Mechanism Analysis And Research Of Digging Process Of Drag Head For Trailing Suction Hopper Dredger

The trailing suction dredger has been widely used in the dredging of waterways,port construction,etc.due to its characteristics of large loading capacity,convenient loading and unloading,flexible maneuverability,adaptability to wind and waves and other harsh construction environments.The traditional construction method of the trailing suction dredger is too dependent on the experience of the construction personnel,resulting in a low level of dredging production.With the widespread application of big data and artificial intelligence technology,intelligent dredging has become the new development direction of the dredging industry;at the same time,the country is building an ecological civilization and taking a sustainable development path.Ecological dredging and green dredging will become the future development trend.In view of this,it is of great significance to carry out intelligent research on dredging technology of trailing suction dredger.This article is supported by the scientific and technological research project of China Communications Construction Co.,Ltd."Ecological Intelligent Dredging Technology Innovation System Construction and Key Technology Research and Application" Researches on modeling of drag head excavation system,preprocessing and feature selection of construction data,prediction of drag head output and evaluation of dredging performance have not been carried out in a way that can truly and objectively reflect the construction status of trailing suction dredgers.The research results provide effective technical support for the improvement of the construction efficiency of the trailing suction dredger,and have a positive significance for improving the intelligence level of the trailing suction dredger in China.The main work of the project is as follows:(1)Mechanism and model analysis of drag head excavation.Firstly,the working principle of the drag head is analyzed,then the drag head production model,drag head cutting model and drag head pressure loss model are established.The wave compensator is used to simulate the impact of wind and waves on the digging process of the drag head,and the mathematical model of the wave compensator is established.(2)Preprocessing and feature selection of construction data.First,pre-process the measured construction data,including: data normalization,standardization,missing value processing,outlier detection,wavelet threshold filter processing.Then the genetic algorithm is used to select the construction parameters that affect the yield of the drag head,and the optimal subset that affects the yield of the drag head is selected.(3)Study on prediction of drag head yield.In view of the problems of low precision and poor timeliness of drag head output prediction,three methods of prediction are put forward: BP neural network,T-S fuzzy neural network and extreme learning machine.Experimental simulation shows that the prediction effect of the extreme learning machine is the best.Then,for the problem that the extreme learning machine randomly initializes the weights and thresholds to weaken the model generalization ability,three optimization schemes are proposed: genetic algorithm(GA),particle swarm optimization(PSO),and bat algorithm(BA).The simulation results show that the prediction performance of the extreme learning machine model optimized by bat algorithm is the best.Then the model is used to simulate the construction process of darg head and analyze the best combination of construction parameters when the output of drag head is the best.Finally,compared with the traditional model of drag head yield,the experimental results show that the prediction performance of the model is greatly improved.(4)Dredging performance evaluation.At present,the construction status of the drag suction dredger depends on the observation of the construction personnel.This method cannot comprehensively and objectively reflect its dredging performance.This paper considers the characteristics of drag head,mud pump and loading characteristics,and constructs a multi-index fusion dredging performance evaluation scheme to solve the above problems.The weight of each dredging performance index is calculated using the entropy method,and the dredging performance is comprehensively evaluated using the measured data of Xiamen Port and Caofeidian.The research shows that this method is not affected by the construction site and the ship size,get rid of the subjective experience interference of the construction personnel,more objectively reflects the construction status of the trailing suction dredger,and has very important engineering application value and market prospects.