中国电机工程学报2025，Vol.45Issue(19)：7510-7524,中插20,16.DOI:10.13334/j.0258-8013.pcsee.240666
基于无监督学习的多断面耦合输电限额条件映射规则生成及其分档切换方法
Unsupervised Learning-based Conditional Mapping Rule Generation and Multi-banding Switching Method for Multiple Power Transfer Inter-corridors' Coupling Limits
摘要
Abstract
System operators usually simplify the dynamic stability constraints into tractable but over-conservative static power transfer limits,curtailing renewable energy export and consumption.To address this,an unsupervised condition mapping rule generation and multi-banding switching method for multiple inter-corridors' coupling power transfer limits is proposed.Firstly,the K-means++is employed to identify various operating patterns of renewable energy.The proof of sample size required for stable clustering is also completed.For each operating pattern,correlation coefficients are used to identify coupled inter-corridor pairs.A grid partitioning algorithm is proposed to build power transfer limits of coupled inter-corridors,thus generating a conditional mapping from operating patterns to coupling transfer limits.A model to match operating patterns and coupling transfer limits is then constructed via distance criteria and the Big-M method.Upon this,penalty functions and Lagrange multipliers are proposed for limit switching to maximize renewable energy export.Case studies on the IEEE 39-bus system and a realistic system show that,compared to traditional methods,the proposed method increases average consumption of renewable energy while reduces operational costs,under the premise of strictly satisfying the pre-fault stability verfications.关键词
多断面耦合输电限额/条件映射规则/无监督学习/预想事故校核/新能源外送消纳Key words
multiple inter-corridors' coupling power transfer limits/conditional mapping rule/unsupervised learning/pre-contingency fault validation/renewable energy export and consumption分类
信息技术与安全科学引用本文复制引用
彭浩晋,邱高,刘友波,刘俊勇,任景,叶希,陈振,王曦,税月..基于无监督学习的多断面耦合输电限额条件映射规则生成及其分档切换方法[J].中国电机工程学报,2025,45(19):7510-7524,中插20,16.基金项目
国家自然科学基金项目(52307124) （52307124）
国家重点研发计划项目(2022YFB2403400).Project Supported by National Natural Science Foundation of China(52307124) （2022YFB2403400）
National Key R&D Program of China(2022YFB2403400). （2022YFB2403400）
