Predictive maintenance for spud condition monitoring in cutter suction dredgers using Gaussian process emulation

Abstract

ABSTRACT The spud system is an integral component of a Cutter suction dredger (CSD) which anchors the hull. In the present study, the spud is modelled as an Euler-Bernoulli beam using finite element analysis. The Soil stiffness is modelled using two spring elements which are identified experimentally. The Dredge hull dynamics is identified using experiments. A numerical model is developed by coupling the identified dredge hull with the spud modelled using finite element analysis. Simulated experimental data are generated by varying the soil stiffness using Monte Carlo simulation. A metamodel for the system is developed based on a Gaussian process emulator (GPE). The maximum response at five different spud locations along the length is input, and the predictor is the soil stiffness. During the validation study, the metamodel accurately predicts the soil stiffness. Therefore, the condition monitoring of the embedment of the spud can be assessed using the accelerometer..