in this thesis, two kinds of load density forecast methods are introduced, classified-divisional load density method and classification method based on artificial neural network and fuzzy theory 本文介紹了兩種負荷密度預(yù)測方法:分類分區(qū)預(yù)測法及基于神經(jīng)網(wǎng)絡(luò)和模糊算法的聚類分析方法。
at present, there are commonly there kinds of method for spatial load forecasting . the first is load density method, which can give the quantity of the unit according to density of classified load and the constitution of sub-districts " area 目前,國內(nèi)外常用的空間負荷預(yù)測方法主要有以下三類:第一、負荷密度法:該方法是通過預(yù)測分類負荷的負荷密度,根據(jù)小區(qū)面積構(gòu)成,計算出各個小區(qū)的負荷值。
allowing for the factors affecting the load forecast in the practical engineering project, i compare the characteristic and their applicability of the above spatial load forecasting methods, and then get the following results . the classified-divisional load density method has tendency to be affected by human factors, but the classification method based on artificial neural network and fuzzy theory can make up with this fault 考慮到實際工程項目中影響負荷預(yù)測的各種因素,我對上述幾種空間負荷預(yù)測方法的特點和適用范圍進行了分析比較,認為分類分區(qū)預(yù)測方法受人為因素影響太多,特別是對于缺乏歷史數(shù)據(jù)的新區(qū)進行負荷密度預(yù)測,往往不能得到滿意的結(jié)果,而基于神經(jīng)網(wǎng)絡(luò)和模糊算法的聚類分析方法可以彌補這一缺陷。