Water level data compiling method and system and computer readable storage medium
By combining an anomaly detection algorithm with multiple strategies and a line-based flow propagation algorithm with intelligent classification, the problems of low efficiency, poor adaptability and insufficient accuracy in water level data compilation methods are solved, and efficient and accurate water level data compilation is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG PONSHINE INFORMATION TECH CO LTD
- Filing Date
- 2026-05-09
- Publication Date
- 2026-06-05
AI Technical Summary
Existing methods for compiling water level data are inefficient, have poor adaptability, and lack accuracy. They also have systemic flaws in terms of model assumptions, data processing logic, and environmental adaptability.
It employs a multi-strategy combination of anomaly detection algorithms, scenario-adaptive interpolation and repair algorithms, and intelligent classification and fusion-based fixed-line streaming algorithms, combined with traditional statistical models and machine learning models, to perform full-process algorithmic processing of data preprocessing, anomaly detection, interpolation and repair, data simplification, and fixed-line streaming.
It achieves efficient and accurate compilation of water level data, improves the efficiency and accuracy of data processing, and enhances the adaptability and logical consistency of the system.
Smart Images
Figure CN122153252A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of hydrological data processing technology, specifically relating to a method, system, and computer-readable storage medium for compiling water level data. Background Technology
[0002] Water level data compilation is a crucial step in hydrological monitoring, with its core objective being to transform raw observation data into systematic and analyzable results. Existing water level data compilation methods suffer from drawbacks such as low efficiency, poor adaptability, and insufficient accuracy. Furthermore, these methods exhibit systemic deficiencies in model assumptions, data processing logic, and environmental adaptability, necessitating optimization and upgrades through the integration of intelligent algorithms and multi-source data fusion technologies. Summary of the Invention
[0003] Based on the aforementioned shortcomings and deficiencies in the prior art, one of the objectives of this invention is to at least solve one or more of the aforementioned problems in the prior art. In other words, one of the objectives of this invention is to provide a water level data compilation method, system, and computer-readable storage medium that meets one or more of the aforementioned requirements.
[0004] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: A method for compiling water level data includes the following steps: S1. Collect raw water level data from different sources and preprocess the raw water level data to obtain standard water level data; S2. Determine if there is an anomaly based on the standard water level data; if yes, proceed to step S3; if no, proceed to step S4. S3. Perform interpolation repair, then proceed to step S4; S4. Data simplification is achieved by using dynamic feature point screening threshold control and feature point screening algorithm to retain key feature points in the water level data and obtain simplified water level data. S5. Based on the simplified water level data, the flow is determined and the compilation results are obtained; the compilation results include water level compilation data and flow rate compilation data.
[0005] As a preferred embodiment, the data preprocessing in step S1 includes the following steps: S21, Based on raw water level data from different sensor protocols Obtain the zero-point deviation parameters of the station and time offset factor The original water level data was calibrated to a reference standard. ; in, The standard water level data at time t is the calibrated value. This refers to the clock deviation between the sensor and standard time. Used to eliminate errors in leveling point measurements; S22, based on Sequence calculation of hydrological change identifier vector for each water level time period : ; in, The rate of change of standard water level data over time: ; Discreteness: ; is the average value of the standard water level data for the water level period, and n is the number of standard water level data sampling points within the water level period; This represents the maximum standard water level data within the specified water level period.
[0006] As a preferred embodiment, in step S2, the anomalies include missing anomalies, jump anomalies, and outlier anomalies.
[0007] As a preferred embodiment, the method for determining the missing anomaly is as follows: judge If the time interval between two adjacent standard water level data sampling points in the sequence exceeds the preset sampling frequency, then there is a missing anomaly, and straight interpolation, process line interpolation, or correlation interpolation is used for interpolation repair; otherwise, it is determined whether there is a jump anomaly.
[0008] As a preferred embodiment, the method for determining the jump anomaly is as follows: calculate Two adjacent standard water level data sampling points in the sequence and Instantaneous water level change rate : ; in, Standard water level data sampling points corresponding Standard water level data at any given time; Determine the instantaneous rate of change of water level Is it greater than the dynamic threshold? If yes, then there is a jump anomaly, which should be removed and repaired using linear interpolation, procedural interpolation, or related interpolation methods; if no, then determine whether there is an outlier anomaly. ; in, This represents the maximum normal fluctuation range of water level data for the target historical period. This is the sensitivity coefficient.
[0009] As a preferred embodiment, the method for determining outliers is as follows: Calculate standard water level data sampling points outlier severity : ; in, For two adjacent data sampling points and The sampling interval duration; Determining outlier severity Is it not less than If yes, then outliers exist, and after removal, they are repaired using linear interpolation, procedural interpolation, or correlation interpolation. If no, then the routing weights are calculated to correct the preset feature point screening threshold. The dynamic feature point filtering threshold is obtained. .
[0010] As a preferred embodiment, in step S4, the data sampling points are calculated. angular deflection rate between its adjacent data sampling points : ; in, adjacent data sampling points and Water level data vector between; adjacent data sampling points and Water level data vector between; judge Is it greater than If so, then retain the data sampling points. The corresponding standard water level data is used as the key feature point; otherwise, it is judged as redundant data and is removed and simplified. ; in, The preset adjustment coefficient, The routing weights are determined based on the complexity of water level fluctuations and the distribution density. will also Standard water level data with a rate of change exceeding a preset threshold are used as key feature points.
[0011] As a preferred approach, the water level data compilation method also includes a quality assessment of the compilation results: Consistency index is calculated based on the flow data and the benchmark flow under historical or similar water conditions, and the consistency index score is obtained. The smoothness index is calculated based on the water level compilation data, and the smoothness index score is obtained; The data integrity rate is obtained by comparing the number of data sampling points after interpolation and repair with the theoretical number of data sampling points. The quality assessment score is obtained by weighted summation of the consistency index score, the smoothness index score, and the data integrity rate. Determine if the quality assessment score is greater than the preset score threshold; if so, the compilation result meets the target requirements.
[0012] The present invention also provides a water level data compilation system, which applies the water level data compilation method described in any of the preceding embodiments, wherein the water level data compilation system includes: The data acquisition module is used to collect raw water level data from different sources; The data preprocessing module is used to preprocess the raw water level data to obtain standard water level data; The anomaly detection module is used to determine whether there are any anomalies based on standard water level data; The interpolation repair module is used for interpolation repair. The data simplification module is used to simplify the data by using dynamic feature point screening threshold control and feature point screening algorithm, retaining key feature points in the water level data to obtain simplified water level data. The alignment and flow module is used to perform alignment and flow based on simplified water level data to obtain the overall results.
[0013] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the water level data compilation method as described in any of the preceding embodiments.
[0014] Compared with the prior art, the beneficial effects of this invention are: This invention achieves efficient and accurate compilation of water level data through a full-process algorithmic processing of data preprocessing, anomaly identification, interpolation and repair, data simplification, and flow positioning. Attached Figure Description
[0015] Figure 1 This is a flowchart of the water level data compilation method according to an embodiment of the present invention; Figure 2 This is a flowchart of data preprocessing according to an embodiment of the present invention; Figure 3 This is a flowchart of anomaly identification according to an embodiment of the present invention; Figure 4 This is a flowchart of the interpolation repair process according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating data simplification in an embodiment of the present invention; Figure 6 This is an overall architecture diagram of the water level data compilation system according to an embodiment of the present invention. Detailed Implementation
[0016] To more clearly illustrate the embodiments of the present invention, specific implementation methods will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.
[0017] This invention employs a multi-strategy combination of anomaly identification algorithm, a scenario-adaptive interpolation and repair algorithm, and an intelligent classification and fusion algorithm for fixed-line streaming. By combining traditional statistical models and machine learning models, it solves the problems of low efficiency, poor adaptability, and insufficient accuracy of existing compilation methods.
[0018] Specifically, such as Figure 1 As shown, the water level data compilation method of this invention includes the following steps:
[0019] (1) Data acquisition; This invention involves deploying water level sensors and other terminals at various monitoring stations in the target area to collect raw water level data; the sensor protocols of different water level sensors may differ.
[0020] (2) Data preprocessing; This invention performs time alignment, zero-point calibration, and format conversion on raw water level data from different sources to eliminate system errors and achieve data standardization. like Figure 2 As shown, a baseline normalization process is first performed to ensure that water level data from different sources are compared on a uniform physical scale, i.e., based on raw water level data from different sensor protocols. Obtain the zero-point deviation parameters of the station and time offset factor A linear correction model was established to perform benchmark calibration on the original water level data to obtain standard water level data. sequence: ; in, The standard water level data at time t is the calibrated value. This refers to the clock deviation between the sensor and standard time. Used to eliminate errors in leveling point measurements; Then, based on Sequence calculation of hydrological change identifier vector for each water level time period This transforms continuous data into discrete feature fragments, providing a semantic basis for subsequent anomaly detection and interpolation path selection. ; in, The rate of change of standard water level data over time: ; Discreteness: ; is the average value of the standard water level data for the water level period, and n is the number of standard water level data sampling points within the water level period; This represents the maximum standard water level data within the specified water level period.
[0021] (3) Anomaly detection; The embodiments of the present invention are based on standard water level data. The sequence employs a multi-strategy combination algorithm to detect missing, jump, and outlier anomalies in the water level data, and performs hierarchical and diversion processing, such as... Figure 3 As shown, the process includes the following steps; a) Missing item exception detection: based on To check the continuity of the sequence timestamps, examine the sampling interval between two adjacent water level data sampling points. ;like Exceeding the preset sampling frequency limit The time period is directly marked as missing and abnormal, and then proceeds to step (4) for imputation and repair; if The preset sampling frequency limit has not been exceeded. If so, continue with the subsequent jump anomaly detection; b) Jump anomaly detection: based on For time-continuous data sampling points in the sequence, calculate the data between two adjacent standard water level sampling points. and Instantaneous water level change rate : ; in, Standard water level data sampling points corresponding Standard water level data at any given time; Determine the instantaneous rate of change of water level Is it greater than the dynamic threshold? If yes, it means that the water level change exceeds the physical limit allowed by the local environment, and it is automatically marked as a jump anomaly. After being removed, it proceeds to step (4) for interpolation and repair. If no, it continues to make subsequent outlier anomaly judgments. ; in, As a static benchmark, it is obtained by statistically analyzing the maximum normal fluctuation range of water level data for a historical target period (such as 10 years). Here, is the sensitivity coefficient, and is a preset empirical constant; c) Outlier Detection: Identifying outliers that have not triggered dynamic thresholds. However, the values at the data sampling points deviate significantly from the local normal trend; based on the hydrological change marker vector... Based on the severity, the following tiered processing path will be automatically executed: Calculate standard water level data sampling points outlier severity ; ; in, For two adjacent data sampling points and The sampling interval duration; Severe outlier refers to the severity of the outlier. Not less than 3 If it is determined to be an outlier, it will be removed and then proceed to step (4) for imputation repair. If the severity of outlier Less than 3 This is considered noise or a minor perturbation; the data is retained, but its weight in subsequent calculations is reduced, i.e., it is adjusted through routing weights. Suppress by calculating route weights And dynamically increase the balance coefficient In the subsequent simplification and streaming stages, noise interference is suppressed to ensure the logical consistency of the results; The above routing weights Adaptive determination: computational complexity of data segment fluctuations With distribution density Based on these two factors, the routing weights for semantic retrieval and structured matching are determined. The aforementioned fluctuation complexity λ is used to characterize the oscillation frequency and nonlinear characteristics of water level data within a preset time window. Its specific calculation process is as follows: Extract m consecutive data sampling points within the window Calculate the sum of the absolute values of the differences between adjacent sampling points and then normalize them. ; in, The larger the value, the more drastic the changes in water conditions, such as flood fluctuations caused by rainfall. In this case, the priority of semantic feature preservation will be increased. The above distribution density The specific calculation process for evaluating the compactness of the distribution of data sampling points in the amplitude space is as follows: Calculate the mean of water level data within the current window. with standard deviation ; Using the Gaussian kernel function to statistically analyze the distribution density of sampled data points : ; in, The higher the value, the more stable the water level is, the higher the data repeatability and the more obvious the structured characteristics. At this time, the weight of structured matching will be increased to achieve a higher data simplification rate. ; Strategy determined based on routing weight: During flood season, water levels fluctuate dramatically and significantly due to physical factors (such as the passage of flood peaks). The water level will rise; in this situation, the spikes generated by the sensor can easily be confused with the actual rise in water level, causing an automatic increase. exist The proportion of data in the data. If the current jump, although dramatic, conforms to the physical slope of flood evolution, i.e., the trend is reasonable, the restrictions on that point will be relaxed, and it will be judged as normal fluctuation rather than noise, thereby avoiding the accidental deletion of real flood peak data.
[0022] (4) Interpolation repair; According to the missing duration and water level change scenario, this invention uses interpolation algorithms, such as linear interpolation, process line interpolation, and correlation interpolation, to fill in the missing data and restore the continuous water level change process. Specifically, such as Figure 4 As shown, this embodiment of the invention performs scene-specific adaptive interpolation restoration based on the duration of the missing data. For scenarios involving water level changes, corresponding interpolation algorithms are used to reconstruct the continuous water level process; specifically, the duration of missing data is determined. The time is greater than the preset target duration T; if If it is a short-term missing element, linear interpolation is used for interpolation repair; if If it is a moderate missing or process line missing, process line interpolation is used for interpolation repair; if Then, the relevant interpolation method is used to fuse historical similar fragments to achieve interpolation repair; The specific process of water level restoration is as follows: ; ; in, Hydrological change identifier vector representing the current incomplete time period With the i-th historical primitive water condition change identifier vector The Euclidean distance between them; ; To repair the water level data, Extracted historical similar water condition feature primitives, The weight coefficients for each primitive are determined by the morphological fit between the known points at both ends of the missing segment and the historical primitives. The offset correction operator is used to compensate for the absolute water level deviation between the historical process line and the current station zero point. Within the time neighborhood before and after the current missing moment, multiple similar historical segments are searched and fused to ensure that the restored water level line conforms to hydraulic characteristics rather than a simple straight line connection in complex scenarios such as rising or receding water periods.
[0023] (5) Data simplification; This invention simplifies data by using dynamic feature point filtering threshold control and feature point filtering algorithm, retaining key feature points such as peaks and valleys and inflection points in the water level data, compressing the data volume, and obtaining simplified water level data. The above dynamic feature point selection threshold control and feature point selection algorithm, such as Figure 5 As shown, it specifically includes: First, through the above routing weights Real-time adjustment of preset feature point filtering threshold Generate dynamic feature point screening threshold ; ; in, This is the preset adjustment coefficient; when the water level fluctuates drastically... When the threshold for dynamic feature point filtering increases, The corresponding reduction improves the sensitivity to capturing minute inflection points and peak-valley features; Then, a second-order curvature constraint operator is introduced, and the data sampling points... angular deflection rate between its adjacent data sampling points : ; in, adjacent data sampling points and Water level data vector between; adjacent data sampling points and Water level data vector between; Finally, key feature point classification and preservation discrimination; Peak and valley point identification: judgment Is it greater than If the current water level value satisfies a local maximum or minimum, then the data sampling point is retained. The corresponding standard water level data are marked as points that must be retained, namely peak and valley points, as key feature points to ensure the integrity of the peak and valley shapes; otherwise, redundant data is identified and removed for simplification. Inflection point identification: Real-time monitoring The rate of change, when If the rate of change exceeds the preset threshold, non-extreme key points representing the turning point of acceleration or deceleration of water conditions are identified, marked as inflection points, and forcibly retained as key feature points. Finally, simplified water level data is obtained, and a simplified sequence is output. .
[0024] (6) Fixed-line flow; This invention employs an intelligent classification algorithm to categorize water level-flow relationship types, matches corresponding flow propagation models to establish water level-flow relationships, and uses simplified sequences. The flow rate data is extrapolated to obtain the compilation results; the compilation results include water level compilation data and flow rate compilation data. Specifically, a classifier is used to identify the perspective category of the current water level-flow relationship, and the corresponding flow propagation model is matched to establish the water level-flow relationship: ; in, The volume of water flowing through a river cross-section per unit time (i.e., flow rate). A comprehensive characteristic constant representing the cross-sectional shape and riverbed roughness. To determine the starting water level, Used to correct deviations caused by changes in water level or surface slope. It is a cross-sectional shape characteristic index; The logic for determining the conflict between backwater and siltation in this embodiment of the invention is as follows: In the calculation Then, by comparing the residual distribution of the current point with the historical alignment standard curve in real time, the following conflict determination is performed: Determination of scour and siltation conflict, when If a material continuously deviates from the standard curve at the same water level and the deviation exceeds the preset physical amplitude threshold, it is determined that there is a scouring and sedimentation conflict; if it deviates towards the water level axis, it is identified as sedimentation, and if it deviates towards the flow rate axis, it is identified as scouring. Backwater conflict determination: when water level It is in an upward phase, but When the growth rate lags significantly behind the normal physical trend and there is a backwater characteristic downstream of the station, it is determined to be a backwater conflict. The conflict handling priority rule in this embodiment of the invention is as follows: If the aforementioned conflict is determined to exist, the rate of change of the real-time standard water level data over time is first identified as a physical constraint; secondly, a correction coefficient is introduced for historically similar water situation primitives. or Secondary smoothing is performed to effectively solve the problem of flow multi-value, ensuring the physical and logical consistency of the flow process line under complex water conditions.
[0025] (7) Quality assessment; This invention embodiment constructs multi-dimensional evaluation indicators to perform quantitative reliability analysis on the compiled water level and flow data, and determines whether to perform automatic backtracking correction based on the scoring results; Specifically, by extracting statistical characteristics from water level and flow rate data, the following evaluation indicators are calculated: 1. Consistency Indicators (Physical logic verification): Used to assess flow The physical consistency with the water level hydrograph and historical trend patterns is calculated using the following formula: ; in, To compile output flow, This is the baseline flow rate under historical or similar water conditions; the smaller this value, the more consistent the physical logic, and N is the total number of sampling points for the target period of the quality assessment. 2. Smoothness index G (morphological feature verification): Used to assess the smoothness of water level hydrographs after data simplification and interpolation, and to identify the presence of unnatural spikes or gaps, its calculation formula is as follows: ; in, This is the second difference of the water level data corresponding to data sampling point i in the simplified sequence; if If the smoothness level is below the preset smoothness threshold, the process line is considered smooth. 3. Data completeness rate : The data integrity rate is obtained by comparing the number of data sampling points after interpolation and repair with the theoretical number of data sampling points. ; in, This represents the actual number of valid data sampling points, i.e., the number of non-missing and non-abnormal data sampling points in the sequence after preprocessing, anomaly identification, and imputation repair. The theoretical number of data sampling points is the total number of data sampling points that should be present within a given time period, calculated based on the set sampling frequency (e.g., one data point every 5 minutes or 1 hour). Then, calculate the comprehensive score S: Integrate the above evaluation indicators and generate a comprehensive score through a weighting operator: ; Where , , are the weights respectively and their sum is 1.
[0026] Quality determination and automatic backtracking mechanism of the embodiment of the present invention: Preset a score threshold R for qualified quality; Determine qualified: If S≥R, it is determined that the current整编 result meets the technical specifications, and the整编 result is directly output, that is, the整编 result table; Determine conflict / unqualified: If S<R, the backtracking mechanism is automatically triggered, and it is determined that there is a backwater, scouring and siltation conflict or algorithm parameter inaccuracy. It will backtrack to step (6) to re-match the flow pushing model, or backtrack to step (3) to re-identify anomalies until the comprehensive score meets the standard.
[0027] Based on the above water level data整编 method, the water level data整编 system of the embodiment of the present invention is as Figure 6 shown, including the following functional modules: data acquisition module, data preprocessing module, anomaly identification module, interpolation and repair module, data reduction module, line determination and flow pushing module, and quality evaluation module; The above data acquisition module is used to acquire raw water level data from different sources; The above data preprocessing module is used to perform data preprocessing on the raw water level data to obtain standard water level data; The above anomaly identification module is used to determine whether there are anomalies based on the standard water level data; The above interpolation and repair module is used to perform interpolation and repair; The above data reduction module is used to perform data reduction through dynamic feature point screening threshold control and feature point screening algorithm, retain the key feature points in the water level data, and obtain reduced water level data; The above line determination and flow pushing module is used to perform line determination and flow pushing based on the reduced water level data to obtain the整编 result; The above quality evaluation module is used to evaluate the quality of the整编 result; The specific processing process of the above functional modules can refer to the detailed description in the above water level data整编 method and will not be elaborated here.
[0028] In the computer-readable storage medium of this embodiment, instructions are stored. When the instructions run on a computer, the computer is made to execute the above water level data整编 method to achieve efficient and accurate整编 of water level data.
[0029] The above description is merely a detailed explanation of preferred embodiments and principles of the present invention. For those skilled in the art, there may be changes in specific implementation methods based on the ideas provided by the present invention, and these changes should also be considered within the scope of protection of the present invention.
Claims
1. A method for compiling water level data, characterized in that, Includes the following steps: S1. Collect raw water level data from different sources and preprocess the raw water level data to obtain standard water level data; S2. Determine if there is an anomaly based on the standard water level data; if yes, proceed to step S3; if no, proceed to step S4. S3. Perform interpolation repair, then proceed to step S4; S4. Data simplification is achieved by using dynamic feature point screening threshold control and feature point screening algorithm to retain key feature points in the water level data and obtain simplified water level data. S5. Based on the simplified water level data, the flow is determined and the compilation results are obtained; the compilation results include water level compilation data and flow rate compilation data.
2. The water level data compilation method according to claim 1, characterized in that, The data preprocessing in step S1 includes the following steps: S21, Based on raw water level data from different sensor protocols Obtain the zero-point deviation parameters of the station and time offset factor The original water level data was calibrated to a reference standard. ； in, The standard water level data at time t is the calibrated value. This refers to the clock deviation between the sensor and standard time. Used to eliminate errors in leveling point measurements; S22, based on Sequence calculation of hydrological change identifier vector for each water level time period : ； in, The rate of change of standard water level data over time: ； Discreteness: ； is the average value of the standard water level data for the water level period, and n is the number of standard water level data sampling points within the water level period; This represents the maximum standard water level data within the specified water level period.
3. The water level data compilation method according to claim 2, characterized in that, In step S2, the anomalies include missing anomalies, jump anomalies, and outlier anomalies.
4. The water level data compilation method according to claim 3, characterized in that, The method for determining the missing anomaly is as follows: judge If the time interval between two adjacent standard water level data sampling points in the sequence exceeds the preset sampling frequency, then there is a missing anomaly, and straight interpolation, process line interpolation, or correlation interpolation is used for interpolation repair; otherwise, it is determined whether there is a jump anomaly.
5. The water level data compilation method according to claim 4, characterized in that, The method for determining the jump anomaly is as follows: calculate Two adjacent standard water level data sampling points in the sequence and Instantaneous water level change rate : ； in, Standard water level data sampling points corresponding Standard water level data at any given time; Determine the instantaneous rate of change of water level Is it greater than the dynamic threshold? If yes, then there is a jump anomaly, which should be removed and repaired using linear interpolation, procedural interpolation, or related interpolation methods; if no, then determine whether there is an outlier anomaly. ； in, This represents the maximum normal fluctuation range of water level data for the target historical period. This is the sensitivity coefficient.
6. The water level data compilation method according to claim 5, characterized in that, The method for determining outliers is as follows: Calculate standard water level data sampling points outlier severity : ； in, For two adjacent data sampling points and The sampling interval duration; Determining outlier severity Is it not less than If yes, then outliers exist, and after removal, they are repaired using linear interpolation, procedural interpolation, or correlation interpolation. If no, then the routing weights are calculated to correct the preset feature point screening threshold. The dynamic feature point filtering threshold is obtained. .
7. The water level data compilation method according to claim 6, characterized in that, In step S4, the data sampling points are calculated. angular deflection rate between its adjacent data sampling points : ； in, adjacent data sampling points and Water level data vector between; adjacent data sampling points and Water level data vector between; judge Is it greater than If so, then retain the data sampling points. The corresponding standard water level data is used as the key feature point; otherwise, it is judged as redundant data and is removed and simplified. ； in, The preset adjustment coefficient, The routing weights are determined based on the complexity of water level fluctuations and the distribution density. will also Standard water level data with a rate of change exceeding a preset threshold are used as key feature points.
8. The water level data compilation method according to any one of claims 1-7, characterized in that, This also includes a quality assessment of the compilation results: Consistency index is calculated based on the flow data and the benchmark flow under historical or similar water conditions, and the consistency index score is obtained. The smoothness index is calculated based on the water level compilation data, and the smoothness index score is obtained; The data integrity rate is obtained by comparing the number of data sampling points after interpolation and repair with the theoretical number of data sampling points. The quality assessment score is obtained by weighted summation of the consistency index score, the smoothness index score, and the data integrity rate. Determine whether the quality assessment score is greater than a preset score threshold; If so, the reorganization result meets the target requirements.
9. A water level data compilation system, employing the water level data compilation method as described in any one of claims 1-8, characterized in that, The water level data compilation system includes: The data acquisition module is used to collect raw water level data from different sources; The data preprocessing module is used to preprocess the raw water level data to obtain standard water level data; The anomaly detection module is used to determine whether there are any anomalies based on standard water level data; The interpolation and repair module is used for interpolation and repair. The data simplification module is used to simplify the data by using dynamic feature point screening threshold control and feature point screening algorithm, retaining key feature points in the water level data to obtain simplified water level data. The alignment and flow module is used to perform alignment and flow based on simplified water level data to obtain the overall results.
10. A computer-readable storage medium storing instructions therein, characterized in that, When the instructions are executed on a computer, the computer performs the water level data compilation method as described in any one of claims 1-8.