A method for processing Doppler three-dimensional data

Abstract

The invention provides a method and software for processing Doppler three-dimensional data. The method for processing Doppler three-dimensional data comprises the following steps: (one) using a robust estimation method to filter the data once; (two) using cubic spline interpolation to interpolate out wild points; (three) using three-dimensional phase The spatial method is used to filter the data twice; (4) The cubic spline interpolation method is used to interpolate the outliers. The software for Doppler three-dimensional data processing includes a median calculation module, a robust middle window module, a wild point assignment module, a non-wild point assignment module, a derivative calculation module, a phase space calculation module, and an operation module. By optimizing the coupling robust estimation method and the three-dimensional phase space method, the present invention can perform three-dimensional simultaneous measurement and is suitable for data with high content of wild points, while avoiding iterations and reducing overprocessing phenomena, while improving the detection accuracy of wild points, it can be applied to fluid dynamics experiments middle.

Description

technical field [0001] The invention relates to a data processing software, in particular to a Doppler three-dimensional data processing method and software. Background technique [0002] Acoustic Doppler current meter is a remote sensing three-dimensional velocity sensor, which can measure instantaneous velocity in three-dimensional space. Due to its wide range, fast response, convenient and easy operation, etc., it is widely used in various laboratories or oceans and lakes. and other field hydrological experiments. Acoustic Doppler current meters and other flow velocity measurement devices based on the acoustic Doppler frequency shift effect are affected by Doppler noise, and outliers (outliers) will be generated during data recording. The principle of Doppler frequency shift is that sound waves undergo a phase shift proportional to the velocity of the object after bouncing off a moving object. This phase shift can only be measured in the range +180° to -180°, any shift ...

AI Technical Summary

Problems solved by technology

[0006] The main disadvantage of the phase space method is the phenomenon of overprocessing, that is, due to the discretization, some valid data points near the outliers may be wrongly marked as outliers and thus eliminated

In addition, some scholars pointed out that the phase space method is suitable for laboratory water flow data with less wild point content, and when the wild point content exceeds 6.7%, the detection accuracy of the phase space method decreases

[0007] The existence of extreme value wild points in the robust estimation method will affect the average and MAD parameters so that its detection threshold is affected by extreme value wild points

When there are outliers with a large difference from the real value, the mean value and MAD are affected by the outliers, and the estimated positioning value is shifted to the outliers and away from the real data, and the estimated scale value increases, causing some outliers to be regarded as real data instead of real data. detected

[0008] In addition, the above two methods are only suitable for the detection of single-dimensional flow velocity data. Generally, the flow velocity measured by the flow meter contains velocity components in three directions, and the three-dimensional data are orthogonal to each other. If any dimension data is affected, it should be associated with it. The other two sets of data are deleted

Images