| As one of the core components of gear measuring center or other precision measuring instruments,3D scanning probe has a key impact on the measurement accuracy.Thus,it is necessary to study the error of 3D scanning probe.In this thesis,the factors affecting the detection error of 3D scanning probe are analyzed and studied in detail.The detection error of 3D scanning probe is measured through specific experiments,and the detection error of 3D scanning probe is predicted and compensated by neural network.The effectiveness of the algorithm is verified on the CNC gear measurement center.The main research contents are as follows.1)Detection error analysis.Through the analysis of the internal structure of the 3D scanning probe,it is concluded that there are parasitic rotation angles and parasitic displacements in the movement process of the guiding mechanism inside the probe,and its influence on the measurement accuracy is obtained through theoretical analysis;the causes of the probe detection error caused by the actual use of the scanning probe are analyzed,and the local deformation of the probe and the bending model of the measuring rod are analyzed Analysis.2)Data acquisition circuit of probe design.Aiming at SP80H 3D scanning probe,a set of data acquisition circuit for CNC gear measurement center is designed.The acquisition circuit takes FPGA and CPLD as the core,realizes the real-time acquisition of the probe data,and communicates with the upper computer of the gear measurement center,which provides the experimental equipment support for the subsequent experiments.3)Measurement of probe detection error.The detection error of the probe is measured in the CNC gear measurement center.By measuring the detection error of different probe indentation,different measuring rod,single point detection and scanning measurement under different measurement conditions,the theoretical analysis of the impact of indentation,measuring rod and measurement mode on the detection error is verified through the analysis of the experimental results.4)Error compensation algorithm of neural network.By collecting the standard ball,the corresponding sample data acquisition method is designed.The neural network is trained in MATLAB.Import new measurement data into the neural network.The effectiveness of the error compensation algorithm is verified by experiments.In conclusion,this article theoretically analyzes the influencing factors of the detection error of the 3D scanning probe,designs the data acquisition circuit of the probe based on FPGA and CPLD as the experimental equipment,and completes the detection error measurement of SP80H scanning probe in different working environments.By using neural network algorithm,the problem that the error of probe is difficult to compensate due to the coupling of error factors is solved. |
