摘要
为实现科学、安全供水,建立精度高、可靠性强的城市日用水量预测模型,分别运用单指数平滑法、灰色预测方法、BP神经网络三种方法,对A市进行城市日用水量预测,并具体分析了各种方法的优缺点及适用范围.通过优化对比分析,当基础数据较完善时,BP神经网络预测模型精度较高,能较好地满足预测要求.
In order to realize a scientific and safe water supply and achieve the accuracy and reliability of the forecasting model of daily urban water consumption,the daily urban water consumption of A city is predicted by using the single exponential smoothing,gray prediction method and BP neural network.The advantages and disadvantages and ranges of application of each method are analyzed in detail.Through comparative analysis by optimizing,when the basic datum is perfect,the BP neural network model has a higher precision and it can preferably meet the forecast requirements.
出处
《青岛理工大学学报》
CAS
2011年第2期74-79,共6页
Journal of Qingdao University of Technology
基金
国家自然科学基金资助项目(50878108)
关键词
城市日用水量
短期预测
单指数平滑法
灰色预测模型
BP神经网络
daily urban water consumption
short-term prediction
single exponential smoo-thing
gray prediction model
BP neural network
