Please use this identifier to cite or link to this item:
Title: Retraceable and online multi-objective active optimal control using belief rule base
Authors: Jiang, Jiang
Chang, Leilei
Zhang, Limao
Xu, Xiaojian
Keywords: Engineering::Civil engineering
Issue Date: 2021
Source: Jiang, J., Chang, L., Zhang, L. & Xu, X. (2021). Retraceable and online multi-objective active optimal control using belief rule base. Knowledge-Based Systems, 233, 107553-.
Project: 04MNP002126C120 
Journal: Knowledge-Based Systems 
Abstract: A new approach that employs the belief rule base (BRB) is proposed in this study with the goal of Retraceable and Online Multi-objective Active (ROMA) optimal control for complex systems, namely ROMA-BRB. Active optimal control means that multiple objectives are controlled by actively identifying and optimizing key factors in the input. The retraceability requirement means that the procedural and final outputs can be traced back to the inputs to provide maximum accountability. The online requirement is met if the procedures are derivable and only adopting deterministic optimization approaches. To meet those requirements, BRB is adopted owing to its derivable procedures as a white box. There are four major steps in the new ROMA-BRB approach. First, multiple BRBs are constructed for multiple outputs. Then, the contribution of each factor in the input made to each output is calculated. Third, key factors are identified by comparing their contributions to multiple outputs. Finally, the identified key factors are optimized for actively controlling multi-objectives by pushing the Pareto frontier in an online manner. A practical tunnel-induced safety control case is studied whose goal is to reduce both the settlement and the building tilt rate (BTR). Case study results validate that both the settlement and BTR are effectively reduced by optimizing only the key operational factors. The validity of the proposed ROMA-BRB approach is also confirmed by comparing the results of pursuing only one objective, as well as using different numbers of key factors and other weight settings.
ISSN: 0950-7051
DOI: 10.1016/j.knosys.2021.107553
Schools: School of Civil and Environmental Engineering 
Rights: © 2021 Elsevier B.V. All rights reserved.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:CEE Journal Articles

Citations 50

Updated on Dec 5, 2023

Page view(s)

Updated on Dec 8, 2023

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.