Study On Dynamic Modeling Of Full-size Water-Lubricated Bearing Under Bias Load Based On Test Information

With the development of large-scale ships,water-lubricated stern bearings(WSBs)are widely used in marine ships under the working condition of low speed and heavy load.WSBs are susceptible to wave impact,variation of loading conditions and other external factors.It leads to a very complex state for the shaft space,which will generate a serious impact on stern bearings under operating conditions.Therefore,the research on the lubrication state and dynamic characteristics of WSBs in service is of great significance.The WSBs are taken as the research object in this paper.The dynamic characteristics tests for the full-size WSB with bias load are carried out,which can obtain the qualitative characteristics and quantitative data of WSB.Accordingly,the flow-solid coupling theory of bearing and machine learning algorithm are used to establish the lubrication model of WSB.The dynamics characteristics research is carried out to reveal the lubrication characteristics and dynamics mechanism of fullsize WSB.The research results contribute an important theoretical basis for the design of WSB structure and have a propelling effect on the development of bearing condition monitoring technology.The main research content and results of this paper are as follows:(1)The experiment of full-size WSB dynamic characteristics under bias load.The effect of hull deformation on the bearing is simulated by the elevation adjustment device of the bearing.The multi-sectional circumferential film pressure and the axis orbit are collected to study the bearing dynamic characteristics under different hull deformation degrees and speeds.The results show that the journal of the WSB represents obvious bidirectional inclination under bias load.The distribution of bearing film pressure shows multiple peaks,which represent the different lubrication subregions and show the distributed dynamics characteristics.The different lubrication sub-regions have different take-off speeds(2)Study on the lubrication characteristics of WSBs considering the bidirectional shaft inclination under bias load.Based on qualitative information obtained from experiments: bidirectional deflection of the journal and elastohydrodynamic lubrication effect,the film thickness equation considering bidirectional journal inclination is derived.The elastohydrodynamic lubrication model of WSB considering bidirectional shaft inclination is established to investigate the effects of horizontal and vertical shaft inclination,eccentricity ratio and bearing structure parameters on the bearing lubrication performance.The results show that the horizontal inclination of the shaft has a greater influence on the lubrication performance of the bearing under the condition of low eccentricity ratio.The variation of the bearing structure parameters significant affect the effect of the bidirectional shaft inclination on the water film of the bearing.(3)Study on the dynamic characteristics of WSB considering bidirectional shaft inclination under bias load.Based on the elastohydrodynamic lubrication model of WSB cosnidering bidirectional shaft inclination,a 32-parameter dynamic characteristic model is derived using the differential solution method to reveal the influence of the bearing vertical shaft inclination horizontal shaft inclination,eccentricity ratio and bearing structure parameters on the bearing dynamic characteristic parameters.The results show that the variation of shaft inclination angle has a great influence on the bending dynamic characteristic of bearing.The magnitude order of bending dynamic characteristic coefficients and radial dynamic characteristics coefficient is always adjacent to or the same order,while the cross dynamic characteristic coefficient is about4 orders of magnitude less than the former two.(4)Research on the lubrication modelling method of WSB based on machine learning algorithm.The quantitative information of experiment(film pressure and axis orbit)is used as the data set.The BP neural network algorithm is adopted to train the bearing lubrication model.The distributed lubrication characteristic of WSB is studied under different working conditions.The results show that the bearing lubrication model established based on BP neural network has good prediction ability and has certain extrapolation ability to the rotational speed under the fixed bearing elevation.With the increase of rotational speed,it prompts the region in both ends of the bearing to transform from the mixed lubrication state to the elastohydrodynamic lubrication state.