An Accuracy Improvement Method of Simulation on Qualitative and Quantitative Hybrid Model with Uncertain Landmarks

カテゴリ: 研究会(論文単位)

論文No: IS10002

グループ名: 【C】電子・情報・システム部門 情報システム研究会

発行日: 2010/04/22

タイトル(英語): An Accuracy Improvement Method of Simulation on Qualitative and Quantitative Hybrid Model with Uncertain Landmarks

著者名: Samejima Masaki(Osaka University),Akiyoshi Masanori(Osaka University),Komoda Norihisa(Osaka University)

著者名(英語): Samejima Masaki(Osaka University),Akiyoshi Masanori(Osaka University),Komoda Norihisa(Osaka University)

要約(日本語): It became possible to simulate qualitative and quantitative hybrid model by dividing ranges of the model’squantitative variables with “landmarks” and by making pairs of divided ranges and kinds of qualitative values each other. Whenlandmarks are uncertain, it is difficult to set appropriate landmarks to keep the accuracy of simulation. In order to improveaccuracy of simulation, uncertain landmarks are estimated based on initial values and reliable landmarks of the hybrid model.First, candidate ranges including landmarks to be supplied are decided as ranges that can not correspond to any other ranges orqualitative values. Next, landmarks are estimated in the candidate ranges by polynomial interpolation. Through the application ofthe proposed method to a hybrid model, the difference between the simulation result on the hybrid model and the result derivedfrom the quantitative model is decreased by 4.7% in comparison to the general method.

原稿種別: 日本語

PDFファイルサイズ: 954 Kバイト