地区分布式光伏发电量日前预测方法探究  

Exploration of Methods for Day-Ahead Forecasting of Distributed Photovoltaic Power Generation at the Regional Level

作　　者：刘诗 王宇飞 李宇龙 LIU Shi;WANG Yufei;LI Yulong(State Grid Jilin Power Supply Company,Jilin,Jilin 132011,China)

机构地区：[1]国网吉林供电公司,吉林吉林132011

出　　处：《东北电力技术》2025年第4期35-38,43,共5页Northeast Electric Power Technology

摘　　要：为探寻地区电网调度分布式光伏发电量预测的实用方法和提高预测的准确性,提出了一种结合数值天气预报和历史数据统计模型的地区分布式光伏发电量日前预测方法。首先,利用机器人流程自动化(robot process automation, RPA)技术和气象服务平台,自动获取光伏发电数据和天气预报信息。其次,通过分析各县域历史发电数据与天气因素之间的关系,建立预测模型,该模型考虑了辐射温度、湿度和风速等主要影响因素,通过回归分析方法进行建立和验证。最后,仿真分析结果表明,该方法相比传统方法有显著改进,能有效提高预测的准确性和可靠性,自动化的数据收集与处理流程不仅提升了工作效率,还降低了人为错误率。该方法对于电网调度优化发电计划,提高可再生能源利用率具有一定的实用价值。To explore practical methods for forecasting regional distributed photovoltaic(PV)power generation and to improve the accuracy of such predictions,it proposes a day-ahead forecasting method that combines numerical weather prediction and historical statistical models.Firstly,it utilizes robotic process automation(RPA)technology and meteorological service platforms to automatically acquire photovoltaic generation data and weather forecast information.Then by analyzing the relationship between historical generation data and weather factors in various counties,it establishes a forecasting model.The model takes into account key influencing factors such as radiation temperature,humidity,and wind speed,and is developed and validated using regression analysis methods.Finally,the results show that this method can enhance the accuracy and reliability of predictions,with improvements compared to traditional methods.The automated data collection and processing procedures not only increase work efficiency but also reduce the possibility of human errors.The method has practical value to optimize the power generation plan and improves the utilization rate of renewable energy.

关 键 词：分布式光伏发电 日前预测 RPA 数值天气预报 

分 类 号：TM615[电气工程—电力系统及自动化]