Analysis of intelligent hybrid fuzzy logic system for multiple crack diagnosis of rotor shaft rotating in fluid environment using vibration signature

• Published in: Sadhana (Bangalore) Vol. 50; no. 3; p. 137
• Main Authors: Yadao, Adik, Mishra, Kundan
• Format: Journal Article
• Language: English
• Published: New Delhi Springer India 02.07.2025; Springer Nature B.V
• Subjects: Algorithms; Amplitudes; Artificial intelligence; Damage; Design; Engineering; Evolutionary algorithms; Evolutionary computation; Finite element analysis; Finite element method; Flow velocity; Fuzzy logic; Fuzzy systems; Genetic algorithms; Hybrid systems; Investigations; Neural networks; Real time; Resonant frequencies; Robust control; Rotating fluids; Rotating shafts; Rotation; Rotors; Structural members; Wavelet transforms; Rotor dynamics; crack depth; differential evolution algorithm; crack location; fuzzy logic; vibration signature; finite element analysis
• ISSN: 0973-7677; 0256-2499; 0973-7677
• DOI: 10.1007/s12046-025-02801-2

From last three decades, fuzzy logic has been applied in various scientific domains. It has become a reliable tool for solving the large and complex problems that would otherwise be difficult to address using traditional methods. However, it does require some human knowledge and expertise. The current study attempts to provide a hybrid optimisation technique for the involuntary generation of fuzzy knowledge-based rules utilising an evolutionary algorithm. This hybridization procedure has been employed in the field of crack recognition and position of damage in damaged structural elements. In this article, we developed a robust fault diagnostic tool based on fuzzy logic (FL) and the Differential Evolution algorithm (DEA). Theoretical and Finite element analysis has been done to model the crack and investigate the impact of the existence of multiple transverse cracks on the alterations of vibrational signature (natural frequencies and amplitude) of a cantilever rotor shaft rotating in fluid environment. The inputs of hybrid Differential Evolution algorithm and fuzzy logic (DEA-FL) system are the relative natural frequencies, relative amplitude at direction of x and y axis of crack and fluid viscosity. The relative crack depths and relative crack locations are the output of hybrid system. Experimental study has been carried out to validate the robustness of the controller and validate the outcomes of the suggested hybrid approach. The total average error of the proposed technique compared with experimental values is less than 3%. Therefore, the recommended Hybridized of Differential Evolution algorithm and fuzzy logic (DEA-FL) technique would establish an effective tool for real-time crack detection in rotor.