{"title":"Real-time small-signal security assessment using graph neural networks","authors":"Glory Justin, Santiago Paternain","doi":"10.1016/j.segan.2024.101469","DOIUrl":null,"url":null,"abstract":"<div><p>Security assessment is one of the most crucial functions of a power system operator. However, growing complexity and unpredictability make this an increasingly complex and computationally difficult task. In recent times, machine learning methods have gained attention for their ability to handle complex modeling applications. Convolutional neural networks (CNNs) in particular, are widely used in literature for their adaptability for classification problems. While CNNs generate promising results and some real-time advantages, they still require long training times and computational resources. This paper proposes a graph neural network (GNN) approach to the small-signal security assessment problem using data from Phasor Measurement Units (PMUs). Using a GNN, the process for small signal security assessment can be optimized, reducing the time needed from minutes, to less than a second, thus allowing for faster real-time application. Also, using graph properties, optimal PMU placement is determined and the proposed method is shown to perform efficiently under partial observability with limited PMU data. Case studies with simulated data from the IEEE 68-bus system and the NPCC 140-bus system are used to verify the effectiveness of the proposed method showing comparisons with the CNN.</p></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"39 ","pages":"Article 101469"},"PeriodicalIF":4.8000,"publicationDate":"2024-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sustainable Energy Grids & Networks","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S235246772400198X","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0
Abstract
Security assessment is one of the most crucial functions of a power system operator. However, growing complexity and unpredictability make this an increasingly complex and computationally difficult task. In recent times, machine learning methods have gained attention for their ability to handle complex modeling applications. Convolutional neural networks (CNNs) in particular, are widely used in literature for their adaptability for classification problems. While CNNs generate promising results and some real-time advantages, they still require long training times and computational resources. This paper proposes a graph neural network (GNN) approach to the small-signal security assessment problem using data from Phasor Measurement Units (PMUs). Using a GNN, the process for small signal security assessment can be optimized, reducing the time needed from minutes, to less than a second, thus allowing for faster real-time application. Also, using graph properties, optimal PMU placement is determined and the proposed method is shown to perform efficiently under partial observability with limited PMU data. Case studies with simulated data from the IEEE 68-bus system and the NPCC 140-bus system are used to verify the effectiveness of the proposed method showing comparisons with the CNN.
期刊介绍:
Sustainable Energy, Grids and Networks (SEGAN)is an international peer-reviewed publication for theoretical and applied research dealing with energy, information grids and power networks, including smart grids from super to micro grid scales. SEGAN welcomes papers describing fundamental advances in mathematical, statistical or computational methods with application to power and energy systems, as well as papers on applications, computation and modeling in the areas of electrical and energy systems with coupled information and communication technologies.