COMBINATION FORECASTING MODEL FOR WATER QUALITY BASED ON OPTIMAL WEIGHTS

摘要：

　　The combination forecasting model was developed based on optimal weights to forecast water quality of the river recharged with reclaimed water.First three different forecasting models which were gray forecasting model, trend extrapolation method, double smoothing method were developed respectively.Furthermore, minimizing the sum of squared prediction error was set as the goal, and the problem of weighting the predictive values was transferred into an optimization problem.The weights of predictive values in three methods could be obtained by solving the optimization problem.Then their results were combined based on optimal weighted.Thus the combination forecasting values were obtained.Water quality of Miyun Reservoir was forecasted based on the combination forecasting model, compared with the simple models.The results showed that the combination forecasting model could balance the bias of all single models.It had better applicability than the single model, and the accuracy was also higher.

关键词： Water Quality Combination Forecasting Model Gray Forecasting Model Trend Extrapolation Model Exponential Smoothing Model

作者: Liu Dongjun Zou Zhihong

作者单位: School of Economics and Management, Beihang University, Beijing 100191, China

会议类型: 国际会议

会议名称: The 11th International Conference on Industrial Management(第十一届工业管理国际会议)

会议地点: 日本

会议语种:英文

页码: 358-362

在线出版日期: 2012-08-29（万方平台首次上网日期，不代表论文的发表时间）