摘要
气温、气压等天气因素决定了人体舒适度。随着社会经济的快速发展,空调和取暖负载在总用电负荷中的比重日益增加,天气对负荷波动的影响越来越明显。提出了一种考虑风速、降水、气压、气温、湿度等天气数据的径向基(RBF)神经网络日负荷预测模型,用实际负荷数据和天气数据进行训练,将预测结果与BP网络模型得到的结果进行比较,表明了该模型的优越性,也介绍了基于该模型和LabVIEW、Matlab的负荷预测虚拟仪器的前面板和流程图设计过程。结果表明,提出的模型算法简单、精度高、稳定性好,用虚拟仪器进行电力负荷预测具有操作简单、直观、节省费用等优点。所介绍的方法可以用于其它类型负荷预测模型的虚拟实现。
Human body amenity is decided by weather factors such as temperature, atmospheric pressure, etc. In the wake of developments in society and economy, the proportion of load for air conditioning and heating in total power consumption load has increased day by day and the influence of weather upon load fluctuation has become more evident. This paper presents a kind of forecasting model for a daily load by taking weather data such as wend speed, precipitation, atmospheric pressure, temperature and humidity etc. into account and using radial basis function （RBF） neural network. After training network using actual load and weather data, the forecasting results, compared with those of BP network, show the model has clear superiority. The front panel and flow diagram of a virtual instrument for load forecasting based on the model, LabVIEW and MATLAB are also introduced, The results indicate that the model presented is simple, accurate, and steady in algorithm, and the forecasting using virtual instrument is easy , visual and inexpensive in operation, The method stated can be used for reference to the virtual realization of other load forecasting models,
出处
《继电器》
CSCD
北大核心
2007年第2期29-32,44,共5页
Relay
基金
湖北省教育厅自然科学研究计划项目(D200513001)
