Evolutionary algorithm-based design of a fuzzy TBF predictive model and TSK fuzzy anti-sway crane control system

The efficiency of material handling system requires an automation on the different levels of control and supervision to keep availability of the material handling devices for fast, safety and precise transferring materials, as well as to reduce the maintenance cost, which is involved by enhancing the productivity of manufacturing process. In this paper, evolutionary-based algorithm for fuzzy logic-based data-driven predictive model of time between failures (TBF) and adaptive crane control system design is proposed. The heuristic searching strategy combining the arithmetical crossover, uniform and non-uniform mutation and deletion/insertion mutation is developed for optimizing the rules base (RB) and tuning the triangular-shaped membership functions to increase the efficiency and accuracy of a fuzzy rule-based system (FRBS). The evolutionary algorithm (EA) was employed to design a fuzzy predictive model based on the historical data of operational states monitored between the failures of the laboratory scaled overhead traveling crane electronic equipment. The fuzzy predictive model of TBF was implemented in the supervisory system created for supporting decision-making process through forecasting upcoming failure and delivering the user-defined maintenance strategies. The effectiveness of EA was also verified through designing a Takagi–Sugeno–Kang (TSK) fuzzy controller in the anti-sway crane control system.