摘要
为建立起城市用水量与其影响因素间的预测模型,以预测的城市用水量趋于合理,针对城市时用水量的特点及影响因素,在考虑充分利用各因素历史观测数据的基础上,利用BP神经网络建立了城市时用水量的时间序列预测与解释性预测组合模型,并对南京市的时用水量进行了预测.预测结果与实际情况具有很好的一致性,预测误差小,能满足供水系统调度的实际需要.可见,本预测组合模型是合理的,为城市时用水量预测提供了一种可行方法.
To establish the relationship between urban water consumption and its affecting factors, according to the characteristics of urban hourly water consumption, a combined urban hourly water consumption prediction model has been developed based on BP neural network. The model has been performed on the historical obser- vation data of Nanjing city. The results show that the forecast error of the developed model is small and meets the practical requirement of water-supply dispatch system. It can be concluded that this model is reasonable and feasible for the forecast of urban hourly water consumption.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2009年第6期197-200,共4页
Journal of Harbin Institute of Technology
基金
国家自然科学基金重点资助项目(50638020)
关键词
时用水量
组合模型
预测
BP神经网络
urban hourly water consumption
combined model
forecasting
BP neural network
