Method, system and electronic equipment for identifying abnormal trends in dam monitoring data

Abstract

The present application relates to a method, system and electronic device for identifying abnormal trends in dam monitoring data, wherein the method for identifying abnormal trends in dam monitoring data includes: acquiring monitoring data of the dam, calculating a first deformation trend component of the dam based on the monitoring data by a statistical regression method; updating parameters of a pre-acquired residual neural network according to the monitoring data, and determining a second deformation trend component of the dam based on the updated parameters; calculating the correlation between the first deformation trend component and the second deformation trend component, and in response to the correlation being less than a preset threshold, determining an abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component by a clustering algorithm.

Description

Technical Field

[0001] The present application relates to the field of dam anomaly monitoring, and in particular to a method, system and electronic equipment for identifying anomaly trends in dam monitoring data. Background Art

[0002] Dam monitoring data is important information for diagnosing the health status of the dam. Identifying trend anomalies in monitoring data is key to preventing dam safety issues. However, dam monitoring data is often affected by environmental changes.

[0003] Currently, trend analysis of dam monitoring data primarily involves constructing regression or reconstruction models and analyzing the residuals between measured and estimated data to identify anomalies. These two methods typically require that measured data points meet a certain regression error or reconstruction residual size before they are considered anomalies. However, when the data exhibits a slowly changing anomaly, traditional methods are unable to directly identify the abnormal trend due to the small estimated residuals in the initial phase. These anomalies can only be identified when the data exhibits significant anomalies, making it difficult to promptly identify potential risks to dam safety.

[0004] Existing technologies are unable to identify abnormal trends in dam monitoring data in a timely and accurate manner. Summary of the Invention

[0005] The embodiments of the present application provide a method, system, and electronic device for identifying abnormal trends in dam monitoring data, so as to at least solve the problem in the related art that abnormal trends in dam monitoring data cannot be identified in a timely and accurate manner.

[0006] In a first aspect, an embodiment of the present application provides a method for identifying abnormal trends in dam monitoring data, comprising:

[0007] Acquiring monitoring data of the dam, and calculating a first deformation trend component of the dam based on the monitoring data using a statistical regression method;

[0008] updating the parameters of the residual neural network obtained in advance according to the monitoring data, and determining the second deformation trend component of the dam based on the updated parameters;

[0009] A correlation between the first deformation trend component and the second deformation trend component is calculated, and in response to the correlation being less than a preset threshold, an abnormal trend in the monitoring data is determined based on a difference between the first deformation trend component and the second deformation trend component through a clustering algorithm.

[0010] In one embodiment, the monitoring data includes dam monitoring data, monitoring time, and upstream water level, and updating the pre-acquired parameters of the residual neural network according to the monitoring data includes:

[0011] Determining a prediction result based on the monitoring time and the upstream water level by a pre-acquired residual neural network;

[0012] The parameters of the residual neural network are updated according to the dam monitoring data and the prediction results.

[0013] In one embodiment, determining the prediction result includes:

[0014] The prediction results are divided into a period component and a trend component, and the prediction results are calculated as follows:

[0015]

[0016] Among them, X output Indicates the preset result, W c represents the weight of the periodic component, represents the periodic component, W t represents the weight of the trend component, Represents the trend component.

[0017] In one embodiment, determining the second deformation trend component includes:

[0018]

[0019] Among them, δ θ2 represents the second deformation trend component, W t represents the weight of the trend component, represents the trend component, which is calculated as follows:

[0020]

[0021]

[0022] in, These are intermediate results. are all weight matrices, are all bias vectors, i=1,2,3, j=1,2,3, relu(*) represents the relu activation function, represents an input parameter, which is determined based on the monitoring time, σ(*) represents a Sigmod activation function, and hardtanh(*) represents a hardtanh activation function.

[0023] In one embodiment, calculating the correlation between the first deformation trend component and the second deformation trend component includes:

[0024]

[0025] Among them, δθ1 represents the first deformation trend component, δ θ2 represents the second deformation trend component, and E(*) represents the mean calculation.

[0026] In one embodiment, the determining the abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component by using a clustering algorithm includes:

[0027] calculating a difference between the first deformation trend component and the second deformation trend component, and classifying the difference into two categories using a clustering algorithm;

[0028] Abnormal data is determined from the two categories according to the quantity of each category, and it is determined that the dam monitoring data corresponding to the abnormal data has an abnormal trend.

[0029] In one embodiment, the calculating the first deformation trend component of the dam based on the monitoring data by using a statistical regression method includes:

[0030] δ=δ H +δ T +δ θ1

[0031]

[0032] δ θ1 =c1θ+c2lnθ

[0033]

[0034] Wherein, δ represents the dam monitoring data, δ H represents the deformation caused by water level change, δ T Indicates deformation caused by temperature change, δ θ1 represents the deformation caused by time change, H represents the upstream water level, a i represents the coefficient to be fitted, n represents the quantity, c1 and c2 are both time coefficients, t represents the monitoring time, which represents the number of days from the current time to the start time of the monitoring data, θ is equal to t / 100, b 1i and b 2i are statistical coefficients, δ θ1 represents the first deformation trend component.

[0035] In one embodiment, the method further includes: in response to the correlation being greater than or equal to a preset threshold, determining that the trend of the monitoring data is normal.

[0036] In a second aspect, an embodiment of the present application provides a system for identifying abnormal trends in dam monitoring data, including:

[0037] A first trend module is used to obtain monitoring data of the dam and calculate a first deformation trend component of the dam based on the monitoring data using a statistical regression method;

[0038] A second trend module is configured to update the parameters of the pre-acquired residual neural network according to the monitoring data, and determine the second deformation trend component of the dam based on the updated parameters;

[0039] Trend judgment module: used to calculate the correlation between the first deformation trend component and the second deformation trend component, and in response to the correlation being less than a preset threshold, determine the abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component through a clustering algorithm.

[0040] In a third aspect, an embodiment of the present application provides a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method for identifying anomaly trends in dam monitoring data as described in the first aspect above is implemented.

[0041] The embodiments of the present application provide a method, system, and electronic device for identifying abnormal trends in dam monitoring data, which have at least the following technical effects.

[0042] This application extracts the first deformation trend component in dam body deformation through a statistical model, extracts the second deformation trend component in dam monitoring data through a residual neural network model, and then screens abnormal processes through correlation analysis between the first deformation trend component and the second deformation trend component. After obtaining the preliminary abnormal process, cluster analysis is performed on the trend components to further identify the abnormal process, thereby more timely and accurately identifying abnormal trends in dam monitoring data, which is conducive to ensuring the safety and stability of the dam.

[0043] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 This is a flow chart of a method for identifying abnormal trends in dam monitoring data provided by an embodiment of the present application;

[0046] Figure 2 is a schematic diagram of residual distribution using a traditional statistical model method according to an exemplary embodiment;

[0047] Figure 3 1 is a schematic diagram of anomaly identification results of monitoring data obtained by using the method of the present application according to an exemplary embodiment;

[0048] Figure 4 1 is a schematic diagram showing a comparison of trend component calculation results obtained by using the method of the present application and a statistical model method according to an exemplary embodiment;

[0049] Figure 5 1 is a schematic structural diagram of a dam monitoring data trend anomaly identification system according to an embodiment of the present application;

[0050] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making any creative efforts are within the scope of protection of this application.

[0052] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0053] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0054] Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by a person of ordinary skill in the technical field to which this application belongs. The words "one", "a", "the" and the like used in this application do not indicate a limit on quantity and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof used in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The word "multiple" used in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0055] In a first aspect, an embodiment of the present application provides a method for identifying abnormal trends in dam monitoring data. Figure 1 This is a flow chart of a method for identifying abnormal trends in dam monitoring data provided by an embodiment of the present application. Figure 1 As shown, the method includes:

[0056] Step S101: Acquire monitoring data of the dam, and calculate a first deformation trend component of the dam based on the monitoring data using a statistical regression method.

[0057] Optionally, the monitoring data includes dam monitoring data, water level upstream of the dam, and monitoring time. Dam monitoring data can be expressed as z = {t, δ}, where δ = (δ1, δ2, ... δ n ) represents the dam monitoring data vectors collected in chronological order, t=(t1, t2, ...t n ) represents the n moments corresponding to the dam monitoring data. The collected dam monitoring data are normalized to convert data of different ranges into a same standard range.

[0058] In one example, calculating the first deformation trend component in step S101 includes:

[0059] δ=δ H +δ T +δθ1

[0060] δ θ1 =c1θ+c2lnθ

[0061]

[0062] Among them, δ represents the dam monitoring data, δ H represents the deformation caused by water level change, δ T Indicates deformation caused by temperature change, δ θ1 represents the deformation caused by time change, H represents the upstream water level, a i represents the coefficient to be fitted, n represents the number, and according to experience, n is usually 3 or 4, c1 and c2 are both time coefficients, t represents the monitoring time, which represents the number of days from the current time to the start time of the monitoring data, θ is equal to t / 100, b 1i and b 2i are statistical coefficients, δ θ1 Represents the first deformation trend component.

[0063] Optionally, the dam deformation represented by the dam monitoring data includes deformation due to water level changes, deformation due to temperature changes, and deformation due to time changes. This application uses a regression method to calculate the deformation due to time changes, i.e., the first deformation trend component. The coefficients in the above formula are calculated using the least squares method based on the monitoring data to calculate the first deformation trend component of the dam monitoring data at different times. The trend component in the regression method is an irreversible component and is used to extract long-term trends in dam deformation to facilitate subsequent analysis.

[0064] Step S102: updating the parameters of the residual neural network acquired in advance according to the monitoring data, and determining the second deformation trend component of the dam based on the updated parameters.

[0065] Optionally, the residual neural network includes, but is not limited to, any one of ResNet, ResNeXt, and DenseNet. The residual neural network is trained based on historical monitoring data to obtain a pre-acquired residual neural network. The parameters of the residual neural network are updated using a backpropagation method for the real-time monitoring data, thereby calculating the second deformation trend component based on the updated parameters.

[0066] In one example, the monitoring data includes dam monitoring data, monitoring time, and upstream water level. Step S102 includes:

[0067] Step S201: Determine a prediction result based on the monitoring time and upstream water level using a pre-acquired residual neural network. Optionally, the prediction result is decomposed into a period component and a trend component to facilitate calculation of a second deformation trend component. The second deformation trend component is a reversible trend component.

[0068] In one example, determining the prediction result in step S201 includes dividing the prediction result into a period component and a trend component. The prediction result is calculated as follows:

[0069]

[0070] Among them, X output Indicates the preset result, W c represents the weight of the periodic component, represents the periodic component, W t represents the weight of the trend component, Represents the trend component.

[0071] The periodic component is calculated as follows:

[0072]

[0073] in, These are intermediate results. are all weight matrices, are all bias vectors, i = 1, 2, 3, j = 1, 2, 3, relu (*) represents the relu activation function, σ (*) represents the sigmoid activation function, Represents the periodic component. Represents input parameters, which are input matrices determined based on upstream water level and monitoring time, including H, H2, H 3 、 and The input matrix is ​​constructed, where H represents the upstream water level, t represents the monitoring time, the monitoring time represents the number of days from the current time to the start time of the monitoring data, and a=1,2.

[0074] Trend Component is calculated as follows:

[0075]

[0076] in, These are intermediate results. are all weight matrices, are all bias vectors, i = 1, 2, 3, j = 1, 2, 3, relu(*) represents the relu activation function, σ(*) represents the sigmoid activation function, and hardtanh(*) represents the hardtanh activation function. represents input parameters, which are determined based on the monitoring time, specifically including θ and ln(θ), where θ=t / 100, and t represents the number of days from the current moment to the start moment of the monitoring data.

[0077] In this way, the Sigmod and Hardtan activation functions are used simultaneously to extract the reversible trend components in the dam monitoring data, making the residual network model more sensitive to both the slow-changing and sudden-changing characteristics of the data, thereby better discovering the changing trend of the data.

[0078] Step S202: Update the parameters of the residual neural network based on the dam monitoring data and the prediction results. Optionally, update the parameters of the residual neural network through back propagation based on the error between the dam monitoring data and the prediction results.

[0079] In one example, determining the second deformation trend component in step S102 includes:

[0080]

[0081] Among them, δ θ2 represents the second deformation trend component, W t In this way, the second deformation trend component is determined based on the updated residual neural network parameters. Compared with traditional algorithms, the residual network used in this application can better discover the changing trend of data and is more sensitive to both the slow-changing and sudden-changing characteristics of data.

[0082] Step S103 , calculating the correlation between the first deformation trend component and the second deformation trend component, and in response to the correlation being less than a preset threshold, determining an abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component through a clustering algorithm.

[0083] Optionally, in response to the correlation between the first deformation trend component and the second deformation trend component being greater than or equal to a preset threshold, the trend of the monitoring data is determined to be normal. When the correlation between the first deformation trend component and the second deformation trend component is less than a preset threshold, the preset threshold can be set according to actual needs. In this application, the preset threshold is generally between 0.8 and 0.9. Clustering algorithms include, but are not limited to, any one of mean clustering, hierarchical clustering, and DBSCAN clustering algorithms.

[0084] In this way, the correlation analysis of the first deformation trend component and the second deformation trend component is performed to identify abnormal trends and quantify the abnormal degree of the trend process, so as to facilitate the timely discovery of data anomalies and ensure the safety and stability of the dam.

[0085] In one example, step S103 includes:

[0086]

[0087] Among them, δ θ1 represents the first deformation trend component, δ θ2 represents the second deformation trend component, E(*) represents mean calculation, and optionally, in this example, the K-means mean clustering method is used to perform correlation calculation.

[0088] In one example, step S103 includes calculating a difference between the first deformation trend component and the second deformation trend component, classifying the difference into two categories using a clustering algorithm, determining abnormal data from the two categories based on the number of data in each category, and determining whether the dam monitoring data corresponding to the abnormal data has an abnormal trend.

[0089] Alternatively, the difference can be expressed as: where δ θ1 represents the first deformation trend component, δ θ2 Represents the second deformation trend component. θ Clustering is performed, d θ All data points in the dataset are divided into two categories, and the category with less data is determined to be abnormal data. The clustering algorithm includes but is not limited to any one of mean clustering, hierarchical clustering, and DBSCAN clustering algorithms.

[0090] In this way, the abnormal process screening is achieved through the correlation analysis of the first deformation trend component and the second deformation trend component. After obtaining the preliminary abnormal process, further abnormal process identification can be carried out through cluster analysis of the trend components, which makes it more sensitive to the slowly changing abnormal process of the data, thereby identifying the abnormal trend in the dam monitoring data more timely and accurately.

[0091] The dam monitoring data trend anomaly identification method proposed in this application is compared with the traditional statistical model method. Figure 2 is a schematic diagram of residual distribution using a traditional statistical model method according to an exemplary embodiment. Figure 2 As shown in the figure, although the traditional statistical model method has large residuals near the abnormal process, it fails to identify the abnormal process of the data due to the small overall residuals. Figure 3 FIG. 1 is a schematic diagram of anomaly identification results of monitoring data obtained by using the method of the present application according to an exemplary embodiment. Figure 3 As shown, the dam monitoring data trend anomaly identification method proposed in this application can better identify the abnormal process of the data, especially the method effectively identifies the stage of rapid increase of the monitoring data in the second half. Figure 4 FIG. 1 is a schematic diagram showing a comparison of trend component calculation results obtained by using the method of the present application and the statistical model method according to an exemplary embodiment. Figure 4As shown, the dam monitoring data trend anomaly identification method proposed in this application can identify more trend anomaly data and has a better identification effect, which is conducive to timely warning dam safety management personnel of anomalies in the data, thereby ensuring the safety and stability of the dam.

[0092] In summary, this application extracts the long-term change trend (first deformation trend component) in the dam body deformation through a statistical model, and uses Sigmod and Hardtan activation functions simultaneously in the residual neural network model to extract the reversible trend component (second deformation trend component) in the dam monitoring data, so that the residual network model can better discover the change trend of the data and is more sensitive to the slow-changing characteristics and sudden change characteristics of the data. Then, the abnormal process is screened through the correlation analysis of the first deformation trend component and the second deformation trend component. After obtaining the preliminary abnormal process, the trend component is clustered for further abnormal process identification, thereby more timely and accurately identifying the abnormal trend in the dam monitoring data, which is conducive to ensuring the safety and stability of the dam.

[0093] In a second aspect, the present application provides a system for identifying abnormal trends in dam monitoring data. Figure 5 FIG. 1 is a structural diagram of a dam monitoring data trend anomaly identification system according to an embodiment of the present application. Figure 5 As shown, the system includes:

[0094] The first trend module 100 is used to obtain monitoring data of the dam and calculate the first deformation trend component of the dam based on the monitoring data using a statistical regression method.

[0095] The second trend module 200 is used to update the parameters of the pre-acquired residual neural network according to the monitoring data, and determine the second deformation trend component of the dam based on the updated parameters.

[0096] Trend judgment module 300: used to calculate the correlation between the first deformation trend component and the second deformation trend component, and in response to the correlation being less than a preset threshold, determine the abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component through a clustering algorithm.

[0097] In one example, the second trend module 200 includes:

[0098] Prediction unit: used to determine the prediction result based on the monitoring time and upstream water level through the pre-acquired residual neural network.

[0099] Update unit: used to update the parameters of the residual neural network based on dam monitoring data and prediction results.

[0100] In one example, the prediction unit includes: dividing the prediction result into a period component and a trend component, and calculating the prediction result as follows:

[0101]

[0102] Among them, X output Indicates the preset result, W c represents the weight of the periodic component, represents the periodic component, W t represents the weight of the trend component, Represents the trend component.

[0103] In one example, determining the second deformation trend component in the second trend module 200 includes:

[0104]

[0105] Among them, δ θ2 represents the second deformation trend component, W t represents the weight of the trend component, Represents the trend component, which is calculated as follows:

[0106]

[0107] in, These are intermediate results. are all weight matrices, are all bias vectors, i=1,2,3, j=1,2,3, relu(*) represents the relu activation function, Represents input parameters, which are determined based on the monitoring time, σ(*) represents the Sigmod activation function, and hardtanh(*) represents the hardtanh activation function.

[0108] In one example, the trend determination module 300 includes:

[0109]

[0110] Among them, δ θ1 represents the first deformation trend component, δ θ2 represents the second deformation trend component, and E(*) represents the mean calculation.

[0111] In one example, the trend determination module 300 includes:

[0112] It is used to calculate the difference between the first deformation trend component and the second deformation trend component, and classify the difference into two categories through a clustering algorithm; determine abnormal data from the two categories based on the number of each category, and determine whether the dam monitoring data corresponding to the abnormal data has an abnormal trend.

[0113] In one example, the first trend module 100 includes:

[0114] δ=δ H +δ T +δ θ1

[0115] δ θ1 =c1θ+c2lnθ

[0116]

[0117] Among them, δ represents the dam monitoring data, δ H represents the deformation caused by water level change, δ T Indicates deformation caused by temperature change, δ θ1 represents the deformation caused by time change, H represents the upstream water level, a i represents the coefficient to be fitted, n represents the quantity, c1 and c2 are both time coefficients, t represents the monitoring time, which represents the number of days from the current moment to the start moment of the monitoring data, θ is equal to t / 100, b 1i and b 2i are statistical coefficients, δ θ1 Represents the first deformation trend component.

[0118] In one example, the system further includes: in response to the correlation being greater than or equal to a preset threshold, determining that the trend of the monitoring data is normal.

[0119] In summary, this application extracts the long-term change trend (first deformation trend component) in the dam body deformation through a statistical model, and uses Sigmod and Hardtan activation functions simultaneously in the residual neural network model to extract the reversible trend component (second deformation trend component) in the dam monitoring data, so that the residual network model can better discover the change trend of the data and is more sensitive to the slow-changing characteristics and sudden change characteristics of the data. Then, the abnormal process is screened through the correlation analysis of the first deformation trend component and the second deformation trend component. After obtaining the preliminary abnormal process, the trend component is clustered for further abnormal process identification, thereby more timely and accurately identifying the abnormal trend in the dam monitoring data, which is conducive to ensuring the safety and stability of the dam.

[0120] In a third aspect, an embodiment of the present application provides an electronic device, Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for identifying abnormal trends in dam monitoring data provided in the first aspect is implemented. Figure 6 The electronic device 60 shown is only an example and should not limit the functions and scope of use of the embodiments of the present application.

[0121] The electronic device 60 may be a general-purpose computing device, such as a server device. Components of the electronic device 60 may include, but are not limited to, the at least one processor 61, the at least one memory 62, and a bus 63 connecting different system components (including the memory 62 and the processor 61).

[0122] The bus 63 includes a data bus, an address bus, and a control bus.

[0123] The memory 62 may include a volatile memory, such as a random access memory (RAM) 621 and / or a cache memory 622 , and may further include a read-only memory (ROM) 623 .

[0124] The memory 62 may also include a program / utility 625 having a set (at least one) of program modules 624, such program modules 624 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0125] The processor 61 executes various functional applications and data processing by running computer programs stored in the memory 62, such as the method for identifying abnormal trends in dam monitoring data provided in the first aspect of the present application.

[0126] The electronic device 60 can also communicate with one or more external devices 64 (e.g., a keyboard, pointing device, etc.). This communication can occur via an input / output (I / O) interface 65. Furthermore, the model-generating device 60 can also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 66. As shown, the network adapter 66 communicates with other modules of the model-generating device 60 via a bus 63. It should be understood that, although not shown, other hardware and / or software modules can be used in conjunction with the model-generating device 60, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.

[0127] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more units / modules described above may be embodied in a single unit / module. Conversely, the features and functions of a single unit / module described above may be further divided and embodied by multiple units / modules.

[0128] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned 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.

[0129] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for identifying abnormal trends in dam monitoring data, characterized in that: include: Acquiring monitoring data of the dam, and calculating a first deformation trend component of the dam based on the monitoring data using a statistical regression method; updating the parameters of the residual neural network obtained in advance according to the monitoring data, and determining the second deformation trend component of the dam based on the updated parameters; A correlation between the first deformation trend component and the second deformation trend component is calculated, and in response to the correlation being less than a preset threshold, an abnormal trend in the monitoring data is determined based on a difference between the first deformation trend component and the second deformation trend component through a clustering algorithm.

2. The method for identifying abnormal trends in dam monitoring data according to claim 1, characterized in that: The monitoring data includes dam monitoring data, monitoring time, and upstream water level. The updating of the pre-acquired parameters of the residual neural network according to the monitoring data includes: Determining a prediction result based on the monitoring time and the upstream water level by a pre-acquired residual neural network; The parameters of the residual neural network are updated according to the dam monitoring data and the prediction results.

3. The method for identifying abnormal trends in dam monitoring data according to claim 2, characterized in that: Determining the prediction result includes: The prediction results are divided into a period component and a trend component, and the prediction results are calculated as follows: Among them, X output Indicates the preset result, W c represents the weight of the periodic component, represents the periodic component, W t represents the weight of the trend component, Represents the trend component.

4. The method for identifying abnormal trends in dam monitoring data according to claim 3, characterized in that: Determining the second deformation trend component includes: Among them, δ θ2 represents the second deformation trend component, W t represents the weight of the trend component, represents the trend component, which is calculated as follows: in, These are intermediate results. are all weight matrices, are all bias vectors, i=1,2,3, j=1,2,3, relu(*) represents the relu activation function, represents an input parameter, which is determined based on the monitoring time, σ(*) represents a Sigmod activation function, and hardtanh(*) represents a hardtanh activation function.

5. The method for identifying abnormal trends in dam monitoring data according to claim 1, characterized in that: The calculating the correlation between the first deformation trend component and the second deformation trend component includes: Among them, δ θ1 represents the first deformation trend component, δ θ2 represents the second deformation trend component, and E(*) represents the mean calculation.

6. The method for identifying abnormal trends in dam monitoring data according to claim 2, characterized in that: The determining of the abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component by using a clustering algorithm includes: calculating a difference between the first deformation trend component and the second deformation trend component, and classifying the difference into two categories using a clustering algorithm; Abnormal data is determined from the two categories according to the quantity of each category, and it is determined that the dam monitoring data corresponding to the abnormal data has an abnormal trend.

7. The method for identifying abnormal trends in dam monitoring data according to claim 2, characterized in that: The calculating the first deformation trend component of the dam based on the monitoring data by using a statistical regression method includes: d=d H +d T +d θ1 d θ1 =c1θ+c2lnθ Wherein, δ represents the dam monitoring data, δ H represents the deformation caused by water level change, δ T Indicates deformation caused by temperature change, δ θ1 represents the first deformation trend component, H represents the upstream water level, a i represents the coefficient to be fitted, n represents the quantity, c1 and c2 are both time coefficients, t represents the monitoring time, which represents the number of days from the current time to the start time of the monitoring data, θ is equal to t / 100, b 1i and b 2i All are statistical coefficients.

8. The method for identifying abnormal trends in dam monitoring data according to claim 1, characterized in that: The method further comprises: In response to the correlation being greater than or equal to a preset threshold, it is determined that the trend of the monitoring data is normal.

9. A dam monitoring data trend anomaly identification system, characterized by: include: A first trend module is used to obtain monitoring data of the dam and calculate a first deformation trend component of the dam based on the monitoring data using a statistical regression method; A second trend module is configured to update the parameters of the pre-acquired residual neural network according to the monitoring data, and determine the second deformation trend component of the dam based on the updated parameters; Trend judgment module: used to calculate the correlation between the first deformation trend component and the second deformation trend component, and in response to the correlation being less than a preset threshold, determine the abnormal trend in the monitoring data based on the difference between the first deformation trend component and the second deformation trend component through a clustering algorithm.

10. An electronic device, characterized in that: include Memory, processor, and A computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for identifying abnormal trends in dam monitoring data according to any one of claims 1 to 8 is implemented.

Images