Premium
Data‐driven optimal predictive control of seismic induced vibrations in frame structures
Author(s) -
Di Girolamo G. D.,
Smarra F.,
Gattulli V.,
Potenza F.,
Graziosi F.,
D'Innocenzo A.
Publication year - 2020
Publication title -
structural control and health monitoring
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.587
H-Index - 62
eISSN - 1545-2263
pISSN - 1545-2255
DOI - 10.1002/stc.2514
Subject(s) - benchmark (surveying) , frame (networking) , vibration control , process (computing) , control theory (sociology) , computer science , model predictive control , vibration , space frame , state space , state space representation , state (computer science) , control (management) , engineering , control engineering , artificial intelligence , algorithm , structural engineering , mathematics , geology , telecommunications , statistics , physics , geodesy , quantum mechanics , operating system
Summary Complex civil engineering structural systems are prone to seismic induced vibrations during which their inherent dynamical features are displayed often causing discrepancies with the prediction of classical idealized models. In the paper, effective control of such systems is pursued by a novel data‐driven modeling and a control methodology based on a technique from machine learning: Regression Trees. A training process to create a state–space model based on partitioning the dataset is illustrated. State‐ and output‐feedback are derived using the recursively identified model in order to reach a suitable performance event for an unknown structure. A benchmark frame structure has been used to demonstrate the effectiveness of the entire procedure.