摘要
为了提高电力负荷预测精度,提出了一种变参数量子粒子群(VPQPSO)算法优化RBF神经网络的短期负荷预测模型(VPQPSO-RBFNN)。首先利用电力负荷的混沌性,对短期负荷时间序列进行相空间重构;然后采用变参数QPSO算法优化RBF神经网络参数对重构后的短期负荷时间序列进行学习,建立短期电力负荷最优预测模型;最后采用对某地区短期电力负荷进行预测。VPQPSO-RBFNN可以准确描述复杂多变的电力负荷变化趋势,提高了电力负荷的预测精度,仿真结果验证了VPQPSO-RBFNN可以用于电力系统负荷预测。
In order to forecast short-term load accurately and quickly,a short load forecasting model is proposed, based on BP neural network optimized by quantum particle swarm optimization algorithm. Firstly,the data of short load are reconstructed by chaotic theory,and then the parameters of BPNN were considered the position vector of quantum particle,the optimal parameters of BPNN are found by quantum particle optimization algorithm to reduce blindness and inefficiency. Lastly, the optimal model for network traffic is built and the performance of mode are tested by short load data. The simulation results show that VPQPSO algorithm has solved the problems of the BP neural network,and the proposed model improved the power load forecasting accuracy.
出处
《电子器件》
CAS
北大核心
2014年第4期782-786,共5页
Chinese Journal of Electron Devices
关键词
电力负荷
RBF神经网络
变参数
量子粒子群算法
相空间重构
power load
forecasting accuracy
quantum particle swarm optimization algorithm
BP neural network
varying parameters
