Research On Online 3D Measurement Technology Of Steel-wheel Based On Multiple Laser-line Sensors Rotating Scanning

With the rapid development of China Railway High-speed(CRH),its high speed and heavy load put forward strict requirements for the moving safety of trains.Cast steel wheels are the most important moving parts of trains,and their manufacturing precision determines the safety and stability of train moving.Therefore,in the production process,the wheel must be fully measured and inspected to ensure its manufacturing accuracy.However,the existing on-line inspection technology of the wheel only performs sampling inspect of the dimensions of a single section or a part of the wheel,and measurement of all dimensions of the wheel cannot be completed.The inspection of the dimensions of cast steel wheels on actual production lines is mainly done by manual calipers.But the manual inspection with poor precision,low efficiency,and high error rate,is difficult to ensure the dimensional accuracy of the wheel when it has been manufactured,thus posing a safety problem to the moving of the high-speed train.In order to achieve high-precision automated on-line inspection of key dimensions of cast steel wheels on the production line,it is necessary to quickly obtain high-precision 3D shape data of the wheels.Therefore,an on-line 3D measurement technology for cast steel wheels based on multiple laser-line sensors rotating scanning is proposed in this paper.The wheel is scanned by multiple laser-line sensors,and the data measured in multi-view is registered to quickly acquire the complete 3D data of the wheel,further the online detection of the full size of the wheel is realized.The research work of this paper is as follows:Firstly,the 3D measurement principle of cast steel wheels is explored.There are two difficulties in the 3D measurement of cast steel wheels.On the one hand,due to the wheel has the large volume,the complex surface and the sharp change of depth,the measured depth of field of the laser-line sensor must be increased to facilitate the acquisition of complete data.On the other hand,the data measured by rotating scanning needs to be accurately registered to obtain the complete 3D data of the wheel.Therefore,based on the study of the principle of 3D reconstruction of laser-line,the relationship between the measured depth of field of the laserline sensor and the depth of field of the camera is analyzed.It is pointed out that the Scheimpflug camera can greatly increase the measured depth of field of the laser-line sensor while ensuring high precision measurement.At the same time,the principle of data registration in multi-sensor rotational scanning process is studied,which lays the foundation for the following research.Secondly,the calibration method of multi-sensor rotating scanning system is studied.Multi-sensor rotating scanning system calibration is the key problem to achieve complete 3D data registration of the wheel.Therefore,a calibration method by directly using high-precision profile data acquired by sensors is proposed for rotating scanning system using multi-sensor of Scheimpflug camera.The first step,key features of the profiles are detected,and the multisensor measurement coordinate system is registered into a coherent global coordinate system.The second step,initial solution of the rotating axis is estimated by the constraint between the laser plane,the target and the rotating axis.Meanwhile,the optimization function is established by the constraint of target radius size,and the precise pose of the rotating axis can be solved iteratively.Then,the experiment verifies that this calibration method has high accuracy and great stability,and a simple calibration process,and is easy to implement an automated process.On the basis of the above,an online 3D measurement system for the cast steel wheels is developed.In the structural design,the automatic positioning of the wheel is completed.In the software design,automated measurement of wheel 3D data,automated system calibration,and automated management are achieved.The on-line measurement results of the production line confirm that the measurement accuracy and stability of the system can meet the requirements of actual production.