摘要
针对智能配电网日益增加的复杂性和动态性,提出了基于态势感知的多目标自适应日前调度策略。首先采用卷积神经网络、长短期记忆网络和自编码器构建态势感知模型,使用在线分析处理技术和流数据处理框架进行电力系统运行状态的实时分析;然后采用支持向量回归、长短期记忆网络分别进行短期负荷预测和可再生能源发电预测,并通过异常检测和风险评估构建态势预警层;最后设计目标函数,采用非支配排序遗传算法Ⅱ构建多目标自适应调度模块。仿真测试结果表明,所提调度策略能够显著提高供电系统的经济性、环保性和可靠性,并可以适应各种复杂多变的电力系统运行环境。
In view of the increasing complexity and dynamics of intelligent distribution networks,a multi-objective adaptive day-ahead scheduling strategy based on situation awareness was proposed.Firstly,a situational awareness model was constructed using convolutional neural network,long short-term memory network,and autoencoders.Real time analysis of the operating status of the power system was carried out using on-line analytical processing technique and streaming data processing framework;then,the support vector regression and long short-term memory network were used for short-term load forecasting and renewable energy generation forecasting,respectively,and a situational warning layer was built through anomaly detection and risk assessment;finally,the objective function was designed and a multi-objective adaptive scheduling module was established using the non-dominated sorting genetic algorithm II.The simulation test results show that the proposed scheduling strategy can significantly improve the economics,environmental friendliness and reliability of the power supply system,and can also adapt to various complex and changing power system operating environments.
作者
杨凡
吕勃翰
高卫东
Yang Fan;Lyu Bohan;Gao Weidong(China Southern Power Grid Co.,Ltd.,Guangzhou Guangdong 510623,China)
出处
《电气自动化》
2025年第3期43-45,49,共4页
Electrical Automation
基金
中国南方电网有限责任公司中国南方电网电力调度控制中心部门项目“南网总调调度实时运行业务辅助校核功能模块加装项目”(000500SZ23030004)。
关键词
智能配电网
日前调度
态势感知
长短期记忆网络
在线分析处理
非支配排序遗传算法Ⅱ
smart distribution network
day-ahead scheduling
situational awareness
long short-term memory network
on-line analytical processing
non-dominated sorting genetic algorithm II
