蔡晨光
@nim.ac.cn
Division of Mechanics and Acoustics Metrology
National Institute of Metrology, China
Chenguang Cai received his Ph.D. degree in Precision Instrumentation and Mechanics from School of Instruments Science and Optoelectronic Engineering, Beihang University, Beijing, China, in 2007. From 2007 to 2009, he worked in Nokia Research Center (Beijing) as a Post doctor. He is currently an associate research fellow in National Institute of Metrology, Beijing, 100029, China. He was a visiting scientist in Physikalisch technische Bundesanstalt (PTB), Germany, in 2016. He is interested in vibration calibration and optics measurement.
EDUCATION
1997-09-01 to 2007-07-31 | Ph.d (School of Instrumentation Science and Opto-electronics Engineering ), Beihang University, Beijing, CN
RESEARCH INTERESTS
transducer, vibration, calibration, metrology, machine vision
Scopus Publications
Scopus Publications
Ming Yang, Haihui Huang, Zhihua Liu, Chenguang Cai, Ying Wang, and Jing Yang
Institute of Electrical and Electronics Engineers (IEEE)
Kai Zhao, Zhihua Liu, Chenguang Cai, Fubing Bao, Chengxu Tu, and Yuxi Qi
Optica Publishing Group
Accurate pose measurement is crucial for parallel manipulators (PM). This study designs a novel integrated 6-DOF motion tracking system to achieve precise online pose measurement. However, the presence of geometric errors introduces imperfections in the accuracy of the measured pose. Based on the displacement information of six grating rulers, measurement pose is obtained through forward kinematics. By comparing the measurement results with the actual pose information captured by stereo vision, measurement errors can be obtained. A closed-loop vector-based kinematic model and an error model are established, and then the geometric errors are identified with the least-squares method. Finally, the geometric calibration experiments are conducted, and the results show that the measurement accuracy has significantly improved, with the average position error decreasing from 3.148 mm to 0.036 mm, and the average orientation error is decreased from 0.225° to 0.022°.
Haihui Huang, Zhihua Liu, Chenguang Cai, Ming Yang, and Deguang Wang
Elsevier BV
Shengnan Zuo, Chenguang Cai, Ming Yang, Zhihua Liu, Deguang Wang, and Ying Wang
Elsevier BV
Ming Yang, Shengnan Zuo, Zhihua Liu, Chenguang Cai, and Ying Wang
Institute of Electrical and Electronics Engineers (IEEE)
The low-frequency vibration sensors have been widely utilized in different applications of precision motion control and measurement. In order to guarantee the performance of these applications, the heterodyne laser interferometry (TLI) recommended by the International Organization for Standard (ISO) is commonly applied to determine the sensitivities of these sensors. However, the original heterodyne interference signal acquisition has to require a high sampling rate, which results in the vibration calibration in low frequency is significantly difficult to be accomplished. Although the direct output displacement signal of the interferometer can be utilized for calibrating the sensitivity magnitude, it involves a phase delay that inevitably reduces the calibration accuracy of the sensitivity phase. In this study, a new TLI that based on the Sony/Philips digital interconnect format (S/PDIF) signal is investigated to simultaneously calibrate the low-frequency analog and digital vibration sensors. This investigated method determines the sensitivity magnitude and phase of the sensors by adopting the high-accuracy S/PDIF decoding and signal alignment methods. Comparison experiments with the current monocular vision (MV) and linear grating (LG) methods as well as the conventional TLI demonstrate that the investigated method has the ability to accurately calibrate the sensitivity magnitude and phase of the analog and digital sensors with a high repeatability in the range from 0.05 to 20 Hz.
Ming Yang, Wenfeng Liu, Zhihua Liu, Chenguang Cai, Ying Wang, and Jing Yang
Institute of Electrical and Electronics Engineers (IEEE)
The long-stroke shaker is essentially required for the calibration of low-frequency vibration transducers, whose performance parameters have significant impact on the calibration accuracy. The accurate measurement of these parameters is the prerequisite to establish a reliable vibration metrology and traceability system. Currently, an optical collimator or a reference accelerometer is applied to get the static parameter, the laser interferometry or triaxial sensor-based method is used to obtain the dynamic parameters. However, the former relies on an extra device which increases the complexity and cost of calibration system, and the latter is always difficult to accomplish the accurate and efficient measurement of these parameters. In this study, a binocular vision-based long-stroke shaker performance measurement method is investigated, which has ability to determine the static and dynamic parameters simultaneously during the calibration. This vision method obtains the shaker's bending by measuring the inclinations at the different positions of its guideway, and achieves the amplitude characteristic, distortion, repeatability as well as transverse ratio measurements by accurately acquiring the spatial displacements at different frequencies. Comparison experiments with the two commonly used inclination estimation methods, and the laser interferometry and sensor-based method demonstrate that the investigated method is able to get the satisfactory accuracies both for the static and dynamic parameters of long-stroke shaker in vibration calibration.
Ming Yang, Sifan Mo, Chenguang Cai, Zhihua Liu, Deguang Wang, and Ying Wang
Optica Publishing Group
The low-frequency vibration exists in building structures, mechanical devices, instrument manufacturing, and other fields, and is the key to modal analysis, steady-state control, and precision machining. At present, the monocular vision (MV) method has gradually become the primary choice to measure the low-frequency vibration because of its distinctive advantages in efficiency, non-contact, simplicity, flexibility, low cost, etc. Although many literature reports have demonstrated that this method has the capability to reach high measurement repeatability and resolution, its metrological traceability and uncertainty evaluation are difficult to be unified. In this study, a novel, to the best of our knowledge, virtual traceability method is presented to evaluate the measurement performance of the MV method for the low-frequency vibration. This presented method achieves traceability by adopting the standard sine motion videos and the precise position error correction model. Simulations and experiments confirm that the presented method can evaluate the amplitude and phase measurement accuracy of the MV-based low-frequency vibration in the frequency range from 0.01 to 20 Hz.
Ming Yang, Zhihua Liu, Ying Wang, Chenguang Cai, and Jing Yang
Institute of Electrical and Electronics Engineers (IEEE)
In this article, the low-frequency linear and angular vibration sensors have been gradually used in many applications of vibration monitoring because they can measure dynamic displacement and angle. These sensors must be calibrated before they are used and after they have been used for a period to ensure their measurement accuracy. Currently, the laser interferometry (LI) and sensor-based comparison method are commonly used to calibrate their sensitivities. However, the former suffers inevitably drawbacks in low-frequency range, while the latter usually has only a limited range, these undoubtedly limit their wide application. In this article, a new monocular vision-based multiparameter calibration method is investigated, which can determine the sensitivities of these sensors in a broad low-frequency range. The monocular vision method with improved line segment detector is applied to measure the linear and angular vibration excitations by extracting the motion feature edges with subpixel accuracy. This investigated method strongly promotes the perfection and unification of linear and angular vibration calibration. Many comparison experiments with the LI, the Earth's graviation method, and the circular grating-based method confirm that the investigated method gets the satisfactory accuracies in the linear and angular vibration calibration with an efficient, flexible, and low-cost system in the low-frequency range.
Lei Fu, Zhihua Liu, Chenguang Cai, Meng Tao, Ming Yang, and Haihui Huang
Elsevier BV
Ming Yang, Jing Zhang, Chenguang Cai, Ying Wang, Zhihua Liu, and Deguang Wang
Institute of Electrical and Electronics Engineers (IEEE)
Lei Fu, Zhihua Liu, Meng Tao, Chenguang Cai, and Ming Yang
Springer Nature Singapore
Shengnan Zuo, Chenguang Cai, Zhihua Liu, Ming Yang, Haihui Huang, and Huinan Gong
IEEE
Noise is always inevitably appeared in images, which directly affects the performance of machine vision applications. Currently, the commonly used denoising methods can be divided into three strategies: the filtering-based, model-based, and deep learning-based methods. However, they are always difficult to get the considerable accuracy and efficiency simultaneously. In this study, a novel denoising method based on autocorrelation function is investigated, which improve the image quality by utilizing the independence of useful periodic information and noise. Simulations and experiments compared with the current denoising methods confirm that the investigated method has a good comprehensive effect on noise reduction and efficiency improvement.
Ming Yang, Chenguang Cai, Deguang Wang, Qinmu Wu, Zhihua Liu, and Ying Wang
Institute of Electrical and Electronics Engineers (IEEE)
Lei Fu, Ming Yang, Zhihua Liu, Meng Tao, Chenguang Cai, and Haihui Huang
Optica Publishing Group
Accuracy is the most important index for the industrial applications of the Stewart platform, which can be guaranteed by the kinematic calibration method to improve the motion orbit performance of this platform. In order to improve the effectiveness of the least squares algorithm and the identified accuracy of the platform’s geometric parameter errors, an applicab-le dimensionless error model based on the structural characteristics of the Stewart platform is investigated. Moreover, a novel stereo vision-based measurement method is proposed, which can measure the 6-degree-of-freedom (DOF) pose of the moving platform. On this basis, an identification simulation is schemed to validate the efficiency of the dimensionless error model, and the kinematic calibration experiment is carried out on a prototype. The experimental results demonstrate that the position error is decreased to 0.261 mm with an improved accuracy of 89.720%, the orientation error is decreased to 0.051° with an improved accuracy of 90.351%.
Ming Yang, Zhihua Liu, Chenguang Cai, Ying Wang, Jing Yang, and Junjie Yang
Institute of Electrical and Electronics Engineers (IEEE)
The low-frequency triaxial vibration sensors have been gradually applied in many engineering fields of vibration monitoring because they can measure the multidirection vibrations simultaneously. The accurate axial and transverse sensitivities, determined by the calibration method, are the prerequisite for ensuring their measurement accuracy. Currently, the laser interferometry (LI) which is based on a single component or a tricomponent linear shaker is usually applied to calibrate these sensitivities. However, the former has to require the multiple reinstallations of the sensor and the latter cannot avoid the motion coupling caused by the shaker, these inevitably increase the calibration uncertainty. In this article, we investigate a monocular vision (MV)-based two-component shaker calibration method, which determines the axial sensitivity based on the time-spatial synchronization and transverse sensitivity at the elliptical orbit excitation. The MV method is used to measure this excitation, and a plane sensitivity model is presented to describe these sensitivities. This investigated method can simultaneously reduce the uncertainties caused by the reinstallations and motion coupling to improve the calibration accuracy. Experimental results compared with the LI and Earth's gravitation method demonstrate that the investigated method obtains the satisfactory accuracies both in axial sensitivity magnitude and phase as well as transverse sensitivity magnitude and direction calibration.
Ying Zhang, Zhihua Liu, Dezhi Zheng, and Chenguang Cai
Elsevier BV
Ran Cheng, Zhihua Liu, Guodong Zhai, Qi Lv, Ming Yang, and Chenguang Cai
MDPI AG
In order to ensure the measurement accuracy of high-acceleration vibration sensors used in engineering applications, it is necessary to calibrate their key performance parameters at high acceleration. The high-acceleration vibration calibration system produces high-acceleration vibration by utilizing the resonance amplification principle; however, the resonance frequency of the resonant beam changes with increasing amplitude, affected by the influences of nonlinear and other factors. In this study, a phase-locked resonance tracking control method based on the phase resonance principle is proposed to accurately and quickly track the resonance frequency of the resonant beam, which can improve the accuracy and stability of resonance control. The resonant beam is able to produce stable vibration with an amplitude exceeding 7500 m/s2 by phase-locking and tracking the resonant frequency. A calibration system built with this method can provide stable vibration with an amplitude of 500–10,000 m/s2 in the range of 80–4000 Hz. Comparison experiments with the commonly used amplitude iteration amplification method demonstrate that the proposed method can give an acceleration stability control index of less than 0.5% and a resonance tracking time of less than 0.1 s.
Hao Cheng, Ying Wang, Kai Wei, Zhihua Liu, Ming Yang, and Chenguang Cai
Optica Publishing Group
Angular vibration calibration is required to determine the sensitivity of sensors such as dynamic inclinometers, gyroscopes, and angular accelerometers, which are used for angular motion measurement in engineering applications. Additionally, the calibration performance depends on the accuracy of the angle measurement by laser interferometry or a circular grating (CG) method that is commonly used in vibration calibration. However, these methods usually own a complex and high-cost system or limited frequency and amplitude ranges. In this study, a novel, to the best of our knowledge, angle measurement method that combines a special visual encoder and an accurate angular position detection method is investigated; the method requires only a simple and flexible telecentric vision measurement system. Comparison experiments with the CG method demonstrate that the investigated method has the maximum measurement deviation of 0.0014° and 0.0138° for static angle measurement in the small-angle range and continuous full circle, respectively. The relative deviation of the angular vibration measurement in the range of 0.1–8 Hz with amplitudes 0–100° is less than 0.173%. Additionally, the relative deviation of calibrated sensitivity of a gyroscope by the investigated and CG methods is less than 0.096%.