The vibration signal of gantry crane gearboxes is often contaminated with noise, making it difficult to extract and identify fault characteristics. Additionally, fault signals in practical operating environments lack label information, making neural network-based fault diagnostic methods impractical for real-world fault diagnosis. To address this issue, this paper designs a gantry crane gearbox condition data acquisition system based on IoT and proposes a gearbox fault diagnosis method using variational mode decomposition (VMD) signal decomposition and an improved Particle Swarm Optimization-Grey Wolf Optimization Algorithm (PSOGWO) under Maximum Correlated Kurtosis Deconvolution (MCKD).This method first applies variational mode decomposition to the original signal to obtain a series of intrinsic mode functions (IMFs), selecting the IMF with the maximum kurtosis as the optimal component. Using kurtosis as the objective function, PSOGWO is employed to find the optimal parameter combination for MCKD on the optimal component. The MCKD with the best parameter combination is then used to denoise the optimal component, highlighting fault impact components. Finally, classification recognition is applied to the denoised signal to complete fault diagnosis.

--

Internet of things,Fault diagnosis,Variational mode decomposition,PSOGWO algorithm,Maximum correlated kurtosis deconvolution

• Other Engineering and Technology Sciences - Engineering and Technology
• Electrical Engineering, Electronic Engineering, Information Engineering - Engineering and Technology