Mechanism Analysis And Research Of Loading Process For Trailing Suction Hopper Dredger

Dredgers are vital for the construction of island and harbor.Under the background of big data and information age,intelligent dredging is the future direction of global dredging industry.Traditional trailing suction hopper dredger(TSHD)operation relies more on the operator to observe dredging instruments and operating conditions to adjust the control methods to maintain the dredging efficiency.This mainly requires the full construction experience of the operator.Intelligent dredging system combines the dredging mechanism and mathematical models to collect,analyze and overcome the working state of the subsystems,which contribute to predict the dredging production,perform rolling optimization and self-learning,and provide the most suitable dredging strategy.As a consequence,it is extremely significance to carry out the intelligent dredging research to improve the dredging efficiency.This thesis relies on the foundation project of CCCC National Engineering Research Center of Dredging Technology and Equipment.In order to overcome the problem of loading optimization of TSHD,this essay does research on the field of sedimentation mechanism modeling,loading parameters estimation,dredging soil estimator design as well as real-time prediction of mud sedimentation process.The research results provide references for dredging optimization of TSHD and have been applied in intelligent dredging system.The main task of this thesis is as follow:(1)Analysis of sedimentation mechanism and establishment of model.Aiming at the loading characteristics of TSHD,the time-varying soil grain diameter is proposed and the dynamical sedimentation model which related to the sedimentation process is constructed.The framework of soil grain diameter estimator is designed and the random walk strategy is used to simulate the time-varying soil properties dynamically.(2)Research on the influence factors of loading production.The important state variables of loading process are established,and the evaluation index of optimization is given.The sensitivity analysis is performed on the hopper size,the volume and density of initial mud in hopper,the concentration of incoming mixture and the soil grain diameter.Simulations show the internal relationship between the effects of soil grain diameter on dredging production and further validate the necessity of prediction of state of dredged soil.(3)Calibration and validation test of real ship loading model.For the difference between the structure of real hopper and ideal model,the construction-data-based calibration is conducted to correct the ideal hopper parameters through the method of numerical curve fitting.(4)Research on loading estimation methods.For the preliminary calibration of soil parameters under different wording conditions,an off-line estimation method based on Pattern Search is given to provide the initial values of the loading estimation.Aiming at the dynamical loading process with soil properties changing obviously,an on-line estimation method based on Bootstrap Particle Filter(BPF)and continuous Feedback Particle Filter(FPF)is presented.Simulations show that FPF the estimate obtained by FPF is more accuracy than BPF and it converges faster.Considering the high frequency of sampling requirement of FPF,the log-homotopy transformation is introduced and the Continuous-discrete Feedback Particle Filter(CD-FPF)is constructed.Results give out that CD-FPF relaxes the sampling interval time limit and enhances the adaptability of FPF,and prove that CD-FPF has the value of engineering application.