Water affair flow burr data detection method, system, device and readable medium

Abstract

The application provides a water flow burr data detection method, system, device and medium, the method comprises the following steps: obtaining water flow data, processing the water flow data according to time sequence to obtain change value data of the water flow data, and obtaining an average fluctuation coefficient according to the change value data; according to the water flow data and the average fluctuation coefficient, processing according to time sequence to obtain the fluctuation multiple of the water flow data; and determining the flow burr data according to the fluctuation multiple. Through time sequence processing and fluctuation multiple calculation on the water flow data, the burr data in the time sequence data of the pipe network flow, pressure and water quality equipment can be detected, and the accuracy is high. The burr data is found and repaired accordingly, and the false alarm rate of the monitoring point is greatly reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of water data processing, and in particular to a water flow burr data detection method, system, device and readable medium. BACKGROUND

[0002] The existing burr data detection methods mainly include the box plot method combined with historical data comparison and the classic algorithm based on density, i.e. the local outlier factor method (LOF). However, these methods have some defects in specific cases.

[0003] Burr data detection belongs to the category of time series data outlier detection, specifically referring to the value of a data point in a time series suddenly changing to the value of another point and then returning to a value similar to the first point. The box plot method uses historical data comparison to determine the upper and lower limits of the threshold of time series data, thereby screening out prominent burr data. However, the threshold selection of this method may not be accurate, resulting in too many burrs being detected, while smaller local burrs may be ignored.

[0004] The local outlier factor method is a density-based algorithm used to quantify the degree of abnormality of each data point. It calculates the relative density of data points and their neighboring points to determine the degree of abnormality, which is suitable for non-uniform density data distribution. However, when the burr is not very discrete and not particularly prominent, the degree of abnormality does not have a linear relationship with the fluctuation amplitude of the burr. This may result in some continuously mutated points being incorrectly identified as burrs. SUMMARY

[0005] An object of the present application is to provide a water flow burr data detection method, system, device and readable medium, at least to solve the problem of inaccurate analysis of monitoring data, find burr data and repair it accordingly.

[0006] To achieve the above object, some embodiments of the present application provide a water flow burr data detection method, which comprises: obtaining water flow data, processing the water flow data according to time series to obtain change value data of the water flow data, and obtaining an average fluctuation coefficient according to the change value data; processing the water flow data and the average fluctuation coefficient according to time series to obtain a fluctuation multiple of the water flow data; and determining flow burr data according to the fluctuation multiple.

[0007] Further, the average fluctuation coefficient is obtained according to the change value data, which includes change values of multiple time points, and the average fluctuation coefficient is obtained by summing and averaging all time point change values.

[0008] Further, the obtaining the fluctuation multiple of the water flow data comprises: a fluctuation multiple multi of a target time point x is numerator / denominator; when numerator<avg, numerator is assigned as avg; numerator=|y(x)-y(x-1)|+|y(x+1)-y(x)|; when denominator<avg, denominator is assigned as avg; denominator=|y(x+1)-y(x-1)|; wherein avg is the average fluctuation coefficient, y(x) is a flow value of the target time point x, y(x+1) is a flow value of a time point of a next collection period of the target time point x, and y(x-1) is a flow value of a time point of a previous collection period of the target time point x.

[0009] Further, the determining the flow glitch data according to the fluctuation multiple comprises: determining a glitch threshold according to a collection time and a collection position of the water flow data; and marking data of a target time point as flow glitch data when the fluctuation multiple of the target time point exceeds the glitch threshold.

[0010] Further, the method further comprises: detecting the flow glitch data by using a pre-trained glitch detection model.

[0011] Further, the glitch detection model is trained by using a LightGBM binary classification algorithm according to the water flow data and the fluctuation multiple.

[0012] Some embodiments of the present application also provide a water flow glitch data detection system, which comprises: a data processing module, configured to acquire water flow data, process the water flow data according to a time sequence, obtain change value data of the water flow data, and obtain an average fluctuation coefficient according to the change value data; a fluctuation multiple module, configured to process the water flow data and the average fluctuation coefficient according to the time sequence, and obtain a fluctuation multiple of the water flow data; and a flow glitch module, configured to determine flow glitch data according to the fluctuation multiple.

[0013] Some embodiments of the present application also provide a water flow glitch data detection device, which comprises: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to execute the method described above.

[0014] Some embodiments of the present application also provide a computer readable medium storing computer program instructions, which can be executed by a processor to implement the water flow glitch data detection method.

[0015] Compared with the prior art, in the scheme provided by the embodiment of the application, flow data is collected from a water monitoring point, and the data is arranged in time sequence; the water flow data is processed in time sequence, and the change value data of each time point, i.e., the difference between each data point and the previous data point, is calculated; the average fluctuation coefficient of the water flow data is calculated by using the change value data, and the fluctuation coefficient can reflect the fluctuation amplitude of the data; the water flow data is processed in time sequence again, and the fluctuation multiple of each time point is calculated according to the average fluctuation coefficient, and the fluctuation multiple is obtained by comparing the fluctuation amplitudes of the current data point and the left and right adjacent data points; according to the fluctuation multiple, it is determined which data points are flow glitch data. And the LightGBM algorithm is used to train the binary classification prediction model to filter out the glitch data points. Overall, by processing the water flow data in time sequence and calculating the fluctuation multiple, the glitch data in the pipe network flow, pressure and water quality equipment time sequence data can be detected, and the accuracy is high. Finding the glitch data and repairing it specifically can greatly reduce the false positive rate of the monitoring point. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of a water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0017] Figure 2 A flowchart of another water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0018] Figure 3 A flowchart of another water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0019] Figure 4 An effect diagram of a water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0020] Figure 5 An effect diagram of another water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0021] Figure 6 An effect diagram of another water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0022] Figure 7 An effect diagram of another water flow glitch data detection method provided by the embodiment of the application is shown in the figure.

[0023] Figure 8 A structural diagram of a water flow glitch data detection device provided by the embodiment of the application is shown in the figure. DETAILED DESCRIPTION

[0024] To make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0025] In actual water monitoring, due to various factors (such as sensor error, equipment failure, data collection anomaly, etc.), some data points may deviate significantly from the normal range, forming burr phenomenon. These burr data may affect the monitoring results, and even interfere with accurate data analysis and prediction.

[0026] To solve the above technical problems, the embodiments of the present application propose a water flow burr data detection method, as shown in Figure 1 The method comprises:

[0027] S101, obtaining water flow data, and processing the water flow data according to time sequence to obtain change value data of the water flow data. The water flow data is preprocessed data, such as collecting flow data at a certain sampling period from a certain field to obtain time sequence arranged water flow data. Taking a certain pressure station as an example, for example, DN150 pressure of XXX road, 1 minute per data (taking the maximum time data every minute), 1440 data per day. Extracting one month of data for training set and test set. The water flow data is processed by first-order difference according to time sequence, and the change value Δy x (x = 0, 1, 2, …) of all time points in a day is obtained. x+1 x

[0028] S102, obtaining an average fluctuation coefficient according to the change value data. Summing the change value data to obtain the total fluctuation value sum. Counting the number of non-zero values (excluding points with no change, i.e. points with change value of 0) in the change value data. Divide the total fluctuation value sum by the number of non-zero values to obtain the average fluctuation value. This average fluctuation value is the average fluctuation coefficient. The average fluctuation coefficient is used to measure the average fluctuation degree of data in a day, and is the basis for subsequent calculation of fluctuation multiple. In actual application, the calculation method may be adjusted or different indicators may be used to measure the fluctuation degree according to specific needs and circumstances.

[0029] Excluding zero values in Δy x , calculating the average fluctuation avg

[0030] ​​

[0031] S103, according to the water flow data and the average fluctuation coefficient, time series processing is carried out, the fluctuation multiple of the water flow data is obtained. As shown in the figure, the fluctuation multiple multi of the target time point x = numerator / denominator; Figure 2

[0032] When numerator < avg, numerator is assigned to avg;

[0033] numerator = |y(x)-y(x-1)|+|y(x+1)-y(x)|;

[0034] When denominator < avg, denominator is assigned to avg;

[0035] denominator = |y(x+1)-y(x-1)|;

[0036] Wherein, avg is the average fluctuation coefficient, y(x) is the flow value of the target time point x, y(x+1) is the flow value of the time point of the next collection period of the target time point x, y(x-1) is the flow value of the time point of the last collection period of the target time point x.

[0037] The calculation of fluctuation multiple is to prevent the numerator and the numerator from being too small, resulting in distortion of fluctuation multiple. When the denominator is less than the average value avg, the denominator is given as avg, and when the numerator is less than the average value avg, the numerator is given as avg. When the numerator and the denominator are too small, the point is not a burr point, and at this time, the numerator and the denominator are both avg, and the multi is 1. The denominator is too small to prevent the fluctuation multiple from being too large, and the given avg does not affect the burr judgment. Such processing makes the fluctuation multiple remain a certain stability and rationality in calculation, which is beneficial to obtain more reliable results in subsequent burr data judgment and detection. Through the correction of the fluctuation multiple, the distortion degree of the fluctuation multiple in extreme cases can be reduced, so that it can be more accurately judged which data points are flow burr data. This is crucial for the effectiveness and reliability of the water flow data detection method.

[0038] S104, according to the fluctuation multiple, the flow burr data is determined. According to the collection time and collection position of the water flow data, the burr threshold is determined; when the fluctuation multiple of the target time point exceeds the burr threshold, the data of the target time point is marked as flow burr data.

[0039] ​The determination of the burr threshold according to the collection time and collection location of the water flow data is to set appropriate fluctuation multiple thresholds to determine whether the flow burr data is according to different time points and location characteristics. This is done to increase the flexibility and adaptability of the algorithm, as different time points and locations may have different data fluctuation conditions.

[0040] In specific implementations, the burr threshold can be determined by the following steps: according to historical data or experimental data, analyze the fluctuation multiple distribution of different time points and locations, it may be found that some time periods or locations have higher fluctuation degree, while some time periods or locations are more stable; according to the analysis result, set the burr threshold of different time points and locations, for example, for high fluctuation time period or location, set a higher burr threshold, while for stable time period or location, set a lower burr threshold; determine the judgment condition of fluctuation multiple exceeding burr threshold, usually, when the fluctuation multiple of the target time point exceeds the set burr threshold, mark the data of the time point as flow burr data. According to the actual data and experimental results, the burr threshold is constantly optimized and adjusted to maintain the accuracy and stability of the method. Through the above steps, appropriate burr threshold can be set for water flow data of different time points and locations according to specific circumstances and needs, so as to more accurately determine the flow burr data. Such flow burr data detection method can provide higher precision and reliability in practical application.

[0041] In some embodiments of the present application, the flow burr data is detected by a pre-trained burr detection model. The burr detection model can use the following algorithms: XGBoost algorithm, Random Forest algorithm, Decision Tree algorithm, GBDT algorithm and LightGBM algorithm, etc. to process massive water flow data and flow burr data.

[0042] In some embodiments of the present application, the burr detection model is trained according to the water flow data and the fluctuation multiple using the LightGBM binary classification algorithm.

[0043] LightGBM is an improved version of the gradient boosting tree (GBDT) algorithm, which has high efficiency in processing massive data. LightGBM algorithm has the following characteristics: histogram algorithm, LightGBM uses histogram algorithm to find the best split point, thereby accelerating the tree growth process; Leaf-wise tree growth strategy, compared with the traditional level-wise tree growth strategy, Leaf-wise strategy selects the leaf node that causes the maximum loss, further improving the training speed; GOSS is a sampling method in LightGBM, which can effectively reduce the sample size while maintaining the gradient information of the sample, further improving the model training efficiency; EFB is a feature selection method in LightGBM, which is used to improve the processing efficiency of categorical features; efficient parallelism, LightGBM supports multi-thread parallel computing, which can better utilize the computing resources of multi-core CPU.

[0044] As shown in Figure 3 , according to the fluctuation multiple multi (feature value) of the site one-month time series data at each collection time point (every minute), all time series data (excluding data with multi equal to 1) are labeled, and the burr data is labeled Y and the non-burr data is labeled N. The LightGBM binary classification model randomly selects 80% of the data as the training set and 20% of the data as the test set, trains the model, and predicts whether the data at the collection time point is burr or not through the feature value multi.

[0045] Since LightGBM algorithm performs well in processing massive data, it is a reasonable choice to use LightGBM as a burr detection model in water flow burr data detection method. Through the pre-trained LightGBM model, it can be applied to the actual data set, so as to quickly and accurately detect the flow burr data, improve the quality and accuracy of water data. Of course, in practical application, other machine learning algorithms or deep learning models can also be tried, and the most suitable model can be selected according to the data characteristics and requirements to detect burr data.

[0046] The implementation details of the water flow burr data detection method of the embodiments of the present application will be described in detail below in conjunction with a specific application example. The following content only provides implementation details for easy understanding, and is not necessary for implementing the scheme.

[0047] As shown in Figure 4 , for the instantaneous flow burr detection of a site, the recorded data is collected from a water department. The data set is recorded from x month x day, 2023 to x month x day, 2023, with a sampling interval of 1 minute, and the data set contains 75418 record data points. The burr point is relatively obvious and sparse.

[0048] Step 1: data preprocessing

[0049] First, the data set is preprocessed, and there is more than one data point per minute for the instantaneous flow point. The maximum time data per minute is selected as the current minute data and is de-duplicated. There are 1440 data points in a day, and a total of 28800 data points in 20 days. The processed effect diagram is shown in Figure 5 .

[0050] Step 2: First-order difference and calculate the fluctuation multiple of each time point

[0051] Loop through the time points, calculate the fluctuation multiple of the time points, and assume that a certain time point n has a value y(n). The previous collection period time point n-1 has a value y(n-1), and the next collection period time point is n+1, which has a value y(n+1). According to the above fluctuation multiple calculation method, the fluctuation multiple of each time point is obtained and saved.

[0052] Step 3: Label the burr points

[0053] In step 2, filter out the data with a fluctuation multiple of 1, because these data are definitely not burrs. In the remaining data, according to the burr threshold, mark the time points with a fluctuation multiple exceeding the burr threshold as Y, and mark the time points with a fluctuation multiple not exceeding the burr threshold as N.

[0054] Step 4: LightGBM binary classification training and verification

[0055] The data set in step 3 is classified according to the 80-20 rule. 80% of the data is used as the training set, and 20% of the data is used as the test set. Print the accuracy and confusion matrix, as shown in Figure 6 , the accuracy is 99%, 820 of the 821 non-burr points in the test set are predicted as non-burr points, and 1 is predicted as a burr point. The 28 burr points are all predicted correctly. The effect diagram after removing the burr points is shown in Figure 7 , which obviously eliminates the burr points and makes the flow data more accurate.

[0056] The application provides a water flow burr data detection method, which can effectively identify and screen out burr points in monitoring data. Burr data is usually an abnormal point or error data that occurs in the monitoring process, which may be caused by sensor failure, transmission problems or other abnormal reasons. By identifying and screening out these burr points, the quality of the data can be significantly improved, making the monitoring data more accurate and reliable. The method adopts a multi-step processing procedure, including first-order difference, fluctuation multiple calculation and LightGBM algorithm binary classification model prediction. Such comprehensive processing can effectively reduce the misjudgment rate and improve the accuracy of burr point detection. After removing the burr points, the monitoring data is more reliable, which helps to improve the reliability of water management and decision-making. Accurate monitoring data can provide more reliable basis for management decision-making, thereby optimizing the operation and maintenance of the water system. The method uses machine learning algorithm LightGBM to predict and classify burr points, so that the entire data processing process can be automated. This can save human resources and time costs, and adapt to the needs of large-scale data processing. By identifying flow burr data in a timely manner, equipment failure or abnormalities can be detected earlier, thereby speeding up the maintenance response speed and reducing the impact of possible losses and failures. Overall, this water flow burr data detection method provides more reliable data support for water management and decision-making by effectively identifying and screening out burr points in monitoring data, improving data quality and accuracy. This has a positive impact on optimizing water system operation, improving equipment maintenance efficiency and supporting decision-making.

[0057] Some embodiments of the application also provide a water flow burr data detection system, which comprises: a data processing module for obtaining water flow data, processing the water flow data according to time sequence to obtain change value data of the water flow data, and obtaining an average fluctuation coefficient according to the change value data; a fluctuation multiple module for processing the water flow data and the average fluctuation coefficient according to time sequence to obtain a fluctuation multiple of the water flow data; and a flow burr module for determining flow burr data according to the fluctuation multiple.

[0058] It is not difficult to find that the embodiments of the application are system embodiments corresponding to the method embodiments, and the implementation details of the embodiments of the application have been described in the method embodiments. To avoid repetition, they will not be described here.

[0059] In addition, the embodiments of the application also provide a water flow burr data detection device, the structure of which is shown as Figure 8 The device comprises a memory 90 for storing computer readable instructions and a processor 100 for executing computer readable instructions, wherein when the computer readable instructions are executed by the processor, the processor executes the method.

[0060] The methods and / or embodiments in this application can be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising program code for executing the methods illustrated by the flowcharts. When the computer program is executed by a processing unit, the above-mentioned functions defined in the methods of the present application are performed.

[0061] It should be noted that the computer readable medium in the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable medium may, for example, be—but is not limited to—electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus or device.

[0062] In the present application, the computer readable signal medium can include a data signal propagated in a baseband or as a part of a carrier wave, in which the computer readable program code is carried. Such propagated data signal can take many forms, including but not limited to, an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can transmit, propagate or transport program for use by or in connection with an instruction execution system, apparatus or device. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the above.

[0063] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0064] The flow diagrams and / or block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams and / or block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the block can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and

[0065] As another aspect, the embodiments of the present application also provide a computer readable medium, which can be contained in the apparatus described in the above embodiments; or can exist separately without being assembled into the apparatus. The above computer readable medium carries one or more computer readable instructions, which can be executed by a processor to implement the steps of the methods and / or the technical solutions of the above embodiments of the present application.

[0066] In a typical configuration of the present application, the apparatus of the terminal and the service network each includes one or more processors (CPU), input / output interface, network interface and memory.

[0067] Memory can include, without being limited to, non- transitory storage in computer-readable media, random access memory (RAM), and / or read-only memory (ROM), such as flash memory, etc. Memory is an example of computer-readable media.

[0068] Computer-readable media includes permanent and non- permanent, removable and non-removable media, implemented by any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Examples of storage media of a computer include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device.

[0069] In addition, the embodiments of the present application further provide a computer program stored in a computer device, so that the computer device executes the method performed by the control code.

[0070] It should be noted that the present application can be implemented in software and / or a combination of software and hardware, for example, can be implemented by using an application specific integrated circuit (ASIC), a general purpose computer or any other similar hardware device. In some embodiments, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) can be stored in a computer readable recording medium, for example, RAM memory, magnetic or optical drive or floppy disk and similar devices. In addition, some steps or functions of the present application can be implemented by hardware, for example, as a circuit cooperating with the processor to perform the respective steps or functions.

[0071] It will be obvious to a person skilled in the art that the application is not limited to the details of the above-described exemplary embodiments but can be implemented in other embodiments without departing from the scope of the application. The application is therefore not limited by the described examples but can vary within the scope of the claims and their equivalents. No reference signs in the claims should be considered as limiting the scope of the claims. Furthermore, it is obvious that the wording "comprising" does not exclude other parts and does not exclude other steps. Singularity is not excluded with respect to plurality and vice versa. Multiple units or devices recited in a device claim can also be implemented by one unit or device by means of software or hardware. The terms first, second and the like do not denote any ordering, but rather serve as names.

Claims

1. A water flow glitch data detection method, characterized in that, The method comprises: obtaining water flow data, processing the water flow data according to a time sequence to obtain change value data of the water flow data, and obtaining an average fluctuation coefficient according to the change value data; processing the water flow data and the average fluctuation coefficient according to a time sequence to obtain a fluctuation multiple of the water flow data; determining flow burr data according to the fluctuation multiple; the average fluctuation coefficient is obtained according to the change value data, which comprises: the change value data comprises change values of multiple time points, and the average fluctuation coefficient is obtained by summing and averaging the change values of all time points; the fluctuation multiple of the water flow data is obtained, which comprises: target time point x, the fluctuation multiple multi = denominator / denominator; When <avg, the assignment is avg; ； when denominator < avg, denominator is assigned as avg; ； wherein avg is the average fluctuation coefficient, y(x) is a flow value of a target time point x, y(x+1) is a flow value of a time point of a next collection period of the target time point x, and y(x-1) is a flow value of a time point of a previous collection period of the target time point x.

2. The method of claim 1, wherein, the flow burr data is determined according to the fluctuation multiple, which comprises: a burr threshold is determined according to a collection time and a collection position of the water flow data; when the fluctuation multiple of a target time point exceeds the burr threshold, data of the target time point is marked as flow burr data.

3. The detection method according to claim 1, characterized in that, The method further comprises: detecting the flow burr data by using a pre-trained burr detection model.

4. The detection method according to claim 3, characterized in that, The burr detection model is trained by using a LightBGM binary classification algorithm according to the water flow data and the fluctuation multiple.

5. A water utility flow glitch data detection system characterized by, The system comprises: a data processing module, configured to obtain water flow data, process the water flow data according to a time sequence to obtain change value data of the water flow data, and obtain an average fluctuation coefficient according to the change value data; the average fluctuation coefficient is obtained according to the change value data, which comprises: the change value data comprises change values of multiple time points, and the average fluctuation coefficient is obtained by summing and averaging the change values of all time points; a fluctuation multiple module, configured to process the water flow data and the average fluctuation coefficient according to a time sequence to obtain a fluctuation multiple of the water flow data; the fluctuation multiple of the water flow data is obtained, which comprises: Target time point x's fluctuation multiple multi = denominator / denominator; When <avg, the assignment is avg; ； when denominator < avg, denominator is assigned as avg; ； wherein avg is the average fluctuation coefficient, y(x) is a flow value of a target time point x, y(x+1) is a flow value of a time point of a next collection period of the target time point x, and y(x-1) is a flow value of a time point of a previous collection period of the target time point x; a flow burr module, configured to determine flow burr data according to the fluctuation multiple.

6. A water flow glitch detection device, characterized by, The device comprises: one or more processors; and a memory storing computer program instructions which, when executed, cause the processor to perform the method of any one of claims 1-4.

7. A computer readable medium having stored thereon computer program instructions executable by a processor to implement the method of any one of claims 1-4.

Images