Loading Model And Control Parameter Optimization Of A Rake Suction Dredger

Dredging operation plays a very important role in the development of economy,and the advantages of the rake suction dredger make it an important type of dredging operation.The loading process is an indispensable part of it.The traditional dredging operation mainly relies on the construction experience of operators,which leads to lower construction efficiency.With the rapid development of artificial intelligence,it has become more and more urgent and necessary to improve the efficiency of dredgers by using artificial intelligence.On the basis of fully grasping the operation principle of the rake suction dredger,the collected construction data are analyzed and processed.By constructing the estimation model of the loading stage of the rake suction dredger,the data which is not easy to be measured directly are estimated,and the construction efficiency is improved by controling construction parameters for optimization.This paper aims at the loading model of the rake suction dredger,the estimation of soil particle size,production and loss of net dry soil tons,and the optimization of multi-objective dredging parameters are studied.The main work is in the following aspects:(1)Research and treatment of loading model: aiming at the mass balance equation of loading model,the productivity and output of net dry soil tons,the loss rate and amount of dry soil tons were studied,the estimation model of loading is built.(2)Data preprocessing: search for wrong data and delete it through DBSCAN clustering algorithm.Through the missing value processing,the deleted data will be filled.Finally,wavelet denoising is used to reduce noise interference and make the original data smoother.(3)Modeling of loading model and data estimation.First is the estimation of the loading mass by Elman neural network optimized by particle swarm optimization.The concept of safe dredging is carried out by this method as an alternative from the safety perspective.The second is to model and estimate the soil particle size,the productivity and output of net dry soil tons,the loss rate and amount of dry soil tons through particle filter of electromagnetic-like mechanism.(4)Optimization of multi-objective dredging parameters: NSGA2 non-dominant sorting genetic algorithm is used to optimize the dredging parameters of rake suction dredger's overall average production efficiency of net dry soil tons and overflow loss of dry soil tons,the optimized construction parameters were obtained,control dredger to specified construction parameters.And compared with the initial working conditions,the production efficiency and dry soil ton storage rate were improved.