PFRC can identify the faulty transistor without help of redundant leg, but the identification time is longer. For inverters equipped with redundant leg already, the PFRUR can achieve faster identification speed, so that inverters in reliability-oriented applications can return to normal operation after fault with short interruption. With average model, this method only requires existing signals sampled for control, thus it can be embedded in system easily without adding extra sensors and diagnosis circuits. The sampling frequency can be as low as switching frequency. Furthermore, this method is suitable for both inverters with L or LCL filters, so that the method can be applied more widely. Finally, experiments are carried out on a 550W/10kHz grid-tied single-phase inverter to verify the effectiveness.