Method, system and equipment for identifying abnormal data of water intake flow of river channel section and storage medium

Abstract

The invention provides an abnormal data identification method, system and device for water intake flow of a river channel interval and a storage medium, and belongs to the technical field of water resource management, and the method comprises the steps: obtaining interval water intake flow historical data and meteorological historical data of a current river channel interval; inputting the interval water intake flow historical data and the meteorological historical data into a pre-trained machine learning model to obtain interval water intake flow prediction data; according to the difference degree between the interval water intake flow prediction data and the interval water intake flow monitoring data, determining an abnormal date when the monitoring data is abnormal; determining any water taking user in the abnormal date and adjacent water taking users of the any water taking user, and taking the water taking flow monitoring value of the any water taking user as a to-be-verified suspected abnormal monitoring value; and performing cooperative verification on the suspected abnormal monitoring value by using the water taking flow monitoring value of the adjacent water taking user so as to identify the water taking abnormity of any water taking user. According to the invention, the efficiency and precision of abnormal data identification are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water resource management, and in particular to a method, system, equipment and storage medium for identifying abnormal data of water intake flow in a river interval. Background Art

[0002] With the development of the economy, the total demand for water resources in society continues to increase, and the amount of water taken from rivers has also increased. Under drought conditions, the river runoff is small, and the proportion of river water intake to runoff is large, which has a great impact on river navigation, power generation, ecology and other functions. Therefore, accurately identifying abnormal values ​​of river water intake and ensuring the accuracy of water intake data are of great significance for objectively analyzing the impact of river water intake on river runoff and supporting river basin management agencies to accurately allocate river water resources.

[0003] In the field of water resources management, river water intake is generally counted by river section according to the river interval, which is called river interval water intake, or "interval water intake" for short. The interval water intake is the sum of the water intake of all water users in the river interval. In the interval water intake statistics, statistical data is often abnormal due to metering facility failure, data transmission link failure, statistical errors, etc., which usually manifests as missing or jumping interval water intake. Abnormal data seriously affects water resource allocation. In order to identify abnormal data of interval water intake, visual analysis and other means are usually used. Visual analysis is to visualize data using charts and manually perform visual analysis and identification. The recognition accuracy depends on manual experience, and the probability of missed recognition and misidentification is high. In addition, due to the large amount of interval water intake data and the high update frequency, it requires a lot of manpower and is inefficient.

[0004] Therefore, a new method for identifying abnormal data of water intake flow in river intervals is urgently needed. Summary of the invention

[0005] The present invention provides a method and system for identifying abnormal data of water intake flow in a river interval, so as to solve the defects of low efficiency and precision of abnormal data identification in the prior art.

[0006] In a first aspect, the present invention provides a method for identifying abnormal data of water intake flow in a river interval, comprising: obtaining historical data of water intake flow in a current river interval and historical meteorological data; wherein the historical data of water intake flow in a current river interval includes daily monitoring values ​​of water intake flow in a preset historical time period, and the historical meteorological data includes daily meteorological information in a preset historical time period; inputting the historical data of water intake flow in a current river interval and the historical meteorological data into a pre-trained machine learning model to obtain predicted data of water intake flow in a current river interval; wherein the predicted data of water intake flow in a current river interval includes daily predicted values ​​of water intake flow in a current time period; determining an abnormal date on which abnormal monitoring data occurs according to the difference between the predicted data of water intake flow in a current river interval and the monitoring data of water intake flow in a current river interval; the monitoring data of water intake flow in a current river interval includes daily monitoring values ​​of water intake flow in a current river interval; determining any water user on an abnormal date and any adjacent water user of the water user, and using the water intake flow monitoring value of any water user as a suspected abnormal monitoring value to be verified; and using the water intake flow monitoring value of the adjacent water user to collaboratively verify the suspected abnormal monitoring value to identify the water intake abnormality of the water user.

[0007] According to the abnormal data identification method of water intake flow in a river section provided by the present invention, the method of training the machine learning model includes: obtaining a bidirectional long short-term memory network to be trained; training the bidirectional long short-term memory network using the training data of the current river section; the training data of the current river section includes historical data samples of water intake flow in the section and historical meteorological data samples, as well as corresponding predicted data samples of water intake flow in the section; and using the trained bidirectional long short-term memory network as the machine learning model.

[0008] According to the abnormal data identification method of river interval water intake flow provided by the present invention, the abnormal date when the monitoring data abnormality occurs is determined according to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data, including: calculating the first absolute value of the difference between the daily interval water intake flow prediction value and the daily interval water intake flow monitoring value; when the first absolute value of the difference is greater than a first preset difference threshold, determining that the corresponding interval water intake flow monitoring value is suspected abnormal data; and taking the date corresponding to the suspected abnormal data as the abnormal date.

[0009] According to the abnormal data identification method of water intake flow in a river section provided by the present invention, the method for determining the adjacent water intake users of any water intake user includes: setting a distance threshold; calculating the distance between any water intake user and other water intake users of the same type in the current river section; and treating other water intake users of the same type whose distance to any water intake user is less than the distance threshold as adjacent water intake users.

[0010] According to the abnormal data identification method of water intake flow in a river section provided by the present invention, the suspected abnormal monitoring value is collaboratively verified using the water intake flow monitoring value of the adjacent water intake household, including: calculating the relative deviation of the water intake flow between the water intake flow monitoring value of the water intake household on the abnormal date and the water intake flow monitoring average value; wherein the water intake flow monitoring average value is the average value of the water intake flow monitoring values ​​in a period of time adjacent to the abnormal date; based on the relative deviation of the water intake flow of the adjacent water intake households, the relative deviation of the water intake flow of any water intake household is collaboratively verified to identify the water intake abnormality of any water intake household.

[0011] According to the abnormal data identification method of water intake flow in a river section provided by the present invention, the relative deviation of water intake flow of any water intake household is collaboratively verified according to the relative deviation of water intake flow of adjacent water intake households, so as to identify the water intake abnormality of any water intake household, including: calculating the average value of the relative deviation of water intake flow of adjacent water intake households; calculating the second absolute value of the difference between the relative deviation of water intake flow of any water intake household and the average value of the relative deviation of water intake flow of adjacent water intake households; when the absolute value of the second difference is greater than the second preset difference threshold, determining that the water intake flow monitoring value of any water intake household is abnormal.

[0012] According to the abnormal data identification method of river interval water intake flow provided by the present invention, the meteorological information includes temperature information and meteorological information.

[0013] In a second aspect, the present invention further provides a system for identifying abnormal data of water intake flow in a river interval, comprising:

[0014] The historical data collection module is used to obtain the historical data of the interval water intake flow and the historical meteorological data of the current river interval; wherein the historical data of the interval water intake flow includes the monitoring value of the interval water intake flow every day within the preset historical time period, and the historical meteorological data includes the meteorological information of every day within the preset historical time period;

[0015] A prediction module, used to input the interval water intake flow historical data and the meteorological historical data into a pre-trained machine learning model to obtain interval water intake flow prediction data; wherein the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period;

[0016] The abnormal data rough detection module is used to determine the abnormal date when the monitoring data abnormality occurs according to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data; the interval water intake flow monitoring data includes the interval water intake flow monitoring value of each day in the current time period;

[0017] A module for determining adjacent water users, used to determine any water user and adjacent water users of any water user on an abnormal date, and to use the water intake flow monitoring value of any water user as a suspected abnormal monitoring value to be verified;

[0018] The abnormal data precision detection module is used to use the water intake flow monitoring values ​​of nearby water users to coordinately verify the suspected abnormal monitoring values, so as to identify the water intake anomaly of any water user.

[0019] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for identifying abnormal data of water intake flow in a river section as described in any one of the above-mentioned methods are implemented.

[0020] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for identifying abnormal data of water intake flow in a river section as described in any of the above-mentioned methods.

[0021] The abnormal data identification method and system of river interval water intake flow provided by the present invention establish a two-stage abnormal data identification scheme based on the data characteristics of river interval water intake flow. First, a machine learning model is used to roughly identify the abnormal data of interval water intake flow, and then an accurate judgment is made on the suspected abnormal data, thereby improving the efficiency and accuracy of abnormal data identification.

[0022] The technical solution provided by the present invention has low requirements on the integrity of monitoring data of water users and has strong universality for different river sections.

[0023] The abnormal data identification process of the present invention greatly reduces the degree of manual participation and improves the identification efficiency; the present invention can also perform rolling optimization and adjustment on the model parameters and the set threshold parameters to improve the recognition accuracy of the abnormal data. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 It is a flow chart of the abnormal data identification method of the water intake flow rate in the river interval provided by the present invention;

[0026] Figure 2 It is a schematic diagram of the prediction of water intake flow rate in the LSTM interval in Example 1 of the present invention;

[0027] Figure 3 This is a schematic diagram of the spatial position relationship of water users in Example 1 of the present invention;

[0028] Figure 4 It is a structural schematic diagram of the abnormal data identification system for water intake flow in a river section provided by the present invention;

[0029] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] It should be noted that, in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0032] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0033] Example 1

[0034] Combine the following Figure 1-Figure 3 The invention describes a method and system for identifying abnormal data of water intake flow in a river section provided by an embodiment of the invention.

[0035] Figure 1 FIG. 1 is a flow chart of the abnormal data identification method of the water intake flow rate in the river interval provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps:

[0036] Step 101: Obtain the interval water intake flow historical data and meteorological historical data of the current river interval; wherein the interval water intake flow historical data includes the interval water intake flow monitoring value of each day within a preset historical time period, and the meteorological historical data includes the meteorological information of each day within the preset historical time period.

[0037] The present invention collects interval water intake flow monitoring values ​​and meteorological information in a preset historical time period in order to obtain input data for a machine learning model. The accuracy and completeness of historical data are crucial to the training and prediction of the model.

[0038] For example, the present invention can query the statistical data of river interval water intake within a preset historical time period (for example, set to nearly 4 weeks) from the database, one point x per day, and replace the missing and abnormal data with the mean of the data of the adjacent two days to construct the interval water intake flow historical data X.

[0039] Optionally,

[0040] X={x1,x2,...,x i , ..., x 28}

[0041] The meteorological historical data in the present invention can be obtained by means of Internet query. For example, the daily measured maximum temperature, minimum temperature, and rainfall in a preset historical period (the past 4 weeks) are queried through a meteorological information query website, and the average of the maximum temperature and the minimum temperature is taken as the average temperature of the day.

[0042] The average temperature series H and rainfall series R are used as historical meteorological data to assist in the prediction of interval water intake flow. Specifically:

[0043] H={h1,h2,...,h i ,...h 28};

[0044] R = {r1, r2, ..., r i ,...r 28};

[0045] Among them, h i represents the average temperature on the i-th day, r i represents the rainfall on the i-th day.

[0046] Step 102: Input the interval water intake flow historical data and meteorological historical data into the pre-trained machine learning model to obtain the interval water intake flow prediction data; wherein the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period.

[0047] The machine learning model in the present invention may be a machine learning model that can process time series data and multivariate inputs, such as a random forest, a support vector machine, a neural network, or a time series specific model.

[0048] As an optional implementation, the method for training and obtaining a machine learning model of the present invention includes: obtaining a bidirectional long short-term memory network (Bi-LSTM) to be trained; using the training data of the current river section to train the bidirectional long short-term memory network; the training data includes historical data samples of interval water intake flow and historical meteorological data samples, as well as corresponding interval water intake flow prediction data samples; and using the trained bidirectional long short-term memory network as the machine learning model.

[0049] Specifically, the input of Bi-LSTM in the present invention is I=[X, H, R], which predicts the daily interval water intake flow in the current time period (such as the current week, which is a future time period compared to the preset historical time period) to obtain the interval water intake flow prediction value.

[0050] For example, the interval water intake flow prediction data is Q = {q 29 ,q 30 ...q 35}, indicating the interval water intake flow forecast value in the current time period.

[0051] Because the interval water intake flow data has a small fluctuation and high data integrity, it is easily affected by the extremely abnormal water intake flow data of individual water users. Therefore, the present invention can use historical data to establish a machine learning model to preliminarily identify suspected abnormal monitoring values. Further explanation is given below.

[0052] Step 103: Determine the abnormal date when the monitoring data is abnormal based on the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data; the interval water intake flow monitoring data includes the interval water intake flow monitoring values ​​of each day in the current time period.

[0053] Among them, the interval water intake flow monitoring value can be the data actually collected by using the monitoring equipment.

[0054] The present invention can use methods such as Z-score, box plot method or customized difference threshold to identify abnormal data and determine the corresponding abnormal date.

[0055] As an optional embodiment, the present invention determines the abnormal date when the monitoring data abnormality occurs based on the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data, including: calculating the first absolute value of the difference between the daily interval water intake flow prediction value and the daily interval water intake flow monitoring value; when the first absolute value of the difference is greater than the first preset difference threshold, determining that the corresponding interval water intake flow monitoring value is suspected abnormal data; and taking the date corresponding to the suspected abnormal data as the abnormal date.

[0056] The first preset difference threshold can be adaptively set and adjusted according to actual needs.

[0057] Step 104: Determine any water user and adjacent water users of any water user on an abnormal date, and use the water intake flow monitoring value of any water user as a suspected abnormal monitoring value to be verified.

[0058] Among them, the adjacent water users may be water users of the same industry type within a preset range of any water user.

[0059] Since there are many water users involved in the river section, the monitoring data of a single water user may have problems of insufficient integrity and large fluctuations in the monitoring data. Modeling and analysis for a single water user is extremely costly and has low accuracy. However, the water intake flow rate changes in the same industry type in the same area often have the same trend. Therefore, the present invention can use the water intake data of nearby water users to coordinately verify the water intake data of water users with large fluctuations.

[0060] As an optional embodiment, the method of the present invention for determining the adjacent water users of any water user includes: setting a distance threshold; calculating the distance between any water user and other water users of the same type (such as the same industry type) in the current river channel interval; and treating other water users of the same type whose distance to any water user is less than the distance threshold as adjacent water users.

[0061] The distance threshold can be adaptively set and adjusted according to actual needs.

[0062] Step 105: Use the water intake flow monitoring values ​​of nearby water users to coordinately verify the suspected abnormal monitoring values, so as to identify the water intake anomaly of any water user.

[0063] The present invention can select comparison analysis, correlation analysis and other methods to improve the accuracy of abnormal identification. For example, when the difference between the suspected abnormal monitoring value and the average value of the water intake flow monitoring value of the adjacent water intake household is greater than the preset difference, the suspected abnormal monitoring value is used as the identified abnormal data.

[0064] In summary, the present invention establishes a two-stage abnormal data identification scheme based on the data characteristics of the water intake flow rate in the river interval. First, a machine learning model is used to roughly identify the abnormal data of the water intake flow rate in the interval, and then an accurate judgment is made on the suspected abnormal data, thereby improving the efficiency and accuracy of abnormal data identification.

[0065] Based on the contents of the above embodiments, as an optional embodiment, the abnormal data identification method of water intake flow in a river section provided by the present invention uses the water intake flow monitoring values ​​of neighboring water intake users to coordinately verify the suspected abnormal monitoring values, including: calculating the relative deviation of the water intake flow between the water intake flow monitoring value of the water intake user on the abnormal date and the water intake flow monitoring average value; wherein the water intake flow monitoring average value is the average value of the water intake flow monitoring values ​​within a period of time adjacent to the abnormal date (for example, within the previous 4 weeks); based on the relative deviation of the water intake flow of neighboring water intake users, the relative deviation of the water intake flow of any water intake user is coordinately verified to identify the water intake abnormality of any water intake user.

[0066] Assume that the water intake flow monitoring value of water user m on an abnormal date is E m The average water flow rate monitored by water user m is E m，agv , the relative deviation of water intake flow is p m .

[0067]

[0068] The water user m may be any water user within the current river channel.

[0069] Optionally, the relative deviation of the water intake flow of any water user is compared with the relative deviation of the water intake flow of adjacent water users. If the relative deviation of the water intake flow of any water user is significantly greater than the relative deviation of the water intake flow of adjacent water users, this may indicate that the monitoring value of the water user is abnormal.

[0070] Based on the content of the above embodiment, as an optional embodiment, the present invention provides a method for collaboratively verifying the relative deviation of the water intake flow of any water user according to the relative deviation of the water intake flow of adjacent water users, so as to identify the water intake anomaly of any water user, including:

[0071] (1) Calculate the average relative deviation of water flow of nearby water users.

[0072] Assuming there are N adjacent water users, the average value of the relative deviation of the water flow of adjacent water users is calculated as follows:

[0073]

[0074] Among them, p avg is the average value of the relative deviation of water intake flow of nearby water users; pn is the relative deviation of the water intake flow of the nth adjacent water user.

[0075] (2) calculating the second absolute value of the difference between the relative deviation of the water intake flow of any water user and the average value of the relative deviation of the water intake flow of adjacent water users;

[0076] (3) When the absolute value of the second difference is greater than a second preset difference threshold, it is determined that an abnormality occurs in the water intake flow monitoring value of any water user.

[0077] The second preset difference threshold can be adaptively set and adjusted according to actual needs.

[0078] This example takes the Yangtze River Basin Three Gorges-Chenglingji section from March 2, 2024 to March 8, 2024 as an example, and uses the Bi-LSTM model to predict the interval water intake flow, with the upper and lower thresholds set at 20m 3 / s, by Figure 2 From Table 1, we can see that March 4 and March 8 exceeded the upper and lower limits respectively. Since the forecast period of March 8 was longer and the prediction accuracy was lower, March 4 was taken as an example for the second stage of anomaly identification.

[0079] Table 1

[0080]

[0081] The water flow deviation of water users in the Three Gorges-Chenglingji area on March 4 was analyzed. It was found that the water flow of a certain water user A on March 4 was much greater than the average daily water flow in the past four weeks. The average relative deviation percentage of the average daily water flow was 36.29%. The spatial data position calculation unit was used to find the neighboring water user B in the same industry as the collaborative identification water user (the position relationship is as follows Figure 3 (As shown in the figure, the average relative deviation percentage of B's ​​daily average water intake flow on March 4 was -1.3%, which was much smaller than the average relative deviation percentage of A's daily average water intake flow. Therefore, it was determined that the water intake monitoring data of water user A on March 4 was suspected to be abnormal.

[0082] In summary, the abnormal data identification method of water intake flow in a river section provided by the present invention has the following beneficial effects:

[0083] (1) The present invention establishes a two-stage abnormal data identification scheme based on the data characteristics of the water intake flow rate in the river interval. First, a machine learning model is used to roughly identify the abnormal data of the water intake flow rate in the interval, and then an accurate judgment is made on the suspected abnormal data, thereby improving the efficiency and accuracy of abnormal data identification.

[0084] (2) The present invention has low requirements on the integrity of the monitoring data of water users and has strong universality for different river sections.

[0085] (3) The abnormal data identification process of the present invention greatly reduces the degree of manual participation and improves the identification efficiency; the present invention can also perform rolling optimization and adjustment on the model parameters and the set threshold parameters to improve the recognition accuracy of abnormal data.

[0086] Example 2

[0087] Figure 4 Schematic diagram of the structure of the abnormal data identification system for water intake flow in a river interval provided by the present invention. Figure 2 As shown, the system includes: a historical data collection module 210, a prediction module 220, an abnormal data rough detection module 230, a nearby water user determination module 240 and an abnormal data fine detection module 250.

[0088] The historical data collection module 210 is used to obtain the interval water intake flow historical data and meteorological historical data of the current river interval; wherein the interval water intake flow historical data includes the interval water intake flow monitoring value of each day within a preset historical time period, and the meteorological historical data includes the meteorological information of each day within a preset historical time period;

[0089] The prediction module 220 is used to input the interval water intake flow historical data and the meteorological historical data into the pre-trained machine learning model to obtain the interval water intake flow prediction data; wherein the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period;

[0090] The abnormal data rough detection module 230 is used to determine the abnormal date when the monitoring data is abnormal according to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data; the interval water intake flow monitoring data includes the interval water intake flow monitoring value of each day in the current time period;

[0091] The adjacent water user determination module 240 is used to determine any water user and adjacent water users of any water user on an abnormal date, and use the water intake flow monitoring value of any water user as a suspected abnormal monitoring value to be verified;

[0092] The abnormal data precise detection module 250 is used to perform collaborative verification on suspected abnormal monitoring values ​​using the water intake flow monitoring values ​​of adjacent water users, so as to identify the water intake abnormality of any water user.

[0093] It should be noted that the abnormal data identification system for water intake flow in a river section provided by an embodiment of the present invention can execute the abnormal data identification method for water intake flow in a river section described in any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0094] Example 3

[0095] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the abnormal data identification method of the water intake flow rate in the river interval, the method comprising: obtaining the interval water intake flow historical data and meteorological historical data of the current river interval; wherein the interval water intake flow historical data includes the interval water intake flow monitoring value of each day within a preset historical time period, and the meteorological historical data includes the meteorological information of each day within the preset historical time period; inputting the interval water intake flow historical data and the meteorological historical data into a pre-trained machine learning model to obtain the interval water intake flow prediction data; wherein the interval water intake flow prediction data includes The predicted value of interval water intake flow for each day in the current time period; determining the abnormal date when the monitoring data abnormality occurs based on the difference between the interval water intake flow predicted data and the interval water intake flow monitoring data; the interval water intake flow monitoring data includes the interval water intake flow monitoring value for each day in the current time period; determining any water intake user and the adjacent water intake users of any water intake user on the abnormal date, and using the water intake flow monitoring value of any water intake user as the suspected abnormal monitoring value to be verified; using the water intake flow monitoring value of the adjacent water intake users to coordinately verify the suspected abnormal monitoring value, so as to identify the water intake abnormality of any water intake user.

[0096] Example 4

[0097] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the abnormal data identification method of the river interval water intake flow provided in the above-mentioned embodiments, the method comprising: obtaining the interval water intake flow historical data and meteorological historical data of the current river interval; wherein the interval water intake flow historical data includes the interval water intake flow monitoring value of each day within a preset historical time period, and the meteorological historical data includes the meteorological information of each day within the preset historical time period; inputting the interval water intake flow historical data and meteorological historical data into a pre-trained complete A machine learning model is formed to obtain interval water intake flow prediction data; wherein the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period; according to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data, the abnormal date when the monitoring data abnormality occurs is determined; the interval water intake flow monitoring data includes the interval water intake flow monitoring value for each day in the current time period; any water intake household and the adjacent water intake households of any water intake household on the abnormal date are determined, and the water intake flow monitoring value of any water intake household is used as the suspected abnormal monitoring value to be verified; the suspected abnormal monitoring value is collaboratively verified using the water intake flow monitoring value of the adjacent water intake household to identify the water intake abnormality of any water intake household.

[0098] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the abnormal data identification method of the river interval water intake flow provided in the above-mentioned embodiments, the method comprising: obtaining the interval water intake flow historical data and meteorological historical data of the current river interval; wherein the interval water intake flow historical data includes the interval water intake flow monitoring value of each day within a preset historical time period, and the meteorological historical data includes the meteorological information of each day within the preset historical time period; inputting the interval water intake flow historical data and the meteorological historical data into a pre-trained machine learning model to obtain the interval water intake flow prediction value. Data; wherein, the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period; according to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data, the abnormal date when the monitoring data abnormality occurs is determined; the interval water intake flow monitoring data includes the interval water intake flow monitoring value for each day in the current time period; any water intake household and the adjacent water intake households of any water intake household on the abnormal date are determined, and the water intake flow monitoring value of any water intake household is used as the suspected abnormal monitoring value to be verified; the suspected abnormal monitoring value is collaboratively verified using the water intake flow monitoring value of the adjacent water intake household to identify the water intake abnormality of any water intake household.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying abnormal data of water intake flow in a river interval, characterized in that: include: Obtain the interval water intake flow historical data and meteorological historical data of the current river interval; wherein the interval water intake flow historical data includes the interval water intake flow monitoring value of each day within a preset historical time period, and the meteorological historical data includes the meteorological information of each day within the preset historical time period; Input the interval water intake flow historical data and meteorological historical data into the pre-trained machine learning model to obtain the interval water intake flow prediction data; wherein the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period; According to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data, the abnormal date when the monitoring data is abnormal is determined; the interval water intake flow monitoring data includes the interval water intake flow monitoring value of each day in the current time period; Determine any water user on an abnormal date and any water user adjacent to the user, and use the water intake flow monitoring value of any water user as a suspected abnormal monitoring value to be verified; The water intake flow monitoring values ​​of nearby water users are used to coordinately verify the suspected abnormal monitoring values ​​to identify the water intake anomaly of any water user.

2. The abnormal data identification method of water intake flow rate in a river section according to claim 1 is characterized in that: The method of training the machine learning model includes: Obtain the bidirectional long short-term memory network to be trained; Using the training data of the current river section, a bidirectional long short-term memory network is trained; the training data of the current river section includes historical data samples of water intake flow in the section and historical meteorological data samples, as well as corresponding predicted data samples of water intake flow in the section; The trained bidirectional long short-term memory network is used as the machine learning model.

3. The abnormal data identification method of water intake flow rate in a river section according to claim 1 is characterized in that: According to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data, determine the abnormal date when the monitoring data is abnormal, including: Calculate the absolute value of the first difference between the daily interval water intake flow prediction value and the daily interval water intake flow monitoring value; When the absolute value of the first difference is greater than the first preset difference threshold, determining that the corresponding interval water intake flow monitoring value is suspected abnormal data; The date corresponding to the suspected abnormal data is taken as the abnormal date.

4. The abnormal data identification method of water intake flow rate in a river section according to claim 1 is characterized in that: Methods for determining the neighbouring water users of any water user include: Set distance threshold; Calculate the distance between any water user and other water users of the same type in the current river section; Other water users of the same type whose distance to any water user is less than the distance threshold are regarded as adjacent water users.

5. The abnormal data identification method of water intake flow rate in a river section according to claim 1 is characterized in that: Use the water intake flow monitoring values ​​of nearby water users to coordinate verification of suspected abnormal monitoring values, including: Calculate the relative deviation of the water intake flow rate between the water intake flow rate monitoring value of the water intake user on the abnormal date and the average water intake flow rate monitoring value; wherein the average water intake flow rate monitoring value is the average of the water intake flow rate monitoring values ​​in a period of time adjacent to the abnormal date; According to the relative deviation of water intake flow of adjacent water users, the relative deviation of water intake flow of any water user is collaboratively verified to identify water intake anomalies of any water user.

6. The abnormal data identification method of water intake flow rate in a river section according to claim 5 is characterized in that: According to the relative deviation of the water intake flow of the adjacent water users, the relative deviation of the water intake flow of any water user is collaboratively verified to identify the water intake anomaly of any water user, including: Calculate the average relative deviation of water flow of nearby water users; Calculate the second absolute value of the difference between the relative deviation of water intake flow of any water user and the average value of the relative deviation of water intake flow of adjacent water users; When the absolute value of the second difference is greater than the second preset difference threshold, it is determined that the water intake flow monitoring value of any water user is abnormal.

7. The abnormal data identification method of water intake flow rate in a river section according to claim 1 is characterized in that: The meteorological information includes temperature information and rainfall information.

8. A system for identifying abnormal data of water intake flow in a river section, characterized in that: include: The historical data collection module is used to obtain the historical data of the interval water intake flow and the historical meteorological data of the current river interval; wherein the historical data of the interval water intake flow includes the monitoring value of the interval water intake flow every day within the preset historical time period, and the historical meteorological data includes the meteorological information of every day within the preset historical time period; A prediction module, used to input the interval water intake flow historical data and the meteorological historical data into a pre-trained machine learning model to obtain interval water intake flow prediction data; wherein the interval water intake flow prediction data includes the interval water intake flow prediction value for each day in the current time period; The abnormal data rough detection module is used to determine the abnormal date when the monitoring data abnormality occurs according to the difference between the interval water intake flow prediction data and the interval water intake flow monitoring data; the interval water intake flow monitoring data includes the interval water intake flow monitoring value of each day in the current time period; A module for determining adjacent water users, used to determine any water user and adjacent water users of any water user on an abnormal date, and to use the water intake flow monitoring value of any water user as a suspected abnormal monitoring value to be verified; An abnormal data precision detection module is used to use the water intake flow monitoring values ​​of adjacent water users to perform collaborative verification on suspected abnormal monitoring values, so as to identify the water intake abnormality of any water user; The abnormal data identification system for water intake flow rate between river channels is used to execute the steps in the abnormal data identification method for water intake flow rate between river channels described in any one of claims 1-7.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the abnormal data identification method of the water intake flow rate between river sections as described in any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the abnormal data identification method of water intake flow in a river section as described in any one of claims 1 to 7 are implemented.

Images