Design of a Performance-Driven Discrete-Time Predictive Functional Control

カテゴリ: 部門大会

論文No: SS2-6

グループ名: 【C】2024年電気学会電子・情報・システム部門大会

発行日: 2024/08/28

タイトル(英語): Design of a Performance-Driven Discrete-Time Predictive Functional Control

著者名: 齊藤 諒介（広島大学）,山本 透（広島大学）

著者名(英語): Ryosuke Saito (Hiroshima University),Toru Yamamoto (Hiroshima University)

キーワード: モデル予測制御|制御性能評価|離散時間システム離散時間システム|Predictive Functional Control|Control performance evaluation|Discrete time system

要約(日本語): Model predictive control is a control method that utilizes a controller's internal target model to predict the future behavior of the control target and determine the appropriate manipulated variable (control input). This paper focuses on Predictive Functional Control (PFC), which is one of the least computationally intensive types of model predictive control. As previously mentioned, since the operation relies on the model to be controlled, the control performance is greatly influenced by the accuracy of the designed model. Therefore, this paper introduces a method that incorporates a mechanism for restructuring the control target model in response to changes in the characteristics of the control target and demonstrates its effectiveness through numerical examples.