基于大数据的分布式光伏接入配电网影响分析与功率预测研究

为提高分布式光伏发电功率预测的精度,满足电网调度和规划的高精度要求,本文利用光伏运行、电能量采集、电网调度等业务系统的海量数据,利用大数据分析方法研究大量分布式光伏接入对配电网负荷特性的影响,并提出基于气象相似日和粒子群算法优化BP神经网络的光伏电站功率预测方法。通过分析光伏发电功率随天气类型、温度、光照强度等气象因素变化规律,运用模糊聚类算法计算确定待预测日的气象相似日序列,选取气象相似日历史数据作为BP神经网络预测模型的输入变量,并采用粒子群算法方法优化BP神经网络的初始值,最终输出分布式光伏各时段发电功率的预测值。实验结果表明该方法可有效提高光伏电站功率预测模型的收敛能力和学习能力,具有较高的预测精度。

Research on influence analysis and power prediction of distributed photovoltaic access to distribution network based on big data

In order to improve the accuracy of distributed photovoltaic power forecasting,and meet the high-precision requirements of dispatching and planning,this paper uses massive data from business systems such as photovoltaic operation,electric energy collection,and grid dispatching.Using big data analysis methods to study the impact of large numbers of distributed PV connections on the load characteristics of the distribution network.The method of PV power prediction based on meteorological similar day and particle swarm algorithm optimized BP neural network is also proposed.By analyzing the variation pattern of PV power generation with weather type,temperature,light intensity and other meteorological factors,using the fuzzy clustering algorithm to calculate and determine the sequence of meteorological similar days for the day to be predicted,selecting the historical data of meteorological similar days as the input variables of the BP neural network prediction model,and using the particle swarm algorithm method to optimize the initial values of the BP neural network.The final output is the predicted power of distributed PV for each time period.The experimental results show that the method can effectively improve the convergence ability and learning ability of the PV power prediction model,and has high prediction accuracy.