Research On Measuring Method Of Micro-inclination Angle Of Equivalent Multi-blade Workpiece

To solve the measurement accuracy failure of the automatic measuring equipment for equivalent multi-blade workpiece,this paper optimizes the precision of the measuring system according to the analysis of the measuring error,the outliers correction of sensor observation,the measuring result model and the calibration of error sources,which could improve the measuring precision of the measuring system.The length measurement is accomplished by four groups of two laser length sensors.The indirect measurement of the micro-obliquity is done by the geometric relationship between the sensors and the workpiece to be measured,so as to determine whether the part meets the machining requirements.The required measurement accuracy of the micro-obliquity is 1’,and the theoretical distribution range of the measurement results is 40’± 5’.However,in the actual measurement process,the accuracy of the measurement system can not meet the requirements,and the measurement results deviate from its proper distribution range.The main contents of this paper are as followed:1.Improving the original positioning method of the part to be measured.According to the improvement of the positioning accuracy for the sensors and the part,the repeating accuracy of the final measurement results could get higher;2.For the length measurement data collected by the laser sensor,based on the characteristics of the data,the outlier detecting and eliminating algorithm is designed to reduce the impact of the outlier;3.By correcting the offset angles and displacement of the sensors and the workpiece in the global coordinate system,the machining accuracy tolerances of the mechanical parts which constitute the measuring system are substituted into the mathematical model of the measuring results,so that the theoretical measuring accuracy is solved;4.Applying the neural network fitting model,the neural network training and forecasting is carried out for the correction of the measurement result.The unknown error factors are corrected by the neural network to improve the measurement accuracy;5.Based on the influence of various error factors on the final measurement results,the tolerances of the components of the existing measurement system are optimized to improve the theoretical measurement accuracy.In order to optimize the measuring accuracy of the measuring system,the improvement of the positioning fixture,the correction of the measuring data and the error calibration are integrated.