Digital Twin for rotating machinery fault diagnosis in smart manufacturing

TLDR

Digital Twin technology enables cloud‑enabled, digitally driven manufacturing, yet its application to rotating machinery is challenged by nonlinear dynamics and uncertainty in degradation processes. This paper introduces a Digital Twin reference model specifically for diagnosing faults in rotating machinery. The model incorporates a parameter‑sensitivity‑based updating scheme and is validated with experimental data from a rotor system exhibiting an unbalance fault, assessing the model’s ability to quantify and localize the fault. Results demonstrate that the Digital Twin rotor model accurately diagnoses unbalance and supports adaptive degradation analysis.

Abstract

With significant advancement in information technologies, Digital Twin has gained increasing attention as it offers an enabling tool to realise digitally-driven, cloud-enabled manufacturing. Given the nonlinear dynamics and uncertainty involved during the process of machinery degradation, proper design and adaptability of a Digital Twin model remain a challenge. This paper presents a Digital Twin reference model for rotating machinery fault diagnosis. The requirements for constructing the Digital Twin model are discussed, and a model updating scheme based on parameter sensitivity analysis is proposed to enhance the model adaptability. Experimental data are collected from a rotor system that emulates an unbalance fault and its progression. The data are then input to a Digital Twin model of the rotor system to investigate its ability of unbalance quantification and localisation for fault diagnosis. The results show that the constructed Digital Twin rotor model enables accurate diagnosis and adaptive degradation analysis.