Maximum Deformation Prediction of the Girder of Gantry Crane Based on the Grey Wolf Optimizer-Stacking Ensemble Learning Surrogate Model

Maximum deformation of the girder of gantry crane can indicate the health status of the girder. Due to the movable loads, the position of the maximum deformation of the girder is changeable, so that measuring maximum deformation is difficult with limited sensors. Numerical simulation is challenging to achieve real-time monitoring owing to high computational complexity and time-consuming computation. To address the aforementioned problems, this paper proposes a stacking ensemble learning surrogate model (SELSM) to predict the maximum deformation of the girder of gantry crane. Moreover, the grey wolf optimizer (GWO) is used to optimize the parameters of the SELSM (GWO-SELSM). The results of experiment show that the GWO-SELSM outperforms the single surrogate models and existing combination of surrogate models for mean square error (MSE) and R-square (R
²
). The proposed model can be used to accurately predict the maximum deformation of the girder of gantry crane online.