Research On Low-frequency Periodic Vibration Control Method Of Fluid-conveying Pipeline Based On Integrating Data And Model

Fluid-conveying pipeline system is the ship’s vessel,which is widely distributed in the ship’s power system.However,its low-frequency periodic vibration will threaten the safety performance and lower the acoustic stealth capability of the ship.In order to solve this problem,this dissertation proposes a vibration performance optimization method for fluidconveying pipelines based on integrating data and models.During construction or maintenance,this method can complete the natural and anti-resonant frequency assignment and periodic vibration control by optimizing the stiffness of elastic support.This method only involves the theoretical dynamics model of the fluid-conveying pipeline and the measured frequency response function data or vibration response data of small samples.It does not need to establish the simulation model or spend a lot of test costs to build a large sample space of measured data.The main work of this dissertation is as follows:1)Aiming at the problem of low-frequency periodic vibration of fluid-conveying pipelines,a method based on the integration of data and model for optimizing its vibration performance is proposed.The receptance method is employed to reduce the dimension of the theoretical dynamics model,and the governing equation for vibration performance optimization is established.The measured data are integrated into the governing equation to construct the objective optimization function so as to realize the integration of the "individual" measured data and the "generic" theoretical model.The optimal stiffness of elastic supports is obtained through the multi-objective optimization algorithm,and the "frequency shifting" and "amplitude suppressing" strategies are realized.The proposed method can not only be used to suppress the vibration response of pipelines under simple source excitation but also can be used to control the vibration of pipelines under complex source excitation.2)In order to solve the inaccuracy problem of optimization results caused by the uncertainty of measured data,this dissertation proposes a vibration performance interval optimization method of a fluid-conveying pipeline based on uncertainty analysis.An interval grey model is introduced to quantify the uncertainty,and interval data,which represent the accurate vibration characteristics and contain the uncertainty of data,is constructed.The interval multi-objective optimization algorithm is employed to solve the objective function based on interval data,and the optimal support stiffness with low sensitivity to uncertainty is obtained.This method can ensure the accuracy and reliability of the calculation results.3)Aiming at the disturbance of optimization results caused by manufacturing errors of support stiffness,this dissertation proposes a robustness evaluation method for support stiffness optimization schemes.This method can be as a post-processing program introduced into the interval optimization method to evaluate the robustness of the support stiffness optimization scheme in obtained Pareto optimal solution set and select the robustoptimal stiffness optimization scheme as the physical implementation scheme of vibration performance optimization.This method can reduce the effect of manufacturing error of stiffness optimization scheme on vibration performance optimization results of a fluidconveying pipeline.Based on the common U-shaped fluid-conveying pipeline model in engineering,the effectiveness of the proposed method is demonstrated from the perspective of numerical verification and experimental study.The results show that the proposed method can effectively solve the problem of resonance and excessive periodic vibration of the fluidconveying pipeline.This dissertation provides a new way to solve the low-frequency periodic vibration problem of the fluid-conveying pipeline,which has vital engineering value and military significance.