Research On Dredging Optimization Of Cutter Suction Dredger

Dredging equipment is widely used in the construction and maintenance of national economic infrastructure,such as ports and waterways.As a kind of dredging vessel,the cutter suction dredger is widely used because it is not restricted by the type of soil excavated,its working principle is simple and its excavation depth is appropriate.In the actual operation of the cutter suction dredger,the construction personnel mainly judge the current dredging status based on the construction experience.Due to the influence of the operating environment and the working status of the operators,the dredging efficiency is often not stable enough.With the development of artificial intelligence technologies such as machine learning and big data,it has become possible to improve dredging efficiency through intelligent technology.Green dredging and intelligent dredging will become the development trend of the dredging industry in the future.Research on the dredging optimization of cutter suction dredgers has great significance.Supported by the Ministry of Industry and Information Technology’s dredging project "Development of Intelligent Operation System for Dredging Ships",this article aims at the difficult problems of the unmeasurable concentration of the cutter suction during the excavation and pipeline transportation of cutter suction dredgers and the low accuracy of the output control of cutter suction dredgers.Research on the mechanism modeling of cutter excavation and mud sand pipeline transportation,real-time engineering data preprocessing of cutter suction dredgers,suction mouth concentration estimation and construction output prediction optimization are carried out.The research results provide technical support for the efficient dredging of cutter suction dredgers.The main work is as follows:(1)Analysis of the reamer excavation and pipeline transportation process.Based on the analysis of the working process of the cutter suction dredger,the analysis of the cutter structure and the study of the cutter cutting model are carried out,and the mud pump model and pipeline model are constructed for the pipeline mud and sand transportation process.(2)Data preprocessing of dredging process.In view of the problems of sensor data disturbance and abnormal condition data of cutter suction dredger,real-time monitoring data cleaning is carried out,including data normalization,abnormal condition data elimination and wavelet threshold filtering,which can effectively improve the quality of data.The parameter selection method combining Lasso regression and Maximum Information Coefficient(MIC)redundant recognition is used to select the data features,which provides suitable model data input for the establishment of production prediction model.(3)Estimation of suction concentration of cutter suction dredger.Aiming at the problem that the suction mouth concentration at the cutter of a cutter suction dredger cannot be directly measured,a data-driven estimation method is used to estimate the suction mouth concentration of a cutter suction dredger.The verification results show that this method can predict the suction mouth concentration value very well.It has better accuracy and real-time performance in engineering applications.(4)Research on production forecasting technology.Aiming at the problem that the output of cutter suction dredgers mainly depends on experience and the real-time performance is poor,the extreme learning machine(ELM)and its improved algorithm are used to predict the output of cutter suction dredgers.The simulation results show that the algorithm has the ability to predict the output and less error.Aiming at the problem that the extreme learning machine’s random initialization weights and thresholds lead to weak model generalization,particle swarm optimization(PSO)is used to optimize the extreme learning machine model.The simulation results show that this method has higher prediction accuracy,faster speed,and improves the prediction ability of cutter suction dredger production.