A Study On Probabilistic Prediction Methods Of Stochastic Parameter Excited Motions In Ship And Ocean Engineering

Parameter excited motions are common in two basic motions--rigid body motions and elastic vibrations.In the field of ship and marine engineering,the parametric roll of ships and the parametric vibration of marine risers are typical representatives of parameter excited motion in the two basic motions.Because the environmental load and other conditions are random,they usually appear as stochastic parameter excited motions.At present,the research on stochastic parameter excited motions is not sufficient,and there is a lack of accurate and efficient methods to predict this kind of motion and its extremum.Therefore,this paper will take ships’ stochastic parametric rolling motion and marine risers’ stochastic parametric vibration as examples to study the probabilistic prediction method of stochastic parameter excited motions,and establishes more accurate and efficient prediction methods.The prediction methods proposed in this paper have certain universality for the stochastic parameter excited motions in ship and ocean engineering.In the study on probabilistic prediction methods of ship stochastic parametric roll,firstly this paper improves the ship roll model proposed by Dostal and puts forward a more reasonable fitting algorithm.Then,the stability of the solution of the improved ship roll model and the stationarity of roll motion process are verified.Finally,the stochastic average method is improved,and a new calculation method of averaging coefficient for high-order nonlinear term is proposed.Based on the improved model and method,a new probabilistic prediction method of ship stochastic parametric roll is established in this paper.By using this method,the stochastic parametric roll of C11 container ship in longitudinal and oblique waves are predicted in this paper.On this basis,this paper also analyzes the probability characteristics of the peak value of ship stochastic parametric roll.This work is very significant to avoid large angle parametric roll for ships.In the study on probabilistic prediction methods of marine riser stochastic parametric vibration,this paper firstly verifies the stationarity of marine riser stochastic parametric vibration process.Then combines the probability density evolution method(PDEM)proposed by Li Jie et al.with the linear external interpolation method proposed by A.Naess,this paper establishes a new probabilistic prediction method based on improved PDEM.By using this method,the stochastic parametric vibration of a top tensioned riser(TTR)is predicted in this paper;meanwhlie,the dynamic reliability of stochastic parametric vibration,the probability characteristics of vibration response and the factors affecting stochastic parametric vibration of the TTR are studied.This work is very significant to avoid large amplitude parametric resonance of marine risers.In the study on probabilistic prediction methods for the extrema of stochastic parameter excited motions,based on the extremum prediction method of narrow-band Gaussian random process based on Markov chain proposed by A.Naess.This paper firstly improves the calculation method of the correlation between two adjacent peaks to improve the accuracy.Then,this paper extends the A.Naess method to the extremum prediction of non-Gaussian nonnarrowband stationary random processes.So that,a new probabilistic prediction method of general stationary random processes is established.By using this method,the extrema of C11 container ship stocashtic parametric roll and the TTR stocashtic parametric vibration are predicted in this paper.Compared with other methods,it is proved that the new method has excellent performance in calculation accuracy and calculation efficiency,which can provide a new way for studying and analyzing the extrema of stochastic parameter excited motion of ships and offshore structures.This work is very significant to evaluate the dynamic safety of ships and offshore structures under parametric excitations.