Fault-tolerant thrust allocation analysis using metaheuristic optimization algorithms

• Published in: Ocean engineering Vol. 299; p. 117269
• Main Authors: Xuebin, Li, Luchun, Yang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2024
• Subjects: Capuchin search algorithm (CapSA); Fault-tolerant; Lévy flight; Metaheuristic optimization; Thrust allocation; Capuchin search algorithm (CapSA); Metaheuristic optimization; Thrust allocation; Fault-tolerant; Lévy flight
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2024.117269

A comprehensive flowchart featuring three optimization models is introduced to ensure compliance with the thrust allocation principle during faults and to improve comprehension of the widely utilized four-part aggregated objective function model in marine control allocation. The evaluation of the system's allocation capability is conducted within Model A. A decision is then made to assess whether the external force falls within the allocation capacity, leading to the creation of two branches: Model B and Model C, respectively. To improve the global searching capability, Lévy flight improvement is integrated into a new metaheuristic optimization method, the Capuchin Search Algorithm (CapSA), to form L-CapSA, addressing the three optimization problems. An example of a platform supply ship is selected to illustrate the flowchart. The performance of L-CapSA in solving fault-tolerant thrust allocation is investigated and compared with other metaheuristic algorithms. Additionally, the application of the four-part aggregated objective function model is studied in detail. The simulation results show that the proposed flowchart achieves an 11.25% reduction in energy consumption and is comparable to previous methods in terms of allocation accuracy throughout the entire process. These findings validate the effectiveness of the flowchart and the application of L-CapSA in improving energy efficiency. •A general flowchart composed of capability judgment and analysis is proposed for thrust re-allocation.•Lévy flight is adopted to improve the global searching capability of the original CapSA.•Performance of L-CapSA and other metaheuristic algorithms are examined for thrust re-allocation.•Comparisons with the four-part aggregated objective function model are provided.