Scholarlink

<p>Power quality monitoring is essential for ensuring the reliability of electrical systems, preventing equipment damage, and guaranteeing stable electricity delivery to end users. Within the context of smart grids, this paper presents an integrated recognition-to-correction framework for the analysis and correction of key disturbances in the electrical network, going beyond conventional detection methods by directly linking fault identification to automatic remediation. Using the time frequency representation obtained with the ST-CSK transform, the proposed approach effectively localizes and characterizes these disturbances, achieving an average classification score of <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(99.75\% \)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>99.75</mn> <mo>%</mo> </mrow> </math></EquationSource> </InlineEquation> with a Random Forest classifier. All detected events are then corrected successfully: voltage-related disturbances are compensated by adjusting reactive power, while frequency-related phenomena are addressed through the synchronizing active power coefficient. Graphic and tabular results confirm the effectiveness of this integrated recognition-to-correction strategy in improving power quality and enhancing the overall stability of smart grid systems under realistic operating conditions.</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

Identification and Correction of PQ Events for Power System Monitoring

  • 期刊论文

  • 发表日期：

    2026年4月15日

查看全文
  • Nouara Ouerk,
  • Redouane Lekhal,
  • Ahmed Amirou,
  • Zahia Zidelmal,
  • Djaffar Ould Abdeslam

摘要

Power quality monitoring is essential for ensuring the reliability of electrical systems, preventing equipment damage, and guaranteeing stable electricity delivery to end users. Within the context of smart grids, this paper presents an integrated recognition-to-correction framework for the analysis and correction of key disturbances in the electrical network, going beyond conventional detection methods by directly linking fault identification to automatic remediation. Using the time frequency representation obtained with the ST-CSK transform, the proposed approach effectively localizes and characterizes these disturbances, achieving an average classification score of \(99.75\% \) 99.75 % with a Random Forest classifier. All detected events are then corrected successfully: voltage-related disturbances are compensated by adjusting reactive power, while frequency-related phenomena are addressed through the synchronizing active power coefficient. Graphic and tabular results confirm the effectiveness of this integrated recognition-to-correction strategy in improving power quality and enhancing the overall stability of smart grid systems under realistic operating conditions.