Water gate intelligent control data real-time processing method based on edge calculation

Abstract

The invention discloses a sluice intelligent control data real-time processing method based on edge calculation, and relates to the technical field of sluice control data analysis, and the method comprises the following steps: obtaining the basic feature data of a sluice, obtaining the sluice opening degree flow relation, and collecting the related data of a water body in real time; in the water gate opening or closing process, the gate passing flow under different opening degrees is collected, data preprocessing is carried out, and basic opening degree flow data are obtained; theoretical opening flow analysis is carried out to obtain theoretical opening flow data; gate blockage analysis is carried out based on the basic opening flow data and the theoretical opening flow data, and the gate blockage condition is obtained; the method is used for solving the problems that when an existing water gate control data analysis technology is used for diagnosing and analyzing the water gate blocking condition through the lockage flow, blocking abnormity and opening abnormity cannot be accurately distinguished according to the deviation between the theoretical lockage flow and the actual lockage flow in the gate opening change process, and meanwhile the blocking condition cannot be judged.

Description

Technical Field

[0001] This invention relates to the field of sluice gate control data analysis technology, specifically a real-time processing method for intelligent sluice gate control data based on edge computing. Background Technology

[0002] Sluice gate control data analysis technology refers to a technical system that collects, processes, and analyzes various monitoring data during the operation of sluice gates, mainly monitoring water level, opening degree, and flow velocity, as well as settlement and seepage pressure. Combining hydraulic engineering theory, data science methods, and control logic, it realizes the assessment of sluice gate operation status, anomaly diagnosis, control optimization, and risk early warning. Its core objective is to transform raw monitoring data into decision support information to improve the safety, efficiency, and intelligence level of sluice gate operation.

[0003] Existing sluice gate control data analysis technologies, when diagnosing sluice gate blockage based on flow rate, often rely solely on the deviation between theoretical and actual flow rates at a fixed valve opening. This neglects the impact of opening changes on flow rate during blockage. The impact of blockage varies across different openings, and the deviation analysis at a fixed opening fails to capture this dynamic pattern, easily leading to missed diagnoses or misjudgments. Furthermore, it cannot distinguish between similar symptoms of sluice gate blockage and abnormal opening. At a fixed opening, blockage and abnormal opening may exhibit the same phenomenon of lower actual flow rates; for example, if the actual opening is less than the set value, both will show the same flow rate deviation at the fixed opening. Relying solely on a single flow rate deviation value can lead to misjudgments of the abnormality type. Therefore, existing sluice gate control data analysis technologies, when diagnosing sluice gate blockage based on flow rate, cannot accurately distinguish between abnormal blockage and abnormal opening based on the deviation between theoretical and actual flow rates during changes in gate opening, nor can they accurately determine the nature of the blockage. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. By acquiring basic characteristic data of the sluice gate, the relationship between the sluice gate opening and flow rate is obtained, and relevant water body data is collected in real time. During the opening or closing of the sluice gate, the flow rate at different opening degrees is collected and preprocessed to obtain basic opening flow rate data. Theoretical opening flow rate analysis is then performed to obtain theoretical opening flow rate data. Finally, gate blockage analysis is conducted to obtain the gate blockage status. This addresses the problem that existing sluice gate control data analysis technology, when diagnosing and analyzing sluice gate blockage based on the flow rate, cannot accurately distinguish between blockage abnormalities and opening abnormalities based on the deviation between the theoretical and actual flow rates during gate opening changes, and cannot simultaneously determine the blockage status.

[0005] To achieve the above objectives, this application provides a real-time data processing method for intelligent control of sluice gates based on edge computing, comprising the following steps:

[0006] Acquire basic characteristic data of the sluice gate, obtain the relationship between sluice gate opening and flow rate, and collect relevant water body data in real time;

[0007] During the opening or closing of the sluice gate, the flow rate at different opening degrees is collected and the data is preprocessed to obtain the basic opening flow rate data.

[0008] Theoretical opening flow analysis was conducted based on relevant water body data and the relationship between sluice gate opening and flow rate to obtain theoretical opening flow rate data;

[0009] Gate blockage analysis is performed based on basic opening flow data and theoretical opening flow data to obtain the gate blockage status.

[0010] Furthermore, acquiring the basic characteristic data of the sluice gate, obtaining the relationship between the sluice gate opening and flow rate, and collecting relevant water body data in real time includes the following sub-steps:

[0011] Obtain the gate structure parameters of the sluice gate and record them as basic characteristic data; based on the principle of energy conservation, combined with the basic characteristic data of the sluice gate, use CFD software to construct a three-dimensional water flow model for numerical simulation, and obtain the theoretical coefficient data of the functional relationship between the gate opening and the flow rate.

[0012] The theoretical coefficient data were corrected based on the actual monitoring data of the sluice gate to obtain the functional relationship between the gate opening and the flow rate, which is denoted as the sluice gate opening-flow rate relationship.

[0013] Furthermore, acquiring the basic characteristic data of the sluice gate, obtaining the relationship between the sluice gate opening and flow rate, and collecting relevant water body data in real time also includes the following sub-steps:

[0014] The water level sensor collects the water level upstream and downstream of the sluice gate in real time and calculates the water level difference between the upstream and downstream sides of the sluice gate, which is recorded as the sluice gate water level difference. At the same time, the flow velocity sensor collects the flow velocity of the water upstream of the sluice gate when it reaches the sluice gate gate, which is recorded as the sluice gate approach velocity. The collection time is recorded, and the sluice gate water level difference and the sluice gate approach velocity are marked as water body related data.

[0015] Furthermore, during the opening or closing of the sluice gate, the flow rate at different opening degrees is collected, and data preprocessing is performed to obtain the basic opening flow rate data, including the following sub-steps:

[0016] The gate opening of the sluice is evenly divided into k1 opening levels, and they are sequentially labeled as opening level 1-k1 in ascending order. Any opening level is designated as the first opening level, where k1 is the number of levels set.

[0017] When the sluice gate opening was at the first opening level, the flow rate through the sluice gate was collected by the flow sensor and recorded as the initial flow rate of the first opening level; at the same time, the current water body related data was acquired and recorded as the water body related data of the first opening level.

[0018] During the opening or closing of the sluice gate, based on the changing trend of the gate opening, the initial flow rate and water-related data corresponding to each opening level are obtained sequentially and recorded as the initial opening flow rate data and the initial opening water body data respectively.

[0019] Furthermore, during the opening or closing of the sluice gate, the flow rate at different opening degrees is collected, and data preprocessing is performed to obtain the basic opening flow rate data. This process also includes the following sub-steps:

[0020] The initial opening flow data are sorted in ascending order according to the corresponding opening level, and denoted as the first opening flow sequence. The first opening flow sequence is divided into k2 opening flow clusters using a density clustering algorithm. For any opening flow cluster, it is sorted in ascending order according to the corresponding opening level, and denoted as the first flow subsequence, where k2 is the number of opening flow clusters.

[0021] Let CQi be any initial flow rate through the gate in the first flow rate subsequence. Based on the initial opening water volume data, obtain the gate water level difference corresponding to CQi, denoted as CHi. Calculate the kinetic energy ratio CR corresponding to CQi, where CR = CQi / CHi.

[0022] Repeatedly calculate the potential energy ratio corresponding to all initial gate flow rates in the first flow subsequence to obtain the potential energy ratio sequence;

[0023] Using the initial flow rate through the gate in the first flow rate subsequence as the horizontal axis and the corresponding dynamic potential energy ratio as the vertical axis, a scatter plot of the initial flow rate through the gate and the dynamic potential energy ratio is plotted in a two-dimensional plane, denoted as the flow rate-energy ratio scatter plot. The flow rate-energy ratio scatter plot is then fitted with a smooth curve to obtain the fitted curve, denoted as RQi=f(ei).

[0024] Calculate the residual from CQi to the fitted curve, denoted as CDi, where CDi = |CQi - RQi|; repeat the calculation of the residuals from all initial gate flow rates in the first flow subsequence to the fitted curve, and calculate the average residual, denoted as PD.

[0025] Furthermore, during the opening or closing of the sluice gate, the flow rate at different opening degrees is collected, and data preprocessing is performed to obtain the basic opening flow rate data. This process also includes the following sub-steps:

[0026] The band-shaped region centered on the fitted curve and bounded by RQi±k3*PD is denoted as the normal trend band, where k3 is the set scaling factor.

[0027] For any point in the flow energy ratio scatter plot, if it is not within the normal trend band, mark the corresponding initial gate flow as abnormal; otherwise, mark it as normal. Repeat the marking process for all initial gate flows in the first flow subsequence.

[0028] For any abnormal initial gate flow, denoted as CQj1, obtain the two nearest normal data on both sides of YQj, denoted as CQj0 and CQj2 respectively, where j0 < j1 < j2. Calculate ZQj1 and replace CQj1, where ZQj = q1 * CQj0 + q2 * CQj2, q1 = (j2 - j1) / (j2 - j0), and q2 = 1 - q1.

[0029] Repeat the replacement process for all abnormal initial gate flows in the first flow subsequence. After completion, obtain the basic flow subsequence corresponding to the first flow subsequence.

[0030] Repeat the processing for all opening-degree flow clusters and merge all basic flow subsequences. After completion, obtain the basic opening-degree flow data.

[0031] Furthermore, based on the water body-related data and the relationship between the gate opening and flow rate, conduct theoretical opening-degree flow analysis. The steps to obtain the theoretical opening-degree flow data are as follows:

[0032] For the water level difference and approach velocity of the gate in the initial opening-degree water body data, perform linear fitting according to the corresponding collection times, and obtain the water level difference-time relationship and velocity-time relationship in sequence.

[0033] Then substitute all the corresponding collection times into the water level difference-time relationship and velocity-time relationship to calculate the water level difference and approach velocity of the gate, and obtain the corrected water level difference data and corrected approach velocity data in sequence.

[0034] Furthermore, based on the water body-related data and the relationship between the gate opening and flow rate, conduct theoretical opening-degree flow analysis. The steps to obtain the theoretical opening-degree flow data are as follows:

[0035] For the first opening-degree level, substitute the corresponding corrected water level difference data and corrected approach velocity data into the relationship between the gate opening and flow rate to calculate the corrected gate flow corresponding to the first opening-degree level. Repeat the process to obtain the corrected gate flows corresponding to all opening-degree levels in the water body-related data, denoted as the corrected opening-degree flow data.

[0036] And substitute the corresponding water body-related data into the relationship between the gate opening and flow rate to calculate the theoretical gate flow corresponding to the first opening-degree level. Repeat the process to obtain the theoretical gate flows corresponding to all opening-degree levels in the water body-related data, denoted as the original opening-degree flow data.

[0037] The corrected opening flow data and the original opening flow data are referred to as the theoretical opening flow data.

[0038] Furthermore, based on the basic opening flow rate data and the theoretical opening flow rate data, gate blockage analysis is performed to obtain the gate blockage status, including the following sub-steps:

[0039] The corrected opening flow data, the original opening flow data, and the theoretical opening flow data are sorted in ascending order according to the opening level, and are respectively denoted as the first theoretical sequence, the second theoretical sequence, and the first actual sequence; any data in the first theoretical sequence, the second theoretical sequence, and the first actual sequence are denoted as LWm, RWm, and SWm in sequence, where m represents the position number;

[0040] Calculate the deviation rate LRm between LWm and RWm, LRm = |LWm - RWm| / LWm; and calculate the deviation rate LSm between LWm and RWm, LSm = |LWm - SWm| / LWm;

[0041] If LSM > k4 * LRm, then SWm is judged to be abnormal. The judgment is repeated on all data in the first actual sequence of SWm to obtain the second actual sequence, where k4 is the set scaling factor.

[0042] If k5% or more of the data in the second actual sequence are abnormal, the sluice gate is judged to be abnormal; otherwise, the sluice gate is judged to be normal. Here, k5% is the set percentage.

[0043] Furthermore, based on the basic opening flow rate data and the theoretical opening flow rate data, the gate blockage analysis to obtain the gate blockage status also includes the following sub-steps:

[0044] In the event of a sluice gate malfunction, the deviation rate of all corresponding data between the first theoretical sequence and the first actual sequence is obtained and denoted as the first deviation rate sequence.

[0045] The first deviation rate sequence is divided into multiple deviation rate clusters using a density clustering algorithm. The proportion of the deviation rate cluster with the highest deviation rate to the total number of the first deviation rate sequence is obtained and denoted as PU0. If PU0 is greater than k6, the sluice gate opening is judged to be abnormal; otherwise, the sluice gate is judged to be blocked, where k6 is the set proportional coefficient.

[0046] When the sluice gate is blocked, calculate the ratio of SWm to LWm, BLm = SWm / LWm, repeatedly obtain the ratio of all corresponding data of the first theoretical sequence and the first actual sequence, and calculate the average value, denoted as BLP;

[0047] If BLP < 10%, the sluice gate is considered slightly blocked; if 10% ≤ BLP < 20%, the sluice gate is considered generally blocked; if 20% ≤ BLP, the sluice gate is considered severely blocked.

[0048] The beneficial effects of this invention are as follows: This invention obtains the basic characteristic data of the sluice gate to determine the relationship between the sluice gate opening and flow rate, and collects relevant water body data in real time; during the opening or closing of the sluice gate, it collects the flow rate at different opening degrees and performs data preprocessing to obtain basic opening degree flow rate data; based on the relevant water body data and the relationship between the sluice gate opening and flow rate, it performs theoretical opening degree flow rate analysis to obtain theoretical opening degree flow rate data; based on the basic opening degree flow rate data and the theoretical opening degree flow rate data, it performs gate blockage analysis to obtain the gate blockage status; when diagnosing and analyzing the sluice gate blockage status through the flow rate, it can accurately distinguish between blockage abnormalities and opening abnormalities based on the deviation between the theoretical flow rate and the actual flow rate during the gate opening change process, and simultaneously determine the blockage status;

[0049] This invention divides the first opening flow sequence into multiple clusters using a density clustering algorithm and performs anomaly detection using the dynamic potential energy ratio, making data analysis more stable and fault-tolerant. It captures abnormal data from the perspective of energy conversion, improving the accuracy of anomaly detection and reducing misjudgments of flow deviations caused by water level fluctuations. It linearly fits the water level difference and approach velocity according to the acquisition time, and then substitutes them into the fitting relationship to calculate the corrected flow through the gate. The advantage is that it can realize a quantitative description of the dynamic state of the water body, eliminate the deviation between the actual flow and the theoretical flow caused by environmental fluctuations, and improve the accuracy of subsequent analysis. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating the steps of the method of the present invention;

[0051] Figure 2 This is a flowchart of the data preprocessing process of the present invention;

[0052] Figure 3 This is a flowchart of the flow analysis of the opening degree of the present invention;

[0053] Figure 4 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1, please refer to Figure 1 As shown, this application provides a real-time data processing method for intelligent control of sluice gates based on edge computing, including the following steps:

[0056] Step S1 involves acquiring basic characteristic data of the sluice gate, obtaining the relationship between the sluice gate opening and flow rate, and collecting relevant water body data in real time. Step S1 includes the following sub-steps:

[0057] Step S101: Obtain the gate structure parameters of the sluice gate and record them as basic characteristic data; based on the principle of energy conservation, combined with the basic characteristic data of the sluice gate, use CFD software to construct a three-dimensional water flow model for numerical simulation, and obtain the theoretical coefficient data of the functional relationship between the gate opening and the flow rate.

[0058] Gate structural parameters generally include gate width, gate top height, gate type, and gate slot shape. These are all related to the flow rate and are the premise for building any model. The relationship between gate opening and flow rate is mainly about how geometric features affect the kinetic energy loss and velocity change of water flow. Without accurate structural data, subsequent numerical simulations and theoretical calculations cannot reflect the actual working conditions on site.

[0059] The theoretical coefficient data are corrected based on the actual monitoring data of the sluice gate to obtain the functional relationship between the gate opening and the flow rate, which is denoted as the sluice gate opening-flow rate relationship. There are often discrepancies between theory and reality. On-site correction can significantly improve the accuracy of the model, making all subsequent flow predictions and blockage determinations based on this relationship more reliable.

[0060] The water level sensor collects the water level upstream and downstream of the sluice gate in real time and calculates the water level difference between the upstream and downstream sides, which is recorded as the sluice gate water level difference. At the same time, the flow velocity sensor collects the flow velocity of the water upstream of the sluice gate when it reaches the gate, which is recorded as the approach velocity. The collection time is recorded, and the sluice gate water level difference and the approach velocity are marked as water body related data. The water level difference directly determines the driving force of the flow rate and is a key input for calculating the flow rate through the gate. The approach velocity can be used to correct local inflow conditions and correct deviations caused by backflow or local disturbances.

[0061] In practical implementation, CFD numerical simulation can reveal the quantitative relationship between gate geometry and flow rate under ideal conditions, providing a scientific initial estimate of the functional relationship between gate opening and flow rate. If the accurate functional relationship between gate opening and flow rate is known, the steps of numerical simulation and subsequent actual monitoring and correction can be omitted.

[0062] Step S2 involves collecting the flow rate at different opening degrees during the opening or closing of the sluice gate, and performing data preprocessing to obtain the basic opening flow rate data. Step S2 includes the following sub-steps:

[0063] Step S201: Divide the gate opening of the sluice gate evenly into k1 opening levels, and label them as opening level 1-k1 in ascending order. Label any one opening level as the first opening level. Here, k1 is the number of levels set. k1 can be set according to the actual application scenario, but it cannot be too small. For example, in this embodiment, the opening of 0-100% is divided into 200 opening levels.

[0064] In step S202, when the gate opening of the sluice is at the first opening level, the flow rate through the sluice is collected by the flow sensor and recorded as the initial flow rate through the first opening level; at the same time, the current water body related data is acquired and recorded as the water body related data for the first opening level; in order to ensure that the measured flow rate corresponds to the first opening level, the gate can be paused for a period of time when the gate opening is at the first opening level before opening and closing to the next opening level.

[0065] Step S203: During the opening or closing of the sluice gate, according to the changing trend of the sluice gate opening, the initial flow rate and water body related data corresponding to each opening level are obtained in sequence and recorded as the initial opening flow rate data and the initial opening water body data respectively.

[0066] For step S204, please refer to... Figure 2 As shown, the initial opening degree flow data are sorted from smallest to largest according to the corresponding opening degree level, and denoted as the first opening degree flow sequence. The first opening degree flow sequence is divided into k2 opening degree flow clusters using a density clustering algorithm. For any opening degree flow cluster, it is sorted from smallest to largest according to the corresponding opening degree level, and denoted as the first flow subsequence, where k2 is the number of opening degree flow clusters. k2 can be set according to the algorithm. Dividing into multiple clusters helps to separate the flow patterns in different opening degree intervals, so that the data within the same cluster has strong homogeneity and reduces the interference of overall data heterogeneity on the fitting.

[0067] Step S205: Denote any initial gate flow rate in the first flow rate subsequence as CQi. Based on the initial opening water body data, obtain the gate water level difference corresponding to CQi, denoted as CHi. Calculate the kinetic energy ratio CR corresponding to CQi, where CR = CQi / CHi.

[0068] When water flows through a gate, the conversion efficiency of potential energy determined by the water level difference and kinetic energy determined by the flow rate will exhibit a regular nonlinear characteristic as the gate opening changes; where potential energy ∝ water level difference * flow rate; kinetic energy ∝ flow rate * flow rate, and all kinetic energy / potential energy ∝ flow rate / water level difference; further simplification, omitting other coefficients does not affect the trend judgment, so all CR = CQi / CHi;

[0069] Step S206: Repeatedly calculate the kinetic potential energy ratio corresponding to all initial gate flow rates in the first flow rate subsequence to obtain the kinetic potential energy ratio sequence;

[0070] Step S207: Taking the initial sluice flow rate in the first flow rate subsequence as the horizontal axis and the corresponding kinetic-potential energy ratio as the vertical axis, plot a scatter diagram of the initial sluice flow rate and the kinetic-potential energy ratio in a two-dimensional plane, denoted as the flow energy ratio scatter diagram, and fit the flow energy ratio scatter diagram with a smooth curve to obtain a fitting curve, denoted as RQi = f(ei); capture the normal trend between the flow rate and the head difference under different opening conditions;

[0071] Step S208: Calculate the residual of CQi to the fitting curve, denoted as CDi, where CDi = |CQi - RQi|; repeat to calculate the residuals of all initial sluice flow rates in the first flow rate subsequence to the fitting curve, and calculate the average residual, denoted as PD; measure the degree of deviation of a single point from the overall trend, providing a statistical basis for subsequent outlier identification and threshold setting;

[0072] Step S209: Denote the strip area centered on the fitting curve and bounded by RQi ± k3*PD as the normal trend band, where k3 is the set proportionality coefficient; in this embodiment, k3 = 1.5; dynamically and adaptively distinguish natural fluctuation points from true outlier points to accurately identify outliers;

[0073] Step S210: For any point in the flow energy ratio scatter diagram, if it is not within the normal trend band, mark the corresponding initial sluice flow rate as abnormal, otherwise mark it as normal, and repeat to mark all initial sluice flow rates in the first flow rate subsequence;

[0074] Step S211: For any abnormal initial sluice flow rate, denoted as CQj1, obtain the two nearest normal data located on both sides of YQj, denoted as CQj0 and CQj2 respectively, where j0 < j1 < j2; calculate ZQj1 and replace CQj1, where ZQj = q1*CQj0 + q2*CQj2, q1 = (j2 - j1) / (j2 - j0), q2 = 1 - q1; smooth the outlier value through the weighted average of adjacent normal points, which not only retains the continuity and monotonicity of the sequence but also eliminates the destructive influence of the outlier on the overall fitting;

[0075] Step S212: Repeat to replace all abnormal initial sluice flow rates in the first flow rate subsequence, and after completion, obtain the basic flow rate subsequence corresponding to the first flow rate subsequence;

[0076] Step S213: Repeat to process all opening flow rate clusters and merge all basic flow rate subsequences, and after completion, obtain the basic opening flow rate data;

[0077] In the specific implementation process, if there are very few data in the divided opening flow clusters, such as only 1 or 2, it means that the cluster is an abnormal cluster. The data in the cluster can be directly marked as abnormal data. The normal trend band constructed based on the residual average can automatically shrink or expand according to the actual data fluctuations. It can not only tolerate natural disturbances within a reasonable range, but also efficiently capture truly outlier anomalies, providing accurate identification for anomalies and avoiding subjective bias caused by manual threshold setting.

[0078] Step S3 involves performing theoretical opening-flow analysis based on relevant water body data and the relationship between sluice gate opening and flow rate to obtain theoretical opening-flow rate data. Step S3 includes the following sub-steps:

[0079] For step S301, please refer to... Figure 3 As shown, for the water level difference and the approach velocity of the sluice gate in the initial opening water body data, linear fitting is performed according to the corresponding acquisition time, and the time relationship of the level difference and the time relationship of the velocity are obtained in sequence.

[0080] In step S302, all corresponding acquisition times are substituted into the position difference time relationship and flow velocity time relationship to calculate the sluice gate water level difference and the sluice gate approach velocity, and the corrected water level difference data and corrected approach velocity data are obtained in sequence. Random fluctuations in the original measurement are removed by fitting, and data anomalies caused by temporary sensor loss or noise are filled in. Under normal circumstances, as the sluice gate opens or closes, the sluice gate water level difference and the sluice gate approach velocity should remain basically unchanged or show a certain slight trend. However, in the actual acquisition process, the data acquired will also fluctuate due to water body fluctuations. Linear fitting can significantly reduce random fluctuations caused by sensor noise, short-term pulsation and external interference, so that the subsequent flow calculation is not misled, thereby improving the robustness and stability of the overall analysis.

[0081] Step S303: For the first opening level, based on the corresponding corrected water level difference data and corrected approach velocity data, substitute them into the sluice gate opening flow relationship to calculate the corrected gate flow corresponding to the first opening level. Repeatedly obtain the corrected gate flow corresponding to all opening levels in the water body related data and record it as the corrected opening flow data. Using the smoothed data, obtain the actual flow that each opening should have under ideal flow conditions.

[0082] Step S304: Based on the corresponding water body data, substitute the sluice gate opening and flow relationship to calculate the theoretical flow rate corresponding to the first opening level. Repeatedly obtain the theoretical flow rate corresponding to all opening levels in the water body data and record it as the original opening flow rate data. Use the unprocessed water body data to obtain the theoretical flow rate, which together with the corrected flow rate constitutes a dual reference to provide a reference for subsequent analysis.

[0083] Step S305: Record the corrected opening flow data and the original opening flow data as the theoretical opening flow data;

[0084] In the specific implementation process, outliers in the collected water data can be corrected by fitting relationships and the collection time. Using the corrected water level difference and approach velocity as inputs, the calculated theoretical flow rate through the gate is closer to the actual ideal water flow behavior. It can effectively distinguish between low flow caused by blockage and transient anomalies caused by instantaneous hydrological fluctuations, thereby improving the sensitivity and accuracy of blockage diagnosis.

[0085] Step S4 involves analyzing gate blockage based on basic opening flow data and theoretical opening flow data to determine the gate blockage status. Step S4 includes the following sub-steps:

[0086] Step S401: Sort the corrected opening flow data, the original opening flow data, and the theoretical opening flow data according to the opening level from small to large, and denot them as the first theoretical sequence, the second theoretical sequence, and the first actual sequence respectively; denot any data in the first theoretical sequence, the second theoretical sequence, and the first actual sequence as LWm, RWm, and SWm respectively, where m represents the position number;

[0087] Step S402: Calculate the deviation rate LRm between LWm and RWm, LRm = |LWm - RWm| / LWm; and calculate the deviation rate LSm between LWm and RWm, LSm = |LWm - SWm| / LWm; LRm measures the normal fluctuation difference between the original theoretical overpass flow and the corrected overpass flow, and is used to judge normal fluctuations and abnormal data; LSm reflects the deviation between the measured flow and the theoretical flow, and is used to identify possible blockages or measurement anomalies;

[0088] Step S403: If LSm > k4 * LRm, then SWm is judged to be abnormal. Repeat the judgment on all data in the first actual sequence of SWm to obtain the second actual sequence, where k4 is the set proportional coefficient; in this embodiment, k4 = 1.5; based on the fluctuation range of the theoretical flow itself, dynamically distinguish which deviations are caused by normal hydrological changes and which are abnormal data that exceed expectations, which may be blockage or abnormal opening, thereby reducing the false alarm rate;

[0089] Step S404: If k5% or more of the data in the second actual sequence is abnormal, the sluice gate is determined to be abnormal; otherwise, the sluice gate is determined to be normal. Here, k5% is a set percentage. In this embodiment, k5% = 85%, and generally k5% is not less than 80%.

[0090] Step S405: In the event of a sluice gate malfunction, obtain the deviation rate of all corresponding data between the first theoretical sequence and the first actual sequence, and record it as the first deviation rate sequence.

[0091] Step S406: The first deviation rate sequence is divided into multiple deviation rate clusters using a density clustering algorithm. The proportion of the deviation rate cluster with the highest deviation rate to the total number of items in the first deviation rate sequence is obtained, denoted as PU0. If PU0 is greater than k6, it indicates that the deviation rates of most points are similar and concentrated, and the sluice gate opening is judged to be abnormal; otherwise, the sluice gate is judged to be blocked. Here, k6 is a set proportional coefficient; in this embodiment, k6 = 0.8.

[0092] Step S407: In the case of sluice gate blockage, calculate the ratio of SWm to LWm, BLm = SWm / LWm, repeatedly obtain the ratio of all corresponding data of the first theoretical sequence and the first actual sequence, and calculate the average value, denoted as BLP; quantify the blockage phenomenon into an accurate level.

[0093] Step S408: If BLP < 10%, the sluice gate is judged to be slightly blocked; if 10% ≤ BLP < 20%, the sluice gate is judged to be generally blocked; if 20% ≤ BLP, the sluice gate is judged to be severely blocked. The classification and threshold of the blockage can be set according to the actual application scenario.

[0094] In the specific implementation process, abnormal gate opening refers to a certain error between the actual gate opening and the set opening. This will cause the deviation rate to be linear or have a stable proportion with the opening. Therefore, most of the deviation rates will be classified into one cluster by the density clustering algorithm. When the gate is blocked, the larger the gate opening, the lower the proportion of the blockage to the flow section. The deviation rate will decrease as the opening increases, and the deviation rate will be classified into different clusters.

[0095] Example 2, please refer to Figure 4 As shown, Figure 4 A schematic diagram of an electronic device is provided, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, which the processor can call. When the processor executes a computer-readable instruction, it performs steps such as those in the real-time data processing method for intelligent control of a sluice gate based on edge computing, to achieve the following functions: acquiring basic characteristic data of the sluice gate, obtaining the sluice gate opening-flow relationship, and collecting relevant water body data in real time; collecting the flow rate at different opening degrees during the sluice gate opening or closing process, and performing data preprocessing to obtain basic opening-flow data; performing theoretical opening-flow analysis based on relevant water body data and the sluice gate opening-flow relationship to obtain theoretical opening-flow data; and performing gate blockage analysis based on the basic opening-flow data and the theoretical opening-flow data to obtain the gate blockage status.

[0096] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] Example 3: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs steps such as those in the real-time data processing method for intelligent control of sluice gates based on edge computing, to achieve the following functions: acquiring basic characteristic data of the sluice gate, obtaining the sluice gate opening-flow relationship, and collecting relevant water body data in real time; collecting the flow rate at different opening degrees during the opening or closing of the sluice gate, and performing data preprocessing to obtain basic opening-flow data; performing theoretical opening-flow analysis based on relevant water body data and the sluice gate opening-flow relationship to obtain theoretical opening-flow data; and performing gate blockage analysis based on basic opening-flow data and theoretical opening-flow data to obtain the gate blockage status.

[0098] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the technical solutions described above, or the parts that contribute to the prior art, can be embodied in the form of software products. These computer software products can be stored in computer-readable storage media, such as ROM / RAM, magnetic disks, optical disks, etc., and include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0099] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A real-time data processing method for intelligent control of sluice gates based on edge computing, characterized in that, Includes the following steps: Acquire basic characteristic data of the sluice gate, obtain the relationship between sluice gate opening and flow rate, and collect relevant water body data in real time; During the opening or closing of the sluice gate, the flow rate at different opening degrees is collected and the data is preprocessed to obtain the basic opening flow rate data. Theoretical opening flow analysis was conducted based on relevant water body data and the relationship between sluice gate opening and flow rate to obtain theoretical opening flow rate data; Gate blockage analysis is performed based on basic opening flow data and theoretical opening flow data to obtain the gate blockage status.

2. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 1, characterized in that, The process of acquiring basic characteristic data of the sluice gate, obtaining the relationship between the sluice gate opening and flow rate, and collecting relevant water body data in real time includes the following sub-steps: Obtain the gate structure parameters of the sluice gate and record them as basic characteristic data; based on the principle of energy conservation, combined with the basic characteristic data of the sluice gate, use CFD software to construct a three-dimensional water flow model for numerical simulation, and obtain the theoretical coefficient data of the functional relationship between the gate opening and the flow rate. The theoretical coefficient data were corrected based on the actual monitoring data of the sluice gate to obtain the functional relationship between the gate opening and the flow rate, which is denoted as the sluice gate opening-flow rate relationship.

3. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 2, characterized in that, Obtaining the basic characteristic data of the sluice gate, obtaining the relationship between the sluice gate opening and flow rate, and collecting relevant water body data in real time also includes the following sub-steps: The water level sensor collects the water level upstream and downstream of the sluice gate in real time and calculates the water level difference between the upstream and downstream sides of the sluice gate, which is recorded as the sluice gate water level difference. At the same time, the flow velocity sensor collects the flow velocity of the water upstream of the sluice gate when it reaches the sluice gate gate, which is recorded as the sluice gate approach velocity. The collection time is recorded, and the sluice gate water level difference and the sluice gate approach velocity are marked as water body related data.

4. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 3, characterized in that, The process of collecting flow rates at different opening degrees during the opening or closing of the sluice gate, and performing data preprocessing to obtain basic opening flow rate data includes the following sub-steps: The gate opening of the sluice is evenly divided into k1 opening levels, and they are sequentially labeled as opening level 1-k1 in ascending order. Any opening level is designated as the first opening level, where k1 is the number of levels set. When the sluice gate opening was at the first opening level, the flow rate through the sluice gate was collected by the flow sensor and recorded as the initial flow rate of the first opening level; at the same time, the current water body related data was acquired and recorded as the water body related data of the first opening level. During the opening or closing of the sluice gate, based on the changing trend of the gate opening, the initial flow rate and water-related data corresponding to each opening level are obtained sequentially and recorded as the initial opening flow rate data and the initial opening water body data respectively.

5. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 4, characterized in that, Collecting flow rates at different opening degrees during the opening or closing of the sluice gate, and performing data preprocessing to obtain basic opening flow rate data, also includes the following sub-steps: The initial opening flow data are sorted from smallest to largest according to the corresponding opening level, and denoted as the first opening flow sequence; the first opening flow sequence is divided into k2 opening flow clusters using a density clustering algorithm; For any openness flow cluster, sort it in ascending order according to the corresponding openness level, and denote it as the first flow subsequence, where k2 is the number of openness flow clusters; Denote any initial flow rate through the sluice in the first flow rate subsequence as CQi. According to the initial opening water body data, obtain the water level difference of the sluice corresponding to CQi, denoted as CHi, and calculate the kinetic-potential energy ratio CR corresponding to CQi, where CR = CQi / CHi; Repeat the calculation of the kinetic-potential energy ratios corresponding to all the initial flow rates through the sluice in the first flow rate subsequence to obtain the kinetic-potential energy ratio sequence; Take the initial flow rate through the sluice in the first flow rate subsequence as the horizontal axis and the corresponding kinetic-potential energy ratio as the vertical axis, and plot a scatter diagram of the initial flow rate through the sluice and the kinetic-potential energy ratio in a two-dimensional plane, denoted as the flow energy ratio scatter diagram, and fit the flow energy ratio scatter diagram with a smooth curve to obtain the fitted curve, denoted as RQi = f(ei); Calculate the residual of CQi to the fitted curve, denoted as CDi, where CDi = |CQi - RQi|; repeat the calculation of the residuals of all the initial flow rates through the sluice in the first flow rate subsequence to the fitted curve, and calculate the average residual, denoted as PD.

6. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 5, characterized in that, Collect the flow rates through the sluice at different openings during the opening or closing process of the sluice and perform data preprocessing. The basic opening flow rate data further includes the following sub-steps: Denote the strip region centered on the fitted curve with RQi ± k3 * PD as the boundary as the normal trend band, where k3 is the set proportionality coefficient; For any point in the flow energy ratio scatter diagram, if it is not within the normal trend band, mark the corresponding initial flow rate through the sluice as abnormal, otherwise mark it as normal, and repeat the marking of all the initial flow rates through the sluice in the first flow rate subsequence; For any abnormal initial flow rate through the sluice, denoted as CQj₁, obtain the two nearest normal data on both sides of YQj, denoted as CQj₀ and CQj₂ respectively, where j₀ < j₁ < j₂; calculate ZQj₁ and replace CQj₁, where ZQj = q1 * CQj₀ + q2 * CQj₂, q1 = (j₂ - j₁) / (j₂ - j₀), q2 = 1 - q1; Repeat the replacement of all the abnormal initial flow rates through the sluice in the first flow rate subsequence. After completion, obtain the basic flow rate subsequence corresponding to the first flow rate subsequence; Repeat the processing of all the opening flow rate clusters and merge all the basic flow rate subsequences. After completion, obtain the basic opening flow rate data.

7. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 6, characterized in that, Conduct theoretical opening flow rate analysis based on the water body related data and the sluice opening flow rate relationship. The theoretical opening flow rate data includes the following sub-steps: For the water level difference of the sluice and the approach velocity of the sluice in the initial opening water body data, perform linear fitting according to the corresponding collection time, and obtain the head difference-time relationship and the velocity-time relationship in sequence; Then substitute all the corresponding collection times into the head difference-time relationship and the velocity-time relationship to calculate the water level difference of the sluice and the approach velocity of the sluice, and obtain the corrected water level difference data and the corrected approach velocity data in sequence.

8. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 7, characterized in that, Conduct theoretical opening flow rate analysis based on the water body related data and the sluice opening flow rate relationship. The theoretical opening flow rate data further includes the following sub-steps: For the first opening level, based on the corresponding corrected water level difference data and corrected approach velocity data, substitute them into the sluice gate opening flow relationship to calculate the corrected gate flow corresponding to the first opening level. Repeatedly obtain the corrected gate flow corresponding to all opening levels in the water body related data and record it as the corrected opening flow data. Based on the relevant water body data, the theoretical flow rate corresponding to the first opening level is calculated by substituting the sluice gate opening and flow rate relationship into the data. The theoretical flow rate corresponding to all opening levels in the relevant water body data is repeatedly obtained and recorded as the original opening flow rate data. The corrected opening flow data and the original opening flow data are referred to as the theoretical opening flow data.

9. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 8, characterized in that, Gate blockage analysis based on basic opening flow data and theoretical opening flow data yields the following sub-steps to determine the gate blockage situation: The corrected opening flow data, the original opening flow data, and the theoretical opening flow data are sorted in ascending order of opening level, and are respectively denoted as the first theoretical sequence, the second theoretical sequence, and the first actual sequence. In sequence, any data point in the first theoretical sequence, the second theoretical sequence, and the first actual sequence is denoted as LWm, RWm, and SWm, where m represents the position number. Calculate the deviation rate LRm between LWm and RWm, LRm = |LWm - RWm| / LWm; and calculate the deviation rate LSm between LWm and RWm, LSm = |LWm - SWm| / LWm; If LSM > k4 * LRm, then SWm is judged to be abnormal. The judgment is repeated on all data in the first actual sequence of SWm to obtain the second actual sequence, where k4 is the set scaling factor. If k5% or more of the data in the second actual sequence are abnormal, the sluice gate is judged to be abnormal; otherwise, the sluice gate is judged to be normal. Here, k5% is the set percentage.

10. The real-time data processing method for intelligent control of sluice gates based on edge computing according to claim 9, characterized in that, Gate blockage analysis based on basic opening flow data and theoretical opening flow data includes the following sub-steps to determine the gate blockage status: In the event of a sluice gate malfunction, the deviation rate of all corresponding data between the first theoretical sequence and the first actual sequence is obtained and denoted as the first deviation rate sequence. The first deviation rate sequence is divided into multiple deviation rate clusters using a density clustering algorithm. The proportion of the deviation rate cluster with the highest deviation rate to the total number of the first deviation rate sequence is obtained and denoted as PU0. If PU0 is greater than k6, the sluice gate opening is judged to be abnormal. Otherwise, the sluice gate is determined to be blocked, where k6 is the set proportional coefficient; When the sluice gate is blocked, calculate the ratio of SWm to LWm, BLm = SWm / LWm, repeatedly obtain the ratio of all corresponding data of the first theoretical sequence and the first actual sequence, and calculate the average value, denoted as BLP; If BLP < 10%, the sluice gate is considered slightly blocked; if 10% ≤ BLP < 20%, the sluice gate is considered generally blocked. If 20% ≤ BLP, the sluice gate is considered to be severely blocked.

Images