摘要
为解决电力系统中因负荷数据混沌特性强、噪声影响多导致多分段短期负荷预测误判率高的问题,引入改进粒子群算法,针对新型电力系统,提出了一种全新的短期负荷自动预测方法。首先,确定电力负荷数据来源,采集负荷数据;其次,对采集的负荷数据进行水平处理,去除噪声数据,修正电力负荷数据中的错误,为后续短期负荷预测提供有力的数据支持;最后,利用改进粒子群算法将短期负荷预测问题转化为多目标优化问题,通过迭代和优化寻找最优解,得到未来短期的负荷预测结果。结果表明,应用该方法后,各时刻的电力负荷预测平均绝对百分比误差最大不超过0.5%,负荷预测值与实际值更接近,预测误判率显著提升。
To solve the problem of high misjudgment rate in multi segment short-term load forecasting caused by strong chaotic characteristics and noise impact of load data in the power system,an improved particle swarm algorithm is introduced to propose a new automatic short-term load forecasting method for the new power system.Firstly,determine the source of power load data and collect load data.Secondly,the collected load data is horizontally processed to remove noise data and correct errors in power load data,providing strong data support for subsequent short-term load forecasting.Finally,using the improved particle swarm algorithm,the short-term load forecasting problem is transformed into a multi-objective optimization problem.Through iteration and optimization,the optimal solution is found to obtain the future short-term load forecasting results.The results show that after applying this method,the average absolute percentage error of power load prediction at each time point does not exceed 0.5%,and the predicted load value is closer to the actual value.The prediction misjudgment rate has been significantly improved.
作者
张立军
王春雷
ZHANG Lijun;WANG Chunlei(State Grid Beijing Tongzhou Power Supply Company,Beijing 101100,China)
出处
《自动化应用》
2024年第10期56-58,共3页
Automation Application
关键词
改进粒子群算法
新型电力系统
短期负荷
improved particle swarm algorithm
new power system
short-term load
