| Metal pipes are widely used in structural load-bearing and gas-liquid transportation in the fields of aerospace,nuclear power,and automobiles due to the hollow structure and excellent mechanical properties.However,due to the elastic-plastic characteristics of the material,the pipe will inevitably spring back after bending,which will affect the accuracy of the pipe.During processing,springback compensation is essential for forming highprecision curved tubes.Aiming at the development requirement for combining the pipe’s springback compensation with cloud platform,this study designs and implements a pipe parameter acquisition system and springback compensation system based on cloud platform.The main work is as follows:First,for the bending process of the pipe,the overall scheme of pipe parameter collection and springback compensation is designed based on the cloud platform.The scheme consists of three parts:the collection system of pipe parameters,the cloud platform and the communication method.The collection system collects pipes’ images and data through the clamping mechanism and hardware circuit,and sends them to the cloud platform.The cloud platform receives images and data,extracts pipes’ parameters,and compensates for springback.The communication mode is used to realize data transmission between the acquisition system and the cloud platform.Secondly,according to the requirements of the image acquisition and cloud platform transmission,build a pipe parameter acquisition system and communication method,and carry out experimental verification.Use 3D printing technology to manufacture a pipe clamping structure,and verify the effectiveness of the clamping function through the test;The hardware circuit of image acquisition mainly includes the selection of the main control chip(STM32),the image acquisition module(OV2640)and the communication module(EC200S)to capture and transmit pipes’ images;The collected images are sent to the cloud platform through the MQTT server,which realizes the data transmission between the acquisition system and the cloud platform.Thirdly,based on the cloud platform,a pipe parameter extraction algorithm based on image recognition is designed to extract pipe parameters,and a springback compensation algorithm is designed to predict the springback compensation.For the collected pipe images,the Canny edge detection algorithm and contour extraction algorithm are used to extract the parameters of the pipe.According to the collected parameters,a least squares model,a decision tree model and a random forest model are established to predict the springback of the pipe,and an angle compensation algorithm is developed to correct the bending angle of the pipe.A database is constructed for data extraction and storage of the program.Finally,the system proposed in this paper is verified by bending experiments.The pipe parameter extraction experiment and the springback compensation experiment are designed respectively.The experimental results show that the pipe parameter extraction algorithm has a higher recognition ability for pipes with larger diameter and wall thickness.Compared with the seamed pipe,the recognition accuracy of the seamless pipe is higher.The overall deviation rate is within 2%.The predicted deviation values of springback prediction model based on random forest is less than 0.75°,and the deviation of the bending angle of the pipe after corrected by the angle compensation program is controlled within 0.6°.Aiming at the development requirement for combining pipe bending springback compensation and cloud platform,this study designs and implements a cloud platform-based pipe parameter acquisition system and springback compensation system.The accuracy of the system is verified by bending experiments.The proposed system provides a practical solution for high-precision bending pipes and cloud manufacturing of bending processing. |
