Research On Identification Of Looseness Of Clamp In Hydraulic Pipeline

As an important fixing device in hydraulic pipelines,clamp usually work in complex environments such as high pressure,high temperature and strong vibration.Long-term operation of the hydraulic pipeline can cause vibration fatigue of the clamp and cause the clamp to loosen,which will seriously affect the healthy operation of the hydraulic pipeline and even lead to major safety accidents.How to timely reinforce or replace the clamp is an urgent problem to be solved.Therefore,it is particularly important to identify the looseness of the clamp.At present,the research on the looseness of the clamp is mainly based on the vibration characteristics of the pipelines for spatial detection.The commonly used methods are based on the natural frequency and mode shape.Although the method has clear physical meaning,the actual engineering operating environment is complex,and the structural modal parameters are difficult to extract,so it is rarely used in practical engineering.In this paper,the clamp of hydraulic pipelines is taken as the research object,and the signal processing method is used to study the identification of the looseness of clamp in hydraulic pipelines and its online monitoring system.The main research contents are as follows:(1)In order to effectively identify the looseness of clamp,a smart clamp based on Fiber Bragg Grating(FBG)sensing technology is proposed.The FBG is embedded in the body of clamp to measure the strain of clamp.At the same time,an analytical model of the relationship between the force of clamp and the strain measured by FBG is established,and the measurement principle of the looseness of clamp is given by the analytical model.Furthermore,the burial position of FBG is optimized by finite element simulation and the analytical model is verified.The calibration experiment reveals that the relationship between the clamp force and the strain measured by FBG is linear,and the correctness of the analytical model is verified from the experimental point of view.Finally,the clamp is applied to a specific industrial piping system and its looseness was tested.(2)Aiming at the problem that the fluid excitation of hydraulic pipelines is complicated and it is difficult to extract the effective feature,a method for identifying the looseness clamp based on 2DCNN is proposed.After merging the distributed FBG strain sensor data on the pipelines,construct the loose state gray map of clamp as the input of the 2DCNN model.According to the specific network structure,adaptive feature extraction,fusion and selection are performed to classify the looseness of clamp.The experimental results of the looseness of clamp on hydraulic straight pipelines show that the method has better flexibility and classification accuracy than the traditional custom feature extraction and recognition algorithm.(3)In order to further verify the robustness and stability of the method for identifying the looseness of clamp based on the 2DCNN,the hydraulic bending pipelines with more complicated fluid action was selected to carry out the experiment of looseness of clamp.The experimental results show that the method still has a high recognition accuracy for the identification of the looseness of clamp on the hydraulic hydraulic bending pipelines.(4)An online monitoring system for hydraulic pipelines was designed and implemented for the status monitoring of hydraulic pipelines.The system acquires FBG sensor data in real time from the acquisition device,performs real-time analysis and processing of data,storage after data preprocessing,and performs real-time diagnosis on current clamp status.