Switch cabinet state monitoring method and related device

Abstract

The invention provides a switch cabinet state monitoring method and a related device, the switch cabinet state monitoring method is applied to edge equipment in a switch cabinet state monitoring system, the switch cabinet state monitoring system further comprises an intelligent switch cabinet and at least one sensor module, and the method comprises the following steps: acquiring first sensor data in a preset time period through a first sensor module, the first sensor module is any one of the at least one sensor module; determining a first model parameter of a first anomaly detection model corresponding to a first detection position of the first sensor module; analyzing the first sensor data to obtain a first analysis result; determining a second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; determining an anomaly detection result according to the second model parameter, the first anomaly detection model and the first sensor data; and performing early warning operation according to an abnormal detection result. According to the embodiment of the invention, the abnormity monitoring accuracy of the intelligent switch cabinet can be improved.

Description

Technical Field

[0001] The present application relates to the field of smart grid technology or electric power technology, and specifically to a switch cabinet status monitoring method and related devices. Background Art

[0002] In actual applications, the smart switch cabinet is an electrical device. The external line of the smart switch cabinet first enters the main control switch inside the cabinet, and then enters the sub-control switch. Each branch is set according to its needs. At present, real-time monitoring and maintenance of the operation process of the smart switch cabinet is very important, but the accuracy of abnormal detection of the smart switch cabinet is low. Therefore, how to improve the accuracy of abnormal monitoring of the smart switch cabinet needs to be solved urgently. Summary of the invention

[0003] The embodiments of the present application provide a switch cabinet status monitoring method and related devices, which can improve the accuracy of abnormality monitoring of an intelligent switch cabinet.

[0004] In a first aspect, an embodiment of the present application provides a switch cabinet state monitoring method, which is applied to an edge device in a switch cabinet state monitoring system, wherein the switch cabinet state monitoring system further includes: an intelligent switch cabinet and at least one sensor module, each sensor module corresponding to a detection position of the intelligent switch cabinet; the method includes: Acquiring first sensor data of a preset time period through a first sensor module, wherein the first sensor module is any sensor module among the at least one sensor module; Determine a first model parameter of a first abnormality detection model corresponding to a first detection position of the first sensor module; Analyze the first sensor data to obtain a first analysis result; Determine a second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Determine an anomaly detection result according to the second model parameter, the first anomaly detection model and the first sensor data, the anomaly detection result including a plurality of anomaly types, each anomaly type corresponding to a probability value; Perform an early warning operation based on the abnormal detection result.

[0005] In a second aspect, an embodiment of the present application provides a switch cabinet state monitoring device, which is applied to an edge device in a switch cabinet state monitoring system, wherein the switch cabinet state monitoring system further includes: an intelligent switch cabinet and at least one sensor module, each sensor module corresponding to a detection position of the intelligent switch cabinet; the device includes: an acquisition unit, a first determination unit, an analysis unit, a second determination unit and an early warning unit, wherein: The obtaining unit is configured to obtain first sensor data for a preset time period through a first sensor module, where the first sensor module is any one of the at least one sensor module; The first determining unit is configured to determine first model parameters of a first anomaly detection model corresponding to a first detection position of the first sensor module; The analysis unit is configured to analyze the first sensor data to obtain a first analysis result; The second determining unit is configured to determine second model parameters of the first anomaly detection model according to the first analysis result and the first model parameters; and determine an anomaly detection result according to the second model parameters, the first anomaly detection model, and the first sensor data, where the anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value; The warning unit is configured to perform a warning operation according to the anomaly detection result.

[0006] In a third aspect, an embodiment of the present application provides an edge device, including a processor, a memory, a communication interface, and one or more programs, where the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for performing the steps in the first aspect of the embodiments of the present application.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program for electronic data exchange, and the computer program causes a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application.

[0008] In a fifth aspect, an embodiment of the present application provides a computer program product, where the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application. The computer program product may be a software installation package.

[0009] Implementing the embodiments of the present application has the following beneficial effects: It can be seen that the switchgear status monitoring method and related devices described in the embodiments of the present application are applied to edge devices in a switchgear status monitoring system. The switchgear status monitoring system further includes: an intelligent switchgear and at least one sensor module. Each sensor module corresponds to a detection position of an intelligent switchgear. The first sensor data for a preset time period is obtained through the first sensor module. The first sensor module is any one of the at least one sensor module. The first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module is determined. The first sensor data is analyzed to obtain a first analysis result. The second model parameter of the first anomaly detection model is determined according to the first analysis result and the first model parameter. The anomaly detection result is determined according to the second model parameter, the first anomaly detection model, and the first sensor data. The anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value. An early warning operation is performed according to the anomaly detection result. On the one hand, since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it is possible to preliminarily detect whether there is an anomaly in the intelligent switchgear. On the other hand, since the first analysis result has preliminarily captured the anomaly to a certain extent, the model parameters are dynamically optimized based on this anomaly, so that the optimized model parameters are more in line with the actual environmental changes, which helps to reduce the probability of model misidentification and also helps to improve the accuracy of anomaly detection. Thus, accurate anomaly detection of the intelligent switchgear can be achieved. Furthermore, it helps to achieve accurate early warning operations and ensure the safety of the intelligent switchgear. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 is a schematic structural diagram of a switchgear status monitoring system provided by an embodiment of the present application; Figure 2 is a schematic structural diagram of another switchgear status monitoring system provided by an embodiment of the present application; Figure 3 is a schematic structural diagram of an intelligent switchgear provided by an embodiment of the present application; Figure 4 is a schematic structural diagram of another switchgear status monitoring system provided by an embodiment of the present application; Figure 5 is a schematic flowchart of a switchgear status monitoring method provided by an embodiment of the present application; Figure 6It is a schematic structural diagram of an edge device provided by an embodiment of the present application; Figure 7 It is a block diagram of functional units of a switch cabinet status monitoring device provided by an embodiment of the present application. Detailed implementation manners

[0012] Terms such as "first" and "second" in the description and claims of the present application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or modules is not limited to the listed steps or modules, but may also include unlisted steps or modules in a possible example, or other steps or modules inherent to these processes, methods, products or devices in a possible example.

[0013] Referring to "embodiment" herein means that a specific feature, structure or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appears in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0014] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0015] In specific implementation, the edge device includes an edge computing device, an edge Internet of Things device, an edge server, etc., which are not limited herein.

[0016] Among them, the intelligent switch cabinet includes at least one of the following: low-voltage switch cabinet, high-voltage switch cabinet, ultra-high-voltage switch cabinet, etc., which are not limited herein.

[0017] Figure 1 It is a schematic structural diagram of a switch cabinet status monitoring system involved in an embodiment of the present application. The switch cabinet status monitoring system includes: an edge device, an intelligent switch cabinet, and at least one sensor module. Among them, the edge device, the intelligent switch cabinet, and at least one sensor module are communicatively connected. Each sensor module corresponds to a detection position of an intelligent switch cabinet. Based on the edge device in the switch cabinet status monitoring system, the following functions can be realized: Obtain first sensor data for a preset time period through a first sensor module, where the first sensor module is any one of the at least one sensor module; Determine the first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module; Analyze the first sensor data to obtain a first analysis result; Determine the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Determine an anomaly detection result according to the second model parameter, the first anomaly detection model, and the first sensor data, where the anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value; Perform a warning operation according to the anomaly detection result.

[0018] Among them, the sensor module may include at least one of the following: temperature and humidity monitoring module, partial discharge monitoring module, mechanical property monitoring module, switch comprehensive status collector, micro water density sensor, etc., which are not limited here. The partial discharge monitoring module may include at least one of the following sensors: partial discharge sensor, partial discharge collector, etc., which are not limited here. The mechanical property monitoring module may include at least one of the following sensors: opening and closing coil sensors, energy storage motor sensors, primary current sensors, etc., which are not limited here. The sensor module can detect at least one type of sensor data, that is, the sensor module can integrate at least one sensor function.

[0019] Among them, the sensor parameters of the temperature and humidity monitoring module may include at least one of the following: temperature measurement accuracy, humidity measurement accuracy, protection level, measurement period, mean time between failures, power supply method, power consumption, etc., which are not limited here.

[0020] Among them, the sensor parameters of the partial discharge sensor may include at least one of the following: detection range, response speed, protection level, housing material, ambient temperature, etc., which are not limited here.

[0021] Among them, the sensor parameters of the partial discharge collector may include at least one of the following: number of channels, acquisition method, communication method, power supply, power consumption, ambient temperature, etc., which are not limited here.

[0022] Among them, the sensor parameters of the opening and closing coil sensors may include at least one of the following: current measurement range, current sampling frequency, number of current acquisition channels, etc., which are not limited here.

[0023] Among them, the sensor parameters of the energy storage motor sensors may include at least one of the following: current measurement range, current sampling frequency, number of current acquisition channels, etc., which are not limited here.

[0024] Among them, the sensor parameters of the primary current sensor may include at least one of the following: measurement current range, AC current sampling rate, etc., which are not limited herein.

[0025] Among them, the sensor parameters of the switch comprehensive status collector may include at least one of the following: operating voltage, number of manageable terminals, wireless transmission distance, operating life, etc., which are not limited herein.

[0026] Among them, the preset time period can be set in advance or be the system default. The preset time period can be understood as a time period with a preset duration before the current moment. This preset time period may or may not include the current moment, and the preset duration can be set in advance or be the system default. The preset time period can be understood as the current time period.

[0027] In specific implementation, as Figure 2 shown, at least one sensor module can be integrated on the intelligent switchgear cabinet, that is, the intelligent switchgear cabinet can include at least one sensor module. The intelligent switchgear cabinet can form a switchgear cabinet status monitoring system with the edge device, and this switchgear cabinet status monitoring system can include the edge device and the intelligent switchgear cabinet. Further, as Figure 3 shown, the intelligent switchgear cabinet can include multiple sensor modules. The sensor modules can be arranged at different positions of the intelligent switchgear cabinet, and the specific positions are determined based on the functions, roles, and circuit layouts of the intelligent switchgear cabinet, which are not limited herein.

[0028] In specific implementation, the sensor data of at least one sensor module can be directly uploaded to the edge device, or the sensor data of at least one sensor module can be first sent to the intelligent switchgear cabinet, and then the intelligent switchgear cabinet sends it to the edge device uniformly.

[0029] In specific implementation, as Figure 4 shown, the switchgear cabinet status monitoring system includes an intelligent switchgear cabinet, an edge device, and a server. The intelligent switchgear cabinet, the edge device, and the server are communicatively connected. The edge device can also send the sensor data to the server for data backup.

[0030] In specific implementation, taking the first sensor module as an example, the first sensor module is any one of the at least one sensor module. The first sensor data for the preset time period can be obtained through the first sensor module. Since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it can be preliminarily detected whether the intelligent switchgear cabinet is abnormal.

[0031] Among them, the first anomaly detection model includes any model used to implement the anomaly detection of intelligent switchgear. Specifically, for example, the first anomaly detection model may include at least one of the following: linear regression model, classifier, neural network model (such as recurrent neural network model, convolutional neural network model, etc.), large model (such as ChatGpt), etc., which are not limited herein.

[0032] In specific implementation, the mapping relationship between the preset detection position and the anomaly detection model can be stored in advance. Furthermore, based on this mapping relationship, the first anomaly detection model corresponding to the first detection position is determined. The mapping relationship between the sensor parameters of the preset sensor module and the model parameters of the first anomaly detection model can also be stored in advance. Based on this mapping relationship, the first model parameters of the first anomaly detection model corresponding to the first detection position of the first sensor module are determined. Among them, the first anomaly detection model is used to implement the anomaly detection function, and the model parameters of the first anomaly detection model are used to control the anomaly detection effect. The anomaly detection effect may include at least one of the following: anomaly detection accuracy, anomaly detection speed, anomaly detection sensitivity, etc., which are not limited herein.

[0033] In specific implementation, the sensor data of the first sensor module at the first detection position can be collected and used as the training set and test set. Various real labels (real anomaly types) and loss functions can also be set. Different sensor data can correspond to different real labels, and each real label corresponds to an anomaly type. The preset anomaly detection model is trained based on this training set to obtain a training result. When the training reaches a certain level, the preset anomaly detection model is tested based on the test set, and the model parameters are dynamically adjusted based on the corresponding test results, real results, and loss functions until the set conditions are met, and the first preset anomaly detection model is obtained. The set conditions can be set in advance or be the system default. The preset anomaly detection model can be set in advance or be the system default.

[0034] Among them, the anomaly type can be set in advance or be the system default. The anomaly types existing at different detection positions can be the same or different. For example, the anomaly type may include at least one of the following: operating mechanism anomaly, electrical fault anomaly, insulation-related anomaly, external force and other anomalies, etc., which are not limited herein. For example, the electrical fault anomaly may include at least one of the following: circuit breaker anomaly, busbar anomaly, secondary circuit anomaly, etc., which are not limited herein. The insulation-related anomaly may include at least one of the following: insulation aging, moisture absorption, flashover, etc., which are not limited herein.

[0035] Next, the first sensor data can be analyzed to obtain a first analysis result. Since the first sensor data is continuously collected data over a period of time, it represents the state change of the intelligent switch cabinet to a certain extent. Based on the first sensor data, the state change of the intelligent switch cabinet can be deeply captured, that is, the first analysis result can be obtained.

[0036] In specific implementation, the second model parameter of the first anomaly detection model can be determined according to the first analysis result and the first model parameter. That is, on the one hand, the first analysis result has initially captured anomalies to a certain extent. On the other hand, based on the first analysis result, the first model parameter is dynamically optimized to obtain the second model parameter, making the optimized model parameter more in line with the actual environmental changes, which helps to reduce the model misidentification probability and also helps to improve the anomaly detection accuracy.

[0037] Next, the first sensor data can be input into the first anomaly detection model configured with the second model parameter to obtain an anomaly detection result. Of course, the first sensor data can also be subjected to feature extraction first to obtain a feature set, and then the feature set is input into the first anomaly detection model configured with the second model parameter to obtain an anomaly detection result. The anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value. Then, a warning operation is performed according to the anomaly detection result. Since the first analysis result has initially captured anomalies to a certain extent, and then the model parameter is dynamically optimized based on this anomaly, making the optimized model parameter more in line with the actual environmental changes, which helps to reduce the model misidentification probability and also helps to improve the anomaly detection accuracy. Thus, accurate anomaly detection of the intelligent switch cabinet can be realized. Furthermore, it helps to realize an accurate warning operation and ensure the safety of the intelligent switch cabinet.

[0038] Optionally, for the above steps of determining the first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module, it can be implemented in the following manner: Determine the first circuit position of the first circuit structure diagram corresponding to the first detection position; Obtain the historical anomaly detection record set of the first circuit position, and the historical anomaly detection record set includes multiple anomaly detection records, and each anomaly detection record corresponds to a working parameter set and a model parameter; Obtain the first working parameter set of the first circuit position at the current moment; Compare the first working parameter set with the working parameter sets corresponding to each anomaly detection record in the multiple anomaly detection records to obtain multiple comparison values; Select the maximum value among the multiple comparison values, and obtain the model parameter corresponding to the maximum value to obtain the first model parameter.

[0039] In specific implementation, the first circuit position of the first circuit structure diagram corresponding to the first detection position can be determined. The first detection position can be the same as or different from the first circuit position. Due to the characteristics of the circuit itself, the loads corresponding to different positions are different, and thus the probabilities of abnormal occurrences also vary.

[0040] In the embodiments of the present application, a historical abnormal detection record set of the first circuit position within a preset historical time period can be obtained. The historical abnormal detection record set includes a plurality of abnormal detection records. Each abnormal detection record corresponds to a set of working parameters and a set of model parameters. The end time of the preset historical time period is earlier than the start time of the preset time period. The set of working parameters can include at least one working parameter corresponding to the first circuit position. The working parameter can include at least one of the following: working current, working voltage, working power, working load, working temperature, etc., which are not limited herein.

[0041] Among them, each abnormal detection record in the plurality of abnormal detection records can include an abnormal detection record with successful abnormal detection, that is, the detection result of the first abnormal detection model is consistent with the actual situation (manual detection result).

[0042] Then, a first set of working parameters of the first circuit position at the current moment can be obtained. The first set of working parameters can also include at least one working parameter. The working parameter can include at least one of the following: working current, working voltage, working power, working load, working temperature, etc., which are not limited herein.

[0043] Furthermore, the first set of working parameters can be compared with the set of working parameters corresponding to each abnormal detection record in the plurality of abnormal detection records to obtain a plurality of comparison values. For example, when both the first set of working parameters and the set of working parameters include only one working parameter, the difference between the working parameters can be determined. The smaller the difference, the larger the comparison value; the larger the difference, the smaller the comparison value. For example, the target difference between the working parameters is determined. The target difference characterizes the difference between the working parameters. A mapping relationship between the preset difference and the comparison value can also be stored in advance. Furthermore, based on this mapping relationship, the comparison value corresponding to the target difference can be determined. Correspondingly, when both the first set of working parameters and the set of working parameters include a plurality of working parameters, the plurality of comparison values corresponding to the plurality of working parameters can be determined in the above manner, and then the plurality of comparison values are weighted and calculated to obtain the final comparison value.

[0044] Finally, the maximum value among multiple comparison values can be selected, and the model parameters corresponding to the maximum value can be obtained to get the first model parameters. In this way, the big data technology can be used to obtain the model parameters (successful experience) of the anomaly detection that is closest to the current situation in the historical anomaly detection experience as the first model parameters. That is, the corresponding model parameters can be determined with the help of historical experience, ensuring the feasibility and adaptability of the model parameters and helping to ensure the accuracy of model detection.

[0045] Optionally, for the above step of analyzing the first sensor data to obtain the first analysis result, it can be implemented as follows: Determine a first fitting straight line and a first fitting curve segment corresponding to the preset time period according to the first sensor data; Determine a first absolute value of the slope of the first fitting straight line; Determine a first standard deviation of the first fitting curve segment; Determine the first analysis result according to the first absolute value and the first standard deviation.

[0046] In the embodiment of the present application, since the first sensor data can collect sensor data at preset time intervals, based on this method, multiple sensor data can be obtained, and each sensor data corresponds to a collection moment. The preset time interval can be set in advance or be the system default. That is, multiple sensor data and the collection moment corresponding to each sensor data can be mapped to a coordinate system to obtain multiple coordinate points, and each sensor data corresponds to a coordinate point. The horizontal axis of this coordinate system is time, and the vertical axis is sensor data.

[0047] Next, straight line fitting can be performed based on these multiple coordinate points to obtain a first fitting straight line. Correspondingly, curve fitting can be performed based on these multiple coordinate points to obtain a first fitting curve segment corresponding to the preset time period.

[0048] Furthermore, the slope of the first fitted straight line can be determined, and then the absolute value is taken based on this slope to obtain the first absolute value. To a certain extent, the first absolute value reflects the change trend of the intelligent switchgear. If there is no abnormality in the intelligent switchgear, the first absolute value approaches 0. If an abnormality occurs in the intelligent switchgear, the first absolute value will be greater than a certain threshold. Based on this, the change of the intelligent switchgear can be initially captured using the sensor data. Correspondingly, a preset number of reference points can be selected from the first fitted curve segment to obtain multiple reference points, and the first standard deviation is obtained based on the operation of these multiple reference points. The preset number can be set in advance or be the system default. The preset number can be positively correlated with the magnitude of the first absolute value, that is, the larger the first absolute value, the larger the preset number. To a certain extent, the first standard deviation reflects the working stability of the intelligent switchgear. If there is no abnormality in the intelligent switchgear, the first standard deviation is smaller. If an abnormality occurs in the intelligent switchgear, the first standard deviation will be greater than a certain threshold. Based on this, the change of the intelligent switchgear can be initially captured using the sensor data.

[0049] Furthermore, the first analysis result can be determined based on the first absolute value and the first standard deviation, that is, the first analysis result can include the first absolute value and the first standard deviation. In this way, not only can the change of the intelligent switchgear be initially captured using the first absolute value and the first standard deviation, but it also helps to dynamically optimize the model parameters using the first absolute value and the first standard deviation, improving the detection accuracy of the model.

[0050] Optionally, the above step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter may include the following steps: According to the first adjustment parameter corresponding to the first absolute value; Determine the first fine-tuning parameter corresponding to the first standard deviation; Determine the second model parameter according to the first adjustment parameter, the first fine-tuning parameter, and the first model parameter.

[0051] In specific implementation, since the first absolute value reflects the change trend of the intelligent switchgear to a certain extent and the first standard deviation reflects the working stability of the intelligent switchgear to a certain extent, furthermore, the first mapping relationship between the preset absolute value and the adjustment parameter can be stored in advance. This first mapping relationship can be obtained based on experience. The second mapping relationship between the preset standard deviation and the fine-tuning parameter can also be stored in advance. Furthermore, the first fine-tuning parameter corresponding to the first standard deviation can be determined based on this second mapping relationship. Among them, the value ranges of the adjustment parameter and the fine-tuning parameter can be set in advance or be the system default.

[0052] Next, adjustment can be performed based on some or all of the first adjustment parameter, the first fine-tuning parameter, and the first model parameter to obtain the second model parameter. For example, the second model parameter = (1 + the first adjustment parameter) * (1 + the first fine-tuning parameter) * the first model parameter. Another example is that the second model parameter = (1 + the first adjustment parameter * (1 + the first fine-tuning parameter)) * the first model parameter. Since the model parameters are adjusted according to the change trend of the intelligent switchgear and the working stability of the intelligent switchgear, the adaptability of the model and the robustness of the model can be improved, and further the accuracy of the model can be ensured.

[0053] Optionally, the following functions can also be implemented: Detect whether the first absolute value is greater than the first preset absolute value; When the first absolute value is greater than the first preset absolute value, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first standard deviation is greater than the first preset standard deviation; When the first standard deviation is greater than the first preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first absolute value is greater than the second preset absolute value and whether the first standard deviation is greater than the second preset standard deviation; When the first absolute value is greater than the second preset absolute value and the first standard deviation is greater than the second preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter.

[0054] Among them, since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it can be preliminarily detected whether there is an anomaly in the intelligent switchgear. For example, both the first absolute value and the first standard deviation can be used to preliminarily detect whether there is an anomaly in the intelligent switchgear.

[0055] Among them, the first preset absolute value can be set in advance or default by the system. The first preset absolute value is a positive number greater than 0.

[0056] Specifically, it is possible to detect whether the first absolute value is greater than the first preset absolute value. When the first absolute value is greater than the first preset absolute value, it indicates that there is an abnormality in the intelligent switchgear, and then the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter can be executed. On the contrary, when the first absolute value is not greater than the first preset absolute value, it indicates that there is no abnormality in the intelligent switchgear or the probability of an abnormality is very small, and then the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter can be not executed. In this way, anomaly detection can be performed only when it is initially detected that there may be an abnormality in the intelligent switchgear, which helps to ensure the accuracy of anomaly detection. On the contrary, when it is initially detected that there may be no abnormality in the intelligent switchgear or the probability of an abnormality is very small, subsequent anomaly detection is not performed, which can reduce power consumption.

[0057] Among them, the first preset standard deviation can be set in advance or default by the system. The first preset standard deviation is a positive number greater than 0.

[0058] In specific implementation, it is possible to detect whether the first standard deviation is greater than the first preset standard deviation. When the first standard deviation is greater than the first preset standard deviation, it indicates that there is an abnormality in the intelligent switchgear, and then the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter can be executed; on the contrary, when the first standard deviation is not greater than the first preset standard deviation, it indicates that there is no abnormality in the intelligent switchgear or the probability of an abnormality is very small, and then the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter can be not executed. In this way, anomaly detection can be performed only when it is initially detected that there may be an abnormality in the intelligent switchgear, which helps to ensure the accuracy of anomaly detection. On the contrary, when it is initially detected that there may be no abnormality in the intelligent switchgear or the probability of an abnormality is very small, subsequent anomaly detection is not performed, which can reduce power consumption.

[0059] Among them, both the second preset absolute value and the second preset standard deviation can be set in advance or default by the system. The second preset absolute value is a positive number greater than 0. The second preset standard deviation is a positive number greater than 0.

[0060] In a specific implementation, it is possible to detect whether the first absolute value is greater than the second preset absolute value and whether the first standard deviation is greater than the second preset standard deviation. When the first absolute value is greater than the second preset absolute value and the first absolute value is greater than the second preset standard deviation, it indicates that there is an abnormality in the intelligent switchgear, and then the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter can be executed; conversely, when the first absolute value is not greater than the second preset absolute value, and / or, the first absolute value is not greater than the second preset standard deviation, it indicates that there is no abnormality in the intelligent switchgear or the probability of an abnormality is very small, and then the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter may not be executed. In this way, anomaly detection can be performed only when it is initially detected that there may be an abnormality in the intelligent switchgear, which helps to ensure the accuracy of anomaly detection. On the contrary, when it is initially detected that there may be no abnormality in the intelligent switchgear or the probability of an abnormality is very small, subsequent anomaly detection is not performed, which can reduce power consumption.

[0061] Optionally, the above step of performing a warning operation according to the anomaly detection result can be implemented as follows: Determine probability thresholds for the multiple anomaly types according to the first absolute value and the first standard deviation, obtaining multiple probability thresholds; Determine multiple probability differences between the multiple probability values corresponding to the multiple anomaly types and the multiple probability thresholds; Select the probability differences greater than 0 among the multiple probability differences, obtaining at least one probability difference; Obtain the anomaly types corresponding to the at least one probability difference, obtaining at least one anomaly type; Perform a warning operation according to the at least one probability difference and the at least one anomaly type.

[0062] In the embodiment of the present application, since the first absolute value reflects the change trend of the intelligent switchgear to a certain extent, and the first standard deviation reflects the working stability of the intelligent switchgear to a certain extent, the first absolute value and the first standard deviation can be used together to characterize the quality of the working state of the intelligent switchgear. Different working states may have different possible abnormal situations. Therefore, the working state of the intelligent switchgear can be evaluated based on the first absolute value and the first standard deviation to obtain an evaluation value, and then the mapping relationship between the preset evaluation value and the set of probability thresholds is pre-stored. Based on this mapping relationship, a set of probability thresholds corresponding to the evaluation value is determined. The set of probability thresholds includes multiple probability thresholds, and each probability threshold corresponds to an anomaly type.

[0063] Next, multiple probability values can be subtracted from the corresponding probability thresholds among multiple probability thresholds to obtain multiple probability differences. Select the probability differences greater than 0 among the multiple probability differences to obtain at least one probability difference. If the probability difference is greater than 0, it indicates that there is an abnormal situation corresponding to the corresponding abnormal type. Then, obtain the abnormal types corresponding to at least one probability difference to obtain at least one abnormal type. Finally, an early warning operation can be performed based on at least one probability difference and at least one abnormal type. For example, the mapping relationship between the preset abnormal type and the early warning parameter set can be stored in advance. Each early warning parameter set can include at least one set of early warning parameters, and each set of early warning parameters corresponds to an early warning level. Specifically, the mapping relationship between the preset probability difference and the early warning level can be stored in advance for each abnormal type. Furthermore, the early warning level corresponding to the probability difference corresponding to each abnormal type can be determined, and then a corresponding set of early warning parameters can be obtained. Based on this method, the early warning parameters corresponding to each abnormal type can be obtained, and an early warning operation can be performed based on these early warning parameters, thereby ensuring the safety of the intelligent switchgear and ensuring the working efficiency of the intelligent switchgear.

[0064] Among them, the early warning parameters can include at least one of the following: early warning method, early warning personnel, early warning frequency, early warning range, early warning location, etc., which are not limited here. The early warning method can include at least one of the following: display method, vibration method, voice method, email method, text message method, etc., which are not limited here. The early warning personnel are understood as the required operation and maintenance personnel.

[0065] Optionally, the above step of determining the probability thresholds of the multiple abnormal types according to the first absolute value and the first standard deviation to obtain multiple probability thresholds can be implemented as follows: Determine the first evaluation value corresponding to the first absolute value; Determine the second evaluation value corresponding to the first standard deviation; Determine the target evaluation value according to the first evaluation value and the second evaluation value; Determine the multiple probability thresholds corresponding to the target evaluation value.

[0066] In specific implementation, the mapping relationship between the preset absolute value and the evaluation value can be stored in advance. Based on this mapping relationship, the first evaluation value corresponding to the first absolute value can be determined. Correspondingly, the mapping relationship between the preset standard deviation and the evaluation value can also be stored in advance. Based on this mapping relationship, the second evaluation value corresponding to the first standard deviation can be determined. A preset weight pair can be obtained. The preset weight pair includes a first weight and a second weight. The first weight is the weight corresponding to the first evaluation value, and the second weight is the weight corresponding to the second evaluation value. The sum of the first weight and the second weight is 1, that is, the target evaluation value = first weight * first evaluation value + second weight * second evaluation value.

[0067] Among them, the preset weight pair can correspond to the detection position, that is, the mapping relationship between the preset detection position and the weight pair is stored in advance, and the preset weight pair corresponding to the first detection position is determined based on this mapping relationship. Of course, the preset weight pair can also correspond to the first absolute value, or the preset weight pair can also correspond to the first standard deviation.

[0068] Next, the mapping relationship between the preset evaluation value and the set of probability thresholds can also be stored in advance, and a set of probability thresholds corresponding to the target evaluation value is determined based on this mapping relationship. The set of probability thresholds includes multiple probability thresholds, and each probability threshold corresponds to an abnormal type. Thus, since the first absolute value reflects the change trend of the intelligent switchgear to a certain extent, and the first standard deviation reflects the working stability of the intelligent switchgear to a certain extent, the first absolute value and the first standard deviation can be used together to characterize the quality of the working state of the intelligent switchgear. Different working states may have different possible abnormal conditions, that is, the corresponding probability thresholds can be dynamically determined based on the working state of the intelligent switchgear. Furthermore, the accuracy of abnormal detection and the accuracy of early warning operations can be further ensured, which helps to ensure the safety of the intelligent switchgear.

[0069] It can be seen that in the switchgear state monitoring system described in the embodiments of the present application, the switchgear state monitoring system further includes: an intelligent switchgear and at least one sensor module. Each sensor module corresponds to a detection position of the intelligent switchgear. The first sensor data for a preset time period is obtained through the first sensor module. The first sensor module is any one of the at least one sensor module. The first model parameter of the first abnormal detection model corresponding to the first detection position of the first sensor module is determined, the first sensor data is analyzed to obtain a first analysis result, the second model parameter of the first abnormal detection model is determined according to the first analysis result and the first model parameter, and the abnormal detection result is determined according to the second model parameter, the first abnormal detection model and the first sensor data. The abnormal detection result includes multiple abnormal types, and each abnormal type corresponds to a probability value. An early warning operation is performed according to the abnormal detection result. On the one hand, since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it can be preliminarily detected whether the intelligent switchgear is abnormal. On the other hand, since the first analysis result has preliminarily captured the abnormality to a certain extent, the model parameters are dynamically optimized based on this abnormality, so that the optimized model parameters are more in line with the actual environmental changes, which helps to reduce the probability of model misidentification and also helps to improve the accuracy of abnormal detection. Thus, accurate abnormal detection of the intelligent switchgear can be realized, and furthermore, accurate early warning operations can be realized to ensure the safety of the intelligent switchgear.

[0070] Please refer to Figure 5 , Figure 5It is a schematic flowchart of a switchgear status monitoring method provided by an embodiment of the present application. This method is applied to an edge device in a switchgear status monitoring system. The switchgear status monitoring system further includes: an intelligent switchgear and at least one sensor module, and each sensor module corresponds to a detection position of the intelligent switchgear. This switchgear status monitoring method includes: 501. Obtain first sensor data for a preset time period through a first sensor module, where the first sensor module is any one of the at least one sensor module.

[0071] 502. Determine first model parameters of a first anomaly detection model corresponding to the first detection position of the first sensor module.

[0072] 503. Analyze the first sensor data to obtain a first analysis result.

[0073] 504. Determine second model parameters of the first anomaly detection model according to the first analysis result and the first model parameters.

[0074] 505. Determine an anomaly detection result according to the second model parameters, the first anomaly detection model, and the first sensor data. The anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value.

[0075] 506. Perform a warning operation according to the anomaly detection result.

[0076] Among them, the specific descriptions of the above steps 501-506 can refer to the descriptions in the above part and will not be repeated here.

[0077] Optionally, for the above step 502, to determine the first model parameters of the first anomaly detection model corresponding to the first detection position of the first sensor module, it can be implemented in the following manner: Determine a first circuit position of a first circuit structure diagram corresponding to the first detection position; Obtain a historical anomaly detection record set of the first circuit position, where the historical anomaly detection record set includes multiple anomaly detection records, and each anomaly detection record corresponds to a set of working parameters and a set of model parameters; Obtain a first set of working parameters of the first circuit position at the current moment; Compare the first set of working parameters with the sets of working parameters corresponding to each anomaly detection record in the multiple anomaly detection records to obtain multiple comparison values; Select the maximum value among the multiple comparison values, and obtain the model parameters corresponding to the maximum value to obtain the first model parameters.

[0078] Among them, the specific description of the above steps can be referred to the description of the above part, and will not be elaborated here.

[0079] Optionally, for the above step 503, analyzing the first sensor data to obtain a first analysis result can be implemented as follows: Determine a first fitting straight line and a first fitting curve segment corresponding to the preset time period according to the first sensor data; Determine a first absolute value of the slope of the first fitting straight line; Determine a first standard deviation of the first fitting curve segment; Determine the first analysis result according to the first absolute value and the first standard deviation.

[0080] Among them, the specific description of the above steps can be referred to the description of the above part, and will not be elaborated here.

[0081] Optionally, for the above step 504, determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter can be implemented as follows: According to a first adjustment parameter corresponding to the first absolute value; Determine a first fine-tuning parameter corresponding to the first standard deviation; Determine the second model parameter according to the first adjustment parameter, the first fine-tuning parameter and the first model parameter.

[0082] Among them, the specific description of the above steps can be referred to the description of the above part, and will not be elaborated here.

[0083] Optionally, the following steps may further be included: Detect whether the first absolute value is greater than a first preset absolute value; When the first absolute value is greater than the first preset absolute value, execute the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first standard deviation is greater than a first preset standard deviation; When the first standard deviation is greater than the first preset standard deviation, execute the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first absolute value is greater than a second preset absolute value and whether the first standard deviation is greater than a second preset standard deviation; When the first absolute value is greater than the second preset absolute value and the first absolute value is greater than the second preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter.

[0084] Among them, the specific description of the above steps can be referred to the description of the above part, and will not be repeated here.

[0085] Optionally, for the above step 506, the warning operation according to the anomaly detection result can be implemented as follows: Determine the probability thresholds of the multiple anomaly types according to the first absolute value and the first standard deviation, and obtain multiple probability thresholds; Determine multiple probability differences between the multiple probability values corresponding to the multiple anomaly types and the multiple probability thresholds; Select the probability differences greater than 0 among the multiple probability differences to obtain at least one probability difference; Obtain the anomaly types corresponding to the at least one probability difference to obtain at least one anomaly type; Perform a warning operation according to the at least one probability difference and the at least one anomaly type.

[0086] Among them, the specific description of the above steps can be referred to the description of the above part, and will not be repeated here.

[0087] Optionally, for the above step of determining the probability thresholds of the multiple anomaly types according to the first absolute value and the first standard deviation to obtain multiple probability thresholds, it can be implemented as follows: Determine the first evaluation value corresponding to the first absolute value; Determine the second evaluation value corresponding to the first standard deviation; Determine the target evaluation value according to the first evaluation value and the second evaluation value; Determine the multiple probability thresholds corresponding to the target evaluation value.

[0088] Among them, the specific description of the above steps can be referred to the description of the above part, and will not be repeated here.

[0089] It can be seen that the switchgear status monitoring method described in the embodiments of the present application is applied to the edge device in the switchgear status monitoring system. The switchgear status monitoring system further includes: an intelligent switchgear and at least one sensor module. Each sensor module corresponds to a detection position of an intelligent switchgear. The first sensor data for a preset time period is obtained through the first sensor module. The first sensor module is any one of the at least one sensor module. The first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module is determined. The first sensor data is analyzed to obtain a first analysis result. The second model parameter of the first anomaly detection model is determined according to the first analysis result and the first model parameter. The anomaly detection result is determined according to the second model parameter, the first anomaly detection model, and the first sensor data. The anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value. An early warning operation is performed according to the anomaly detection result. On the one hand, since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it can be preliminarily detected whether there is an anomaly in the intelligent switchgear. On the other hand, since the first analysis result has preliminarily captured the anomaly to a certain extent, and then the model parameters are dynamically optimized based on this anomaly, the optimized model parameters are more in line with the actual environmental changes, which helps to reduce the probability of model misidentification and also helps to improve the accuracy of anomaly detection. Thus, accurate anomaly detection of the intelligent switchgear can be achieved. Furthermore, it helps to achieve accurate early warning operations and ensure the safety of the intelligent switchgear.

[0090] Consistent with the above embodiments, please refer to Figure 6 , Figure 6 FIG. is a schematic structural diagram of an edge device provided by an embodiment of the present application. The edge device includes a processor, a memory, a communication interface, and one or more programs. Among them, the above one or more programs are stored in the above memory and are configured to be executed by the above processor. In the embodiments of the present application, the edge device is applied to a switchgear status monitoring system. The switchgear status monitoring system further includes: an intelligent switchgear and at least one sensor module. Each sensor module corresponds to a detection position of the intelligent switchgear. The above programs include instructions for performing the following steps: Obtain first sensor data for a preset time period through the first sensor module. The first sensor module is any one of the at least one sensor module; Determine the first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module; Analyze the first sensor data to obtain a first analysis result; Determine the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Determine an anomaly detection result based on the second model parameter, the first anomaly detection model, and the first sensor data, where the anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value; Perform a warning operation based on the anomaly detection result.

[0091] Optionally, in terms of determining the first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module, the above program includes instructions for performing the following steps: Determine the first circuit position of the first circuit structure diagram corresponding to the first detection position; Obtain a historical anomaly detection record set of the first circuit position, where the historical anomaly detection record set includes multiple anomaly detection records, and each anomaly detection record corresponds to a set of working parameters and a model parameter; Obtain the first set of working parameters of the first circuit position at the current moment; Compare the first set of working parameters with the sets of working parameters corresponding to each anomaly detection record in the multiple anomaly detection records to obtain multiple comparison values; Select the maximum value among the multiple comparison values, obtain the model parameter corresponding to the maximum value, and obtain the first model parameter.

[0092] Optionally, in terms of analyzing the first sensor data to obtain a first analysis result, the above program includes instructions for performing the following steps: Determine a first fitting line and a first fitting curve segment corresponding to the preset time period according to the first sensor data; Determine the first absolute value of the slope of the first fitting line; Determine the first standard deviation of the first fitting curve segment; Determine the first analysis result according to the first absolute value and the first standard deviation.

[0093] Optionally, in terms of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter, the above program includes instructions for performing the following steps: According to the first adjustment parameter corresponding to the first absolute value; Determine the first fine-tuning parameter corresponding to the first standard deviation; Determine the second model parameter according to the first adjustment parameter, the first fine-tuning parameter, and the first model parameter.

[0094] Optionally, the above program further includes instructions for performing the following steps: Detect whether the first absolute value is greater than a first preset absolute value; When the first absolute value is greater than the first preset absolute value, execute the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first standard deviation is greater than the first preset standard deviation; When the first standard deviation is greater than the first preset standard deviation, execute the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first absolute value is greater than the second preset absolute value and whether the first standard deviation is greater than the second preset standard deviation; When the first absolute value is greater than the second preset absolute value and the first standard deviation is greater than the second preset standard deviation, execute the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter.

[0095] Optionally, in terms of performing a warning operation according to the anomaly detection result, the above program includes instructions for performing the following steps: Determine the probability thresholds of the multiple anomaly types according to the first absolute value and the first standard deviation, and obtain multiple probability thresholds; Determine multiple probability differences between the multiple probability values corresponding to the multiple anomaly types and the multiple probability thresholds; Select the probability differences greater than 0 from the multiple probability differences to obtain at least one probability difference; Obtain the anomaly types corresponding to the at least one probability difference to obtain at least one anomaly type; Perform a warning operation according to the at least one probability difference and the at least one anomaly type.

[0096] Optionally, in terms of determining the probability thresholds of the multiple anomaly types according to the first absolute value and the first standard deviation to obtain multiple probability thresholds, the above program includes instructions for performing the following steps: Determine a first evaluation value corresponding to the first absolute value; Determine a second evaluation value corresponding to the first standard deviation; Determine a target evaluation value according to the first evaluation value and the second evaluation value; Determine the multiple probability thresholds corresponding to the target evaluation value.

[0097] It can be seen that the edge device described in the embodiments of the present application is applied to a switch cabinet status monitoring system. The switch cabinet status monitoring system further includes: an intelligent switch cabinet and at least one sensor module. Each sensor module corresponds to a detection position of an intelligent switch cabinet. The first sensor data for a preset time period is obtained through the first sensor module. The first sensor module is any one of the at least one sensor module. The first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module is determined. The first sensor data is analyzed to obtain a first analysis result. The second model parameter of the first anomaly detection model is determined according to the first analysis result and the first model parameter. The anomaly detection result is determined according to the second model parameter, the first anomaly detection model, and the first sensor data. The anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value. An early warning operation is performed according to the anomaly detection result. On the one hand, since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it can be preliminarily detected whether there is an anomaly in the intelligent switch cabinet. On the other hand, since the first analysis result has preliminarily captured the anomaly to a certain extent, the model parameters are dynamically optimized based on this anomaly, so that the optimized model parameters are more in line with the actual environmental changes, which helps to reduce the probability of model misidentification and also helps to improve the accuracy of anomaly detection. Thus, accurate anomaly detection of the intelligent switch cabinet can be achieved. Furthermore, it helps to achieve an accurate early warning operation and ensure the safety of the intelligent switch cabinet.

[0098] Figure 7 It is a functional unit composition block diagram of a switch cabinet status monitoring device 700 involved in the present application. The switch cabinet status monitoring device 700 is applied to an edge device in a switch cabinet status monitoring system. The switch cabinet status monitoring system further includes: an intelligent switch cabinet and at least one sensor module. Each sensor module corresponds to a detection position of the intelligent switch cabinet. The switch cabinet status monitoring device 700 includes: an acquisition unit 701, a first determination unit 702, an analysis unit 703, a second determination unit 704, and an early warning unit 705. Among them, The acquisition unit 701 is configured to obtain first sensor data for a preset time period through the first sensor module. The first sensor module is any one of the at least one sensor module. The first determination unit 702 is configured to determine the first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module. The analysis unit 703 is configured to analyze the first sensor data to obtain a first analysis result. The second determination unit 704 is configured to determine the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; determine an anomaly detection result according to the second model parameter, the first anomaly detection model, and the first sensor data, where the anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value; The warning unit 705 is configured to perform a warning operation according to the anomaly detection result.

[0099] Optionally, in terms of determining the first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module, the first determination unit 702 is specifically configured to: Determine the first circuit position of the first circuit structure diagram corresponding to the first detection position; Obtain a historical anomaly detection record set of the first circuit position, where the historical anomaly detection record set includes multiple anomaly detection records, and each anomaly detection record corresponds to a set of working parameters and a model parameter; Obtain the first set of working parameters of the first circuit position at the current moment; Compare the first set of working parameters with the set of working parameters corresponding to each anomaly detection record in the multiple anomaly detection records to obtain multiple comparison values; Select the maximum value from the multiple comparison values, obtain the model parameter corresponding to the maximum value, and obtain the first model parameter.

[0100] Optionally, in terms of analyzing the first sensor data to obtain a first analysis result, the analysis unit 703 is specifically configured to: Determine a first fitting straight line and a first fitting curve segment corresponding to the preset time period according to the first sensor data; Determine the first absolute value of the slope of the first fitting straight line; Determine the first standard deviation of the first fitting curve segment; Determine the first analysis result according to the first absolute value and the first standard deviation.

[0101] Optionally, in terms of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter, the second determination unit 704 is specifically configured to: According to the first adjustment parameter corresponding to the first absolute value; Determine the first fine-tuning parameter corresponding to the first standard deviation; Determine the second model parameter according to the first adjustment parameter, the first fine-tuning parameter, and the first model parameter.

[0102] Optionally, the switchgear status monitoring device 700 is further specifically configured to: Detect whether the first absolute value is greater than a first preset absolute value; When the first absolute value is greater than the first preset absolute value, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first standard deviation is greater than a first preset standard deviation; When the first standard deviation is greater than the first preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first absolute value is greater than a second preset absolute value and whether the first standard deviation is greater than a second preset standard deviation; When the first absolute value is greater than the second preset absolute value and the first standard deviation is greater than the second preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter.

[0103] Optionally, in terms of performing a warning operation according to the anomaly detection result, the warning unit 705 is specifically configured to: Determine probability thresholds for the multiple anomaly types according to the first absolute value and the first standard deviation, to obtain multiple probability thresholds; Determine multiple probability differences between the multiple probability values corresponding to the multiple anomaly types and the multiple probability thresholds; Select the probability differences greater than 0 among the multiple probability differences, to obtain at least one probability difference; Obtain the anomaly types corresponding to the at least one probability difference, to obtain at least one anomaly type; Perform a warning operation according to the at least one probability difference and the at least one anomaly type.

[0104] Optionally, in terms of determining the probability thresholds for the multiple anomaly types according to the first absolute value and the first standard deviation to obtain multiple probability thresholds, the warning unit 705 is specifically configured to: Determine a first evaluation value corresponding to the first absolute value; Determine a second evaluation value corresponding to the first standard deviation; Determine a target evaluation value according to the first evaluation value and the second evaluation value; Determine the multiple probability thresholds corresponding to the target evaluation value.

[0105] It can be seen that the switchgear status monitoring device described in the embodiments of the present application is applied to the edge device in the switchgear status monitoring system. The switchgear status monitoring system further includes: an intelligent switchgear and at least one sensor module. Each sensor module corresponds to a detection position of an intelligent switchgear. The first sensor data for a preset time period is obtained through the first sensor module. The first sensor module is any one of the at least one sensor module. The first model parameter of the first anomaly detection model corresponding to the first detection position of the first sensor module is determined, the first sensor data is analyzed to obtain a first analysis result, the second model parameter of the first anomaly detection model is determined according to the first analysis result and the first model parameter, and the anomaly detection result is determined according to the second model parameter, the first anomaly detection model and the first sensor data. The anomaly detection result includes multiple anomaly types, and each anomaly type corresponds to a probability value. An early warning operation is performed according to the anomaly detection result. On the one hand, since the first sensor data is data for a period of time, based on the continuity of the first sensor data, it can be preliminarily detected whether there is an anomaly in the intelligent switchgear. On the other hand, since the first analysis result has preliminarily captured the anomaly to a certain extent, the model parameters are dynamically optimized based on this anomaly, so that the optimized model parameters are more in line with the actual environmental changes, which helps to reduce the model misidentification probability and also helps to improve the anomaly detection accuracy. Thus, accurate anomaly detection of the intelligent switchgear can be achieved. Furthermore, it helps to achieve an accurate early warning operation and ensure the safety of the intelligent switchgear.

[0106] It can be understood that the functions of the various program modules of the switchgear status monitoring device in this embodiment can be specifically implemented according to the methods in the above method embodiments. The specific implementation process can refer to the relevant descriptions in the above method embodiments and will not be elaborated here.

[0107] The embodiments of the present application further provide a computer storage medium. The computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute some or all of the steps of any of the methods described in the above method embodiments. The above computer includes an edge device.

[0108] The embodiments of the present application further provide a computer program product. The above computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the above computer program is operable to enable a computer to execute some or all of the steps of any of the methods described in the above method embodiments. The computer program product can be a software installation package, and the above computer includes an edge device.

[0109] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0110] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0111] In the several embodiments provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the above module division is only a logical function division. In actual implementation, there can be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or modules can be in electrical or other forms.

[0112] The modules described as separate components above may or may not be physically separated. The components displayed as modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] In addition, in each embodiment of this application, the various functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0114] If the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the above methods in various embodiments of this application. The aforementioned memory includes: various media such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), external hard drives, magnetic disks, or optical discs that can store program codes.

[0115] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable memory. The memory can include: flash drives, read-only memories (English: Read-Only Memory, abbreviated: ROM), random access memories (English: Random Access Memory, abbreviated: RAM), magnetic disks, or optical discs, etc.

[0116] The above has introduced the embodiments of this application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A switchgear state monitoring method, characterized in that, An edge device applied to a switch cabinet status monitoring system, the switch cabinet status monitoring system further comprising: an intelligent switch cabinet and at least one sensor module, each sensor module corresponding to a detection position of the intelligent switch cabinet; the method comprising: Obtaining first sensor data for a preset time period through a first sensor module, the first sensor module being any one of the at least one sensor module; Determining first model parameters of a first anomaly detection model corresponding to a first detection position of the first sensor module; Analyzing the first sensor data to obtain a first analysis result; Determining second model parameters of the first anomaly detection model according to the first analysis result and the first model parameters; Determining an anomaly detection result according to the second model parameters, the first anomaly detection model and the first sensor data, the anomaly detection result including multiple anomaly types, each anomaly type corresponding to a probability value; Performing a warning operation according to the anomaly detection result.

2. The method according to claim 1, wherein The determining first model parameters of a first anomaly detection model corresponding to a first detection position of the first sensor module includes: Determining a first circuit position of a first circuit structure diagram corresponding to the first detection position; Obtaining a historical anomaly detection record set of the first circuit position, the historical anomaly detection record set including multiple anomaly detection records, each anomaly detection record corresponding to a set of working parameters and a set of model parameters; Obtaining a first set of working parameters of the first circuit position at the current moment; Comparing the first set of working parameters with the sets of working parameters corresponding to each anomaly detection record in the multiple anomaly detection records to obtain multiple comparison values; Selecting the maximum value among the multiple comparison values, obtaining the model parameters corresponding to the maximum value, and obtaining the first model parameters.

3. The method according to claim 2, wherein The analyzing the first sensor data to obtain a first analysis result includes: Determining a first fitting straight line and a first fitting curve segment corresponding to the preset time period according to the first sensor data; Determining a first absolute value of the slope of the first fitting straight line; Determining a first standard deviation of the first fitting curve segment; Determining the first analysis result according to the first absolute value and the first standard deviation.

4. The method according to claim 3, characterized in that, The determining second model parameters of the first anomaly detection model according to the first analysis result and the first model parameters includes: According to a first adjustment parameter corresponding to the first absolute value; Determining a first fine-tuning parameter corresponding to the first standard deviation; Determining the second model parameters according to the first adjustment parameter, the first fine-tuning parameter and the first model parameters.

5. The method according to claim 3 or 4, characterized in that, The method further comprises: Detecting whether the first absolute value is greater than a first preset absolute value; When the first absolute value is greater than the first preset absolute value, performing the step of determining second model parameters of the first anomaly detection model according to the first analysis result and the first model parameters; Or, Detecting whether the first standard deviation is greater than a first preset standard deviation; When the first standard deviation is greater than the first preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; Or, Detect whether the first absolute value is greater than a second preset absolute value and whether the first standard deviation is greater than a second preset standard deviation; When the first absolute value is greater than the second preset absolute value and the first absolute value is greater than the second preset standard deviation, perform the step of determining the second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter.

6. The method according to any one of claims 1-4, characterized in that The performing a warning operation according to the anomaly detection result includes: Determine probability thresholds for the multiple anomaly types according to the first absolute value and the first standard deviation, obtaining multiple probability thresholds; Determine multiple probability differences between the multiple probability values corresponding to the multiple anomaly types and the multiple probability thresholds; Select the probability differences greater than 0 from the multiple probability differences, obtaining at least one probability difference; Obtain the anomaly types corresponding to the at least one probability difference, obtaining at least one anomaly type; Perform a warning operation according to the at least one probability difference and the at least one anomaly type.

7. The method according to claim 6, wherein The determining probability thresholds for the multiple anomaly types according to the first absolute value and the first standard deviation, obtaining multiple probability thresholds, includes: Determine a first evaluation value corresponding to the first absolute value; Determine a second evaluation value corresponding to the first standard deviation; Determine a target evaluation value according to the first evaluation value and the second evaluation value; Determine the multiple probability thresholds corresponding to the target evaluation value.

8. A switchgear status monitoring device, characterized in that, Applied to an edge device in a switchgear status monitoring system, the switchgear status monitoring system further includes: an intelligent switchgear and at least one sensor module, each sensor module corresponding to a detection position of the intelligent switchgear; the device includes: an acquisition unit, a first determination unit, an analysis unit, a second determination unit, and a warning unit, wherein, The acquisition unit is configured to acquire first sensor data for a preset time period through a first sensor module, the first sensor module being any one of the at least one sensor module; The first determination unit is configured to determine a first model parameter of a first anomaly detection model corresponding to a first detection position of the first sensor module; The analysis unit is configured to analyze the first sensor data to obtain a first analysis result; The second determination unit is configured to determine a second model parameter of the first anomaly detection model according to the first analysis result and the first model parameter; determine an anomaly detection result according to the second model parameter, the first anomaly detection model, and the first sensor data, the anomaly detection result including multiple anomaly types, each anomaly type corresponding to a probability value; The warning unit is configured to perform a warning operation according to the anomaly detection result.

9. An edge device, characterized in that, The edge device includes a processor and a memory. The memory is used to store one or more programs and is configured to be executed by the processor. The programs include instructions for performing the steps in the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, A computer program for electronic data interchange is stored, wherein the computer program causes a computer to execute the method according to any one of claims 1-7.

Images