2024-03-29T01:46:46Zhttps://ipsj.ixsq.nii.ac.jp/ej/?action=repository_oaipmhoai:ipsj.ixsq.nii.ac.jp:001780562023-04-27T10:00:04Z01164:02036:09049:09119
電力需要データに特化した特徴抽出方法と背景情報推定方法の提案Data Mining and Private Information Detection Method using Power Demandjpnディペンダビリティhttp://id.nii.ac.jp/1001/00177968/Technical Reporthttps://ipsj.ixsq.nii.ac.jp/ej/?action=repository_action_common_download&item_id=178056&item_no=1&attribute_id=1&file_no=1Copyright (c) 2017 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG.慶應義塾大学理工学部慶應義塾大学理工学部慶應義塾大学理工学部吉田, 将大今西, 智哉西, 宏章今日,スマートメータの普及に伴い,建物の詳細な消費電力データが取得可能になった.今後スマートメータが更に普及すると,取得環境の異なる膨大なデータが蓄積されるため, これらを有効に活用するにはデータの取得環境に対して汎用的,かつ低計算コストな特徴抽出方法が必要である.そこで本報告では,電力データに特化した特徴量抽出手法を提案する.更に,抽出した特徴量から居住者人数と延べ床面積を推定し,提案手法をその推定精度,計算コスト,汎用性の面から提案手法を評価した.結果,既存手法を用いた場合と比べて推定精度は向上した. また,取得環境の異なる電力データに対して提案手法を適応する事で,提案手法の汎用性を確認した.Recently, detailed power demand information of the building is being aggregated, accompanied by the spread of smart meters. Continuous spread of smart meter will aggregate large amount of data from area in several conditions. In order to utilize these data effectively, feature extraction method that is generic, and law computational cost is required. Therefore, the feature value extraction method for power demand information is proposed in this report. Furthermore, we estimate family structure and floor space from the extracted feature value, and evaluate the proposal method through its estimation accuracy, computational cost, and generality. As the result, we have managed to improve the estimation accuracy compared to the existing method. We have also verified its' generality by adopting the proposal method to data aggregated from different conditions.AA11451459研究報告システムとLSIの設計技術(SLDM)2017-SLDM-17949162017-03-022188-86392017-03-01