Scholarlink

<p>The dynamic characteristics of doubly fed induction generator (DFIG)-based wind farms, together with the variable reactance introduced by a thyristor-controlled series capacitor (TCSC) compensation, significantly alter fault current profiles and apparent line impedance, which may lead to maloperation of conventional transmission line protection schemes. In this paper, an intelligent fault detection and classification approach for a TCSC-compensated transmission line integrated with a DFIG-based wind power generation system is proposed. The method used the Wavelet Scattering Transform (WST) for robust feature extraction with a feed-forward back propagation neural network (BPNN) for accurate classification. The proposed approach exploits the inherent stability and invariance properties of WST to extract discriminative features from transmission line current signals under dynamic operating conditions. Its performance is evaluated through extensive simulation studies involving 3240 fault cases covering variations in fault type, location, inception angle, wind-farm operating conditions, and TCSC compensation levels. A comparative analysis with discrete wavelet transform (DWT)-based features is conducted in terms of detection latency, robustness, computational efficiency, and classification accuracy. Simulation results demonstrate that the WST-based approach outperforms conventional DWT-based methods, achieving high classification accuracy up to 100% with strong robustness to noise and operating variations, while maintaining a response time suitable for practical protection applications. These results confirm the effectiveness of the proposed scheme for modern series-compensated transmission systems with high renewable energy penetration. All simulations are carried out using MATLAB.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel ANN-based approach for fault detection and classification in modern TCSC-compensated transmission lines integrated with DFIG-based wind farms utilizing WST

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月20日

查看全文
  • Eyad S. Oda,
  • Abdelrahman Mamdouh M. Habib,
  • Basem E. Elnaghi,
  • Naema M. Mansour

摘要

The dynamic characteristics of doubly fed induction generator (DFIG)-based wind farms, together with the variable reactance introduced by a thyristor-controlled series capacitor (TCSC) compensation, significantly alter fault current profiles and apparent line impedance, which may lead to maloperation of conventional transmission line protection schemes. In this paper, an intelligent fault detection and classification approach for a TCSC-compensated transmission line integrated with a DFIG-based wind power generation system is proposed. The method used the Wavelet Scattering Transform (WST) for robust feature extraction with a feed-forward back propagation neural network (BPNN) for accurate classification. The proposed approach exploits the inherent stability and invariance properties of WST to extract discriminative features from transmission line current signals under dynamic operating conditions. Its performance is evaluated through extensive simulation studies involving 3240 fault cases covering variations in fault type, location, inception angle, wind-farm operating conditions, and TCSC compensation levels. A comparative analysis with discrete wavelet transform (DWT)-based features is conducted in terms of detection latency, robustness, computational efficiency, and classification accuracy. Simulation results demonstrate that the WST-based approach outperforms conventional DWT-based methods, achieving high classification accuracy up to 100% with strong robustness to noise and operating variations, while maintaining a response time suitable for practical protection applications. These results confirm the effectiveness of the proposed scheme for modern series-compensated transmission systems with high renewable energy penetration. All simulations are carried out using MATLAB.