A novel class of non-Gaussian system performance assessment and controller parameter tuning methods.

Traditional variance-based control performance assessment (CPA) and controller parameter tuning (CPT) methods tend to ignore non-Gaussian external disturbances. To address this limitation, this study proposes a novel class of CPA and CPT methods for non-Gaussian single-input single-output systems, denoted as data Gaussianization (inverse) transformation methods. The idea of quantile transformation is used to transform the non-Gaussian data with the goal of maximizing mutual information into virtual Gaussian data.

Research on the zero-calibration gauge device.

To meet the demand for the track's geometric parameter detection equipment for train speed and high-speed aerodynamics tests, a zero-calibration gauge device is designed with the centering limit, height adjustment and horizontal display function in this paper. The bending situation of the zero-calibration gauge is analyzed and the processing technology is studied, which ensures the rationality and realizability of the design of zero-calibration gauge. Then the gauge zero value and the parallelism of the working surface of zero-calibration gauge have been experimentally tested.