Research On Strain Data Analysis Method Of Ship Lock Miter Gate

As an important hydro-junction in inland river,ship lock plays an important role in ensuring navigation efficiency and safety.However,with the increase of shipping volume,frequent opening and closing of miter gates,combined with the pressure effect of water level difference between upstream and downstream of the gate body and the action of hydrodynamic load,make the stress concentration and release on the surface of the gate body constantly occur.After a long period of accumulation,the surface of the gate body will appear cracks and other factors that destroy the structure of the gate body,which have an impact on the health and service life of the gate body and threaten the opening of the gate body.Aviation safety.However,due to the long maintenance period of the lock,it is impossible to judge the health status of the gate in real time.Therefore,it is of great practical significance to establish an expert system for strain detection and to detect early faults in time and effectively,which can ensure the daily operation of the lock and the safety of navigation.In this paper,the miter gate downstream of Gezhouba 3#lock is taken as the research object.The stress and energy changes of the actual signal collected by the gate body are analyzed by using the theory of wavelet packet analysis.The structure of the miter gate is judged by time-domain and frequency-domain analysis,which provides theoretical support for the construction of stress monitoring expert system.The main research work of this paper is as follows:(1)Finite element analysis of the door body is carried out by ANSYS-Workbench software,and the stress/strain change nephogram of the door body is obtained.According to the nephogram analysis,the main stress concentration position of the door body is obtained.Fiber Bragg Grating sensor is selected for laying and collecting the actual strain signal.Sensor data of three parts with the most obvious stress change are selected,and a complete navigation cycle is adopted to focus on its analysis.The time-domain parameters change,and the time-domain analysis results show that they are within a reasonable range.(2)The data are smoothed by wavelet threshold denoising to highlight the changing trend and reduce external noise interference.By comparing and analyzing the data processing results of four threshold selection rules and soft,hard and two new threshold functions,the superiority of the improved threshold function and heuristic threshold rule in processing gate body strain signals is illustrated.(3)The strain signals of three parts are analyzed by wavelet packet,and the reconstructed signals of wavelet node coefficients at each stage of each part are obtained.The analysis shows that the strain signals of the portal body are mainly concentrated in the low frequency band.According to the characteristics of the energy distribution histogram of each node of the signal reconstructed by wavelet coefficients and the judgement of the energy deviation parameters,the energy change of the gate body is within a reasonable range,but there are hidden dangers in the bottom water-stop part.The energy change of the gate body is quantitatively analyzed by using the wavelet energy entropy based on Shannon entropy,and the characteristic parameters of the strain signal are measured,and the trend of energy change is obtained.The prediction and analysis of portal health status provide theoretical support for fault diagnosis.