Scholarlink

<p>In the present technological landscape of smart grids, where renewable energy-based distributed generation units inject power at various distribution nodes, ensuring optimal operation and preventing voltage instability require autonomous and intelligent actuation at distribution level. Fast and reliable algorithms for computing power flow are crucial in numerical simulations that aid the design of closed-loop actuation strategies, contributing to the advancement of smart grid technologies. In this context, we present a unified complex variables formulation of distribution circuits that allows a comprehensive analytic comparison between various classic and contemporary power flow methods, including the traditional Newton–Raphson, Gauss–Seidel, and Iwamoto methodologies, as well as modern successive current approximations and fixed-point strategies. As a result of this comparative process, we are able to contribute novel alternatives that avoid costly matrix operations, thus improving the algorithmic performance in terms of convergence speed and iterations count. Extensive numerical tests, using open-source software and standard desktop hardware to run realistic dynamic simulations, show that the proposed algorithms surpass classic and modern methods in terms of convergence speed, saving up to a <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="202_2025_2994_Article_IEq1.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="31" /> </InlineMediaObject> <EquationSource Format="TEX">\(90\%\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>90</mn> <mo>%</mo> </mrow> </math></EquationSource> </InlineEquation> of effective computation time.</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

On convergence performance of classic and novel power flow algorithms for dynamic simulation of distribution smart grids with renewable sources

  • 期刊论文

  • 发表日期：

    2025年2月17日

查看全文
  • Miguel Parada Contzen

摘要

In the present technological landscape of smart grids, where renewable energy-based distributed generation units inject power at various distribution nodes, ensuring optimal operation and preventing voltage instability require autonomous and intelligent actuation at distribution level. Fast and reliable algorithms for computing power flow are crucial in numerical simulations that aid the design of closed-loop actuation strategies, contributing to the advancement of smart grid technologies. In this context, we present a unified complex variables formulation of distribution circuits that allows a comprehensive analytic comparison between various classic and contemporary power flow methods, including the traditional Newton–Raphson, Gauss–Seidel, and Iwamoto methodologies, as well as modern successive current approximations and fixed-point strategies. As a result of this comparative process, we are able to contribute novel alternatives that avoid costly matrix operations, thus improving the algorithmic performance in terms of convergence speed and iterations count. Extensive numerical tests, using open-source software and standard desktop hardware to run realistic dynamic simulations, show that the proposed algorithms surpass classic and modern methods in terms of convergence speed, saving up to a \(90\%\) 90 % of effective computation time.