Design Of Data Analysis And Optimization Decision System For Dredging Operation Of Trailing Suction Dredgers

With the advent of a new era,more and more technological innovations have emerged in our vision,data mining,machine learning,deep learning,artificial intelligence,and so on.The dredging industry began to advocate "green dredging," "environmental dredging," and "smart dredging," and dredging dredgers as the main ship type for dredging ships should assume the historic mission of building a leader of "smart dredging." Due to the complex dredging operation environment,the large number of control parameters and the coupling of different operators,the dredging efficiency is not high,and the neural network combined with the expert system is used to optimize the dredging process.This paper relies on the fund project of China National Engineering Research Center for Dredging Technical Equipment to analyze and study the optimization of dredging and loading of dredger dredgers.In this paper,an on-line extreme learning machine algorithm was used to build a neural network model to predict the suction density of the suction head of the trailing suction dredger and the overflow loss of the cabin.The neural network expert system was selected as the solution to optimize the dredging operation.A set was designed and developed.Expert System Software for Dredging Optimization Decision of Dredging Dredgers.The main work is as follows:(1)Study the dredging equipment parameters under different conditions,and study the application of wavelet filtering,first-order low-pass filter and limiting filter in dredging data,and select the corresponding filtering method for different data types.(2)The inhalation density of trailing suction dredgers cannot be known in time and there is no effective measurement method for overflow loss.The limit learning algorithm and BP neural network algorithm are studied.The online extreme learning machine algorithm is used to build a neural network model for dredging.The data is predicted.Experiments were conducted using dredging data from the Yangtze River estuary.Simulation results show that the suction density of the suction head of the trailing suction dredger is accurately predicted and the predicted overflow density of the sludge compartment can be verified by the loading amount and can be provided to the operator as a reference.(3)According to the optimization of the loading problem,the three performanceindicators of dry soil tonnage productivity,sediment storage rate,and mass ratio of dry soil to tonnage were analyzed and differently optimized in different dredging stages.Through the study of case-based expert systems and neural network-based expert systems,a solution based on expert system for optimal decision-making of dredging operations for dredger dredgers was proposed.This combines offline and online optimization measures to ensure the effectiveness and real-time performance of the expert system.(4)Visual Stdio2008 was used to design and develop a set of optimization decision-making software for dredging operations of trailing suction dredgers.The function was tested and the experimental results showed that the design goals were achieved.