Fault detection method and system for relay protection loop

Abstract

The invention discloses a fault detection method and system for a relay protection loop, and the method comprises the steps: monitoring and obtaining the state information of a power system, and enabling the state information of the power system to comprise an electrical parameter, an equipment state parameter, and a protection motion parameter; the state information of the power system is input into the trained first fault detection model for identification, a first state of the power system is obtained, and the first state comprises a normal state or a fault state; if the first state is a fault state, inputting the state information of the power system into a trained second fault detection model for identification to obtain a second state of the power system; wherein the second state comprises fault time, a fault position and a fault type; and controlling a relay protection loop to carry out fault maintenance on the power system through the fault time, the fault position and the fault type so as to realize fault removal of the power system. The state information of the power system is identified twice, and the fault detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of relay protection circuit detection, and particularly to a fault detection method and system for a relay protection circuit. Background Art

[0002] A relay protection circuit is an automatic measure that can reflect the occurrence of faults or abnormal operating states of electrical components in a power system and act on circuit breakers or send signals. It is a kind of anti-accident automation measure in the power system, mainly used to protect the power system and its components (such as generators, transformers, transmission lines, busbars, etc.) from damage. Once a fault is detected, the relay protection circuit will quickly start, and according to the pre-set logical relationships and action criteria, send a warning signal to the on-duty operator or directly control the circuit breaker to trip to terminate the development of the fault event.

[0003] However, information such as fault types and fault locations in the power system is usually judged and analyzed based on human experience, delaying the rapid response of the relay protection circuit, which undoubtedly reduces the efficiency of fault handling.

[0004] Aiming at the technical problem of low efficiency caused by the artificial analysis of fault information in the relay protection circuit in the prior art, there is currently no effective solution. Summary of the Invention

[0005] The purpose of the present invention is to provide a fault detection method, system, device and readable storage medium for a relay protection circuit, which can solve the technical problem of low efficiency caused by the artificial analysis of fault information in the relay protection circuit in the prior art.

[0006] One aspect of the present invention provides a fault detection method for a relay protection circuit, the method comprising: monitoring and obtaining the state information of the power system, wherein the state information of the power system includes electrical parameters, equipment state parameters and protection action parameters; inputting the state information of the power system into a trained first fault detection model for identification to obtain the first state of the power system, wherein the first state includes a normal state or a fault state; if the first state is a fault state, inputting the state information of the power system into a trained second fault detection model for identification to obtain the second state of the power system; wherein the second state includes the fault time, fault location and fault type; controlling the relay protection circuit to perform fault repair on the power system through the fault time, fault location and fault type to achieve the elimination of the fault of the power system.

[0007] Optionally, the relay protection circuit is controlled by the fault time, fault location, and fault type to perform fault repair on the power system, including: dividing the second states with the same fault location into a set; sequentially extracting the fault removal strategies corresponding to the fault types in the set according to the occurrence order of the fault time; and controlling the relay protection circuit to perform fault repair on the power system through the fault removal strategies.

[0008] Optionally, the relay protection circuit is controlled by the fault time, fault location, and fault type to perform fault repair on the power system, including: dividing the second states with the same fault type into a set; sorting the second states in the same set according to the occurrence order of the fault time, and calculating the first time required to solve the faults in this order; determining the shortest path passing through all the fault locations in the same set, and calculating the second time required to solve all the faults in this set according to the shortest path; comparing the magnitudes of the first time and the second time, and extracting the fault removal strategies of the fault types in the same set according to the order corresponding to the minimum time, and controlling the relay protection circuit to perform fault repair on the power system.

[0009] Optionally, the first fault detection model is trained in the following manner: obtaining the historical state texts of a batch of power systems, where the historical state texts include normal state texts and fault state texts, and each type of historical state text contains the state information of the corresponding power system; using the historical state texts of the power system as input data, and the normal state and the fault state as output data to input into the first initial fault detection model for training. The input data is calculated through each layer of neural network to obtain the first prediction result; using the defined loss function to calculate the difference between the first prediction result and the true label to generate the first loss value; performing backpropagation according to the first loss value, calculating the gradients of each layer of neural network, and using the optimizer algorithm to update the model parameters; repeating the above steps until the predetermined number of training times or the convergence condition is reached, and the first fault detection model is trained.

[0010] Optionally, the second fault detection model is trained as follows: Obtain the historical state texts of a batch of power systems, where the historical state texts include normal state texts and fault state texts, and each type of historical state text contains the state information of the corresponding power system; Use the historical state texts of the power system as input data, and use the fault time, fault location, and fault type as output data to input into the second initial fault detection model for training. The input data is calculated through each layer of neural network to obtain a second prediction result; Use the defined loss function to calculate the difference between the second prediction result and the true label to generate a second loss value; Perform backpropagation based on the second loss value, calculate the gradients of each layer of neural network, and use the optimizer algorithm to update the model parameters; Repeat the above steps until the predetermined number of training times or convergence conditions are reached, and the second fault detection model training is completed.

[0011] Optionally, before monitoring and obtaining the state information of the power system, the method further includes: Obtain the operation image of the power system taken in real time; Identify the electrical parameters, equipment state parameters, and protection action parameters from the operation image; Parse the identified information through regular expressions and convert the information into a structured data format; Fill the structured data format into a preset template to generate the state information of the power system.

[0012] Optionally, after controlling the relay protection circuit to perform fault repair on the power system through the fault time, fault location, and fault type, the method further includes: Set different alarm types according to the severity of the fault type, where the alarm types can include any one of the following: audible and visual alarm, indicator light alarm, and digital display alarm.

[0013] Another aspect of the present invention provides a fault detection system for a relay protection circuit, the system includes: A monitoring module, configured to monitor and obtain the state information of the power system, where the state information of the power system includes electrical parameters, equipment state parameters, and protection action parameters; A first identification module, configured to input the state information of the power system into the trained first fault detection model for identification to obtain the first state of the power system, where the first state includes a normal state or a fault state; A second identification module, configured to, if the first state is a fault state, input the state information of the power system into the trained second fault detection model for identification to obtain the second state of the power system; Wherein, the second state includes the fault time, fault location, and fault type; A fault elimination module, configured to control the relay protection circuit to perform fault repair on the power system through the fault time, fault location, and fault type to achieve fault elimination of the power system.

[0014] Another aspect of the present invention provides a computer device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. It is characterized in that when the processor executes the computer program, it implements the fault detection method of the relay protection circuit in any of the above embodiments.

[0015] Another aspect of the present invention provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the fault detection method of the relay protection circuit in any of the above embodiments. Further, the computer-readable storage medium may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function, etc.; the data storage area may store data created according to the use of blockchain nodes, etc.

[0016] In the present invention, the state information of the power system is first obtained, and the state information of the power system is identified through two progressive fault detection models to obtain the specific fault elements of the power system, realizing the accurate positioning of the faults in the power system. Finally, based on the specific fault elements, the relay protection circuit is controlled to process the faults, improving the fault detection efficiency of the relay protection circuit. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 An optional flowchart showing the fault detection method of the relay protection circuit provided in Embodiment 1 of the present invention is shown; Figure 2 A structural block diagram showing the fault detection system of the relay protection circuit provided in Embodiment 2 of the present invention is shown; and Figure 3 A block diagram of a computer device suitable for implementing the fault detection method of the relay protection circuit provided in Embodiment 3 of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] It should be noted that in this text, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the presence of another identical element in the process, method, article or device including that element. Embodiment

[0020] This embodiment provides a fault detection method for a relay protection circuit. Figure 1 The flowchart showing the fault detection method for this relay protection circuit is as Figure 1 shown. The fault detection method for this relay protection circuit may include steps S101 to S104, where: Step S101: Monitor and obtain the state information of the power system, where the state information of the power system includes electrical parameters, equipment status parameters, and protection action parameters; Specifically, the electrical parameters may be data such as current, voltage, and power. The equipment status parameters may be the physical and electrical characteristics of the relay protection device itself, such as action time, action area, etc. The protection action parameters may be action current, action voltage, reclosing parameters, frequency protection parameters. It should be noted that the state information of the electrical system is not limited to the above-mentioned examples, and may also be other information of the same type, which is not limited here.

[0021] Step S102: Input the state information of the power system into the trained first fault detection model for identification to obtain the first state of the power system, where the first state includes a normal state or a fault state; The state information of the power system in the normal state may be: the voltage is within ±5% of the rated voltage; the standard frequency is 50 Hz or 60 Hz; the power factor is close to 1, etc. The state parameters of the power system in the fault state may be: overloaded load; overvoltage or undervoltage; periodic voltage swing, etc. The first fault detection model may be a convolutional neural network or other network models that can perform text recognition, which is not limited here.

[0022] Step S103: If the first state is a fault state, input the state information of the power system into the trained second fault detection model for identification to obtain the second state of the power system; where the second state includes the fault time, fault location, and fault type; The fault types can be: voltage and current transformer faults; relay contact faults; deformation faults of riveted parts in the electromagnetic system; internal faults of protection devices; incorrect wiring in secondary circuits; control circuit faults; DC grounding faults.

[0023] The second fault detection model can be a convolutional neural network or other network models capable of text recognition, which is not limited here.

[0024] It should be noted that the recognition processes of the first fault detection model and the second fault detection model have a progressive relationship. The first fault detection model classifies the state information of the power system into normal states and fault states, enabling the second fault detection model to only process the data in the fault state, improving the efficiency of data processing in the fault detection of the relay protection circuit. Further, by identifying the specific parameters of the fault through the second fault detection model, it is convenient for the precise positioning of power system faults and improves the accuracy of the fault detection results of the relay protection circuit.

[0025] Step S104, control the relay protection circuit to perform fault repair on the power system through the fault time, the fault location, and the fault type, so as to eliminate the faults of the power system.

[0026] The recognition of the two neural network models gives specific fault information, making the fault location and cause in the power system more accurate and contributing to the rapid elimination of faults in the power system.

[0027] Preferably, the controlling the relay protection circuit to perform fault repair on the power system through the fault time, the fault location, and the fault type includes steps S1041 to S1043, where: Step S1041, divide the second states with the same fault location into one set; Step S1042, sequentially extract the fault elimination strategies corresponding to the fault types in the set according to the occurrence order of the fault time; Step S1043, control the relay protection circuit to perform fault repair on the power system through the fault elimination strategy.

[0028] Classify the second states according to the fault location and process the faults of the same type in chronological order, improving the fault detection efficiency of the power system.

[0029] Preferably, the controlling the relay protection circuit to perform fault repair on the power system through the fault time, the fault location, and the fault type includes steps S1041’ to S1044’, where: Step S1041’, divide the second states with the same fault type into one set; Step S1042’, sort the second states of the same set according to the occurrence order of the fault time, and calculate the first time required to solve the faults in this order. Step S1043’, determine the shortest path passing through all the fault positions of the same set, and calculate the second time required to solve all the faults in this set according to the shortest path. Among them, the shortest path can be calculated by the Dijkstra algorithm or other types of path algorithms, and there is no limitation here.

[0030] Step S1044’, compare the magnitudes of the first time and the second time, extract the fault removal strategies of the fault types in the same set according to the order corresponding to the minimum time, and control the relay protection circuit to perform fault repair on the power system.

[0031] Among them, the order corresponding to the minimum time can be: when the first time is less than the second time, select to process the faults in turn according to the occurrence order of the fault time in the set; when the second time is less than the first time, select to process the faults in turn according to the order of the fault positions in the shortest path; when the second time is equal to the first time, select one of the fault time and the fault position to process the faults according to the corresponding order.

[0032] Through the above steps, rapid processing of faults can be achieved, and the situation where faults cannot be processed in time due to the long occurrence time of faults can be reduced, improving the fault detection efficiency of the relay protection circuit.

[0033] Preferably, the first fault detection model is trained through the following steps A1 to A5: Step A1, obtain the historical state texts of a batch of power systems, where the historical state texts include normal state texts and fault state texts, and each type of historical state text contains the state information of the corresponding power system. Step A2, input the historical state texts of the power system as input data, and the normal state and the fault state as output data into the first initial fault detection model for training. The input data is calculated through each layer of neural network to obtain the first prediction result. The first initial fault detection model can be a convolutional neural network model, a recurrent neural network model, or other types of text recognition neural network models, and there is no limitation here.

[0034] Step A3, use the defined loss function to calculate the difference between the first prediction result and the true label, and generate the first loss value. Step A4: Perform backpropagation based on the first loss value, calculate the gradients of each layer of the neural network, and update the model parameters using an optimizer algorithm; Step A5: Repeat the above steps until a predetermined number of training times or convergence conditions are reached, and the first fault detection model is trained.

[0035] Using the historical state text of the power system as the basic data framework, which encompasses all possible fault scenarios in the power system, input the historical state text into the first initial fault detection model for identification to obtain a basic judgment of the power system fault scenario, reducing the reprocessing of normal state data in subsequent identification processes and improving the fault detection efficiency of the relay protection circuit.

[0036] Preferably, the second fault detection model is trained through the following steps B1 to B5: Step B1: Obtain the historical state text of a batch of power systems, where the historical state text includes normal state text and fault state text, and each type of historical state text contains the corresponding state information of the power system; Step B2: Use the historical state text of the power system as input data, and the fault time, fault location, and fault type as output data to input into the second initial fault detection model for training. The input data is calculated through each layer of the neural network to obtain a second prediction result; Step B3: Use the defined loss function to calculate the difference between the second prediction result and the true label to generate a second loss value; Step B4: Perform backpropagation based on the second loss value, calculate the gradients of each layer of the neural network, and update the model parameters using an optimizer algorithm; Step B5: Repeat the above steps until a predetermined number of training times or convergence conditions are reached, and the second fault detection model is trained.

[0037] After obtaining the fault result identified by the first fault detection model, the second fault detection model further identifies the state information of the power system in the fault scenario to obtain more specific fault elements such as the fault time, fault location, and fault type, making an accurate positioning of the fault information of the power system and improving the efficiency of subsequent relay protection circuits for fault handling.

[0038] Preferably, before monitoring and obtaining the state information of the power system, the method further includes steps C1 to C4, where: Step C1: Obtain the running image of the power system taken in real time; Step C2: Identify and obtain the electrical parameters, equipment status parameters, and protection action parameters from the running image; Among them, the recognition of the operation image can be achieved through a convolutional neural network or other neural network models capable of image recognition, which is not limited herein.

[0039] Step C3: Parse the recognized information through regular expressions and convert the information into a structured data format. Step C4: Fill the structured data format into a preset template to generate the status information of the power system.

[0040] During the normal operation process of the power system, the obtained power system information is generally irregular information, which is not conducive to the direct processing of the neural network model. By parsing and extracting the information in the operation image of the power system, and further generating standardized information through regular expressions and preset templates, the detection accuracy of the subsequent model is improved, and the fault detection efficiency of the relay protection circuit is also enhanced.

[0041] It should be noted that this embodiment can also obtain the log table information of the power system, identify electrical parameters, equipment status parameters, and protection action parameters from the log table information; further parse the recognized information through regular expressions and convert the information into a structured data format; fill the structured data format into a preset template to generate the status information of the power system.

[0042] Preferably, after controlling the relay protection circuit to perform fault repair on the power system through the fault time, the fault location, and the fault type, the method further includes step D1: Step D1: Set different alarm types according to the severity of the fault type, where the alarm type can include any one of the following: audible and visual alarm, indicator light alarm, and digital display alarm.

[0043] Specifically, when the severity of the fault type is high, set an audible and visual alarm; when the severity of the fault type is medium, set an indicator light alarm; when the severity of the fault type is low, set a digital display alarm.

[0044] In this embodiment, the status information of the power system is first obtained, and the status information of the power system is recognized through two progressive fault detection models to obtain the specific fault elements of the power system, realizing the accurate positioning of the faults of the power system. Finally, based on the specific fault elements, the relay protection circuit is controlled to handle the faults, improving the fault detection efficiency of the relay protection circuit. Embodiment

[0045] Embodiment 2 of the present invention further provides a fault detection system for a relay protection circuit. This fault detection system for the relay protection circuit corresponds to the fault detection method for the relay protection circuit provided in Embodiment 1 above. The corresponding technical features and technical effects will not be elaborated in detail in this embodiment, and the relevant parts can refer to Embodiment 1 above. Specifically, Figure 2 shows a structural block diagram of the fault detection system for the relay protection circuit. As Figure 2 shown, the fault detection system 200 for the relay protection circuit includes a monitoring module 201, a first identification module 202, a second identification module 203, and a fault elimination module 204, where: The monitoring module 201 is configured to monitor and obtain the state information of the power system. Among them, the state information of the power system includes electrical parameters, equipment state parameters, and protection action parameters; The first identification module 202 is connected to the monitoring module 201 and is configured to input the state information of the power system into the trained first fault detection model for identification to obtain the first state of the power system. Among them, the first state includes a normal state or a fault state; The second identification module 203 is connected to the first identification module 202 and is configured to, if the first state is a fault state, input the state information of the power system into the trained second fault detection model for identification to obtain the second state of the power system; among them, the second state includes the fault time, fault location, and fault type; The fault elimination module 204 is connected to the second identification module 203 and is configured to control the relay protection circuit to perform fault repair on the power system through the fault time, fault location, and fault type to achieve fault elimination of the power system.

[0046] Optionally, the fault elimination module is further configured to: divide the second states with the same fault location into a set; sequentially extract the fault elimination strategies corresponding to the fault types in the set according to the occurrence order of the fault time; and control the relay protection circuit to perform fault repair on the power system through the fault elimination strategies.

[0047] Optionally, the fault elimination module is further configured to: divide the second states with the same fault type into a set; sort the second states in the same set according to the occurrence order of the fault time, and calculate the first time required to solve the faults in this order; determine the shortest path passing through all the fault locations in the same set, and calculate the second time required to solve all the faults in this set according to the shortest path; compare the magnitudes of the first time and the second time, and extract the fault elimination strategies of the fault types in the same set according to the order corresponding to the minimum time, and control the relay protection circuit to perform fault repair on the power system.

[0048] Optionally, the device further includes a first training module, configured to: obtain historical status texts of a batch of power systems, where the historical status texts include normal status texts and fault status texts, and each type of historical status text contains status information of the corresponding power system; use the historical status texts of the power system as input data, and the normal status and the fault status as output data to input into a first initial fault detection model for training. The input data is calculated through each layer of neural network to obtain a first prediction result; use a defined loss function to calculate the difference between the first prediction result and the true label, and generate a first loss value; perform backpropagation according to the first loss value, calculate the gradients of each layer of neural network, and use an optimizer algorithm to update the model parameters; repeat the above steps until a predetermined number of training times or convergence conditions are reached, and the first fault detection model is trained.

[0049] Optionally, the device further includes a second training module, configured to: obtain historical status texts of a batch of power systems, where the historical status texts include normal status texts and fault status texts, and each type of historical status text contains status information of the corresponding power system; use the historical status texts of the power system as input data, and the fault time, fault location, and fault type as output data to input into a second initial fault detection model for training. The input data is calculated through each layer of neural network to obtain a second prediction result; use a defined loss function to calculate the difference between the second prediction result and the true label, and generate a second loss value; perform backpropagation according to the second loss value, calculate the gradients of each layer of neural network, and use an optimizer algorithm to update the model parameters; repeat the above steps until a predetermined number of training times or convergence conditions are reached, and the second fault detection model is trained.

[0050] Optionally, the device further includes a preprocessing module, configured to: obtain an operation image of the power system captured in real time; identify electrical parameters, equipment status parameters, and protection action parameters from the operation image; parse the identified information through regular expressions and convert the information into a structured data format; fill the structured data format into a preset template to generate status information of the power system.

[0051] Optionally, the device further includes an alarm module, configured to: set different alarm types according to the severity of the fault type, where the alarm types may include any one of the following: audible and visual alarm, indicator light alarm, and digital display alarm. Embodiment

[0052] Figure 3The block diagram of the computer device suitable for implementing the fault detection method of the relay protection circuit provided in the third embodiment of the present invention is shown. In this embodiment, the computer device 300 may be a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server or a cabinet server (including an independent server or a server cluster composed of multiple servers) that executes a program, etc. As Figure 3 shown, the computer device 300 of this embodiment at least includes, but is not limited to, a memory 301, a processor 302, and a network interface 303 that can be communicatively connected to each other through a system bus. It should be noted that Figure 3 only the computer device 300 with components 301-303 is shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0053] In this embodiment, the memory 301 at least includes one type of computer-readable storage medium. The readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disc, etc. In some embodiments, the memory 301 may be an internal storage unit of the computer device 300, such as the hard disk or memory of the computer device 300. In other embodiments, the memory 301 may also be an external storage device of the computer device 300, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 300. Of course, the memory 301 may also include both the internal storage unit and the external storage device of the computer device 300. In this embodiment, the memory 301 is generally used to store the operating system and various application software installed in the computer device 300, such as the program code of the fault detection method of the relay protection circuit, etc.

[0054] In some embodiments, the processor 302 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 302 is generally used to control the overall operation of the computer device 300. For example, it executes control and processing related to data interaction or communication with the computer device 300. In this embodiment, the processor 302 is used to run the program code of the steps of the fault detection method of the relay protection circuit stored in the memory 301.

[0055] In this embodiment, the fault detection method for the relay protection circuit stored in the memory 301 can also be divided into one or more program modules and executed by one or more processors (processor 302 in this embodiment) to complete the present invention.

[0056] The network interface 303 may include a wireless network interface or a wired network interface, which is generally used to establish a communication link between the computer device 300 and other computer devices. For example, the network interface 303 is used to connect the computer device 300 to an external terminal through a network, and establish a data transmission channel and a communication link between the computer device 300 and the external terminal. The network can be an enterprise internal network (Intranet), the Internet, the Global System of Mobile communication (abbreviated as GSM), Wideband Code Division Multiple Access (abbreviated as WCDMA), 4G network, 5G network, Bluetooth, Wi-Fi and other wireless or wired networks. Embodiment

[0057] This embodiment also provides a computer-readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disc, server, App application mall, etc., on which a computer program is stored, and the steps of the fault detection method for the relay protection circuit are implemented when the computer program is executed by a processor.

[0058] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the embodiments of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device, so that they can be stored in the storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order from here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. Thus, the embodiments of the present invention are not limited to any specific combination of hardware and software.

[0059] It should be noted that the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments.

[0060] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described example methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation.

[0061] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied to other related technical fields, shall equally be included in the patent protection scope of the present invention.

Claims

1. A fault detection method for a relay protection circuit, characterized in that: The method comprises: Monitoring and obtaining status information of the power system, wherein the status information of the power system includes electrical parameters, equipment status parameters and protection action parameters; Inputting the state information of the power system into a trained first fault detection model for identification to obtain a first state of the power system, wherein the first state includes a normal state or a fault state; If the first state is a fault state, inputting the state information of the power system into the trained second fault detection model for identification to obtain the second state of the power system; wherein the second state includes the fault time, fault location and fault type; The fault time, the fault location and the fault type are used to control the relay protection circuit to perform fault repair on the power system, so as to eliminate the fault of the power system.

2. The method according to claim 1, characterized in that The controlling the relay protection circuit to perform fault repair on the power system according to the fault time, the fault location and the fault type includes: Grouping the second states with the same fault location into a set; Extracting the fault elimination strategies corresponding to the fault types in the set in order of occurrence of the fault times; The fault relief strategy is used to control the relay protection circuit to perform fault repair on the power system.

3. The method according to claim 1, characterized in that The controlling the relay protection circuit to perform fault repair on the power system according to the fault time, the fault location and the fault type includes: Grouping the second states with the same fault type into a set; sorting the second states of the same set according to the order of occurrence of the failure time, and calculating the first time required to resolve the failure according to the order; Determine the shortest path passing through all fault locations in the same set, and calculate the second time required to solve all faults in the set according to the shortest path; The first time and the second time are compared, and the fault elimination strategy of the fault type in the same set is extracted according to the sequence corresponding to the minimum time, and the relay protection circuit is controlled to perform fault repair on the power system.

4. The method according to claim 1, characterized in that: The first fault detection model is trained in the following manner: Acquire historical status texts of a batch of power systems, wherein the historical status texts include normal status texts and fault status texts, and each type of historical status text contains status information of the corresponding power system; Inputting the historical status text of the power system as input data, and the normal state and the fault state as output data into a first initial fault detection model for training, the input data is calculated through each layer of neural network to obtain a first prediction result; Calculate the difference between the first prediction result and the true label using the defined loss function to generate a first loss value; Perform back propagation according to the first loss value, calculate the gradient of each layer of the neural network, and use the optimizer algorithm to update the model parameters; The above steps are repeated until a predetermined number of training times or convergence conditions are reached, and the training of the first fault detection model is completed.

5. The method according to claim 1 or 4, characterized in that: The second fault detection model is trained in the following manner: Acquire historical status texts of a batch of power systems, wherein the historical status texts include normal status texts and fault status texts, and each type of historical status text contains status information of the corresponding power system; The historical status text of the power system is used as input data, and the fault time, fault location and fault type are used as output data to be input into the second initial fault detection model for training, and the input data is calculated through each layer of the neural network to obtain a second prediction result; Calculate the difference between the second prediction result and the true label using the defined loss function to generate a second loss value; Perform back propagation according to the second loss value, calculate the gradient of each layer of the neural network, and use the optimizer algorithm to update the model parameters; Repeat the above steps until a predetermined number of training times or convergence conditions are reached, and the training of the second fault detection model is completed.

6. The method according to claim 1, characterized in that Before monitoring and obtaining the status information of the power system, the method further includes: Obtain real-time images of the operation of the power system; Identify and obtain electrical parameters, equipment status parameters and protection action parameters from the operation image; Parse the identified information through regular expressions and convert the information into a structured data format; The structured data format is filled into a preset template to generate status information of the power system.

7. The method according to claim 1, characterized in that After controlling the relay protection circuit to perform fault repair on the power system according to the fault time, the fault location and the fault type, the method further includes: Different alarm types are set according to the severity of the fault type, wherein the alarm type may include any one of the following: sound and light alarm, indicator light alarm and digital display alarm.

8. A fault detection system for a relay protection circuit, characterized in that: The system comprises: A monitoring module, used to monitor and obtain status information of the power system, wherein the status information of the power system includes electrical parameters, equipment status parameters and protection action parameters; A first identification module is used to input the state information of the power system into a trained first fault detection model for identification, so as to obtain a first state of the power system, wherein the first state includes a normal state or a fault state; A second identification module is used for inputting the state information of the power system into a trained second fault detection model for identification if the first state is a fault state, so as to obtain a second state of the power system; wherein the second state includes a fault time, a fault location and a fault type; The fault removal module is used to control the relay protection circuit to perform fault repair on the power system according to the fault time, the fault location and the fault type, so as to remove the fault of the power system.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images