Scholarlink

<p>This paper proposes the extension of the geometrically-based bad data processing to fast-decoupled state estimation formulation. The novelty relies upon the geometric interpretation of decoupled residuals in association with both the use of <i>complex</i> per-unit normalization, to detect and identify gross measurement errors in real-time modelling of distribution systems, and the generalized decoupled state estimation, to process modelling errors in transmission systems. While <i>cpu</i> normalization enables the use of a fast-decoupled approach to the distribution system (DS), a decoupled bus-section modelling formulation is adopted for the transmission system (TS). The paper demonstrates that the geometric interpretation of the decoupled residuals can be as effective as that devised for Lagrange multipliers in restricted state estimation while bringing the efficiency and computational superiority of decoupled methods to the detection and identification of gross measurement and network topology errors. The proposed method eliminates the usual successive state re-estimation required by conventional algorithms, as is the case with the well-known largest normalized residual test. The fast-decoupled approach allied to the <i>cpu</i> and the geometric test significantly reduces the required computational burden, which is crucial for enabling bad data processing in large-scale networks, as for real-size DS and TS modelled down to the bus-section level. Simulations on the 136-node feeder and the IEEE 30-bus system are used to demonstrate the accuracy and effectiveness of the proposed methodology.</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

Bad Data Processing in Fast-Decoupled State Estimation via Geometric Approach

  • 期刊论文

  • 发表日期：

    2025年1月8日

查看全文
  • Andressa L. Monteiro,
  • Ellen M. M. Nogueira,
  • Elizete M. Lourenço,
  • Odilon L. Tortelli

摘要

This paper proposes the extension of the geometrically-based bad data processing to fast-decoupled state estimation formulation. The novelty relies upon the geometric interpretation of decoupled residuals in association with both the use of complex per-unit normalization, to detect and identify gross measurement errors in real-time modelling of distribution systems, and the generalized decoupled state estimation, to process modelling errors in transmission systems. While cpu normalization enables the use of a fast-decoupled approach to the distribution system (DS), a decoupled bus-section modelling formulation is adopted for the transmission system (TS). The paper demonstrates that the geometric interpretation of the decoupled residuals can be as effective as that devised for Lagrange multipliers in restricted state estimation while bringing the efficiency and computational superiority of decoupled methods to the detection and identification of gross measurement and network topology errors. The proposed method eliminates the usual successive state re-estimation required by conventional algorithms, as is the case with the well-known largest normalized residual test. The fast-decoupled approach allied to the cpu and the geometric test significantly reduces the required computational burden, which is crucial for enabling bad data processing in large-scale networks, as for real-size DS and TS modelled down to the bus-section level. Simulations on the 136-node feeder and the IEEE 30-bus system are used to demonstrate the accuracy and effectiveness of the proposed methodology.