配电网无功优化的数据驱动方法研究
Reactive power optimization of distribution network based on data driven and enhanced convolutional neural networks
张晋豪 1叶剑华 1杨耿煌 1罗凤章2
作者信息
- 1. 天津职业技术师范大学自动化与电气工程学院,天津 300222;天津职业技术师范大学天津市信息传感与智能控制重点实验室,天津 300222
- 2. 天津大学智能电网教育部重点实验室,天津 300072
- 折叠
摘要
大批分布式新能源接入,使得配电网运行环境日趋复杂,对无功优化提出了更高要求.为应对该挑战,提出了一种数据驱动的无功优化新方法.使用带残差结构的改进卷积神经网络提取配电网负荷和新能源出力数据的高维特征;进行特征位置编码并引入自注意力机制进一步处理,建立数据驱动模型.利用历史运行数据和无功优化最优策略库训练所提无功优化模型,发掘配电网节点数据和无功设备补偿策略间的非线性映射规律.通过IEEE 33 节点算例验证了本文所提方法的有效性.
Abstract
The operational complexity of distribution networks is increasingly exacerbated due to the integration of numerous distributed new energy sources,which imposes more stringent requirements for reactive power optimization.To address this challenge,this paper proposes a new data-driven approach for reactive power optimization.Initially,an enhanced convolu-tional neural network with a residual structure is employed to extract high-dimensional features from the load data and new energy output of the distribution network.Subsequently,encoding for feature location is performed,and a self-attention mechanism is integrated to refine these features,thereby establishing a data-driven model.The proposed reactive power op-timization model is trained using historical operating data and a repository of optimal reactive power optimization strategies,enabling it to discern the intricate nonlinear relationship between distribution network node data and reactive power equip-ment compensation strategies.The validity of the proposed method is demonstrated through a case study involving the IEEE 33-node test system.
关键词
无功优化/数据驱动/改进卷积神经网络/配电网/注意力机制Key words
reactive power optimization/data-driven/enhanced convolutional neural network/distribution network/self-attention mechanism引用本文复制引用
基金项目
天津市科技计划项目(22JCZDJC00710)
出版年
2024
NETL
NSTL
万方数据