エネルギー分野へのAI技術の適用動向

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2023/02/01

タイトル(英語): A Review of Applications of Artificial Intelligence Technologies to Energy Fields

著者名: 福山　良和（明治大学総合数理学部ネットワークデザイン学科）

著者名(英語): Yoshikazu Fukuyama (Department of Network Design, School of Interdisciplinary Mathematical Sciences, Meiji University)

キーワード: エネルギープラント，最適運用，進化計算，Maximum Correntropy Criterion　　energy plant，optimal operation，evolutionary computation，Maximum Correntropy Criterion

要約(英語): Recently, many countries focus on achieving carbon neutral by 2050, and researches on optimal operation of energy plants in factories and large commercial buildings have been conducted for energy saving and reduction of CO2 emissions. Since the optimal operation problem of energy plants can be formulated as a mixed integer nonlinear optimization problem, evolutionary computation techniques, which is one of artificial intelligence (AI) techniques, have been applied. On the other hand, in energy facilities, when AI techniques are required, the Least Squares Method (LSM) has been applied in various problems. One of the main purposes to apply the LSM is to solve the problems only using measured data without experts' knowledge. However, in practical fields, outliers may be included in the measured data because of setting errors of sensors, radio interference, and so on. When LSM is applied with the outliers, inappropriate results may be obtained. Maximum Correntropy Criterion (MCC) is one of the technology trends to tackle the challenge. This paper reviews researches on applications of evolutionary computation techniques to optimal operation of energy plants and presents applications of MCC with AI techniques to energy problems.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.143 No.2 （2023） 特集：エネルギー分野へ適用されたAI・IoT技術

本誌掲載ページ: 104-107 p

原稿種別: 解説／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/143/2/143_104/_article/-char/ja/