Loading [MathJax]/extensions/MathZoom.js
A Faulty Submodule Mathematical Model-Based Localization Strategy for Switch Open-Circuit Fault of Module Multilevel Converter | IEEE Journals & Magazine | IEEE Xplore

A Faulty Submodule Mathematical Model-Based Localization Strategy for Switch Open-Circuit Fault of Module Multilevel Converter


Abstract:

A faulty submodule (SM) mathematical model-based switch open-circuit fault diagnosis and localization (FDL) method for module multilevel converter (MMC) is investigated i...Show More

Abstract:

A faulty submodule (SM) mathematical model-based switch open-circuit fault diagnosis and localization (FDL) method for module multilevel converter (MMC) is investigated in this article, which can diagnosis the faulty switch in time. Initially, the mathematical model is established through the modified switching function of the faulty SM, which can express the real switching state of the faulty SM accurately. Based on the proposed mathematical model, an innovative FDL method for MMC is proposed. The faulty switch is located by comparing the capacitor voltage calculated through the proposed mathematical model with the sampled capacitor voltage value. To enhance the accuracy and anti-interference ability of the model, the capacitance observation method is proposed in the capacitor precharge stage. The proposed FDL method can diagnose and locate faulty switch in any state of the SM, even if the operating state of the faulty SM is the same as that of the normal SM. Single or multiple switch open-circuit faults can be accurately located by this method under either light or heavy load conditions. Finally, the mathematical model and FDL method are verified through simulation and experiment.
Published in: IEEE Transactions on Power Electronics ( Volume: 38, Issue: 3, March 2023)
Page(s): 3899 - 3916
Date of Publication: 30 November 2022

ISSN Information:

Funding Agency:

No metrics found for this document.

I. Introduction

Since the module multilevel converter (MMC) was proposed by Lesnicar and Marquardt [1], it is widely researched by scholars and industry because of the advantages, such as modular structure, low switching frequency, and high output voltage. So far, it has been widely used in high voltage direct current (HVdc) transmission systems, motor drive, and static reactive power compensation [2], [3], [4]. Nonetheless, dozens or even hundreds of power switches are installed in MMC for medium and high voltage applications, which are considered to be the most vulnerable devices because 31% of system failures are caused by power semiconductor devices [5]. Consequently, the switch open-circuit (OC) fault of the submodule (SM) is unavoidable in the MMC operation, which makes the current path in the faulty SM alter under certain conditions. In this case, the capacitor voltage of the faulty SM rises rapidly and the SM output voltage is abnormal. The rapid rise of the capacitor voltage not only increases the voltage stress of the device, which is easy to cause secondary faults, but also distorts the output voltage, and generates odd harmonic components in the circulating current. If the switch OC fault is not dealt with in time, the MMC even stops running or is damaged by the fault [5], which is not allowed in the HVdc transmission system. While, the OC faults are not recognized promptly in the early stages of their occurrence. Therefore, high sensitivity is inevitably necessitated in switch open-circuit fault diagnosis and localization (FDL) methods. In addition, the FDL methods should have the advantages of high computational efficiency, noise isolation, simple detection circuit, and low cost, to meet the requirements of swift and accurate positioning, decent robustness and strong anti-interference ability. Hence, numerous FDL methods have been presented to handle the faults [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35]. These methods can be broadly categorized as hardware-based [6], [7], [8], [9], [10], artificial intelligent-based [11], [12], [13], [14], [15], [16], [17], and model-based [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35].

Usage
Select a Year
2025

View as

Total usage sinceNov 2022:914
051015202530JanFebMarAprMayJunJulAugSepOctNovDec231820000000000
Year Total:61
Data is updated monthly. Usage includes PDF downloads and HTML views.

Contact IEEE to Subscribe

References

References is not available for this document.