乗り心地と省エネルギーを考慮したマルコフ決定過程による列車走行曲線最適化

カテゴリ: 部門大会

論文No: OS5-12

グループ名: 【C】平成25年電気学会電子・情報・システム部門大会講演論文集

発行日: 2013/09/04

タイトル(英語): Markov decision process-based run curve optimization for energy saving and comfort

著者名: 藤井 彩恵(三菱電機),吉本 剛生(三菱電機),上田 健詞(三菱電機),高橋 理(三菱電機),Daniel Nikovski(Mitsubishi Electric Research Laboratory)

著者名(英語): Sae Fujii(Mitsubishi Electric Corp.),Koki Yoshimoto(Mitsubishi Electric Corp.),Kenji Ueda(Mitsubishi Electric Corp.),Satoru Takahashi(Mitsubishi Electric Corp.),Daniel Nikovski(Mitsubishi Electric Research Laboratory)

キーワード: 列車走行曲線最適化|マルコフ決定過程|動的計画法|省エネルギー|乗り心地|Run curve optimization|Markov decision process|Dynamic programming|Energy saving|Comfort

要約(日本語): This paper addresses train run curve optimization problem considering both energy saving and comfort. In our previous work, we proposed a Markov decision process-based run curve optimization (MDP-RCO). In this paper, we propose two extended methods of MDP-RCO, MDP-RCO-JERK and MDP-RCO-COUNT, in order to reduce run mode fluctuation for comfort. MDP-RCO-JERK uses an evaluation function defined by the weighted sum of time, energy consumption and jerk. Also, MDP-RCO-COUNT prohibits the state transition more than some times in a small section. We have confirmed that MDP-RCO-COUNT would derive run curves reducing the mode fluctuation almost without losing optimality.

PDFファイルサイズ: 6,709 Kバイト