Research On Axis Trajectory Reconstruction Of Rotating Machinery Based On Photogrammetry

With the continuous improvement of science and technology and the degree of automation,rotating machinery equipment in the current and future social production will play an increasingly important role,and its application will be more and more extensive.In order to ensure the normal operation of rotating machinery equipment,the measurement and monitoring of its axis trajectory is essential to ensure the safety of rotating machinery operation is of great significance.Although the traditional methods of measuring the Axis trace of eddy current displacement sensor and laser micro-sensor have certain research value and significance in theory and laboratory,they need to be equipped with special bracket and fixed probe,and it is limited by the collection environment.These auxiliary equipment ins tallation and environmental constraints,and other factors may introduce errors to the measurement results,all these adverse factors will make the Axis track accuracy and clarity become very complex,and sometimes even distortion.Compared with the traditional axial trajectory measurement method,the photographic measurement-based method has the characteristics of non-contact,visualization and high measurement accuracy.Therefore,this paper puts forward a scheme of Axis track reconstruction based on photogrammetry,and introduces the measuring method of the scheme in detail,at the same time,the methods of camera calibration parameter optimization,feature object detection,Axis track 3D reconstruction and heterogeneous measurement data fusion are studied.The experimental results show that the proposed method is more accurate and reliable than the traditional methods.Therefore,the research results of this paper have important theoretical and practical application value,and provide strong support for the monitoring and maintenance of rotating machinery equipment.The specific research work consists of the following parts:1.In view of the importance of Axis locus in engineering and the sharp contradiction that it is difficult to solve in theory,this paper proposes a scheme of axis locus reconstruction for rotating machinery with high stability and strong adaptability.Combining the advantages of photogramnietry,a photogrammetry-based platform for measuring the axis trajectory of rotating machinery was built,which bypassed the problem of solving Reno equation and solved the theoretical calculation problem of axis trajectory by hardware measurement technology.2.In order to solve the problem of low precision due to visual distortion in the imaging process of vision measurement equipment,a hybrid optimization algorithm based on improved differential evolution particle swarm optimization is proposed to optimize camera calibration parameters,this method can reduce distortion and improve camera calibration accuracy.According to the imaging distortion model,four algorithms are used to optimize the camera calibration parameters,and the validity and reliability of the improved algorithms are verified.3.Aiming at the problem of target detection and recognition in vision measurement,this paper proposes SURF-HSV target detection and Canny operator edge processing method for target recognition,the ellipse fitting method and image weighted centroid method are used to extract the center coordinates of the mark points and improve the accuracy of Axis detection.4.In response to the problem of axis trajectory reconstruction,based on the theory and methods of binocular stereo vision,the spatial information of the field of view is obtained through stereo calibration.The linear triangle method and the forward intersection of stereo images arc used to reconstruct the three-dimensional axis traj ectory,and the effectiveness of the three-dimensional axis traj ectory reconstruction method is verified.5.In order to sllve the problem of heterogeneous measurement data fusion,the linear weighted average method,convex combination data fusion method and covariance cross fusion algorithm are studied,an improved covariance crossover data fusion algorithm based on hunter-prey optimization is proposed.The experimental results show that the improved covariance cross fusion method based on hunter-prey optimization has the best fusion effect,the least error and the best performance.6.A software for Axis track reconstruction of rotating machinery based on photogrammetry is developed,and the corresponding data processing functions are solidified in the software.The software system includes data storage module,camera calibration module,image processing module,trajectory processing module,measurement data fusion module,result display module and so on.