Disassembly sequence planning and application using simplified discrete gravitational search algorithm for equipment maintenance in hydropower station. (1st December 2022)

• Record Type: Journal Article
• Authors: Wu, Panqi; Wang, Huanhe; Li, Bailin; Fu, Wenlong; Ren, Jie; He, Qiang
• Is Part Of: Expert systems with applications. Volume 208(2022)
• Publication Date: 2022-12-01
• Subjects: Simplified discrete gravitational search algorithm -- Disassembly sequence planning -- Virtual operation instruction -- Equipment maintenance; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.118046
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: A model is established to more reasonably evaluate maintenance process. A new simplified discrete gravitational search algorithm is proposed. Two operators are proposed to balance global search and local search. A method integrating DSP technology and three-dimensional visualization is proposed. Abstract: Disassembly sequence planning (DSP) is a crucial way to optimize equipment maintenance process for hydropower equipment (HE). As a discrete combinatorial optimization problem with complex disassembly precedence constraints, however, the serious combination explosion caused by a large number of HE components will make the solution process difficult. Thus, a simplified discrete gravitational search algorithm (SDGSA) is proposed based on the core idea of the gravitational search algorithm (GSA). In the proposed algorithm, fast feasible solution generator (FFSG) is used to generate the initial population, precedence preservative operator (PPO) is utilized to generate the next generation efficiently, multipoint optimization operator (MOO) is applied to guide a solution to move to the better neighbor solution, and escaping operator (EO) is employed to prevent population falling into local optimum prematurely and improve the ability to find the optimal solution. In this study, the performance comparison experiments are carried out among SDGSA, simplified swarm optimization (SSO) algorithm, genetic algorithm (GA), simplified teaching-learning-based optimization (STLBO), and Team-Based Genetic Algorithm (TBGA). The results of the three maintenance tasks with different complexity in the flat plate water seal (FPWS) shows that in the solution of case 1-case 3, the proportion of the optimal sequence found by the SDGSA is 13.3 %, 70 %, and 70 % higher than comparison algorithms, respectively. The convergence speed and optimization ability are also better than other algorithms. Finally, the proposed method has been successfully applied to the automatic generation of virtual operation instruction (VOI) for equipment maintenance.