Calculation Method of Intermediate Bearing Displacement Value for Multisupported Shafting Based on Neural Network

Abstract

The mutual influence between the bearings of a ship's multisupport shafting makes its installation and alignment very difficult. This article addresses the problem of the calculation of the precise displacement value of each intermediate bearing and proposes a method for fitting the shafting characteristic function by using the GA-BP (genetic algorithm-back propagation) neural network. The neural network uses the intermediate bearing reaction as input to calculate the theoretical height of the bearing, thereby accurately calculating the displacement value. Taking the installation and alignment of a ro-ro ship's propulsion shafting as an application example, a neural network of the ship's shafting is established with training samples based on finite element simulation, and the effect of network training is discussed. The accuracy of the method is verified by a comparative analysis with the measured data of the ship's shafting. The calculation results of this method are used as a guide for the installation and alignment of the ship's shafting and have passed the delivery inspection of the classification society.