An outlier processing method suitable for non-constant flow ADV data

Abstract

This invention discloses an outlier processing method for non-steady flow ADV data, relating to the field of post-processing technology for acoustic Doppler current meter velocity data. The method includes: acquiring an instantaneous velocity data sequence; calculating a unit aliasing velocity as a threshold; constructing local velocity variation features and setting discrimination conditions; identifying and generating an outlier index set; iterating until no new outliers are found; uniformly interpolating and replacing outliers; comparing and filtering the interpolated sequence with the original sequence to obtain the final velocity sequence. This method has effective velocity data protection, solves the problem of accidental deletion of valid values, maximizes the integrity of the velocity sequence, and has strong adaptability and reliability.

Description

Technical Field

[0001] This invention relates to the field of post-processing technology for acoustic Doppler velocity (ADV) data, and in particular to a method for processing outliers in non-steady flow ADV data. Background Technology

[0002] Acoustic Doppler current meters (ADVs) are widely used in turbulence measurement, hydrodynamics research, river and coastal engineering, and marine environmental observation due to their high time resolution and measurement accuracy. However, ADV velocity data is susceptible to Doppler aliasing, reduced signal correlation, and measurement noise during measurement, resulting in outliers that deviate significantly from the measured values. Outliers can significantly affect velocity statistics, turbulent kinetic energy spectra, and flow field structure analysis results; therefore, outlier processing of ADV velocity data is usually necessary.

[0003] Current outlier processing methods for Advance Velocity Flow (ADV) mainly include the Phase Space Method (PST), Kernel Density Estimation Algorithm (KDE), and Improved Kernel Density Estimation Algorithm (MKDE). These methods are all based on the steady flow assumption, designing and constructing outlier discrimination criteria through overall statistical characteristics to remove outliers from velocity data. However, in actual hydrological and marine environments, ADV measurement data often contains significant fluctuations and non-steady flow characteristics, such as tidal currents, wave currents, unsteady turbulent currents, and complex hydrodynamic processes. Under non-steady flow conditions, the mean velocity, probability density distribution, and local statistical characteristics change over time, making it impossible for existing outlier processing algorithms to construct outlier identification conditions based on overall statistics. Consequently, they cannot determine the effective velocity range and achieve post-processing of ADV velocity data. Furthermore, in outlier discrimination using velocity data, most outlier processing algorithms require calculating the first derivative of the velocity, leading to some effective velocities being identified as outliers, thus compromising the integrity of the original fluctuating velocity signal.

[0004] Therefore, there is an urgent need for an outlier processing method suitable for non-steady flow ADV data to improve the reliability of ADV velocity data post-processing under complex flow conditions. Summary of the Invention

[0005] The purpose of this invention is to provide a method for processing outliers in non-constant flow ADV data, aiming to solve or improve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following solution: An outlier processing method for non-constant flow ADV data includes: Step 1: Obtain the instantaneous velocity data sequence of the acoustic Doppler current meter (ADV) under non-steady flow conditions; Step 2: Calculate the unit aliasing velocity based on the acoustic parameters of the acoustic Doppler current meter (ADV) and use it as the discrimination threshold. Step 3: Based on the instantaneous velocity data sequence, calculate the difference between velocities at adjacent time points to construct local velocity variation characteristics; Step 4: Based on the discrimination threshold, set the discrimination conditions for local velocity change characteristics, identify outliers in the instantaneous velocity data sequence, and generate an outlier index set; Step 5: Remove the flow rate data corresponding to the outlier index set from the instantaneous flow rate data sequence. Repeat steps 3 and 4 until no new outliers are identified during the loop. Finally, generate a total location index set for all outliers. Step 6: Perform uniform interpolation replacement on the outliers corresponding to the total location index set to generate a flow velocity sequence; Step 7: Compare the flow velocity sequence with the original sequence before interpolation according to preset constraints, and obtain the final flow velocity sequence after filtering.

[0007] Furthermore, the expression for the unit aliasing velocity is: In the formula, Unit aliasing velocity; The speed of sound in water; The frequency of sound wave emission; The interval between sound wave pulses.

[0008] Furthermore, the local variation characteristics of the flow velocity are expressed as follows: In the formula, The local velocity variation characteristics at time i; Let be the instantaneous flow velocity value at time i.

[0009] Furthermore, the discrimination condition is expressed as: In the formula, The local velocity variation characteristics at time i; To determine the proportionality coefficient.

[0010] Furthermore, determine the proportionality coefficient. The value ranges from 0.7 to 0.9.

[0011] Furthermore, outliers in the instantaneous flow velocity data sequence are identified, and an outlier index set is generated, including: The velocity values ​​corresponding to the local velocity change features that meet the judgment criteria in the instantaneous velocity data sequence are identified as outliers, and an outlier index set is generated.

[0012] Furthermore, the limiting condition is expressed as follows: In the formula, The value is the i-th instantaneous velocity in the original sequence before interpolation; The i-th instantaneous velocity value in the velocity sequence after outlier interpolation replacement; This is an empirical coefficient; Let be the i-th instantaneous flow velocity in the final flow velocity sequence.

[0013] Furthermore, the empirical coefficient The value ranges from 0.2 to 0.3.

[0014] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: This invention discloses an outlier processing method for non-steady flow ADV data. The method does not rely on overall statistics and is applicable not only to wave currents, tidal currents and other non-steady flow conditions, but also to steady flow conditions, thus having strong adaptability. By constructing a discrimination criterion using local variation characteristics and using the unit aliasing velocity as the discrimination threshold feature value, the discrimination criterion is clarified, ensuring the scientific nature of the processing method while maintaining its innovation. It has an effective flow rate data protection function, which can solve the problem of accidental deletion of valid values ​​and can ensure the integrity of the flow rate sequence to the maximum extent, thus demonstrating significant reliability. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic flowchart of the method of the present invention; Figure 2 This is a schematic diagram comparing the constant flow ADV velocity sequence before and after processing in this embodiment; Figure 3 This is a schematic diagram comparing the turbulent kinetic energy spectrum before and after the constant flow ADV velocity sequence processing in this embodiment; Figure 4 This is a schematic diagram showing the comparison of ADV velocity sequence processing before and after the fluctuation process in this embodiment; Figure 5 This is a schematic diagram showing the comparison of the ADV velocity sequence before and after the wave process superimposed on the tidal process in this embodiment. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The purpose of this invention is to provide a method for processing outliers in non-constant flow ADV data, aiming to solve or improve at least one of the above-mentioned technical problems.

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] like Figure 1 As shown, this invention provides a method for processing outliers in non-constant flow ADV data, comprising: Step 1: Obtain the instantaneous velocity data sequence of the acoustic Doppler current meter (ADV) under non-steady flow conditions; Step 2: Calculate the unit aliasing velocity based on the acoustic parameters of the acoustic Doppler current meter (ADV), using it as the discrimination threshold, including: The unit aliasing velocity is expressed as: In the formula, Unit aliasing velocity; The speed of sound in water; The frequency of sound wave emission; The interval between sound wave pulses.

[0021] Step 3: Based on the instantaneous velocity data sequence, calculate the difference between velocities at adjacent time points to construct local velocity variation characteristics, including: The characteristics of local velocity variation are expressed as follows: In the formula, The local velocity variation characteristics at time i; Let be the instantaneous flow velocity value at time i.

[0022] Step 4: Based on the discrimination threshold, set the discrimination conditions for local velocity variation characteristics, identify outliers in the instantaneous velocity data sequence, and generate an outlier index set, including: The judgment condition, expressed as: In the formula, The local velocity variation characteristics at time i; To determine the scaling factor, this embodiment determines the scaling factor. The value is less than 1, and the range is 0.7 to 0.9; The velocity values ​​corresponding to the local velocity change features that meet the judgment criteria in the instantaneous velocity data sequence are identified as outliers, and an outlier index set is generated.

[0023] Step 5: Remove the flow rate data corresponding to the outlier index set from the instantaneous flow rate data sequence. Repeat steps 3 and 4 until no new outliers are identified during the loop. Finally, generate a total location index set for all outliers. Step 6: Perform uniform interpolation replacement on the outliers corresponding to the total location index set to generate a flow velocity sequence; The interpolation method used is linear interpolation.

[0024] Step 7: Compare the flow velocity sequence with the original sequence before interpolation according to preset constraints, and obtain the final flow velocity sequence after filtering, including: The limiting condition, expressed as: In the formula, The value is the i-th instantaneous velocity in the original sequence before interpolation; The i-th instantaneous velocity value in the velocity sequence after outlier interpolation replacement; This is an empirical coefficient, and in this embodiment, its value ranges from 0.2 to 0.3. To determine the threshold; Let be the i-th instantaneous flow velocity in the final flow velocity sequence.

[0025] like Figure 2 As shown, the ADV instantaneous velocity data contained a large number of outliers before processing. After processing by the method proposed in this invention, the outliers in the velocity sequence were removed, and the instantaneous velocity fluctuated within a limited range around 0.5 m / s.

[0026] According to the appendix Figure 2 This indicates that the method of the present invention is also applicable to processing constant-flow data.

[0027] Appendix Figure 3 In this context, the formula for calculating the turbulent energy spectrum is: In the formula, Pulsation velocity Time-dependent function, pulsating flow velocity , The average flow rate is denoted as .

[0028] The turbulent kinetic energy spectrum was obtained by using an improved periodogram method, which divided the data into 16 sub-parts with 50% overlap, and added a Hanning window to avoid spectral leakage.

[0029] From the appendix Figure 3 As can be seen, the slope of the turbulent energy spectrum of the processed velocity sequence in the inertial sub-region is quite consistent with the slope of Kolmogorov -5 / 3, which demonstrates the effectiveness of the present invention in processing steady flow ADV turbulent velocity data.

[0030] Appendix Figure 4 In this study, the ADV velocity data were obtained from a research experiment on the scouring effect of a marine monopile under tidal currents. The ADV sampling frequency in the experiment was 50 Hz, and the tidal period was approximately 100 min. It can be seen that the processing method proposed in this invention can effectively handle outliers in ADV velocity sequences containing undulating processes.

[0031] Appendix Figure 5 In the experiment, the sampling frequency of ADV was 100 Hz, and the time-averaged flow velocity was 6.67 × 10⁻⁶ within 0–550 s. -4 m 2 The acceleration increases linearly from 0 m / s to 0.36 m / s, while within 550 s to 1100 s, the flow velocity decreases linearly from 0.36 m / s to 0 m / s with the opposite acceleration. The superimposed wave process has a wave frequency of 20 Hz and an amplitude of 0.2 m / s. The results show that, based on the superposition of wave and tidal processes, the method proposed in this invention can still effectively remove outliers from the flow velocity sequence.

[0032] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0033] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for processing outliers in non-steady-flow ADV data, characterized in that, include: Step 1: Obtain the instantaneous velocity data sequence of the acoustic Doppler current meter (ADV) under non-steady flow conditions; Step 2: Calculate the unit aliasing velocity based on the acoustic parameters of the acoustic Doppler current meter (ADV) and use it as the discrimination threshold. Step 3: Based on the instantaneous velocity data sequence, calculate the difference between velocities at adjacent time points to construct local velocity variation characteristics; Step 4: Based on the discrimination threshold, set the discrimination conditions for local velocity change characteristics, identify outliers in the instantaneous velocity data sequence, and generate an outlier index set; Step 5: Remove the flow rate data corresponding to the outlier index set from the instantaneous flow rate data sequence. Repeat steps 3 and 4 until no new outliers are identified during the loop. Finally, generate a total location index set for all outliers. Step 6: Perform uniform interpolation replacement on the outliers corresponding to the total location index set to generate a flow velocity sequence; Step 7: Compare the flow velocity sequence with the original sequence before interpolation according to preset constraints, and obtain the final flow velocity sequence after filtering.

2. The outlier processing method for non-constant flow ADV data according to claim 1, characterized in that, The expression for the unit aliasing velocity is: In the formula, Unit aliasing velocity; The speed of sound in water; The frequency of sound wave emission; The interval between sound wave pulses.

3. The outlier processing method for non-constant flow ADV data according to claim 1, characterized in that, The expression for the local velocity variation characteristic is: In the formula, The local velocity variation characteristics at time i; Let be the instantaneous flow velocity value at time i.

4. The outlier processing method for non-constant flow ADV data according to claim 1, characterized in that, The discrimination condition is expressed as follows: In the formula, The local velocity variation characteristics at time i; To determine the proportionality coefficient.

5. The outlier processing method for non-constant flow ADV data according to claim 4, characterized in that, The discrimination ratio coefficient The value ranges from 0.7 to 0.

9.

6. The outlier processing method for non-constant flow ADV data according to claim 1, characterized in that, The process of identifying outliers in the instantaneous flow velocity data sequence and generating an outlier index set includes: The velocity values ​​corresponding to the local velocity change features that meet the judgment criteria in the instantaneous velocity data sequence are identified as outliers, and an outlier index set is generated.

7. The outlier processing method for non-constant flow ADV data according to claim 1, characterized in that, The limiting condition is expressed as follows: In the formula, The value is the i-th instantaneous velocity in the original sequence before interpolation; The i-th instantaneous velocity value in the velocity sequence after outlier interpolation replacement; This is an empirical coefficient; Let be the i-th instantaneous flow velocity in the final flow velocity sequence.

8. The outlier processing method for non-constant flow ADV data according to claim 7, characterized in that, The empirical coefficient The value ranges from 0.2 to 0.3.

Images