CN115730201A

CN115730201A - Fault arc monitoring method and device, computer equipment and storage medium

Info

Publication number: CN115730201A
Application number: CN202211431561.8A
Authority: CN
Inventors: 王桂光; 韦秋花
Original assignee: Kangtijia Intelligent Technology Shenzhen Co ltd
Current assignee: Kangtijia Intelligent Technology Shenzhen Co ltd
Priority date: 2022-11-15
Filing date: 2022-11-15
Publication date: 2023-03-03

Abstract

The application relates to a method, a system, a computer device and a storage medium for monitoring a fault arc. The method comprises the following steps: carrying out data processing according to the detected electric signals to acquire harmonic characteristic information corresponding to the electric signals; inputting the harmonic characteristic information into a first recognition model which is trained in advance, so that the first recognition model outputs a first recognition result which is used for indicating whether the harmonic characteristic information is recognized as that a fault arc exists or not; if the first recognition result indicates that the fault arc exists, inputting harmonic characteristic information corresponding to the fault arc into a second recognition model which is trained in advance, so that the second recognition model outputs a second recognition result for indicating whether misjudgment exists; and executing a corresponding coping control strategy according to the second identification result. By adopting the method, the accuracy of fault arc identification can be improved, and the reliability and the safety of the power utilization manager are further improved.

Description

Fault arc monitoring method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of fault arc identification technologies, and in particular, to a fault arc monitoring method and apparatus, a computer device, and a storage medium.

Background

With the continuous development of electricity utilization technology, the electricity utilization manager is widely applied to various electricity utilization occasions. The power utilization manager is used for controlling the power on or off of the power utilization circuit. During the electricity utilization process, if a fault arc exists in a power utilization circuit, the circuit fault (such as short circuit, ignition phenomenon, fire and the like) can be caused. However, in the current technical solution, the power manager monitors the fault arc, the fault arc cannot be accurately identified, and misjudgment is easy to occur, which results in the misoperation of the power manager. Therefore, how to improve the accuracy of identifying the fault arc and further improve the reliability and the safety of the power utilization manager becomes a technical problem to be solved urgently.

Disclosure of Invention

In view of the above, it is necessary to provide a method, an apparatus, a computer device, and a storage medium for monitoring a fault arc, which can improve the accuracy of identifying the fault arc and further improve the reliability and safety of the power manager.

A method of monitoring a fault arc, the method comprising:

carrying out data processing according to the detected electric signals to acquire harmonic characteristic information corresponding to the electric signals;

inputting the harmonic characteristic information into a first recognition model which is trained in advance, so that the first recognition model outputs a first recognition result for indicating whether the harmonic characteristic information is recognized as that a fault arc exists;

if the first recognition result indicates that the fault arc exists, inputting harmonic characteristic information corresponding to the fault arc into a second recognition model which is trained in advance, so that the second recognition model outputs a second recognition result for indicating whether misjudgment exists;

and executing a corresponding coping control strategy according to the second identification result.

In one embodiment, inputting the harmonic feature information into a first recognition model trained in advance so that the first recognition model outputs a first recognition result of whether a fault arc exists or not includes: inputting the harmonic characteristic information into a first identification model which is trained in advance, so that the first identification model compares the harmonic characteristic information with current fingerprint information of known fault arcs in a fault arc database which is stored in advance; if an information segment which is matched with the current fingerprint information of the known fault arc exists in the harmonic characteristic information, the first identification model outputs a first identification result which is used for indicating that the harmonic characteristic information is identified to exist the fault arc, and the first identification result comprises an information segment which corresponds to the fault arc.

In one embodiment, after inputting the harmonic feature information into a first identification model which is trained in advance, so that the first identification model compares the harmonic feature information with current fingerprint information of known fault arcs in a fault arc database which is stored in advance, the method further comprises the following steps: if the harmonic characteristic information does not have an information segment matched with the current fingerprint information of the known fault arc, judging whether the harmonic characteristic information has the fault arc or not based on a preset fault arc judgment rule; if the harmonic characteristic information is judged to have the fault arc based on the fault arc judgment rule, extracting an information segment corresponding to the fault arc in the harmonic characteristic information as current fingerprint information of the fault arc to be learned; and learning and training the current fingerprint information of the fault arc to be learned by using the first recognition model based on a pre-trained deep neural network or convolutional neural network, and storing the learned current fingerprint information in a fault arc database.

In one embodiment, outputting a corresponding coping control strategy according to the second recognition result includes: and if the second identification result indicates that no misjudgment exists, outputting corresponding warning information and/or tripping instructions.

In one embodiment, before the harmonic feature information is input into a first recognition model which is trained in advance so that the first recognition model outputs a first recognition result indicating whether the harmonic feature information is recognized as the existence of the fault arc, the method further includes: determining whether a first identification model corresponding to the current environment exists or not according to the parameter information of the current environment; and if the current fingerprint information exists, the corresponding first identification model is obtained, and if the current fingerprint information does not exist, the current fingerprint information corresponding to the fault arc of the current environment is trained to construct the first identification model corresponding to the parameter information of the current environment.

In one embodiment, the data processing according to the detected electrical signal to obtain harmonic characteristic information corresponding to the electrical signal includes: processing the detected electric signal to obtain input parameters corresponding to the electric signal, wherein the input parameters comprise a high-frequency signal or a low-frequency signal; and carrying out harmonic analysis processing on the input parameters based on a Fourier transform algorithm to obtain corresponding harmonic characteristic information.

In one embodiment, before inputting harmonic feature information corresponding to a fault arc into a second recognition model trained in advance if the fault arc exists, so that the second recognition model outputs a second recognition result indicating whether a misjudgment is made, the method further includes: and acquiring the higher harmonic characteristic information generated by the current circuit, and training the second recognition model.

A device for monitoring a fault arc, the device comprising:

the acquisition module is used for carrying out data processing according to the detected electric signals and acquiring harmonic characteristic information corresponding to the electric signals;

the first identification module is used for inputting the harmonic characteristic information into a first identification model which is trained in advance so as to enable the first identification model to output a first identification result which is used for indicating whether the harmonic characteristic information is identified to be that a fault arc exists or not;

the second identification module is used for inputting the harmonic characteristic information corresponding to the fault arc into a second identification model which is trained in advance if the fault arc exists, so that the second identification model outputs a second identification result for indicating whether misjudgment is performed or not;

and the processing module is used for outputting a corresponding coping control strategy according to the second identification result.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

inputting the harmonic characteristic information into a first recognition model which is trained in advance, so that the first recognition model outputs a first recognition result which is used for indicating whether the harmonic characteristic information is recognized as that a fault arc exists or not;

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

One of the above technical solutions has the following advantages and beneficial effects:

according to the method, the device, the computer equipment and the storage medium for monitoring the fault arc, data processing is carried out according to the detected electric signals, harmonic characteristic information corresponding to the electric signals is obtained, the harmonic characteristic information is input into a first recognition model which is trained in advance, so that the first recognition model outputs a first recognition result for indicating whether the harmonic characteristic information is recognized as the fault arc, if the first recognition result indicates that the fault arc exists, the harmonic characteristic information corresponding to the fault arc is input into a second recognition model which is trained in advance, so that the second recognition model outputs a second recognition result for indicating whether misjudgment exists, and a corresponding coping control strategy is executed according to the second recognition result. Therefore, the first recognition model is matched with the second recognition model to recognize whether the fault arc exists or not, and the second recognition model verifies the output result of the first recognition model, so that the accuracy of the recognition result of the fault arc is ensured, and the reliability and the safety of the power utilization manager are improved.

Drawings

Fig. 1 is a schematic flow chart of a fault arc monitoring method in an embodiment of the present application.

Fig. 2 is a schematic flowchart of step S120 of the method for monitoring a fault arc in fig. 1 in the embodiment of the present application.

Fig. 3 is a schematic flowchart of a process of acquiring the first identification model and/or the second identification model, which is further included in the monitoring method for a fault arc in the embodiment of the present application.

Fig. 4 is a block diagram of a device for monitoring a fault arc according to an embodiment of the present invention.

Fig. 5 is an internal structural diagram of a computer device in the embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

Fig. 1 shows a schematic flow diagram of a method of monitoring a fault arc according to an embodiment of the present application. Referring to fig. 1, the method at least includes steps S110 to S140, and the following steps are described in detail:

in step S110, data processing is performed based on the detected electrical signal, and harmonic characteristic information corresponding to the electrical signal is acquired.

In this embodiment, the power manager may monitor the power circuit in real time to obtain the electrical signal generated by the power circuit. According to the acquired electric signal, the electricity manager can perform data processing (such as digital-to-analog conversion) on the electric signal, so as to acquire harmonic characteristic information corresponding to the electric signal for subsequent identification.

In another embodiment, the electricity manager may also obtain the electrical signal generated by the electricity circuit at predetermined intervals, for example, every 1S or 3S. It should be noted that, those skilled in the art may balance the electrical safety and the calculation performance or resource loss according to the actual implementation needs, and set the time length of the corresponding predetermined interval, which is not particularly limited.

In step S120, the harmonic feature information is input into a first recognition model trained in advance, so that the first recognition model outputs a first recognition result indicating whether the harmonic feature information is recognized as the existence of the fault arc.

In this embodiment, the first identification model may be a model for identifying whether there is harmonic characteristic information corresponding to a fault arc in the input harmonic characteristic information. In one example, one skilled in the art can build the first recognition model based on a convolutional neural network or a deep neural network, and train the built first recognition model with known harmonic feature information of the fault arc, so that the first recognition model can accurately output the recognition result.

In use, after the acquired harmonic characteristic information is input to the first recognition model, the first recognition model can output a first recognition result indicating whether the harmonic characteristic information is recognized as the existence of the fault arc. In an example, if the input harmonic characteristic information has harmonic characteristic information corresponding to a fault arc, the output first identification result may indicate that the fault arc exists, and output the harmonic characteristic information of the identified fault arc, and if the fault arc does not exist, the first identification result indicates that the fault arc does not exist.

In step S130, if the first recognition result indicates that a fault arc exists, the harmonic feature information corresponding to the fault arc is input to a second recognition model that is trained in advance, so that the second recognition model outputs a second recognition result indicating whether the fault arc is erroneously determined.

In this embodiment, it should be understood that in a complex circuit application environment, there are some factors that may cause a false alarm behavior of a fault arc to occur in the power utilization manager, for example, when a circuit generates a higher harmonic feature, it is easy to cause the first recognition model to judge the higher harmonic feature as a fault arc, thereby causing the false alarm behavior to occur. Thus, the second identification model can be used for identification according to the harmonic characteristic information of the fault arc in the first identification result so as to determine whether misjudgment occurs or not.

In an example, a person skilled in the art may build the second recognition model based on a convolutional neural network or a deep neural network, and train the built second recognition model with known harmonic feature information where misjudgment occurs, so that the second recognition model can accurately output a recognition result.

By calling the second recognition model, the second recognition model can output a second recognition result for indicating whether the first recognition result is misjudged or not, namely, the first recognition result is verified, so that the situation that the power utilization manager misjudges is avoided.

In step S140, a corresponding countermeasure control strategy is executed according to the second recognition result.

In this embodiment, the coping control strategy may be a protection strategy for the power consuming circuit preset by a person skilled in the art. For example, if the second identification result indicates that there is a false determination, the processing may not be performed, and if the second identification result indicates that there is no false determination, that is, it indicates that there is a fault arc, a protection strategy of the power utilization circuit may be correspondingly adopted, for example, a trip instruction is output to disconnect the circuit, or early warning information is generated.

Thus, in the embodiment shown in fig. 1, data processing is performed based on the detected electric signal to acquire harmonic feature information corresponding to the electric signal, the harmonic feature information is input to the first recognition model trained in advance, so that the first recognition model outputs a first recognition result indicating whether the harmonic feature information is recognized as the presence of a fault arc, if the first recognition result indicates the presence of a fault arc, the harmonic feature information corresponding to the fault arc is input to the second recognition model trained in advance, so that the second recognition model outputs a second recognition result indicating whether misjudgment is performed, and a corresponding countermeasure control strategy is executed based on the second recognition result. Therefore, the first recognition model is matched with the second recognition model to recognize whether the fault arc exists or not, and the second recognition model verifies the output result of the first recognition model, so that the accuracy of the recognition result of the fault arc is ensured, and the reliability and the safety of the power utilization manager are improved.

In one embodiment, if the second recognition model outputs a second recognition result indicating that there is a false determination, a new round of fault arc monitoring may be started.

Based on the embodiment shown in fig. 1, fig. 2 shows a schematic flow diagram of step S120 in the monitoring method of the fault arc of fig. 1 according to an embodiment of the present application. Referring to fig. 2, step S120 at least includes steps S210 to S220, which are described in detail as follows:

in step S210, the harmonic feature information is input into a first identification model trained in advance, so that the first identification model compares the harmonic feature information with current fingerprint information of known fault arcs in a fault arc database stored in advance.

In this embodiment, a skilled person may pre-build a fault arc database, which may include pre-collected current fingerprint information for known fault arcs. When the first identification model identifies, the harmonic characteristic information of the input electric signal can be compared with the current fingerprint information of the known fault arc in the fault arc database, so as to determine whether the two match.

In step S220, if there is an information segment matching the current fingerprint information of the known fault arc in the harmonic feature information, the first identification model outputs a first identification result indicating that the harmonic feature information is identified as that there is a fault arc, and the first identification result includes an information segment corresponding to the fault arc.

In this embodiment, when there is an information segment in the input harmonic feature information that matches the current fingerprint information of a known fault arc, then the first recognition model may output a first recognition result indicating that the harmonic feature information is recognized as the presence of a fault arc, and accordingly, the first recognition result may include an information segment corresponding to the recognized fault arc.

If the input harmonic characteristic information does not have the information segment matched with the current fingerprint information of the known fault arc, the fault arc is not present, and therefore the first identification model can output a first identification result for indicating that the fault arc is not present.

Therefore, in the embodiment shown in fig. 2, the first identification model matches the harmonic characteristic information of the input electrical signal with the current fingerprint information of known fault arcs in the fault arc database established in advance, so as to determine whether fault arcs exist, and the accuracy of fault arc identification can be ensured.

It should be noted that, in order to ensure the richness of the data in the fault arc database, when the current fingerprint information of a new fault arc is identified, the current fingerprint information may be added to the fault arc database, so as to enrich the data in the fault arc database.

Based on the embodiments shown in fig. 1 and fig. 2, in an embodiment of the present application, after the harmonic feature information is input into a first recognition model that is trained in advance, so that the first recognition model compares the harmonic feature information with current fingerprint information of known fault arcs in a fault arc database stored in advance, the method further includes:

if the harmonic characteristic information does not have an information segment matched with the current fingerprint information of the known fault arc, judging whether the harmonic characteristic information has the fault arc or not based on a preset fault arc judgment rule;

if the harmonic characteristic information is judged to have the fault arc based on the fault arc judgment rule, extracting an information segment corresponding to the fault arc in the harmonic characteristic information as current fingerprint information of the fault arc to be learned;

and learning and training the current fingerprint information of the fault arc to be learned by using the first recognition model based on a pre-trained deep neural network or convolutional neural network, and storing the learned current fingerprint information in a fault arc database.

In this embodiment, the first identification model may compare the harmonic characteristic information of the acquired electrical signal with the current fingerprint information in the fault arc database, and when there is no matched information segment, the first identification model may determine whether there is a fault arc in the harmonic characteristic information based on a preset fault arc determination rule, and if it is determined that there is a fault arc in the harmonic characteristic information according to the fault arc determination rule, extract an information segment corresponding to the fault arc in the harmonic characteristic information as the current fingerprint information of the fault arc to be learned. And the current fingerprint information of the fault arc to be learned is learned and trained by utilizing the first recognition model based on the pre-trained deep neural network or convolutional neural network, so that the richness of the training data of the first recognition model is improved, and the accuracy of recognition is ensured. And adding the current fingerprint information of the fault arc to be learned into a fault arc database for storage so as to be inquired next time.

Therefore, the two-layer identification mode of the fault arc database and the fault arc judgment rule is combined, the harmonic characteristic information is compared with the information in the fault arc database, the identification accuracy can be improved, the fault arc judgment rule is adopted to identify the harmonic characteristic information under the condition of no matching result, the accuracy of the identification result can be further ensured, and missing detection is prevented.

In an embodiment, before the harmonic feature information corresponding to the fault arc is input to a second recognition model that is trained in advance if the first recognition result indicates that the fault arc exists, so that the second recognition model outputs a second recognition result indicating whether the fault arc is determined incorrectly, the method further includes:

and acquiring the higher harmonic characteristic information generated by the current circuit, and training the second recognition model.

In the embodiment, the second recognition model can be learned and trained according to the high-order harmonic characteristic information generated by the current circuit and acquired in the using process of the power utilization manager, the data volume of the second recognition model is enriched, and the judgment capability of misjudgment characteristics is improved. Meanwhile, the adaptability of the second recognition model and the current power utilization circuit is ensured, and the accuracy of the recognition result of the second recognition model is ensured.

Based on the embodiment shown in fig. 1, in an embodiment of the present application, executing a corresponding coping control strategy according to the second recognition result includes:

and if the second identification result indicates that no misjudgment exists, outputting corresponding warning information and/or tripping instructions.

In this embodiment, if the second identification result indicates that no misjudgment exists, it indicates that a fault arc exists, and therefore, the power utilization manager may generate and output corresponding warning information and/or a tripping instruction, so as to remind a manager to perform corresponding protection in time. And the output tripping instruction can also control the power utilization circuit to be disconnected, thereby ensuring the safety of the power utilization circuit, protecting the power utilization circuit before the circuit fault occurs and suppressing the circuit fault.

Based on the foregoing embodiments, fig. 3 shows a schematic flowchart of obtaining the first identification model and/or the second identification model, which is further included in the monitoring method for a fault arc according to an embodiment of the present application. Referring to fig. 3, the obtaining of the first recognition model at least includes steps S310 to S320, which are described in detail as follows:

in step S310, it is determined whether there is a first recognition model and/or a second recognition model corresponding thereto according to parameter information of the current environment.

In this embodiment, the parameter information of the current environment may be regional information (e.g., longitude and latitude, altitude, etc.), electricity utilization place information (e.g., power distribution network, industrial electricity, household electricity, etc.). It should be understood that due to different environments, the current fingerprint information corresponding to the fault arc of each device may have a certain difference. Thus. The skilled person can establish and train the corresponding first recognition model and/or the second recognition model in advance according to the parameter information of different environments, and store the corresponding relationship between the parameter information of the environments and the first recognition model and/or the second recognition model.

In an actual use process, before the first recognition model and/or the second recognition model is called, parameter information of a current environment may be acquired, and whether the first recognition model and/or the second recognition model corresponding to the parameter information exists or not may be determined.

In step S320, if the current fingerprint information exists, the corresponding first recognition model and/or second recognition model is obtained, and if the current fingerprint information does not exist, the current fingerprint information of the current environment is trained to construct the first recognition model and/or second recognition model corresponding to the parameter information of the current environment.

In this embodiment, if there is a first recognition model corresponding to the parameter information of the current environment, the first recognition model may be directly invoked. If the current fingerprint information does not exist, the first identification model and/or the second identification model can be trained according to the current fingerprint information identified by the current power utilization circuit, the current fingerprint information can be the current fingerprint information of the fault arc, the higher harmonic characteristic information and the like, so that the first identification model and/or the second identification model used subsequently can be matched with the current environment, and the accuracy of the identification result is ensured.

After the training of the first recognition model and/or the second recognition model is completed, the first recognition model and/or the second recognition model can be stored in association with the parameter information of the current environment so as to be called when the same use environment is encountered subsequently, and the recognition efficiency of the fault arc is improved.

Based on the embodiment shown in fig. 1, in an embodiment of the present application, performing data processing according to a detected electrical signal to acquire harmonic feature information corresponding to the electrical signal includes:

processing the detected electric signal to obtain input parameters corresponding to the electric signal, wherein the input parameters comprise high-frequency signals and/or low-frequency signals;

and carrying out harmonic analysis processing on the input parameters based on a Fourier transform algorithm to obtain corresponding harmonic characteristic information.

In this embodiment, the power utilization manager may monitor the power utilization circuit in real time to obtain the electrical signal, and perform signal processing on the electrical signal, and specifically, the power utilization manager may perform digital-to-analog conversion, amplification, filtering, and other processing on the electrical signal in sequence, so as to obtain the input parameter corresponding to the electrical signal, where the input parameter includes a high-frequency signal and/or a low-frequency signal. And then, carrying out harmonic analysis processing on the acquired input parameters based on a Fourier transform algorithm to obtain corresponding harmonic characteristic information, thereby ensuring the accuracy of acquiring the harmonic characteristic information and further ensuring the accuracy of a subsequent fault arc identification result.

In an embodiment, if the first identification model does not match the corresponding information segment in the arc fault database, but identifies the arc fault according to the arc fault determination rule, and the second identification model determines that no misjudgment exists, the current fingerprint information corresponding to the newly identified arc fault can be stored in the arc fault database, so as to enrich the data in the arc fault database for the next identification.

It should be understood that although the various steps in the flow charts of fig. 1-3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-3 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least some of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 4, there is provided a fault arc monitoring apparatus comprising:

an obtaining module 410, configured to perform data processing according to the detected electrical signal, and obtain harmonic characteristic information corresponding to the electrical signal;

a first identification module 420, configured to input the harmonic feature information into a first identification model that is trained in advance, so that the first identification model outputs a first identification result indicating whether the harmonic feature information is identified as having a fault arc;

a second identification module 430, configured to, if the first identification result indicates that a fault arc exists, input harmonic feature information corresponding to the fault arc into a second identification model that is trained in advance, so that the second identification model outputs a second identification result used for indicating whether to perform a misjudgment;

and the processing module 440 is configured to execute a corresponding coping control strategy according to the second identification result.

In one embodiment, the first identification module 420 is configured to: inputting the harmonic characteristic information into a first identification model which is trained in advance, so that the first identification model compares the harmonic characteristic information with current fingerprint information of known fault arcs in a fault arc database which is stored in advance; if an information segment which is matched with the current fingerprint information of the known fault arc exists in the harmonic characteristic information, the first identification model outputs a first identification result which is used for indicating that the harmonic characteristic information is identified to exist the fault arc, and the first identification result comprises an information segment which corresponds to the fault arc.

In one embodiment, the first identification module 420 is further configured to: if the harmonic characteristic information does not have an information segment matched with the current fingerprint information of the known fault arc, judging whether the harmonic characteristic information has the fault arc or not based on a preset fault arc judgment rule; if the harmonic characteristic information is judged to have the fault arc based on the fault arc judgment rule, extracting an information segment corresponding to the fault arc in the harmonic characteristic information as current fingerprint information of the fault arc to be learned; and learning and training the current fingerprint information of the fault arc to be learned by using the first identification model based on a pre-trained deep neural network or convolutional neural network, and storing the learned current fingerprint information in a fault arc database.

In one embodiment, the processing module 440 is configured to: and if the second identification result indicates that no misjudgment exists, outputting corresponding warning information and/or tripping instructions.

In an embodiment, the processing module 440 is further configured to: determining whether a first recognition model and/or a second recognition model corresponding to the current environment exist or not according to the parameter information of the current environment; if the current fingerprint information exists, the corresponding first recognition model and/or the second recognition model are/is obtained, and if the current fingerprint information does not exist, training is carried out based on the current fingerprint information of the current environment so as to construct the first recognition model and/or the second recognition model corresponding to the parameter information of the current environment.

In an embodiment, the obtaining module 410 is configured to: processing the detected electric signal to obtain input parameters corresponding to the electric signal, wherein the input parameters comprise high-frequency signals and/or low-frequency signals; and carrying out harmonic analysis processing on the input parameters based on a Fourier transform algorithm to obtain corresponding harmonic characteristic information.

In one embodiment, the second identification module 430 is further configured to: and acquiring the high-order harmonic characteristic information generated by the current circuit so as to train the second recognition model.

The specific definition of the fault arc monitoring device can be referred to the definition of the fault arc monitoring method in the foregoing, and is not described herein again. The modules in the device for monitoring the fault arc can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal (e.g., a power manager, etc.), and its internal structure diagram may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of fault arc monitoring. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the configuration shown in fig. 5 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

if the first recognition result indicates that a fault arc exists, inputting harmonic characteristic information corresponding to the fault arc into a second recognition model trained in advance, so that the second recognition model outputs a second recognition result used for indicating whether misjudgment exists or not;

In one embodiment, the processor, when executing the computer program, further performs the steps of:

inputting the harmonic characteristic information into a first recognition model which is trained in advance, so that the first recognition model compares the harmonic characteristic information with the current fingerprint information of the known fault arc in a fault arc database which is stored in advance; if an information segment which is matched with the current fingerprint information of the known fault arc exists in the harmonic characteristic information, the first identification model outputs a first identification result which is used for indicating that the harmonic characteristic information is identified to exist the fault arc, and the first identification result comprises an information segment which corresponds to the fault arc.

if the harmonic characteristic information does not have an information segment matched with the current fingerprint information of the known fault arc, judging whether the harmonic characteristic information has the fault arc or not based on a preset fault arc judgment rule; if the harmonic characteristic information is judged to have the fault arc based on the fault arc judgment rule, extracting an information segment corresponding to the fault arc in the harmonic characteristic information as current fingerprint information of the fault arc to be learned; and learning and training the current fingerprint information of the fault arc to be learned by using the first identification model based on a pre-trained deep neural network or convolutional neural network, and storing the learned current fingerprint information in a fault arc database.

determining whether a first recognition model and/or a second recognition model corresponding to the current environment exist or not according to the parameter information of the current environment; if the current fingerprint information exists, the corresponding first recognition model and/or the second recognition model are/is obtained, and if the current fingerprint information does not exist, training is carried out based on the current fingerprint information of the current environment so as to construct the first recognition model and/or the second recognition model corresponding to the parameter information of the current environment.

In one embodiment, the processor when executing the computer program further performs the steps of:

and acquiring the high-order harmonic characteristic information generated by the current circuit so as to train the second recognition model.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

processing the detected electric signal to obtain input parameters corresponding to the electric signal, wherein the input parameters comprise a high-frequency signal and/or a low-frequency signal; and carrying out harmonic analysis processing on the input parameters based on a Fourier transform algorithm to obtain corresponding harmonic characteristic information.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the claims. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of monitoring a fault arc, comprising:

2. The method for monitoring a fault arc according to claim 1, wherein inputting the harmonic feature information into a first recognition model trained in advance so that the first recognition model outputs a first recognition result indicating whether the harmonic feature information is recognized as the presence of the fault arc comprises:

inputting the harmonic characteristic information into a first identification model which is trained in advance, so that the first identification model compares the harmonic characteristic information with current fingerprint information of known fault arcs in a fault arc database which is stored in advance;

if an information segment which is matched with the current fingerprint information of the known fault arc exists in the harmonic characteristic information, the first identification model outputs a first identification result which is used for indicating that the harmonic characteristic information is identified to exist the fault arc, and the first identification result comprises an information segment which corresponds to the fault arc.

3. The method for monitoring a fault arc according to claim 2, wherein the harmonic characteristic information is inputted into a first recognition model trained in advance, so that after the first recognition model compares the harmonic characteristic information with current fingerprint information of known fault arcs in a fault arc database stored in advance, the method further comprises:

4. The method according to claim 1, wherein the step of executing a corresponding coping control strategy according to the second identification result comprises:

5. The method according to any one of claims 1 to 4, wherein before inputting the harmonic feature information into a first recognition model trained in advance so that the first recognition model outputs a first recognition result indicating whether the harmonic feature information is recognized as the presence of a fault arc, the method further comprises:

determining whether a first recognition model and/or a second recognition model corresponding to the current environment exist or not according to the parameter information of the current environment;

if the current fingerprint information exists, the corresponding first recognition model and/or the second recognition model are/is obtained, and if the current fingerprint information does not exist, training is carried out based on the current fingerprint information of the current environment so as to construct the first recognition model and/or the second recognition model corresponding to the parameter information of the current environment.

6. The method for monitoring a fault arc according to claim 1, wherein the step of performing data processing according to the detected electrical signal to obtain harmonic characteristic information corresponding to the electrical signal comprises:

processing the detected electric signal to obtain input parameters corresponding to the electric signal, wherein the input parameters comprise a high-frequency signal and/or a low-frequency signal;

7. The method according to claim 1, wherein before the harmonic feature information corresponding to the fault arc is input to a second recognition model trained in advance if the first recognition result indicates that the fault arc exists, so that the second recognition model outputs a second recognition result indicating whether the fault arc is judged incorrectly, the method further comprises:

8. A device for monitoring a fault arc, the device comprising:

the second identification module is used for inputting the harmonic characteristic information corresponding to the fault arc into a second identification model which is trained in advance if the first identification result indicates that the fault arc exists, so that the second identification model outputs a second identification result which is used for indicating whether misjudgment exists or not;

and the processing module is used for executing a corresponding coping control strategy according to the second identification result.

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program performs the steps of the method according to any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.