Search

White-light interferometry (WLI) is an established method for surface measurement. In WLI, surface recovery algorithm from series of vertical scanning white light interferogram is pivotal technique and has been researched widely. However, the effectiveness of existing surface recovery algorithms is easy to be affected by some conditions like mechanical vibrations, low-reflection of the test surface and phase change caused by the reflection. In this paper, a new recovery algorithm is presented, in which correlation analysis of WLI envelop curves and a multi reference position based phase solution method are employed for robust and high precision surface recovery. Mathematical derivation of the algorithm is carried out, and simulation and experimental testing and comparison experiments are conducted, which show that the new algorithm is effective. [ABSTRACT FROM AUTHOR]

• Double Magnetic coils controller to ensure a constant stiffness of 10 mm range. • Bearing capacity considering anti-deflection torque and stiffness. • Finite difference method combining mass flow theory and boundary conditions. • Dynamic characteristics and critical velocity verification. • 2D/3D texture parameters compared to a commercial conventional stylus profilometry. Stylus profilometry is extensively employed in practice due to its wide measurement range and praiseworthy stability. However, its lever structure will cause nonlinear errors and mitigated precision. Given this context, this paper proposes a new aerostatic bearing stylus sensor (ABSS) for surface measurement, which inherits the stability and large-scale measuring range while avoiding the structural nonlinear errors of the conventional stylus profilometry. The designed ABSS is composed of a magnetic levitation control system and an aerostatic bearing device in the vertical and horizontal directions, respectively. Furthermore, the bearing capacity and the radial dynamic characteristics are analyzed to ensure stylus' scanning stability and rapid response. Template experiments with different scanning velocities of the stylus are performed to verify the natural frequency and obtain the critical velocity. Finally, the surface two/three-dimensional topographies reconstruction validates the repeatability and accuracy of the proposed ABSS compared to a commercial conventional stylus profilometry. [ABSTRACT FROM AUTHOR]

• A coarse-to-fine denoising method is proposed for improving the calibration signal quality. • An adaptive method is proposed for estimating the optimal mode number of VMD. • Two indicators are introduced to identify the relevant modes of decomposition results. Dynamic calibration is an essential way to characterize the measurement performance of pressure sensors under dynamic environment. However, it is difficult to guarantee the reliability of calibration results because the calibration signal is inevitably contaminated by noises. In this paper, a coarse-to-fine denoising method is proposed to improve the quality of calibration signals based on adaptive mode decompositions. Firstly, variational mode decomposition (VMD) is used to decompose the calibration signal into several band-limited intrinsic mode functions (BLIMFs). The optimal mode number is estimated based on their center frequency spacing and mutual information of BLIMFs. The coarsely denoised signal is then reconstructed by the relevant BLIMFs based on the ringing energy loss ratio indicator. Subsequently, the coarsely denoised signal is decomposed as a series of IMFs with empirical mode decomposition (EMD). By introducing a clustering indicator named as comprehensive weighted correlative degree, the ringing and trend IMFs are extracted for obtaining the finally denoised result. The performance of the proposed method is validated by both simulated and actual dynamic calibration signals. Results show that the SNR of denoised result with the proposed method is 33.91, which is obviously larger than that obtained by EMD (SNR = 25.03) and VMD (SNR = 22.67) for simulated signal. Furthermore, comparative experiments also demonstrate the superiority of the proposed method over the existing approaches in both denoising ability and signal integrity. [ABSTRACT FROM AUTHOR]