The Development Of A Digital Twin System For Quality Control Of Robotic Drilling

Robot drilling is an automated machining technology widely used in fields such as aerospace,automotive,shipbuilding,etc.It can improve production efficiency with its high degree of freedom and flexibility.However,during the machining process,due to various factors such as the dynamic stiffness and natural frequency of multi joint connecting rod robots,chatter is prone to occur between the cutting tool and the workpiece,greatly affecting the dimensional accuracy and surface roughness of the machining holes.The influencing factors are multi degree of freedom vibration coupling,etc.Traditional single variable or multivariate analysis cannot accurately analyze the specific changes in the machining process.Digital twin is a technology that maps physical entity and its digital model,which can realize the state perception,behavior simulation and optimization decision of physical entity.This article aims to achieve real-time monitoring and intelligent quality control of the robot drilling process.A digital twin system for robot drilling quality control is designed and developed to address the deviation between burrs and apertures.The main research content of this article includes:(1)establishing a digital twin system architecture for robot drilling quality control,and analyzing the composition and functions of the five dimensional model of the digital twin.This article takes the Yaskawa GP-110 robot as the research object,designs various functional modules based on the operational characteristics of its drilling system,constructs a digital twin system architecture,and designs a quality control operation mechanism based on digital twin technology.(2)Establish a digital twin model for robot drilling.For the digital twin model,solidworks and 3dsmax software are used to build the geometric modeling of the hole making system,establish the physical mathematical models of the vibration,stiffness and drilling force of the hole making system,as well as the mathematical models of the robot synchronous movement and end drilling behavior.Each model is combined and packaged,and the unity3 d engine is used to realize the rendering and driving of the model.At the same time,sensors are used to collect processing status data during the robot drilling process,and filtering and feature extraction are carried out.By attaching scripts to different parts,synchronous simulation of digital twin models is achieved.(3)Establish a deep learning based decision-making method for robot drilling quality optimization.Construct a quality control process and establish a static mathematical model and a dynamic correction fusion prediction model for burr types;Utilizing deep learning to analyze vibration signals and achieve aperture deviation prediction;Establish an optimization model for spindle speed and feed rate using deep learning;Using genetic algorithm to establish an optimization model for robot pose and achieve quality control of robot drilling.(4)Develop human-computer interactive operation interface.Encapsulate the above model,and develop an interactive operation interface based on the proposed digital twin system architecture of robot drilling quality control,to achieve synchronous processing control,and use the data processing module to process the processing state data,to achieve quality prediction and processing parameter optimization.