Research On Coaxiality Detection System Of Armhole Group Of Loading Maneuver Based On Machinevision

The boom arm,as the core component of a loader’s working mechanism,bears continuous varying external loads during shovel operations.Wear and failure often occur at the joints of the components,making it necessary to ensure the strength and rigidity of the boom arm while controlling the coaxiality error of the boom arm holes.Based on research,most companies rely on manual sampling using a coordinate measuring machine for coaxiality inspection.Although this method provides high accuracy,the inspection process is cumbersome,inefficient,and incurs high equipment maintenance costs.To address the issue of low efficiency in detecting coaxiality errors in large-span and large-sized non-coplanar through-holes,this study proposes a machine vision-based detection system for measuring the coaxiality of loader boom arm holes,aiming to achieve fast,accurate,and automated online inspections.The main contributions of this paper are as follows:(1)Conducting research on coaxiality error detection methods and the application of vision inspection techniques in coaxiality measurement.Proposing a measurement approach based on vision inspection techniques and establishing a coaxiality error model based on the structural characteristics of the loader boom arm.Seeking an error evaluation method that satisfies the minimum inclusion condition.(2)Designing the overall detection system for the coaxiality error of the boom arm holes.Selecting components based on the actual measurement environment and constructing the detection system platform.Integrating Lab VIEW vision modules with the HALCON vision algorithm library to develop the detection system interface.The system interface includes modules for image acquisition and display,motion control,and data output and storage.Furthermore,the detection process of the system is described using the example of the boom arm chassis holes,based on the design of the software and hardware systems.(3)Performing camera calibration on the detection system to obtain the internal and external parameters of the industrial camera,thereby establishing the relationship between image pixel dimensions and physical dimensions.Conducting research on fundamental image processing algorithms and proposing a morphological-based Canny edge detection algorithm that combines traditional Canny edge detection with morphology.This approach aims to extract more complete and clearer sub-pixel edges of the boom arm holes.Experimental measurements are conducted to validate the feasibility of the detection method.(4)Analyzing the overall error sources of the detection system based on experimental results.In addition to the unavoidable manufacturing errors of key components,external environmental factors,and human factors,the influence of assembly size deviations on the detection platform is considered the most significant.Using 3D deviation analysis software 3DCS,this issue is analyzed from the perspective of size deviation transfer and accumulation.The three main sources of deviation are treated as optimization variables.A polynomial response surface methodology is employed to establish an approximate function model between independent variables and target values.The parameter combination that minimizes the assembly size rejection rate is obtained.The accuracy of this parameter combination is validated through assembly simulation.Comparative experiments before and after size optimization show that the coaxiality error remains stable at 0.8±0.03 mm,meeting the production requirements of keeping it below 1mm set by the enterprise.Moreover,the detection accuracy improves by 32.8% after size optimization.