Water intake monitoring abnormal data identification method, device and equipment and storage medium

Abstract

The invention discloses a water taking monitoring abnormal data identification method, device and equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing of the daily water taking amount historical monitoring data of a water taking user and the to-be-detected data of the daily water taking amount, and obtaining a daily water taking amount sequence and a daily water taking amount change rate sequence; performing iterative training based on the daily water intake quantity sequence and the daily water intake quantity change rate sequence to determine an optimal target parameter; and clustering the daily water intake quantity sequence and the daily water intake quantity change rate sequence based on the optimal target parameter to identify abnormal water intake data. According to the method, the daily water taking amount sequence and the daily water taking amount change rate sequence are determined, the requirement for historical data is low, the optimal target parameters are further calculated, manual recognition is not needed, missing detection possibly caused by manual recognition is avoided, the optimal target parameters are automatically determined, and therefore water taking abnormal data are subjected to clustering recognition, and the efficiency is improved. And the identification result of the abnormal data has relatively high accuracy.

Description

Technical Field

[0001] The present application relates to the technical field of water resource management, and more specifically, to a method, device, equipment and storage medium for identifying abnormal data in water intake monitoring. Background Art

[0002] Online monitoring and metering of water withdrawals by water users is a crucial tool for strengthening refined water resource management and building a water-saving society. Currently, water withdrawal monitoring data is prone to abnormal data during data collection, data transmission, and data processing due to equipment failures, link failures, and human factors, seriously impacting water withdrawal supervision. Therefore, accurately, efficiently, and quickly identifying abnormal water withdrawal monitoring data and ensuring its authenticity and validity are crucial for strengthening water resource supervision and improving water resource utilization efficiency.

[0003] Currently, the identification of abnormal data in national water withdrawal monitoring is primarily based on manual analysis and judgment. If the daily water withdrawal exceeds a certain percentage of the annual permitted total, it is considered abnormal data. To improve identification accuracy, research is conducted from two aspects: cluster analysis and time series prediction. The K-Means algorithm is used to cluster historical water withdrawal data to form multiple clusters. The target data is then verified to determine whether it belongs to an abnormal cluster to determine whether the data is abnormal. A least squares support vector machine model using adaptive inertia function and particle swarm optimization can also be used to reconstruct and recover abnormal data. The ARIMA model can also be used to fit the daily water withdrawal curve, and the 3σ criterion is used to identify abnormal data.

[0004] However, in the existing methods for identifying abnormal data in water intake monitoring, the accuracy of manual rule analysis depends on manual experience. The same set of rules is difficult to adapt to different types of water users, and the probability of missed identification is high; K-Means-based clustering analysis requires the pre-input of the number of categories, making it difficult to perform automatic identification; and the method based on time series prediction has high requirements for the quality of historical data, and the accuracy is poor when the water intake fluctuates greatly or the sequence is short. Summary of the Invention

[0005] In response to at least one defect or improvement need in the prior art, the present invention provides a method, device, equipment and storage medium for identifying abnormal data in water intake monitoring, which are used to solve the problems in the prior art that the abnormal data identification method for water intake monitoring relies on manual experience, has a high probability of missed identification, cluster analysis requires the pre-input of the number of categories, and is difficult to perform automatic identification, is based on time series prediction and has high requirements on the quality of historical data, and has poor accuracy when the water intake volume fluctuates greatly or the sequence is short.

[0006] To achieve the above objectives, according to a first aspect of the present invention, a method for identifying abnormal data in water intake monitoring is provided, comprising: The historical monitoring data of daily water intake of water users and the data of water intake to be tested on the same day are preprocessed to obtain the daily water intake sequence and the daily water intake change rate sequence; Iterative training is performed based on the daily water intake sequence and the daily water intake change rate sequence to determine the optimal target parameters; Based on the optimal target parameters, the daily water intake series and the daily water intake change rate series are clustered to identify abnormal water intake data.

[0007] In one possible implementation, the historical monitoring data of daily water intake of water users and the data to be detected of the water intake on the current day are preprocessed to obtain a daily water intake sequence and a daily water intake change rate sequence, including: Obtain the data on the water intake for the day to be tested, and conduct an initial inspection on the data on the water intake for the day to be tested; The historical monitoring data of daily water intake and the data to be tested of water intake on the current day are combined into a time series in chronological order and normalized to obtain a daily water intake series; The daily water intake change rate sequence is calculated based on the historical monitoring data of daily water intake and the water intake data to be detected on the current day.

[0008] In a possible implementation, obtaining the data to be tested of the water intake on the day and performing a preliminary inspection on the data to be tested of the water intake on the day include: If the data of water intake to be tested on that day is null, it will be recorded that the water intake on that day was not reported; If the water intake data to be detected on that day is greater than the preset daily water intake threshold, the water intake data to be detected on that day is recorded as being extremely large on that day.

[0009] In one possible implementation, the daily water intake change rate sequence is calculated based on the historical monitoring data of the daily water intake and the data to be detected of the water intake on the current day, including: Calculate the change rate weights for the historical monitoring data of daily water intake within a preset number of days in chronological order; The daily water intake change rate sequence is calculated based on the historical monitoring data of daily water intake, the data of water intake to be detected on the current day and the change rate weight.

[0010] In one possible implementation, iterative training is performed based on the daily water intake sequence and the daily water intake change rate sequence to determine the optimal target parameters, including: Randomly generate several initial solution sets within the preset value range of the target parameter; Calculate the particle fitness based on the initial solution set, daily water intake sequence and daily water intake change rate sequence; The particle speed and position are updated and iteratively trained to a preset number of iterations to determine the optimal target parameters.

[0011] In one possible implementation, the particle fitness is calculated based on the initial solution set, the daily water intake sequence, and the daily water intake change rate sequence, including: The initial particle fitness is calculated based on the silhouette coefficient evaluation function, the initial solution set, the daily water intake sequence and the daily water intake change rate sequence; The initial particle fitness is mapped based on the mapping rule to determine the target particle fitness.

[0012] In one possible implementation, the particle velocity and position are updated and iteratively trained to a preset number of iterations to determine the optimal target parameters, including: Update the speed and position of the particles and perform iterative training to determine the optimal solution of the particle swarm; Calculate the target fitness corresponding to rounding up and rounding down the optimal solution of the particle swarm; The optimal target parameters are determined based on the target fitness.

[0013] According to a second aspect of the present invention, there is also provided a device for identifying abnormal data of water intake monitoring, comprising: A preprocessing module is configured to preprocess the historical monitoring data of daily water intake of water users and the data to be detected of water intake on the current day to obtain a daily water intake sequence and a daily water intake change rate sequence; an iterative training module, configured to perform iterative training based on a daily water intake sequence and a daily water intake change rate sequence to determine an optimal target parameter; The clustering identification module is configured to cluster and identify abnormal water intake data on the daily water intake sequence and the daily water intake change rate sequence based on the optimal target parameters.

[0014] According to the third aspect of the present invention, a water intake monitoring abnormal data identification device is also provided, which includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit executes any step of the above-mentioned water intake monitoring abnormal data identification method.

[0015] According to the fourth aspect of the present invention, a storage medium is also provided, which stores a computer program that can be executed by an access authentication device. When the computer program runs on the access authentication device, the access authentication device executes any step of the above-mentioned water intake monitoring abnormal data identification method.

[0016] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art: The present invention provides a method for identifying abnormal data in water intake monitoring. The method first pre-processes the acquired historical monitoring data of daily water intake of water users and the data of water intake on the same day to be detected, determines the daily water intake sequence and the daily water intake change rate sequence, and uses them as the input values ​​of the optimal target parameters for clustering. The method can clearly and intuitively understand the specific values ​​of the daily water intake and the daily water intake change rate, and by processing the data, reduces the requirements for the quality of historical data. On this basis, the optimal target parameters are further calculated through iterative training, and the optimal target parameters can be automatically determined without manual identification, avoiding the missed detection phenomenon that may be caused by manual identification. Cluster identification is then performed based on the optimal target parameters determined by iterative training, thereby identifying the abnormal water intake data in the data of water intake on the same day to be detected, and the identification result of the abnormal data has high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic flow chart of an embodiment of a method for identifying abnormal data in water intake monitoring provided by the present invention; Figure 2 The present invention provides Figure 1 A flow chart of an embodiment of step S100; Figure 3 The present invention provides Figure 1 A flow chart of an embodiment of step S200; Figure 4 The present invention provides Figure 3 A flow chart of an embodiment of step S230; Figure 5 A schematic structural diagram of an embodiment of a device for identifying abnormal data in water intake monitoring provided by the present invention; Figure 6 A schematic diagram of the structure of a water intake monitoring abnormal data identification device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0020] The terms "first," "second," "third," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0021] The present invention provides a method, device, equipment and storage medium for identifying abnormal data in water intake monitoring, which are described below respectively.

[0022] See also Figure 1 , Figure 1 This is a flow chart of an embodiment of a method for identifying abnormal data in water intake monitoring provided by the present invention. In a specific embodiment of the present invention, a method for identifying abnormal data in water intake monitoring is disclosed, comprising: S100, pre-processing the historical monitoring data of daily water intake of water users and the data to be tested on the water intake of the current day to obtain a daily water intake sequence and a daily water intake change rate sequence; S200, performing iterative training based on the daily water intake sequence and the daily water intake change rate sequence to determine the optimal target parameters; S300: Clustering the daily water intake sequence and the daily water intake change rate sequence based on the optimal target parameters to identify abnormal water intake data.

[0023] In the above embodiment, historical daily water withdrawal monitoring data for all water users is collected from the water withdrawal monitoring system. This data typically includes key information such as timestamps, user IDs, and daily water withdrawals. Simultaneously, daily (or real-time) water withdrawal data for testing is obtained to ensure data integrity and timeliness. The collected data is cleaned to remove duplicate records, erroneous data (such as negative values, abnormally large values), and incomplete records. Data validation rules, such as threshold checks based on physical laws, further ensure data legitimacy. The cleaned data is converted into a form suitable for analysis, such as calculating a daily water withdrawal series. Furthermore, the data is normalized to eliminate the influence of different dimensions and magnitudes, making the data comparable across water users. Normalization methods can include Z-score normalization and minimum-maximum normalization. Based on the daily water withdrawal series, the rate of change of water withdrawal between two consecutive days is calculated to form a daily water withdrawal rate series. This step helps capture trends and patterns in water withdrawal changes and is particularly important for identifying abnormal changes. It should be noted that the daily water intake sequence is a record of the water intake of water users every day, and the daily water intake change rate sequence is the change ratio of water intake in adjacent days.

[0024] Based on the characteristics of the daily water withdrawal and rate of change series, a clustering algorithm was selected for anomaly detection. Iterative training was performed using historical data, and the clustering algorithm's performance in identifying anomalies was optimized by adjusting its parameters. This process involved techniques such as cross-validation and grid search to find the optimal combination of target parameters.

[0025] Clustering groups similar water withdrawal data, while outliers are individually clustered or labeled as abnormal due to their significant differences from normal data patterns. Based on the clustering results, data points that do not fall into the primary clusters are identified as outliers. These outliers may indicate illegal water withdrawal, equipment failure, data entry errors, and other issues during the water withdrawal process. The results of the identified outliers are transmitted to relevant departments or personnel for timely verification and resolution.

[0026] Compared with the prior art, the present embodiment provides a method for identifying abnormal data in water intake monitoring, which first pre-processes the acquired historical monitoring data of daily water intake of water users and the data to be detected for water intake on the same day, determines the daily water intake sequence and the daily water intake change rate sequence, and uses them as the input values ​​of the optimal target parameters for clustering. This method can clearly and intuitively understand the specific values ​​of daily water intake and daily water intake change rate, and by processing the data, reduces the requirements for the quality of historical data. On this basis, the optimal target parameters are further calculated through iterative training, and the optimal target parameters can be automatically determined without manual identification, avoiding the missed detection phenomenon that may be caused by manual identification. Cluster identification is then performed based on the optimal target parameters determined by iterative training, thereby identifying the abnormal water intake data in the data to be detected for water intake on the same day, and the identification results of the abnormal data have high accuracy.

[0027] See also Figure 2 , Figure 2 The present invention provides Figure 1 FIG. 1 is a flow chart of an embodiment of step S100 in FIG. 1 . In some embodiments of the present invention, the historical monitoring data of daily water intake of water users and the data to be detected of the water intake on the current day are preprocessed to obtain a daily water intake sequence and a daily water intake change rate sequence, including: S110, obtaining the data on the water intake volume to be tested on the day, and performing an initial inspection on the data on the water intake volume to be tested on the day; S120, combining the historical monitoring data of daily water intake and the data to be tested of the water intake on the current day into a time series in chronological order and performing normalization processing to obtain a daily water intake series; S130. Calculate a daily water intake change rate sequence based on the daily water intake historical monitoring data and the current day water intake data to be detected.

[0028] In the above embodiment, real-time daily water withdrawal data for each water user is collected through an automated monitoring system or manual recording. This data typically includes basic information such as withdrawal time, user ID, and withdrawal volume. A preliminary check is performed on the daily withdrawal data to be tested to identify and eliminate potential errors or unreasonable data. This initial check may include checking the correct data format, whether the withdrawal volume is within a reasonable range (e.g., avoiding negative values ​​or abnormally large values), and whether there are duplicate records. A series of data validation rules are established to ensure data accuracy and validity.

[0029] The collected historical monitoring data on daily water withdrawals and the pending water withdrawal data after the initial inspection on the same day are sorted and combined in chronological order (e.g., date) to form a complete time series. This series shows the changes in daily water withdrawals by water users over a period of time. Because water withdrawals can vary significantly among different water users, the data in the time series must be normalized to eliminate these dimensional effects and make the data comparable. There are various normalization methods, but commonly used ones include minimum or maximum normalization (scaling the data to the interval [0, 1]) or Z-score normalization (converting the data to a distribution with a mean of 0 and a standard deviation of 1).

[0030] After obtaining the normalized daily water withdrawal sequence, the next step is to calculate the daily water withdrawal rate of change. The rate of change reflects the increase or decrease in water withdrawal relative to the previous day and is a key indicator for identifying water withdrawal anomalies. The rate of change can be calculated by dividing the difference between the current and previous day's water withdrawals by the previous day's withdrawals. It can also be expressed as a percentage. The daily water withdrawal rate of change at each time point is arranged in order to form a daily water withdrawal rate sequence. This sequence intuitively demonstrates the dynamic trends in water withdrawal, providing an important basis for subsequent anomaly detection.

[0031] In some embodiments of the present invention, obtaining the data to be detected of the water intake on the day and performing a preliminary inspection on the data to be detected of the water intake on the day include: If the data of water intake to be tested on that day is null, it will be recorded that the water intake on that day was not reported; If the water intake data to be detected on that day is greater than the preset daily water intake threshold, the water intake data to be detected on that day is recorded as being extremely large on that day.

[0032] In the above example, the data for all water users' water withdrawals for the day is traversed to check for null values ​​(i.e., missing data). Null values ​​may indicate that data was not reported due to data collection failure, sensor malfunction, or human negligence. If a null value is found in the data for the day's water withdrawals, the system immediately records the situation as "water withdrawals not reported for the day." Furthermore, based on business rules, further action may be required, such as notifying the relevant water users to submit additional data or checking the proper functioning of data collection equipment.

[0033] As a preferred embodiment, the preset daily water withdrawal threshold in the present invention is 10% of the annual approved water withdrawal amount. This preset daily water withdrawal threshold is determined based on historical data, industry standards, policy requirements, and other factors. The daily water withdrawal data to be tested is then compared with the preset threshold. If the daily water withdrawal data to be tested exceeds the preset daily water withdrawal threshold, the data is recorded as "extremely large for the day." This marking indicates that the daily water withdrawal data is abnormal.

[0034] In some embodiments of the present invention, the daily water intake change rate sequence is calculated based on the daily water intake historical monitoring data and the daily water intake data to be detected, including: Calculate the change rate weights for the historical monitoring data of daily water intake within a preset number of days in chronological order; The daily water intake change rate sequence is calculated based on the historical monitoring data of daily water intake, the data of water intake to be detected on the current day and the change rate weight.

[0035] In the above embodiment, the preset number of days is typically set based on analysis requirements, data availability, and business rules, and is used to specify the range of historical data to be considered when calculating the rate of change weight. As a preferred embodiment, the preset number of days in the present invention is three. For each of the preset days of historical water intake monitoring data, the water intake change rate is calculated in chronological order (typically, by date, from earliest to latest). A weight is assigned to each day's water intake based on the chronological order.

[0036] After obtaining the weights for the rate of change of historical daily water withdrawal monitoring data over a preset number of days, these weights are used to comprehensively assess the changing trend of daily water withdrawal. This weighted estimate then calculates a daily water withdrawal rate of change sequence, including the pending water withdrawal data for that day. Arranging all these elements in chronological order creates a complete daily water withdrawal rate of change sequence. This sequence not only reflects the daily fluctuations in water withdrawal but, through the introduction of rate of change weights, more accurately reveals the dynamic characteristics and trends of water withdrawal.

[0037] In a specific embodiment of the present invention, the daily water intake sequence is , is the sequence length. In order to reflect the changing trend of water intake compared with the water intake of the adjacent days, the weighted moving average change rate is used as the clustering feature. That is, the change rate of the water intake on the current day and the weighted water intake of the previous three days is used as a type of feature input. The daily water intake change rate sequence is calculated based on the historical monitoring data of daily water intake, the data to be tested on the current day's water intake, and the change rate weight. The calculation formula is as follows: , ; Where Y represents the daily water intake change rate sequence.

[0038] See also Figure 3 , Figure 3 The present invention provides Figure 1 FIG. 1 is a flow chart of an embodiment of step S200 in FIG. 1 . In some embodiments of the present invention, iterative training is performed based on a daily water intake sequence and a daily water intake change rate sequence to determine the optimal target parameter, including: S210, randomly generating several initial solution sets within a preset value range of the target parameter; S220, calculating the particle fitness based on the initial solution set, the daily water intake sequence, and the daily water intake change rate sequence; S230: Update the speed and position of the particle and perform iterative training to a preset number of iterations to determine the optimal target parameters.

[0039] In the above embodiment, the target parameter to be calculated is the area radius and the minimum number of points within the radius of the core object area The preset value range of the target parameter is based on the comprehensive consideration of factors such as the problem background, historical data analysis and actual operation feasibility. The preset value range of the target parameter is and( ),exist and( ) Randomly generate multiple initial solutions within the interval Using a random number generator, a number of initial values ​​are randomly generated for each target parameter within a preset range, forming a set of initial solutions. These initial solutions serve as the starting point for subsequent iterative optimization, aiming to approach the optimal solution through continuous exploration. To increase the likelihood of finding a global optimal solution, the initial solution set should be diverse, meaning that each solution should be significantly different to avoid prematurely falling into a local optimum.

[0040] Each initial solution (or "particle") is substituted into the model, and training and prediction are performed using the daily water withdrawal sequence and the daily water withdrawal rate change sequence as input data. These two sequences not only contain the absolute values ​​of water withdrawal but also reflect the changing trends of water withdrawal, which is crucial for improving prediction accuracy. A fitness function is designed to quantify the performance of each solution. This function may take into account multiple factors, such as prediction accuracy (such as mean squared error (MSE) and mean absolute error (MAE), model complexity (controlled by regularization terms), and prediction stability (such as the accuracy of rate of change predictions). The fitness value of each particle is calculated and recorded. This value will be used for subsequent particle updates and comparisons to determine which particles are closer to the optimal solution.

[0041] In a specific embodiment of the present invention, the particle swarm optimization (PSO) algorithm uses the particle's current position and velocity, as well as information about the global and individual optimal solutions, to update the velocity and position of each particle. This process simulates the food-seeking behavior of a group of organisms in nature, optimizing the search direction through information sharing and mutual learning. The particle velocity and position updates are repeated, and the particle fitness is recalculated after each iteration. As the iterations proceed, the particles gradually approach the optimal solution until a preset number of iterations is reached. At the end of the iteration, the optimal particle is selected based on its fitness, and its corresponding target parameters become the desired optimal target parameters. These parameters are used in the final model construction and prediction task.

[0042] In some embodiments of the present invention, the particle fitness is calculated based on the initial solution set, the daily water intake sequence, and the daily water intake change rate sequence, including: The initial particle fitness is calculated based on the silhouette coefficient evaluation function, the initial solution set, the daily water intake sequence and the daily water intake change rate sequence; The initial particle fitness is mapped based on the mapping rule to determine the target particle fitness.

[0043] In the above example, for each particle's solution (i.e., a set of target parameters), this set of parameters is used to construct a model and predict the daily water withdrawal sequence. While the external dissimilarity in the traditional silhouette coefficient refers to the distance between the sample and other clusters, it can be understood as the difference between the predicted value sequence and the actual value change rate sequence (such as the rate of change component of the mean squared error or mean absolute error), which reflects the ability to capture the changing trend of water withdrawal. Combining internal similarity and external dissimilarity, a comprehensive silhouette coefficient value is calculated using a weighted method, which serves as the initial fitness of the particle.

[0044] The silhouette coefficient (SC) is used to evaluate the clustering effect. The larger the silhouette coefficient, the higher the similarity within the cluster, and the lower the similarity between clusters, the better the clustering effect. The particle fitness of PSO is: ; for Continuous values ​​within ( , are the upper and lower bounds of the domain radius), for Integers within , so the optimization problem is a mixed integer nonlinear programming problem, in order to meet For integer constraints, the evolved solutions are usually rounded when the PSO algorithm is used to solve them. The rounding operation often causes the particles to move away from the optimal solution.

[0045] In order to enable particles to evolve in a continuous solution space and improve the stability and convergence of the algorithm, the present invention performs interval mapping on the silhouette coefficient evaluation function so that the fitness function can accept non-integer Parameters. The mapping rules are as follows: ; in, for Inner real numbers, 、 for Round up or down to an integer.

[0046] See also Figure 4 , Figure 4 The present invention provides Figure 3 FIG. 1 is a flow chart of an embodiment of step S230 in FIG. 1 . In some embodiments of the present invention, updating the velocity and position of the particle and performing iterative training to a preset number of iterations to determine the optimal target parameters includes: S231, updating the speed and position of the particles and performing iterative training to determine the optimal solution of the particle swarm; S232, calculating the target fitness corresponding to rounding up and rounding down the particle swarm optimal solution; S233. Determine optimal target parameters according to the target fitness.

[0047] In the above embodiment, the particle velocity and position update formula is: ; ; in, For the Particle No. The speed of the round evolution; For the The weight of wheel speed update, also known as inertia factor; 、 is the learning factor, 、 for A random number between For the Particle No. Position during wheel evolution; For the The best historical position of each particle; is the historical best position of all particles.

[0048] The present invention adopts a linear decreasing weight strategy to increase the particle's global search capability in the early stage and local optimization capability in the later stage by dynamically adjusting the weight. The weight update formula is as follows: ; in, 、 is the maximum and minimum weight; is the preset number of iterations.

[0049] The particle size was calculated using the silhouette coefficient evaluation function. Fitness ,like ,but ,like ,but .

[0050] Output the optimal solution by cyclic training to the preset number of iterations ,calculate correspond The fitness corresponding to rounding up and down 、 ,like ,but ,on the contrary .

[0051] In order to better implement the water intake monitoring abnormal data identification method in the embodiment of the present invention, based on the water intake monitoring abnormal data identification method, correspondingly, please refer to Figure 5 , Figure 5 This is a schematic structural diagram of an embodiment of a device for identifying abnormal data for water intake monitoring provided by the present invention. The embodiment of the present invention provides a device 500 for identifying abnormal data for water intake monitoring, comprising: A preprocessing module 510 is configured to preprocess the historical monitoring data of daily water intake of water users and the data to be detected of the water intake on the current day to obtain a daily water intake sequence and a daily water intake change rate sequence; An iterative training module 520 is configured to perform iterative training based on a daily water intake sequence and a daily water intake change rate sequence to determine an optimal target parameter; The cluster identification module 530 is configured to cluster and identify abnormal water intake data on the daily water intake sequence and the daily water intake change rate sequence based on the optimal target parameter.

[0052] It should be noted here that the device 500 provided in the above embodiment can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can be found in the corresponding contents in the above method embodiments, which will not be repeated here.

[0053] See also Figure 6 , Figure 6 This is a schematic diagram of the structure of a device for identifying abnormal water intake monitoring data, provided in an embodiment of the present invention. Based on the aforementioned method for identifying abnormal water intake monitoring data, the present invention also provides a device for identifying abnormal water intake monitoring data. The device can be a computing device such as a mobile terminal, desktop computer, notebook, PDA, or server. The device 600 includes a processor 610, a memory 620, and a display 630. Figure 6 Only some components of the synchronous tracking flying welding equipment for real-time measurement of battery height are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components may be implemented instead.

[0054] In some embodiments, the memory 620 can be an internal storage unit of the water intake monitoring abnormal data identification device 600, such as the hard drive or memory of the water intake monitoring abnormal data identification device 600. In other embodiments, the memory 620 can also be an external storage device of the water intake monitoring abnormal data identification device 600, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped with the water intake monitoring abnormal data identification device 600. Furthermore, the memory 620 can include both the internal storage unit of the water intake monitoring abnormal data identification device 600 and an external storage device. The memory 620 is used to store application software installed in the water intake monitoring abnormal data identification device 600 and various data, such as the program code for installing the water intake monitoring abnormal data identification device 600. The memory 620 can also be used to temporarily store data that has been output or is about to be output. In one embodiment, the memory 620 stores a water intake monitoring abnormal data identification program 640, which can be executed by the processor 610 to implement the water intake monitoring abnormal data identification method of each embodiment of the present application.

[0055] In some embodiments, the processor 610 may be a central processing unit (CPU), a microprocessor, or other data processing chip, configured to execute program codes or process data stored in the memory 620, such as executing a method for identifying abnormal water intake monitoring data.

[0056] In some embodiments, the display 630 can be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 630 is used to display information on the water intake monitoring abnormality data identification device 600 and to display a visual user interface. Components 610-630 of the water intake monitoring abnormality data identification device 600 communicate with each other via a system bus.

[0057] In one embodiment, when the processor 610 executes the water intake monitoring abnormal data identification program 640 in the memory 620 , the steps in the above water intake monitoring abnormal data identification method are implemented.

[0058] This embodiment further provides a computer-readable storage medium on which a water intake monitoring abnormal data identification program 640 is stored. When the water intake monitoring abnormal data identification program is executed by a processor, the following steps are implemented: The historical monitoring data of daily water intake of water users and the data of water intake to be tested on the same day are preprocessed to obtain the daily water intake sequence and the daily water intake change rate sequence; Iterative training is performed based on the daily water intake sequence and the daily water intake change rate sequence to determine the optimal target parameters; Based on the optimal target parameters, the daily water intake series and the daily water intake change rate series are clustered to identify abnormal water intake data.

[0059] The present invention provides a method for identifying abnormal data in water intake monitoring. The method first pre-processes the acquired historical monitoring data of daily water intake of water users and the data of water intake on the same day to be detected, determines the daily water intake sequence and the daily water intake change rate sequence, and uses them as the input values ​​of the optimal target parameters for clustering. The method can clearly and intuitively understand the specific values ​​of the daily water intake and the daily water intake change rate, and by processing the data, reduces the requirements for the quality of historical data. On this basis, the optimal target parameters are further calculated through iterative training, and the optimal target parameters can be automatically determined without manual identification, avoiding the missed detection phenomenon that may be caused by manual identification. Cluster identification is then performed based on the optimal target parameters determined by iterative training, thereby identifying the abnormal water intake data in the data of water intake on the same day to be detected, and the identification result of the abnormal data has high accuracy.

[0060] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a microdrive, a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0061] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0062] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0063] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0064] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0065] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0066] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk, or optical disk, etc., various media that can store program code.

[0067] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0068] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure herein, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art that are not recorded in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

[0069] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0070] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying abnormal data in water intake monitoring, characterized in that: include: Preprocess the historical monitoring data of daily water intake of water users and the data to be tested on the water intake of the day to obtain the daily water intake sequence and the daily water intake change rate sequence; Performing iterative training based on the daily water intake sequence and the daily water intake change rate sequence to determine the optimal target parameter; Based on the optimal target parameter, the daily water intake sequence and the daily water intake change rate sequence are clustered to identify abnormal water intake data.

2. The method for identifying abnormal data of water intake monitoring according to claim 1, characterized in that: The preprocessing of the historical monitoring data of the daily water intake of the water users and the data to be detected of the water intake of the day to obtain the daily water intake sequence and the daily water intake change rate sequence includes: Acquire the data of water intake on the day to be detected, and perform a preliminary inspection on the data of water intake on the day to be detected; The daily water intake historical monitoring data and the current day water intake to be detected data are combined into a time series in chronological order and normalized to obtain a daily water intake series; The daily water intake change rate sequence is calculated based on the daily water intake historical monitoring data and the current day water intake data to be detected.

3. The method for identifying abnormal data of water intake monitoring according to claim 2, characterized in that: The obtaining of the data to be detected of the water intake on the day and performing a preliminary inspection on the data to be detected of the water intake on the day includes: If the data to be detected for the water intake on that day is null, then the water intake on that day is recorded as not being reported; If the data to be detected of the water intake on that day is greater than the preset daily water intake threshold, the data to be detected of the water intake on that day is recorded as being extremely large on that day.

4. The method for identifying abnormal data of water intake monitoring according to claim 2, characterized in that: The calculating of the daily water intake change rate sequence based on the daily water intake historical monitoring data and the daily water intake to be detected data includes: Calculate the change rate weight of the historical monitoring data of daily water intake within a preset number of days in chronological order; The daily water intake change rate sequence is calculated based on the daily water intake historical monitoring data, the daily water intake data to be detected and the change rate weight.

5. The method for identifying abnormal data of water intake monitoring according to claim 1, characterized in that: The iterative training based on the daily water intake sequence and the daily water intake change rate sequence to determine the optimal target parameter comprises: Randomly generate several initial solution sets within the preset value range of the target parameter; Calculating particle fitness based on the initial solution set, the daily water intake sequence and the daily water intake change rate sequence; The particle speed and position are updated and iteratively trained to a preset number of iterations to determine the optimal target parameters.

6. The method for identifying abnormal data of water intake monitoring according to claim 5, characterized in that: The calculating of particle fitness based on the initial solution set, the daily water intake sequence and the daily water intake change rate sequence comprises: Calculate the initial particle fitness based on the silhouette coefficient evaluation function, the initial solution set, the daily water intake sequence and the daily water intake change rate sequence; The initial particle fitness is mapped based on a mapping rule to determine the target particle fitness.

7. The method for identifying abnormal data of water intake monitoring according to claim 5, characterized in that: The updating of the particle speed and position and iterative training to a preset number of iterations to determine the optimal target parameters include: Update the speed and position of the particles and perform iterative training to determine the optimal solution for the particle swarm; Calculate the target fitness corresponding to rounding up and rounding down the optimal solution of the particle swarm; The optimal target parameter is determined according to the target fitness.

8. A device for identifying abnormal data of water intake monitoring, characterized in that: include: A preprocessing module is configured to preprocess the historical monitoring data of the daily water intake of the water users and the data to be detected of the daily water intake to obtain a daily water intake sequence and a daily water intake change rate sequence; An iterative training module, configured to perform iterative training based on the daily water intake sequence and the daily water intake change rate sequence to determine an optimal target parameter; A clustering identification module is configured to cluster the daily water intake sequence and the daily water intake change rate sequence to identify abnormal water intake data based on the optimal target parameter.

9. A water intake monitoring abnormal data identification device, characterized in that: It comprises at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit executes the steps of the water intake monitoring abnormal data identification method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: It stores a computer program executable by an access authentication device. When the computer program runs on the access authentication device, the access authentication device executes the steps of the water intake monitoring abnormal data identification method according to any one of claims 1 to 7.

Images