基于自助法的中国水电能资源开发利用Logistic预测  

Logistic Prediction Model of Hydroelectric Power Resources Utilization in China Based on Bootstrap Method

作　　者：汪哲荪[1] 金菊良[1] 魏一鸣[2] 周玉良[1] 杨希凡[1] 

机构地区：[1]合肥工业大学土木与水利工程学院,安徽合肥230009 [2]北京理工大学管理与经济学院,北京100081

出　　处：《水电能源科学》2010年第10期151-153,共3页Water Resources and Power

基　　金：国家自然科学基金资助项目(70941032,70733005)

摘　　要：基于近20 a中国水电装机容量序列,采用加速遗传算法构建了Logistic预测模型,并运用自助法对Logistic预测模型的误差进行重抽样,提出了基于自助法的中国水电能资源开发利用Logistic预测模型(BLP-HPU),预测了2015、2020年中国水电能资源开发利用预测值的置信区间。实例结果表明,该模型物理概念清晰、计算简便、预测精度高,在风能等可再生能源资源开发利用预测中具有推广应用价值。Based on the time series of hydropower installed capacity in China in the past years from 1990 to 2009,a logistic prediction model of hydroelectric power resources utilization is established by using accelerating genetic algorithm.Bootstrap method is used to generate the samples with the logistic prediction model errors.And then bootstrap method based logistic prediction model of hydroelectric power resources utilization in China,named BLP-HPU for short,is put forward.The confidence intervals of hydroelectric power resources utilization in China in 2015 and 2020 years are predicted by BLP-HPU.The instance results show that the physical concept of BLP-HPU is clear;its computation process is convenient;it has high prediction precision.Thus,BLP-HPU can be widely applied to predict development and utilization of renewable energy resources such as wind energy.

关 键 词：自助法 中国 电能 资源开发利用 LOGISTIC 预测模型 BOOTSTRAP Method Based China RESOURCES Utilization Power 水电 加速遗传算法 可再生能源 装机容量 置信区间 预测精度 应用价值 概念清晰 重抽样 

分 类 号：TK79[动力工程及工程热物理—流体机械及工程] TV7[水利工程—水利水电工程]