计及可再生能源时空特征量灵敏度分析的储能电站规划研究

It is the basic problem to rationally determine the spatial and temporal characteristics, acquisition size and the overall span, that need to be solved in the design and planning of energy renewable storage power station. This project focused on the key technology of energy storage station under large data scenarios for renewable energy storage through the statistics analysis of historical data from wind/photovoltaic station. The acquisition size and the overall span are extracted by data mining methods to revealing the relationship among the data acquisition size, overall span and energy storage system capacity. On the basic of science and reasonable selection for data acquisition size and the overall span, a comprehensive analysis of the relationship among battery energy storage capacity, the battery charging and discharging depth and the service life of the battery energy storage system. The data mining algorithm is used to improve the prediction accuracy of the SOC in the energy storage system. On the premise of ensuring the state of charge (SOC) and safety operation of the energy storage system, according to different types of energy storage characteristic of system, based on the economic perspective and maximize returns, game theory is used to solving the key technology of the planning for multi types battery energy storage station and the optimizing capacity allocation of hybrid energy storage system to ensure the stability and economic operation of renewable energy storage system as to provide a theoretical basis to further enhance the ability to access the grid.

合理确定可再生能源时空特征量采集粒度和样本总体跨度是储能电站设计与规划需要解决的基本问题。本项目围绕大数据应用场景下用于可再生能源的储能电站规划关键技术进行研究，通过对可再生能源历史数据进行统计分析，研究提取数据采集粒度和总体跨度的数据挖掘方法，揭示数据采集粒度和总体跨度与储能系统容量间的关系。在科学、合理选择数据采集粒度和跨度基础上，全面分析电池储能容量、充放电深度与电池使用寿命之间的关联性规律，研究提高电池储能系统SOC预测精度的数据挖掘方法；在保证储能系统荷电状态SOC和大容量储能系统安全运行基础上，根据不同类型储能系统所具有的特点，基于经济视角和收益最大化，研究博弈论挖掘方法在解决多类型电池储能电站规划以及混合储能系统容量优化配置的关键技术，以保证含有可再生能源的储能系统稳定、经济运行，为进一步提升可再生能源接入电网的能力提供理论依据。

合理确定可再生能源时空特征量采集粒度和样本总体跨度是储能电站设计与规划需要解决的基本问题。本项目围绕大数据应用场景下用于可再生能源的储能电站规划关键技术进行研究，通过对可再生能源历史数据进行统计分析，建立用于可再生能源多尺度时空特征量灵敏度分析理论体系，为合理、科学的选择储能容量规划所需的数据样本粒度和跨度提供依据。该项目在国内外首次提出了通过构造光伏输出功率持续变动三角形确定最佳粒度的挖掘方法，建立了基于多目标优化的光伏电站采集粒度标定模型。通过对样本信息熵的分析，提出了基于信息熵理论的光伏电站跨度标定方法，分析了样本数据采集粒度和跨度与储能系统容量的关系。搭建了储能系统仿真实验平台，完成了储能系统运行特性的数据挖掘，实现了储能系统SOC的精确估计。针对风电/光伏功率平滑应用场景，提出了基于博弈论的含可再生能源多类型储能电站规划及其协调优化控制方法，实现了风/光储联合发电系统经济性评估，保障了风电/光伏-储能联合发电系统经济、高效运行。