水火风光多源发电调度系统大数据平台架构及关键技术

Big Data Platform Architecture and Key Techniques of Power Generation Scheduling for Hydro-thermal-wind-solar Hybrid System

水风光等可再生能源高速发展使得电力调度数据出现爆发式增长,并呈现出多源、异构、高维等大数据典型特点,如何应对电力大数据的集成管理及高效应用是我国电网运行的重大技术挑战之一。为此,围绕水火风光复杂发电调度系统,解析超大规模电站群调度大数据特征及相互关系,设计调度管理功能体系,构建发电调度大数据平台架构,提出满足不同业务场景共性需求的多源数据校验和多平台协同存储技术、面向发电调度分析场景的大数据融合处理技术以及面向发电计划编制场景的大数据分析决策技术,实现电力大数据的采集、存储、分析及知识提取等一体化功能。以云南省调400余座大中型电站为工程背景,应用大数据平台架构和关键技术,构建超大规模多源发电系统调度软件,通过电网月度计划跟踪分析、新能源功率预测偏差确定、梯级水电站群发电调度计划编制3个核心业务场景的典型应用,表明电力大数据技术确实可以为复杂发电调度系统高效和实用化运行提供新的解决途径。

The rapid development of renewable energy sources such as hydropower, wind power and solar power leads to an explosive growth of the power dispatching data, and presents the typical big data features such as multiple sources, heterogeneous and high dimension. How to deal with the integrated management and efficient application of the electric power big data is one of major technology challenges facing power grid operation in this country. Focusing on the complex hydro-thermal-wind-solar power generation scheduling system, this paper analyzed the scheduling big data characteristics and the mutual relations of the ultra-large-scale power station groups. Based on the data analysis, the scheduling management functional system and a power generation scheduling big data platform architecture were constructed. Multi-source data check techniques and multi-platform cooperative storage technology were developed to meet the common needs for different operation scenarios. Meanwhile, big data fusion processing technology and analysis decision technology were proposed to serve for analyzing operation plans and making generation schedule, respectively. These techniques can help to realize the integrated function of electric power big data collection, storage, analysis and knowledge extraction. Taking more than 400 large and medium-sized power stations in Yunnan power grid as the background, an ultra-large-scale multi-source power generation scheduling software system was constructed based on the big data platform architecture and related key technologies. Typical applications in practicalengineering such as tracking monthly generation schedule, determining new energy power forecast error, and scheduling cascaded hydropower plants were presented. The results show that electric power big data technology can indeed provide new solutions for the efficient and practical operation of complex power generation scheduling system.