Methodology And Application Of Construction Data Deep Mining And Productivity Optimization Analysis For Trailing Suction Hopper Dredgers

Trailing suction hopper dredger is one of the most widely used dredgers in dredging engineering construction.Its productivity analysis and optimization is an important way to achieve efficient dredging.However,under the influence of complex construction conditions,huge multi-system construction machinery and massive construction data,the characteristics of dredging construction equipment vary nonlinearly with the operation status and operation environment,and the relationship between the control parameters in the operation process and the output parameters in the actual process is not clear.Therefore,it is difficult to analyze and optimize the influence factors of production productivity from three aspects: the overall mechanical performance of the trailing suction hopper dredger,the small-scale mud transportation and the large-scale earthwork loading.In recent years,with the continuous development of computer,sensor,artificial intelligence and other advanced technologies,it provides a new technical method and method for analyzing the multi-sensor data,multi system model,multi-temporal and spatial scales,multi probability events and other aspects of simulation and optimization of trailing suction hopper dredger productivity.In this paper,based on the physical model and mathematical expression of the productivity influence mechanism model and mathematical expression of the dredging system of the trailing suction hopper dredger,and based on the data of dredging construction monitoring,the artificial intelligence method is introduced and improved.From the overall optimization of equipment performance,increase the utilization rate of equipment time,enhance the real-time optimization control of production productivity from different scales and maximize the loading cycle productivity,which provides scientific and reliable decision-making basis for the safe,stable and efficient dredging construction of trailing suction hopper dredger.Based on the construction data of large-scale trailing suction hopper dredger in Tianjin Port waterway dredging project,combined with artificial intelligence theory and algorithm,in-depth mining and intelligent analysis are carried out.The main research contents and research results are as follows:(1)Taking the dredging construction capacity as the guidance,this paper puts forward the analysis model and mathematical expression model of the capacity influence mechanism of each system of the trailing suction hopper dredger construction,and constructs the monitoring data processing model of the trailing suction dredger to eliminate the time delay and noise,which provides a guarantee for the accuracy and effectiveness of the basic analysis data.On the one hand,in principle,the factors influencing the process such as the drag head system model,dredge pump pipeline system model,loading and overflow system model,power ship system model and soil system model are analyzed in depth.According to the influence relationship of dredged soil in each system,the number of influencing factors on capacity optimization is established It provides a model basis for the subsequent analysis of practical problems.On the other hand,starting from the data structure,the integral delay elimination method is used to eliminate the time-delay effect of the monitoring data of the trailing suction ship,from the data noise,the overall smooth optimal denoising model is proposed,and the Savitzky-Golay filter is used to filter different types of noise data,which improves the reliability of the input data.(2)A virtual sensor technology for instrument detection and mechanical fault monitoring of trailing suction hopper dredger is proposed.Based on intelligent algorithm,the influence of sensor fault and mechanical fault on productivity is deeply explored,and the theoretical model and solution based on virtual sensor design are constructed.Through the proposed feature selection method,the key factors affecting the velocity sensor are obtained,and then the virtual velocity sensor with reliable performance is obtained by using intelligent model training.On the one hand,the virtual sensor can diagnose the performance of monitoring instrument,on the other hand,it can make temporary replacement when sensor fails,which provides safety support for the operation optimization of trailing suction hopper dredger.According to the virtual sensor generation framework,the "torsional vibration" virtual sensor is obtained,and the warning threshold is obtained according to the historical statistical data.The combination of the two can early warning and monitor the mechanical failure of the trailing suction ship,and avoid the reduction of time utilization and productivity efficiency caused by the mechanical failure of the trailing suction hopper dredger maintenance and mechanical damage.(3)For small-scale mud transportation,a real-time productivity analysis model and method of trailing suction dredger based on stacking strategy is proposed.Considering the small-scale mud pump transportation,starting from the real-time display and optimization control of the production capacity and the drag head and the mud pump pipeline,this model can eliminate the inherent defects of the sensor equipment installation,display the production efficiency in real time,and obtain the influence on the production capacity through the feature selection method the key controllable monitoring features of the ship are conducive to real-time optimization of control by ship operators to improve construction efficiency.(4)Aiming at large-scale earthwork loading,a prediction model for the fitting of the earthwork curve and an optimization model for the loading period are constructed.Considering the large-scale earthwork loading curve,mathematical modeling prediction is made on the cumulative curve of the loading earthwork.According to the predicted earthwork curve,combining geometric coordinate model can predict the optimal loading time,avoiding more loading overflows Ineffective power and time wasted,the construction capacity is optimized,which is conducive to improving dredging efficiency and dredging quality,and reducing energy consumption.