A method and system for compensating data from a mobile ADCP test.

Abstract

This invention relates to the field of hydrological fluid measurement and testing technology, and particularly to a method and system for compensating data from a mobile ADCP (Advanced Hydrological Computational Processing) test. The method includes the following steps: acquiring raw hydrological test data using a mobile ADCP, identifying the failure areas of the raw hydrological test data; selecting interpolation types according to the failure area range to fill the failure data in the raw hydrological test data, verifying the rationality of the interpolation results, and obtaining preliminary completed velocity data; constructing a correlation between bank distance and velocity attenuation based on historical effective velocity data of the complete cross-section, extracting continuous velocity data near the boundary from the preliminary completed velocity data, and calculating the velocity in the cross-section blind zone based on the attenuation correlation, thus forming velocity data for the completed blind zone. This invention, through data association and fusion and data completion technology adapted to complex hydrological scenarios, achieves accurate completion of velocity blind zones and water depth anomaly intervals in complex scenarios, and reconstructs the cross-sectional flow regime and accurately calculates the flow rate.

Description

Technical Field

[0001] This invention relates to the field of hydrological fluid measurement and testing technology, and in particular to a method and system for compensating data from a mobile ADCP test. Background Technology

[0002] Flow measurement is one of the core components of hydrological work. Its importance is reflected in many key areas such as water resources management, flood control and disaster reduction, water ecological protection, and water conservancy project planning and scheduling. It is of great significance for ensuring water security and promoting sustainable economic and social development.

[0003] As a core piece of equipment in the field of hydrological surveying, the mobile ADCP (Acoustic Doppler Current Profiler) is widely used to acquire key parameters such as cross-sectional velocity, water depth, and flow rate due to its non-contact measurement and efficient synchronous acquisition features. It provides crucial technical support for both routine hydrological monitoring and data acquisition in special hydrological scenarios. However, in practical applications, the presence of complex factors such as high river drop, high flow velocity, and high sediment content can lead to the absorption and scattering of acoustic signals by suspended matter such as sediment particles. This increases signal noise, causes signal interruptions, and shortens the effective observation distance of the transducer, resulting in a significant reduction in the observed water depth. Furthermore, "bubbles" in rapid river currents can also reflect acoustic signals, which may be incorrectly identified by the ADCP, leading to a large number of false velocity data.

[0004] These external interferences can cause transducer signal distortion, thus affecting normal testing. Current processing technologies for mobile ADCP test data lack specific solutions for handling these faulty data, failing to compensate for critical data such as velocity and depth measurements in the affected areas. This makes it difficult to generate reliable flow measurement results under extreme scenarios such as high flood seasons. This situation impacts high flood and above-standard flood measurements within the hydrological industry, hinders hydrological data collection, delays the monitoring and reporting of first-hand hydrological test results under special hydrological conditions, affects hydrological assessment, and significantly impacts flood defense in river basins and regions. Summary of the Invention

[0005] Based on this technical background, it is necessary to provide a method and system for compensating mobile ADCP test data to solve at least one of the above-mentioned technical problems, thereby obtaining the complete flow rate of the entire cross section.

[0006] To achieve the above objective, a method for compensating for underway ADCP test data is provided, the method comprising the following steps: Step S1: Collect raw hydrological measurement data using a mobile ADCP and identify areas of failure in the raw hydrological measurement data; Step S2: Collect effective water depth data associated with the cross-sectional velocity dataset; based on the effective water depth data, fuse the water depth data obtained by ADCP and other depth measurement methods to complement each other in cross-sectional morphology, or combine with the corresponding river topographic features to complete abnormal or missing water depth intervals, so as to complete the compensation operation of the original hydrological survey data. Step S3: Select the faulty data to fill the original hydrological test data according to the faulty area range, verify the rationality of the interpolation results, and obtain the preliminary completed flow velocity data; Step S4: Based on the historical effective velocity data of the complete cross-section, construct the correspondence between the distance to the shore and the velocity attenuation. Extract continuous velocity data near the boundary from the initially completed velocity data. Combine the attenuation correspondence to estimate the velocity in the blind zone of the cross-section, forming the velocity data of the completed blind zone. Extract the velocity data of the local blind zone at the boundary from the velocity data of the completed blind zone and fit the velocity change characteristic curve. Supplement the velocity data of the local blind zone at the boundary according to the velocity change characteristic curve to form a continuous cross-sectional velocity dataset. Step S5: When the effective cross-sectional velocity data cannot reconstruct the complete cross-sectional flow pattern, the method of extracting the cross-sectional velocity vertical line is adopted. The average velocity of the vertical line of the supplemented water depth data and the effective velocity measurement vertical line is associated with the same cross-sectional position to reconstruct the cross-sectional velocity distribution characteristics, so as to complete the compensation operation of the full cross-sectional flow calculation.

[0007] The present invention also provides a compensation system for walk-through ADCP test data, used to perform the above-described compensation method for walk-through ADCP test data, the compensation system for walk-through ADCP test data comprising: The data acquisition module is used to collect raw hydrological measurement data through a mobile ADCP and identify areas of failure in the raw hydrological measurement data. The water depth completion module, based on various types of invalid water depth, and according to the two depth measurement modes of ADCP, uses reliable water depth data at the breakpoint as anchor points, and completes the cross-sectional shape through linear interpolation or cross-sectional trend fitting; it uses the latitude and longitude position of the cross-sectional trajectory to determine the location, and uses the effective water depth data as a benchmark, combined with the river channel topographic features to eliminate abnormal or missing water depth intervals, and performs location matching with historical cross-sectional measurement results or subsequent remeasurement cross-sectional results to complete the compensation operation of hydrological measurement water depth data; The velocity interpolation supplementation module is used to initially fill the velocity failure data in the original hydrological test data by selecting the interpolation type according to the failure area range and failure type, and to verify the rationality of the interpolation results to obtain the preliminary supplemented velocity data. The boundary velocity completion module is used to construct the attenuation relationship between shore distance and velocity based on the historical valid test data of the complete cross section. It extracts continuous velocity data near the boundary from the initially completed velocity data, and calculates the velocity in the blind area on both sides of the cross section by combining the attenuation relationship, thus forming the velocity data of the completed blind area. The vertical line extraction module extracts velocity vertical lines at specific locations on the cross-section according to preset rules, extracts the average flow velocity of the velocity vertical line, and matches it with the corresponding water depth of the velocity vertical line on the cross-section to calculate the cross-sectional flow rate.

[0008] The beneficial effects of this invention are as follows: On the one hand, by adapting to the systematic processing logic of complex hydrological scenarios, it achieves accurate completion of velocity data; on the other hand, it constructs velocity attenuation correlation based on complete cross-sectional data, integrates terrain constraints and boundary effective velocity unit data, accurately calculates the velocity in the cross-sectional blind zone, and avoids data mutations; it adaptively selects the completion method for different ranges of failure areas, embeds a rationality verification mechanism, ensures the continuous and uninterrupted velocity distribution, significantly improves the integrity and accuracy of velocity data in complex scenarios, and lays a high-quality data foundation for flow calculation.

[0009] On the other hand, relying on the short-term stability characteristics of cross-sectional morphology and cross-validation of dual-source bathymetry data, the abnormal intervals are segmented and supplemented with effective water depth as the anchor point to form continuous full-section water depth data. For extreme scenarios, the layout of velocity measuring verticals is dynamically optimized through sub-interval division and hydraulic characteristic analysis to accurately match the characteristics of flow velocity changes. Overcoming technical bottlenecks such as bathymetry failure during high flood season and insufficient effective data ratio, reliable reconstruction of cross-sectional flow velocity distribution and accurate flow estimation are achieved. The adaptability and timeliness of hydrological measurements are improved, providing reliable technical support for obtaining real flow data under special hydrological conditions and optimizing measurement efficiency. Attached Figure Description

[0010] Figure 1 A schematic diagram of the steps involved in a method for compensating data from a mobile ADCP test. Figure 2 This is a schematic diagram of the characteristic curve of flow velocity change; Figure 3 A diagram of the operating interface of the compensation system for mobile ADCP test data; The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0011] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0012] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0013] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0014] In all respects, the embodiments should be regarded as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, it is intended that all variations falling within the meaning and scope of the equivalents of the application be included within the present invention.

[0015] To achieve the above objectives, please refer to Figures 1 to 3 A method for compensating data from a mobile ADCP test, the method comprising the following steps: Step S1: Collect raw hydrological measurement data using a mobile ADCP and identify areas of failure in the raw hydrological measurement data; In this embodiment, a mobile acoustic Doppler current profiler (ADCP) is used as the core measuring device. This device integrates a dual depth sounding module and a current velocity detection module. It is mounted on a hydrological survey vessel and travels along a preset test section route, with the travel speed stably controlled within 3 m / s. After the ADCP is started, it simultaneously activates vertical beam sounding, bottom tracking sounding, and multi-depth layer current velocity detection functions, and links with the GPS positioning module to collect data. The current velocity detection frequency is set to 2 Hz, the water depth sampling interval is 0.5 seconds, the current velocity measurement range is 0-5 m / s with an accuracy of ±0.01 m / s, and the water depth measurement range is 0.5-100 m with an accuracy of ±0.05 m. It collects water velocity data, dual-source water depth data, and corresponding location coordinate data within the section. All raw data are stored with dual tags of collection timestamp and coordinate information to form a complete hydrological survey raw dataset.

[0016] The original dataset was categorized and organized according to data type, and flow velocity data sequences, vertical beam depth data sequences, bottom tracking depth data sequences, and coordinate data sequences were constructed respectively. The flow velocity and depth data sequences were reordered from the starting bank to the ending bank according to the cross-sectional position coordinates, and the spatial location of different types of data was accurately matched by timestamps.

[0017] It should be noted that for signal loss during the test, the missing values ​​in the process data are directly identified. For other variable data, combined with the measurement characteristics of complex scenarios such as water flow turbulence, floating object interference, and high sediment content, the data validity and reliability judgment criteria are set: For cross-sections with historical test results, the reasonable range of flow velocity data is defined as 2.0 times the maximum measured flow velocity and 0.5 times the minimum historical flow velocity, and the reasonable range of water depth data is defined as 3 times and 0.3 times the historical average water depth. Data outside the range is marked as invalid. At the same time, the gradient of adjacent data changes is calculated. Data with a flow velocity gradient exceeding 0.1 m / (s·m) and a water depth gradient exceeding 0.5 m / m are also marked as invalid. The cross-section location interval in all invalid data is the failure area. The start and end points and span of the failure area are accurately marked by coordinates.

[0018] Step S1 includes the following steps: Identify the core data items from the test results files of different brands of ADCPs as follows: timestamp, GPS latitude and longitude, ship speed / heading, water depth, beam current velocity, east / north / vertical current velocity, surface / bottom current velocity, sampling interval, depth cell size, blind zone distance, effectiveness, bottom tracking status, and beam status.

[0019] Step S2: Collect effective water depth data associated with the cross-sectional velocity dataset; based on the effective water depth data, fuse the water depth data obtained by ADCP and other depth measurement methods to complement each other in cross-sectional morphology, or combine with the corresponding river topographic features to complete abnormal or missing water depth intervals, so as to complete the compensation operation of the original hydrological survey data. The effective water depth data associated with the cross-sectional velocity dataset collected in step S2 includes: The mobile ADCP system employs both vertical beam and bottom tracking depth measurement methods, synchronously acquiring dual-source depth data at 1-meter intervals at various locations along the cross-section. It identifies the effective ranges and failure points of both methods, eliminating abnormal data that exceed the reasonable fluctuation range of the cross-section's historical average depth. Cross-validation is performed on the effective dual-source depth data, calculating the data deviation value. Consistent data with deviation values ​​within a preset allowable threshold are extracted and marked as effective depth data.

[0020] In this embodiment, the vertical beam echo sounding and bottom tracking echo sounding functions of the mobile acoustic Doppler current profiler (ADCP) are activated. The same route and moving speed as the cross-section current velocity data acquisition are maintained, and dual-source water depth data at each location of the cross-section are collected synchronously at fixed intervals of 0.5m, namely vertical beam water depth data and bottom tracking water depth data. Each set of data includes the corresponding cross-section location coordinates, acquisition timestamp, and water depth measurement value. The water depth measurement accuracy is controlled within ±0.05m, and the measurement range covers 0.5-100m.

[0021] Retrieve the most recent historical water depth measurement data for this cross section, calculate the historical average water depth value, and set the reasonable fluctuation range of the historical average water depth of the cross section to be 0.3 to 3 times the historical average water depth. Values ​​exceeding this range in the water depth data collected by both vertical beam and bottom tracking methods are identified as abnormal data and removed. At the same time, failure breakpoints are retrieved through continuous data sequence. When data of one of the sounding methods is missing at a certain location or when data of three or more consecutive measurement points are abnormal, that location is identified as the failure breakpoint of the corresponding sounding method, and the effective range of each of the two sounding methods is determined.

[0022] It should be noted that cross-validation is performed on the water depth data within the effective range of the two sounding methods. Vertical beam depth data and bottom-tracking depth data are matched one-to-one according to the cross-sectional coordinates. The absolute deviation value of each set of matched data is calculated, i.e., the absolute value of the difference between the vertical beam depth value and the bottom-tracking depth value. A preset allowable deviation threshold of 0.2m is set. Matching data with an absolute deviation value ≤ 0.2m are extracted as consistent data. These consistent data are sorted according to the cross-sectional coordinates and associated with the continuous cross-sectional velocity dataset obtained in step S4 through a coordinate reference, and marked as effective water depth data corresponding one-to-one with the cross-sectional velocity data.

[0023] The specific data mentioned below are for illustrative purposes only, and this invention does not impose any specific limitations. Assuming the historical average water depth of the cross-section is 10m, and the reasonable fluctuation range is set to 3m-30m, at a cross-section location X=10m, the vertical beam water depth measurement value is 10.2m, the bottom tracking water depth measurement value is 10.1m, and the absolute deviation value is 0.1m. Within the allowable threshold range, this set of data is marked as valid water depth data. At X=15m, the vertical beam water depth measurement value is 2.5m, exceeding the lower limit of the reasonable fluctuation range, and is therefore judged as abnormal data and discarded. Only the water depth measurement value closest to the historical average water depth is retained for subsequent verification.

[0024] Step S2, based on the effective water depth data, involves supplementing abnormal or missing water depth intervals by combining the corresponding river channel topographic features. Collect topographic data of the complete cross section measured before the flood or the cross section remeasured after the flood, and use it as a reference frame; register the reference frame with the current cross section coordinates; use the effective water depth data as anchor points to identify the missing or abnormal water depth intervals in the current cross section; Based on the topographic undulation trend of the corresponding water depth range of the benchmark frame, the data is supplemented by segmented matching and adjustment. The gradient of water depth change between adjacent anchor points is controlled within a preset reasonable range to form supplemented water depth data consistent with the topographic features of the benchmark frame.

[0025] In this embodiment, topographic data of a pre-flood measured complete cross-section or a post-flood re-measured cross-section is collected using river topographic mapping equipment. This data includes the coordinates of each position in the transverse direction of the cross-section, elevation information, and riverbed undulations, serving as a reference framework for completing the water depth. The data scale is set to 1:1000, and the coordinate accuracy is controlled within ±0.1m. The cross-sectional coordinates of the reference framework are aligned with the coordinate system of the current test cross-section using coordinate translation and rotation registration methods, ensuring that the reference framework and the current cross-section are aligned under the same coordinate reference, thus ensuring that the topographic features at corresponding locations can be accurately referenced.

[0026] Using effective water depth data as anchor points, the anchor point data includes the cross-sectional location coordinates and the corresponding effective water depth values. All positions of the current cross-section are traversed in coordinate order. Intervals with no water depth data or exceeding the reasonable fluctuation range of historical average water depth are identified as missing or abnormal water depth intervals, and the start and end coordinates and span of the interval are determined.

[0027] Referring to the topographic relief trend of the corresponding water depth interval in the reference framework, the missing or abnormal intervals are divided into several segments according to the distribution of effective water depth anchor points. Each segment has effective water depth data as fixed anchor points at both ends. The data is supplemented by segmented matching adjustment. Based on the water depth change pattern of the corresponding segment in the reference framework, the water depth supplement value at each position in the current interval is calculated. At the same time, the reasonable range of water depth change gradient between adjacent anchor points is set to 0.1-0.5m / m.

[0028] The specific data mentioned below are for illustrative purposes only, and this invention does not impose any specific limitations. Assume that the current cross-section X=20m-30m represents a missing water depth range. The effective water depth at the left anchor point (X=20m) is 8.5m, and the effective water depth at the right anchor point (X=30m) is 9.2m. The corresponding range in the baseline frame shows a gentle upward trend. Dividing the area into 10 completion nodes at 1m intervals, based on the trend of the baseline frame, the calculated water depth completion values ​​are 8.6m at X=22m, 8.8m at X=25m, and 9.0m at X=28m. The water depth change gradient at each node is controlled between 0.07m / m and 0.08m / m, which is within the preset reasonable range. After all segments are completed, the effective water depth data and the completed water depth data are integrated to form complete completed water depth data that is consistent with the topographic features of the baseline frame, without any missing or abnormal data.

[0029] Most importantly, step S2, which uses effective water depth data as a benchmark and combines it with the corresponding river topographic features to fill in abnormal or missing water depth intervals, also includes: selecting the water depth data with the smallest deviation as a reliable anchor point at the failure breakpoint of both vertical beam and bottom tracking bathymetry methods; determining the river topography type through topographic survey data and determining the appropriate fitting method based on the river topography type; gradually extending the effective data of a single bathymetry method to the failure interval of the other method according to the selected method; and filling in the water depth data gaps in the failure interval in 0.5m steps.

[0030] In this embodiment, the complete data sequences of both vertical beam echo sounding and bottom tracking echo sounding are sorted out, and the failure points of each are located. Within the effective data range adjacent to the failure point (3 measurement points before and after), 3 sets of water depth data corresponding to the two echo sounding methods are extracted, and the difference between each set of data is calculated. The set of water depth data with the smallest difference is selected as a reliable anchor point. The anchor point data contains clear cross-sectional location coordinates and water depth measurement values.

[0031] Topographic data of the current test section is obtained using river topographic survey equipment (such as a multibeam echo sounder). This data includes riverbed elevation values ​​at 0.5m intervals laterally. The ratio of the elevation difference to the horizontal distance between adjacent locations is calculated to obtain the segmented slope. A slope fluctuation threshold of ±0.05m / m is set. If the difference between all segmented slopes and the average slope of the section is within this threshold and there is no significant undulation, it is judged as a gentle terrain. If the segmented slopes continuously show a change of first increasing and then decreasing or first decreasing and then increasing, and the maximum elevation difference exceeds 10% of the average water depth of the section, forming an arc-shaped profile, it is judged as a curved terrain. If there are at least 3 alternating steep sections (slope ≥0.2m / m) and gentle sections (slope ≤0.05m / m), and the length of both steep and gentle sections is ≥2m, it is judged as a stepped terrain.

[0032] It should be noted that the appropriate fitting method is determined according to different river topography types: for gently sloping terrain, a linear fitting method is used, with the water depth value of a reliable anchor point and the water depth value of the adjacent valid data point as endpoints, and the water depth at each position in the failure interval is calculated according to the linear relationship between the two points; for curved terrain, a parabolic fitting method is used, selecting the water depth values ​​of a reliable anchor point and two valid data points before and after it, constructing a parabolic equation, and calculating the water depth in the failure interval based on the equation; for stepped terrain, a trapezoidal fitting method is used, dividing the terrain into segments according to the steepness, calculating the water depth within each segment using a linear relationship, and maintaining the continuity of water depth between segments.

[0033] Starting with reliable anchor points as the baseline, the effective data from a single sounding method is gradually extended to the failure range of another sounding method according to the selected fitting method. During the extension process, a check point is set every 0.5m to verify the matching degree between the calculated water depth and the corresponding elevation in the benchmark topographic data, ensuring that the deviation is ≤0.3m. Filling nodes are divided within the failure range at fixed step sizes of 0.5m, with each node corresponding to a unique cross-sectional coordinate. The water depth filling value is calculated node by node according to the extension pattern.

[0034] The specific data mentioned below are for illustrative purposes only, and this invention does not impose any specific limitations. Assume that the vertical beam echo sounding experiences a failure point at X=30m, with the adjacent reliable anchor point at X=29m (vertical beam depth 8.2m, bottom tracking depth 8.1m, minimum difference of 0.1m). Topographic data shows that within the X=25m-35m range, the difference between the slope of each segment and the average slope (0.03m / m) is ≤0.02m / m, indicating a gentle terrain. A linear fitting method is used to extend the failure range to X=30m-35m. Ten filling nodes are defined with a 0.5m step size. The water depth filling value at X=30.5m is calculated as (8.1m + (30.5m - 29m) × 0.03m / m), resulting in 8.145m, and at X=31m it is 8.19m. All nodes are filled sequentially, ultimately forming continuous full-section water depth data.

[0035] Most importantly, in step S2, after completing the abnormal or missing water depth intervals, the formation of continuous full-section water depth data is specifically as follows: the completed water depth data is arranged in the order of the cross-section position, the water depth difference between adjacent data points is calculated, and abnormal connection areas where the difference exceeds the reasonable range are identified; a smoothing correction is performed on the abnormal areas, while keeping the original value of the effective water depth data unchanged during the correction process; the continuity of the corrected data is verified through a sliding window to form continuous full-section water depth data.

[0036] In this embodiment, the completed water depth data (including the original effective water depth data and the completed water depth data) are arranged in an orderly manner from the starting bank to the ending bank according to the cross-sectional position coordinates, ensuring that each coordinate point corresponds to a unique water depth value. The data arrangement interval is consistent with the measurement interval, which is uniformly 0.5m. The water depth difference between two adjacent coordinate points is calculated point by point, and the reasonable range of the water depth difference is set to ±0.3m. This range is determined with reference to the short-term relative stability characteristics of the river channel topography and the accuracy of water depth measurement. Adjacent data pairs that exceed this range are marked as connection anomalies, and the cross-sectional coordinate interval in which they are located is the connection anomaly area.

[0037] For the abnormal area, a smooth correction is performed. The original values ​​of the three consecutive valid water depth data before and after the abnormal area are kept unchanged. The supplementary water depth data in the abnormal area is corrected by step-by-step adjustment in segments. The adjustment is made point by point from both ends of the abnormal area to the middle. For each data point adjusted, the water depth difference between it and the adjacent fixed or corrected data points is calculated to ensure that the difference after adjustment is ≤0.3m, until the difference between all adjacent data points in the abnormal area is within a reasonable range.

[0038] It should be noted that a sliding window was used to verify the continuity of the corrected data. The sliding window contained 5 consecutive data points, and the window moved sequentially along the cross-sectional coordinates in 0.5m increments. After each window movement, the standard deviation of the water depth of the 5 data points within the window was calculated. The standard deviation threshold was set at 0.05m. When the standard deviation of all windows was ≤0.05m, the data continuity was deemed to meet the requirements.

[0039] The specific data mentioned below are for illustrative purposes only, and this invention does not impose any specific limitations. Assuming the corrected cross-section X=25m-27m represents the original area of ​​anomalous connection, the corrected water depths are 8.3m at X=25m, 8.4m at X=25.5m, 8.5m at X=26m, 8.6m at X=26.5m, and 8.7m at X=27m. The difference between adjacent data points is 0.1m, which is within a reasonable range. The sliding window includes X=25m-27m and one data point before and after it. The calculated standard deviation is 0.087m, exceeding the threshold. Further fine-tuning of the X=26m water depth to 8.55m is required. The standard deviation of the window is recalculated to 0.05m, meeting the requirements. After all windows pass verification, continuous full-section water depth data is generated, showing continuous water depth changes without anomalous connections and conforming to the river channel topography.

[0040] Step S3: Select the faulty data to fill the original hydrological test data according to the faulty area range, verify the rationality of the interpolation results, and obtain the preliminary completed flow velocity data; Step S3 includes the following steps: The blank area values ​​of the failed regions are calculated. When the blank area value is ≤5㎡, the interpolation type suitable for local data completion, namely the inverse distance weighted interpolation (IDW, distance weight = 4 or above), is selected to fill the failed data point by point. When the blank area value is between 5-20㎡, the interpolation type suitable for medium-range data completion, namely the ordinary kriging or inverse distance weighted interpolation (IDW, distance weight = 2-3), is selected to fill the failed data in a targeted manner. When the blank area value is >20㎡, the interpolation type suitable for large-scale data reconstruction, namely the universal kriging, is selected to fill the failed data as a whole.

[0041] After the data filling of the failed area is completed, the gradient changes of adjacent valid data are compared to verify the interpolation results and abnormal interpolation data with abrupt gradient changes are removed. For the failed sub-regions corresponding to the abnormal data, a new appropriate interpolation type is selected to perform the filling process, and finally, preliminary completed flow velocity data without obvious data gaps is obtained.

[0042] In this embodiment, based on the determined start and end coordinates of the failure area, combined with the cross-sectional water depth data and lateral span, a two-dimensional plane coordinate system is established with the cross-sectional position coordinates as the horizontal axis and the water depth as the vertical axis. The failure area appears as a polygon in this coordinate system. The lateral and vertical boundaries of the polygon are divided at 0.1m intervals to form several tiny rectangular calculation units. Each tiny rectangle has a length and width of 0.1m. By counting the number of tiny rectangles and multiplying them by the area of ​​a single rectangle (0.01㎡), the blank area value of the failure area is accurately calculated, ensuring that the area calculation error is ≤0.1㎡.

[0043] It should be noted that the filling method is selected according to the blank area value. When the blank area value is ≤5㎡, the four nearest continuous effective flow velocity data around the failure area are used as references. Weights are assigned according to the distance between each reference data and the failure location, with the closer the distance, the higher the weight. The filling value of each failure location is calculated by weighted summation to achieve precise point-to-point filling. When the blank area value is between 5-20㎡, the failure area is divided into filling units according to a 0.2m×0.2m grid. The filling value is calculated unit by unit based on the effective flow velocity data around each unit and the flow velocity change trend of the cross section where the unit is located, to complete targeted filling. When the blank area value is >20㎡, the filling data of the failure area is calculated as a whole based on the distribution trend of the effective flow velocity data of the entire cross section.

[0044] After the data is filled, the difference between the filled data and the adjacent effective flow velocity data is calculated point by point, and then divided by the spatial distance between the two points to obtain the change gradient. The flow velocity gradient threshold is set to 0.05 m / (s・m). Filled data with gradients exceeding this threshold are marked as abnormal interpolation data, and the range of the failure sub-region where the abnormal data is located is recorded.

[0045] The blank area values ​​of the failed sub-regions are recalculated, and the appropriate filling method is matched again according to the corresponding area range. The above filling process is repeated, and the gradient change is checked again after filling until the gradients of all filled data and adjacent valid data are within the set threshold range. Finally, the preliminary completed flow velocity data is obtained. In this step, the preliminary filling estimation of the blank cross-sectional area except for the riverbank boundary can be realized. At the same time, the filling interval can be smoothly calculated by using the flow velocity gradient change threshold to achieve the stabilization of flow velocity changes.

[0046] In step S4, the specific steps for constructing the correspondence between shore distance and velocity attenuation based on the historical effective velocity data of the complete cross-section are as follows: Collect flow velocity measurements corresponding to different bank distances from the historical effective flow velocity data of each complete cross section to form bank distance data pairs; perform fitting operations on the bank distance data pairs to establish a correspondence between bank distance and flow velocity attenuation that reflects the law of flow velocity decreasing with bank distance, and verify the reliability of the correspondence between bank distance and flow velocity attenuation through mean square error.

[0047] In this embodiment, multiple sets of complete historical effective flow velocity data from the same period over the past three years were collected for the test section. Invalid and abnormal data were removed to ensure that the effectiveness of each set of historical data met the standard. For each set of historical data, flow velocity measurements were extracted at fixed shore distance intervals to form several sets of "shore distance-flow velocity" shore distance data pairs. This ensured that the data pairs covered the entire shore distance range of the cross-section, and a sufficient number of data pairs were extracted from each set of historical data to ensure fitting accuracy.

[0048] All historical shore distance data pairs were aggregated, and outlier data pairs with excessive deviations were removed. Using shore distance as the independent variable and flow velocity as the dependent variable, a nonlinear fitting operation was used to construct the attenuation relationship of flow velocity as shore distance decreases. A suitable fitting model was selected and the relevant parameters were solved to determine the final attenuation relationship.

[0049] The reliability of the attenuation correspondence is verified using mean square error. Effective shore distance data are substituted into the fitted correspondence, and a verification index is calculated and compared with a preset threshold. If the verification index meets the threshold requirement, the attenuation correspondence is deemed reliable and can be used for subsequent blind zone velocity estimation; otherwise, historical data is re-screened, and the fitting model is adjusted until the reliability requirement is met.

[0050] In step S4, the flow velocity in the blind zone of the cross-section is calculated based on the attenuation correspondence, and the flow velocity data for completing the blind zone is specifically as follows: The effective velocity unit data is used as input value and substituted into the attenuation correspondence with constraints. The theoretical velocity values ​​at each position of the cross-section blind zone are calculated step by step at preset fixed intervals. The theoretical velocity values ​​are aligned with the preliminary completed velocity data according to the cross-section position. Velocity data with deviations exceeding the reasonable range in overlapping areas are removed, the velocity in that area is recalculated, and all effective velocity data are integrated to form the velocity data for the completed blind zone.

[0051] In this embodiment, based on the preliminary completed velocity data obtained in step S3, continuous and anomaly-free effective velocity unit data within the non-blind zone of the cross section are extracted and used as input values. These data are then substituted into the established and verified reliable correspondence between shore distance and velocity attenuation with topographic constraints to ensure that the input data matches the parameter range of the correspondence.

[0052] Calculation nodes are divided at preset fixed intervals, with the start and end coordinates of the cross-sectional blind zone as the boundary. Each node corresponds to a unique shore distance and cross-sectional position coordinate. The shore distance of each node is substituted into the attenuation correspondence with constraints, and the theoretical velocity value is calculated for each node. The velocity correction coefficient of the corresponding area is then used to correct the theoretical velocity value, which conforms to the actual terrain conditions.

[0053] All theoretical velocity values ​​are sorted according to cross-sectional coordinates and aligned with the initially supplemented velocity data. The overlapping area is identified, and the deviation between the theoretical velocity values ​​and the initially supplemented velocity values ​​within the overlapping area is calculated. Abnormal data points with deviations exceeding a reasonable range are removed. For abnormal areas, relevant data are retrieved again, and the attenuation correspondence is substituted to recalculate until the deviation meets the standard. Finally, all valid velocity data are integrated to form supplemented blind zone velocity data that covers the entire cross-sectional blind zone and has no data gaps.

[0054] Step S4 extracts the flow velocity data of the boundary local blind zone from the flow velocity data of the completed blind zone, and fits the flow velocity change characteristic curve, including: The start and end positions of the local blind zone at the boundary are located by cross-sectional coordinates, and the width range of the blind zone is determined. Measured data from three continuous effective velocity units before and after the local blind zone are extracted to form a velocity trend dataset. Trend fitting is performed on the velocity trend dataset to obtain the slope parameter of velocity change with position. A continuous velocity change characteristic curve is constructed based on the slope parameter.

[0055] In this embodiment, based on the completed blind zone velocity data, the boundary local blind zone is retrieved and located using the cross-sectional position coordinates. Using the cross-sectional boundary as a reference, the starting and ending coordinates of the local blind zone are determined according to the coordinate range centrally distributed in the invalid data. The horizontal distance between the two points is calculated as the blind zone width, which is controlled within the range of 1-5m. Measured data from three consecutive effective velocity units in front of and behind the local blind zone are extracted from the completed blind zone velocity data. The effective data in front consists of the three data points adjacent to the left of the blind zone's starting coordinate, and the effective data in the rear consists of the three data points adjacent to the right of the blind zone's ending coordinate. All data contain complete cross-sectional position coordinates and corresponding velocity values, and are arranged in ascending order of cross-sectional position coordinates to form a velocity trend dataset.

[0056] For example, suppose a local blind zone is formed near the right boundary of the cross-section due to floating debris. The starting coordinates are determined to be X=48.0m and the ending coordinates to be X=50.0m, with a blind zone width of 2.0m. The flow velocities at X=46.5m, 47.0m, and 47.5m in front of the blind zone are extracted to be 0.9m / s, 0.8m / s, and 0.7m / s, respectively, and the flow velocities at X=50.5m, 51.0m, and 51.5m behind the blind zone are extracted to be 0.6m / s, 0.5m / s, and 0.4m / s, respectively, forming six sets of "cross-section location coordinates - flow velocity" trend datasets. The specific coordinates, flow velocity data, and calculation processes mentioned above are for illustrative purposes only and do not constitute a specific limitation of this invention.

[0057] It should be noted that a coordinate system is established with the cross-sectional location coordinates as the horizontal axis and the flow velocity as the vertical axis. The six sets of data in the flow velocity trend dataset are sequentially marked on the coordinate system. The ratio of the change in flow velocity to the change in coordinates between adjacent data points is calculated for each set to obtain the local slope of each segment. The average of all local slopes is taken as the overall slope parameter of the flow velocity as a function of position. Based on this overall slope parameter, a continuous flow velocity change characteristic curve is constructed. Starting from the last valid data point in front of the blind zone, the flow velocity value of each coordinate point within the blind zone is calculated point by point according to the overall slope. For example, the flow velocity value at X=48.5m = the flow velocity value of the last valid data point in front + (48.5m - the coordinates of the last valid data point in front) × the overall slope. The flow velocity value at X=49.5m is calculated using the same logic. All calculated data are smoothly connected with the valid data points before and after to form a continuous flow velocity change characteristic curve.

[0058] Please see Figure 2 The flow velocity change characteristic curve is shown, illustrating the fitting process of the flow velocity change characteristic curve in the local blind zone at the boundary. Figure 2 The blind zone's starting point (48m) and ending point (50m) are marked by red dashed lines, with a width of 2.0m. Green dots represent measured effective velocity data, distributed at three locations each in front of (46.5-47.5m, velocity 0.7-0.9m / s) and behind (50.5-51.5m, velocity 0.4-0.6m / s). Yellow dots represent velocity data within the blind zone (48-50m) calculated based on slope parameters, with the velocity decreasing from 0.65m / s to 0.35m / s. The blue continuous curve represents the velocity variation characteristic curve, constructed by calculating the overall slope parameter from six effective data points. This achieves a smooth connection between the effective data before and after the blind zone, fully reconstructing the velocity distribution characteristics within the 46-52m range of the cross-section, providing a scientific basis for forming a continuous cross-sectional velocity dataset.

[0059] Based on the flow velocity change characteristic curve, the flow velocity allocation nodes of the local blind zone are divided at preset fixed intervals, and the flow velocity supplement value at each location is calculated sequentially according to the nodes. In this embodiment, based on the velocity change characteristic curve and using the cross-sectional position coordinates as a reference, velocity allocation nodes are used to divide the boundary local blind zone at preset fixed intervals of 0.2m. Each node corresponds to a unique cross-sectional position coordinate. Taking the local blind zone as the object, starting from the node corresponding to the starting coordinate of the blind zone, the velocity supplement value of each node is calculated sequentially according to the slope parameter and trend law of the velocity change characteristic curve. The calculation process strictly follows the change logic of the curve to ensure the consistency of the trend of the supplement value with the effective velocity data before and after.

[0060] For each calculated velocity supplement value for a node, the difference between this supplement value and the adjacent determined velocity data (including valid data before and after or calculated supplement values) is immediately calculated. A difference verification threshold of ±0.3 m / s is set, referencing the error standard between the boundary extension data and the actual shoreline velocity in the disclosure materials. If the difference is within the threshold range, the velocity supplement value is retained; if the difference exceeds the threshold range, the characteristic curve parameters corresponding to the node are rechecked, the calculation logic is adjusted, and the calculation is repeated until the difference meets the verification requirements.

[0061] Following the above process, the velocity of all nodes in the blind zone is supplemented sequentially. The supplemented velocity values ​​are then integrated with the valid data in the original blind zone velocity data according to the cross-sectional coordinates to form a preliminary overall velocity dataset containing complete data of the local blind zone. Moving average smoothing is then performed on the preliminary overall velocity dataset. The moving window is set to contain 5 consecutive data nodes. Starting from the beginning of the dataset, the arithmetic mean of the 5 data points in the window is calculated for each node. This mean is used to replace the original data value of the center node of the window. The smoothing process of the entire dataset is completed sequentially, and finally a complete cross-sectional velocity dataset with continuous, uninterrupted, and abrupt velocity changes is formed.

[0062] The specific coordinates, velocity data, and calculation processes mentioned below are for illustrative purposes only, and this invention does not impose any specific limitations. Assuming the starting coordinate of the local blind zone is X=48.0m and the ending coordinate is X=50.0m, and 10 velocity distribution nodes are obtained by dividing the zone at 0.2m intervals, corresponding to coordinates X=48.2m, 48.4m…50.0m. Based on the characteristic curve with a slope of -0.2m / (s・m), the calculated velocity supplement value for the X=48.2m node is 0.66m / s, and the difference between this and the effective velocity value of 0.7m / s at the adjacent X=47.5m node is 0.04m / s, meeting the verification requirements. The velocity supplement value for the X=49.8m node is 0.34m / s, and the difference between this and the effective velocity value of 0.6m / s at X=50.5m is 0.26m / s, meeting the threshold requirements. After all nodes were added, a moving average smoothing process was used to smoothly transition the flow velocity in the X=48.0m to X=50.0m interval from 0.7m / s to 0.6m / s, forming a continuous cross-sectional flow velocity dataset.

[0063] Step S4 involves extracting continuous velocity data near the boundary from the initially completed velocity data, including: selecting 3-5 consecutive effective velocity units from the initially completed velocity data, sequentially from the cross-sectional boundary inwards; collecting river channel topographic mapping data and extracting the cross-sectional slope and riverbed roughness topographic parameters from the river channel topographic mapping data; fusing the cross-sectional slope and riverbed roughness topographic parameters into velocity correction coefficients, and integrating the velocity correction coefficients into the corresponding relationships to construct constraint conditions.

[0064] In this embodiment, based on the preliminary completed velocity data obtained in step S3, effective velocity data is sequentially retrieved from the boundary positions on the left and right banks of the cross section towards the inside of the cross section, and 3-5 continuous and unbroken effective velocity units are selected on each boundary side. Each effective velocity unit includes the corresponding cross section position coordinates, bank distance and velocity value. During the selection process, it is ensured that the cross section position interval between adjacent units is kept at 0.5m, which is consistent with the measurement interval of historical data.

[0065] The topographic mapping data of the current test section at a scale of 1:1000 was obtained using river topographic mapping equipment. This data includes a record of elevation information at various transverse locations of the section. The section slope and riverbed roughness parameters were extracted from the topographic mapping data. The section slope was obtained by calculating the ratio of the elevation difference between adjacent locations to the horizontal distance, and the average value was taken for each segment at 0.5m intervals. The riverbed roughness was determined by consulting relevant hydrological specifications based on the riverbed material composition. The roughness of gravel riverbeds was 0.03-0.04, and the roughness of sediment riverbeds was 0.015-0.025.

[0066] According to the weighted calculation rules, the cross-sectional slope and riverbed roughness are integrated into a velocity correction coefficient. The correction weight for cross-sectional slope is set to 0.4, and the correction weight for riverbed roughness is set to 0.6. The comprehensive correction coefficient is obtained by weighted summation, and the value of the correction coefficient is controlled between 0.8 and 1.2. This velocity correction coefficient is embedded into the established correspondence between bank distance and velocity, clarifying the correlation logic between the correction coefficient and bank distance. That is, the estimated velocity value corresponding to different bank distances is multiplied by the correction coefficient of the corresponding region, forming a correspondence between bank distance and velocity attenuation with topographic constraints.

[0067] Taking four continuous effective velocity units near the left boundary of the cross section as an example, their bank distances are 0.5m, 1.0m, 1.5m, and 2.0m, respectively, with corresponding velocity values ​​of 0.8m / s, 1.2m / s, 1.5m / s, and 1.7m / s. The terrain correction coefficient for this area is 1.05. These data are used as input values ​​and substituted into the correspondence between bank distance and velocity attenuation under terrain constraints to clarify the matching logic between input parameters and the correspondence.

[0068] Calculation nodes are divided within the start and end coordinate range of the cross-sectional blind zone at a preset fixed interval of 0.5m. Each calculation node corresponds to a unique cross-sectional position coordinate and distance from the shore. The starting shore blind zone (distance from the shore 0-0.5m) can be divided into 2 nodes, and the ending shore blind zone (distance from the shore 45-50m) can be divided into 10 nodes. The theoretical velocity value is calculated node by node through the attenuation correspondence. For example, the theoretical velocity value of the 0.25m node on the left is calculated as (0.8m / s ÷ 0.5m) × 0.25m × 1.05, which is 0.42m / s. The 45.5m node on the right is calculated as 0.6m / s according to the attenuation law of the corresponding area. All theoretical velocity values ​​are sorted according to the cross-sectional position coordinates and aligned with the preliminary completed velocity data obtained in step S2 according to the same coordinate reference to ensure that the spatial positions correspond one-to-one.

[0069] The difference between the theoretical velocity value and the preliminary completed velocity data within the overlapping area is calculated, with a reasonable range set at ±0.1 m / s. If the preliminary completed velocity at a certain location in the middle of the cross-section (20 m from the shore) is 1.6 m / s and the calculated theoretical velocity value is 1.8 m / s, the difference of 0.2 m / s exceeds the reasonable range and is marked as abnormal data. For this abnormal interval, the terrain correction coefficient and effective velocity unit data for the corresponding area are retrieved again, and the attenuation correspondence with constraints is substituted to recalculate until the difference between the result and the surrounding effective data is within a reasonable range. Finally, all the theoretical velocity values ​​without abnormalities, the preliminary completed velocity data, and the recalculated velocity data are integrated and arranged in an orderly manner from the starting shore to the ending shore according to the cross-section location coordinates to form a completed velocity data for the blind area that covers the entire cross-section and has no data gaps.

[0070] For each supplementary value calculated, a difference verification is performed with the velocity data of adjacent nodes; the velocity supplementation of all nodes in the entire blind zone is completed in sequence, and the overall velocity data after supplementation is smoothed by moving average to form a continuous cross-sectional velocity dataset.

[0071] In step S4, the flow velocity in the blind zone of the cross section is calculated by combining the correspondence between the shore distance and the flow velocity to form the flow velocity data for completing the blind zone. Specifically, the effective flow velocity unit data is used as the input value and substituted into the attenuation correspondence with constraints. The theoretical flow velocity values ​​at each position of the blind zone of the cross section are calculated step by step at preset fixed intervals. The theoretical flow velocity values ​​are aligned with the preliminary completed flow velocity data according to the cross section position. Flow velocity data with deviations exceeding a reasonable range in overlapping areas are removed, the flow velocity in that area is recalculated, and all effective flow velocity data are integrated to form the flow velocity data for completing the blind zone.

[0072] In this embodiment, the effective flow velocity unit data, number of computing nodes, and flow velocity estimation examples mentioned below are all illustrative and the present invention does not impose any specific limitations.

[0073] The compensation operation of the original hydrological survey data in step S5 is as follows: First, determine the effective data ratio of the continuous cross-sectional velocity dataset. When the effective data ratio is <70% and the complete cross-sectional flow pattern cannot be reconstructed, divide the cross-sectional velocity dataset into sub-intervals at intervals of 5-10m according to the cross-sectional width, and count the number of effective data points in each sub-interval. Select sub-intervals with ≥3 effective data points as potential vertical line candidate areas, and mark these potential vertical line candidate areas as mandatory candidate areas based on preset cross-sectional hydraulic characteristics. Determine the total number of initial vertical lines based on the cross-sectional width. Add one auxiliary vertical line to the region where the velocity gradient in the cross-sectional velocity dataset is >0.05m / (s・m). Merge two adjacent vertical lines in the region where the velocity gradient in the cross-sectional velocity dataset is <0.01m / (s・m), ultimately determining 8-25 effective velocity measurement vertical lines. Associate the completed water depth data with the velocity data of the effective velocity measurement vertical lines according to the same cross-sectional location to reconstruct the cross-sectional velocity distribution characteristics, thus completing the compensation operation for cross-sectional flow calculation.

[0074] In this embodiment, the validity of the continuous cross-sectional velocity dataset is statistically analyzed, calculated based on the proportion of the cross-sectional area covered by valid data to the entire cross-section. When the proportion is <70%, sub-intervals are divided at 5-10m intervals. For a 60m wide cross-section, 10 sub-intervals can be selected at 6m intervals. The number of valid data points is counted for each sub-interval. Sub-intervals with ≥3 valid data points are selected as potential vertical line candidate areas. Combining the measured geometry of the cross-section with historical observation records, key areas such as the main flow area, the cross-sectional contraction section, and the deep channel area are identified. The sub-intervals containing these areas are listed as mandatory candidate areas, regardless of the number of valid data points. The total number of initially selected vertical lines is determined according to the cross-sectional width: 8 lines for ≤30m, 12 lines for 30-50m, 16 lines for 50-80m, and 2 lines for every 20m increase above 80m. The velocity gradient is obtained by calculating the ratio of the velocity difference between adjacent data points to the spatial distance segment by segment. For regions with a gradient > 0.05 m / (s·m), an auxiliary vertical line is added at the center. For regions with a gradient < 0.01 m / (s·m), two adjacent initially selected vertical lines are merged, and the average velocity of the two lines is taken as the merged data. Finally, 8-25 effective velocity measurement vertical lines are determined.

[0075] The following data are for illustrative purposes only and are not intended to limit the scope of this invention. For a 60m wide cross-section, 16 vertical lines were initially selected. One additional line was added in the 0-6m interval with a gradient of 0.06m / (s・m). Two lines were merged in the 12-18m interval with a gradient of 0.009m / (s・m). The original vertical lines were retained in the 30-36m deep trench area. Finally, 15 effective velocity-measuring vertical lines were determined.

[0076] The measured geometric morphology data of the cross-section were retrieved, and the continuous full-section water depth data was precisely correlated with 15 vertical lines according to coordinates. Each vertical line corresponds to a unique water depth, elevation, and morphological parameters. Taking the vertical line velocity as the core, the velocity at each position of the cross-section was calculated point by point according to the velocity change trend of adjacent vertical lines, combined with water depth and morphological characteristics, to reconstruct the complete velocity distribution. Using the velocity-area method, the cross-section was divided into units at 0.5m intervals. The product of the unit area and the average velocity was calculated and summed to obtain the total flow rate of the cross-section, thus completing the data compensation.

[0077] Please see Figure 3 This is the interface diagram of the mobile ADCP test data compensation system. Specifically, this interface demonstrates the failure area identification function of the mobile ADCP test data compensation system. The top statistics area shows that the system collected a total of 2456 data points, of which 1823 were valid, and 4 failure areas were identified. The current processing progress is 68%, and the data quality reaches 85.2%. The left side lists the specific information of the 4 failure areas: the area ranges from 1.8m² to 25.8m². According to the size of the failure range (≤5m² is local, 5-20m² is medium, and >20m² is large), linear interpolation, spline interpolation, or data reconstruction methods are used for repair. The bar chart on the right visually shows the area distribution of each type of failure area, providing a decision basis for the interpolation verification module and ensuring the scientificity and accuracy of data compensation.

[0078] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for compensating for mobile ADCP test data, characterized in that, Includes the following steps: Step S1: Collect raw hydrological measurement data using a mobile ADCP and identify the failure areas of the raw hydrological measurement data; Step S2: Collect effective water depth data associated with the cross-sectional velocity dataset; based on the effective water depth data, fuse the water depth data obtained by ADCP and other depth measurement methods to complement each other in cross-sectional morphology, or combine with the corresponding river topographic features to complete abnormal or missing water depth intervals, so as to complete the compensation operation of the original hydrological survey data. Step S3: Select the faulty data to fill the original hydrological test data according to the faulty area range, verify the rationality of the interpolation results, and obtain the preliminary completed flow velocity data; Step S4: Based on the historical effective velocity data of the complete cross-section, construct the correspondence between the distance to the shore and the velocity attenuation. Extract continuous velocity data near the boundary from the initially completed velocity data. Combine the attenuation correspondence to estimate the velocity in the blind zone of the cross-section, forming the velocity data of the completed blind zone. Extract the velocity data of the local blind zone at the boundary from the velocity data of the completed blind zone and fit the velocity change characteristic curve. Supplement the velocity data of the local blind zone at the boundary according to the velocity change characteristic curve to form a continuous cross-sectional velocity dataset. Step S5: When the effective cross-sectional velocity data cannot reconstruct the complete cross-sectional flow pattern, the method of extracting the cross-sectional velocity vertical line is adopted. The average velocity of the vertical line of the supplemented water depth data and the effective velocity measurement vertical line is associated with the same cross-sectional position to reconstruct the cross-sectional velocity distribution characteristics, so as to complete the compensation operation of the full cross-sectional flow calculation.

2. A compensation system for mobile ADCP test data, characterized in that, For performing the compensation method for underway ADCP test data as described in claim 2, the compensation system for underway ADCP test data includes: The data acquisition module is used to collect raw hydrological measurement data through a mobile ADCP and identify areas of failure in the raw hydrological measurement data. The water depth completion module, based on various types of invalid water depth, and according to the two depth measurement modes of ADCP, uses reliable water depth data at the breakpoint as anchor points to complete the cross-sectional shape through linear interpolation or cross-sectional trend fitting. By fixing the latitude and longitude position of the cross-sectional trajectory, and using the effective water depth data as a benchmark, it combines river topographic features to eliminate abnormal or missing water depth intervals, and performs position matching with historical cross-sectional measurement results or subsequent remeasurement cross-sectional results to complete the compensation operation of hydrological measurement water depth data. The velocity interpolation supplementation module is used to initially fill the velocity failure data in the original hydrological test data by selecting the interpolation type according to the failure area range and failure type, and to verify the rationality of the interpolation results to obtain the preliminary supplemented velocity data. The boundary velocity completion module is used to construct the attenuation relationship between shore distance and velocity based on the historical valid test data of the complete cross section. It extracts continuous velocity data near the boundary from the initially completed velocity data, and calculates the velocity in the blind area on both sides of the cross section by combining the attenuation relationship, thus forming the velocity data of the completed blind area. The vertical line extraction module extracts velocity vertical lines at specific locations on the cross-section according to preset rules, extracts the average flow velocity of the velocity vertical line, and matches it with the corresponding water depth of the velocity vertical line on the cross-section to calculate the cross-sectional flow rate.

3. The compensation method for mobile ADCP test data according to claim 1, characterized in that, Step S1 includes the following steps: Identify the core data items from the test results files of different brands of ADCPs as follows: timestamp, GPS latitude and longitude, ship speed / heading, water depth, beam current velocity, east / north / vertical current velocity, surface / bottom current velocity, sampling interval, depth cell size, blind zone distance, effectiveness, bottom tracking status, and beam status.

4. The compensation method for mobile ADCP test data according to claim 2, characterized in that, In step S2, based on the effective water depth data, two complementary bathymetry methods are fused to construct the full cross-sectional morphology, and the abnormal or missing water depth intervals are filled in by combining the corresponding river channel topographic features. Specifically: For the two depth sounding methods of the underway ADCP in the test, vertical beam and bottom tracking, dual-source depth data were collected synchronously at 1m intervals at various locations of the cross section; the effective range and failure points of the two depth sounding methods were identified, and abnormal data that exceeded the reasonable fluctuation range of the historical average depth of the cross section were removed; cross-validation was performed on the effective depth data of the dual sources, the data deviation value was calculated, and consistent data with data deviation values ​​within the preset allowable threshold were extracted and marked as effective depth data; For areas where both types of sounding signals fail, topographic data of the pre-flood measured complete cross section or the post-flood remeasured cross section are collected and used as a reference frame. The reference frame is then registered with the current cross section coordinates. Using the effective water depth data as anchor points, missing or abnormal water depth intervals in the current cross section are identified. Based on the topographic undulation trend of the corresponding water depth interval of the reference frame, segmented matching is used to adjust and complete the data, controlling the water depth change gradient between adjacent anchor points within a preset reasonable range, thus forming complete water depth data consistent with the topographic features of the reference frame.

5. The compensation method for mobile ADCP test data according to claim 3, characterized in that, Step S3 includes the following steps: The blank area values ​​of the failed regions are statistically analyzed. When the blank area value is ≤5㎡, an interpolation type suitable for local data completion is selected to fill the failed data point by point. When the blank area value is between 5-20㎡, an interpolation type suitable for medium-range data completion is selected to fill the failed data in a targeted manner. When the blank area value is >20㎡, an interpolation type suitable for large-scale data reconstruction is selected to fill the failed data as a whole. After the data filling of the failed regions is completed, the gradient changes of adjacent valid data are compared to verify the interpolation results, and abnormal interpolation data with abrupt gradient changes are removed. For the failed sub-regions corresponding to abnormal data, a new suitable interpolation type is selected to execute the filling process, and finally, preliminary completed flow velocity data without obvious data gaps is obtained.

6. The compensation method for mobile ADCP test data according to claim 4, characterized in that, In step S4, the specific steps for constructing the correspondence between shore distance and velocity attenuation based on the historical effective velocity data of the complete cross-section are as follows: Collect flow velocity measurements corresponding to different bank distances from the historical effective flow velocity data of each complete cross section to form bank distance data pairs; perform fitting operations on the bank distance data pairs to establish a correspondence between bank distance and flow velocity attenuation that reflects the law of flow velocity decreasing with bank distance, and verify the reliability of the correspondence between bank distance and flow velocity attenuation through mean square error.

7. The compensation method for mobile ADCP test data according to claim 4, characterized in that, In step S4, the continuous velocity data near the boundary extracted from the initially completed velocity data includes: From the initially completed velocity data, select 3-5 consecutive effective velocity units sequentially from the cross-sectional boundary inwards; collect river topographic mapping data and extract the cross-sectional slope and riverbed roughness topographic parameters from the river topographic mapping data; integrate the cross-sectional slope and riverbed roughness topographic parameters into velocity correction coefficients, and incorporate the velocity correction coefficients into the attenuation correspondence to construct constraint conditions.

8. The compensation method for mobile ADCP test data according to claim 4, characterized in that, In step S4, the flow velocity in the blind zone of the cross-section is calculated based on the attenuation correspondence, and the flow velocity data for completing the blind zone is specifically as follows: The effective velocity unit data is used as input value and substituted into the attenuation correspondence with constraints. The theoretical velocity values ​​at each position of the cross-section blind zone are calculated step by step at preset fixed intervals. The theoretical velocity values ​​are aligned with the preliminary completed velocity data according to the cross-section position. Velocity data with deviations exceeding the reasonable range in overlapping areas are removed, the velocity in that area is recalculated, and all effective velocity data are integrated to form the velocity data for the completed blind zone.

9. The compensation method for mobile ADCP test data according to claim 4, characterized in that, Step S4 extracts the flow velocity data of the boundary local blind zone from the flow velocity data of the completed blind zone, and fits the flow velocity change characteristic curve, including: The start and end positions of the local blind zone at the boundary are located by cross-sectional coordinates, and the width range of the blind zone is determined. Measured data from three continuous effective velocity units before and after the local blind zone are extracted to form a velocity trend dataset. Trend fitting is performed on the velocity trend dataset to obtain the slope parameter of velocity change with position. A continuous velocity change characteristic curve is constructed based on the slope parameter.

10. The compensation method for mobile ADCP test data according to claim 5, characterized in that, The compensation operation for the raw hydrological survey data in step S5 is as follows: To determine the effective data percentage of continuous cross-sectional velocity datasets, when the effective data percentage is <70% and a complete cross-sectional flow pattern cannot be reconstructed, sub-intervals are divided according to the cross-sectional width of the velocity dataset at intervals of 5-10m. The number of effective data points in each sub-interval is counted. Sub-intervals with ≥3 effective data points are selected as potential vertical line candidate areas, and these potential vertical line candidate areas are marked as mandatory candidate areas based on preset cross-sectional hydraulic characteristics. The total number of initial vertical lines is determined based on the cross-sectional width. One auxiliary vertical line is added to the region where the velocity gradient in the cross-sectional velocity dataset is >0.05m / (s・m). For the region where the velocity gradient in the cross-sectional velocity dataset is <0.01m / (s・m), two adjacent vertical lines are merged, and finally 8-25 effective velocity measurement vertical lines are determined. The completed water depth data and the velocity data of the effective velocity measurement vertical lines are associated according to the same cross-sectional location to reconstruct the cross-sectional velocity distribution characteristics, thereby completing the compensation operation of the original hydrological survey data.

Images