Method and system for intelligently detecting fault and health state of device

Abstract

The invention discloses a method and system for intelligently detecting device faults and health states, and the method comprises the steps: collecting the current data of each module in a relay protection device, the current data comprising the working information, the current alarm information, the current protection function state and the current protection action information of each board card or chip in the relay protection device; according to the current data and historical data of each module and the fault and health diagnosis logic in the knowledge base, reasoning the current fault, the fault reason, the current health state and the reason influencing the current health state of the relay protection device in real time; wherein the fault and health diagnosis logic is stored in the knowledge base in a regular manner. According to the invention, the current fault and the fault reason thereof, the current health state and the reason influencing the current health state of the device can be output in real time, the fault board card or chip can be quickly positioned, and the accuracy of signal transmission of the relay protection device is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of relay protection of power systems. Specifically, it relates to a method and system for intelligently detecting the faults and health status of a device. Background Art

[0002] A relay protection device is an automated measure and equipment that can send warning signals to on-duty operators in a timely manner when a power component (such as a generator, line, etc.) or the power system itself in the power system fails and endangers the safe operation of the power system, or directly send a tripping command to the controlled circuit breaker to terminate the development of these events.

[0003] The reliability of the signals sent by the relay protection device is directly related to whether the power system can operate safely. Therefore, the health status of the relay protection device itself is crucial. With the development of the power grid, the functions of relay protection devices are becoming increasingly diverse and the structures are becoming increasingly complex. When a software or hardware fault occurs in the device, it may cause incorrect signals to be sent, resulting in misoperation of the protection action. Therefore, to ensure the correctness of the signals sent by the relay protection device, it is necessary to monitor the health status of the device. Currently, the substation master station collects a large number of alarm messages of relay protection devices, and it is impossible to clearly locate the faulty circuit board of the faulty device. After a fault occurs, in order to eliminate the fault as soon as possible, it is necessary to locate the fault to the circuit board level or chip level of the device, analyze the cause of the fault, and give fault handling measures.

[0004] Prior art document 1 (CN112580712A) discloses a method and system for auxiliary decision-making in relay protection device fault handling, including: obtaining real-time alarm data of the relay protection device, inputting the data into a trained risk logic tree classification model, and outputting alarm nature, fault location, and alarm cause data; based on the output result information, matching with the data in the expert knowledge base to obtain the fault handling method for the corresponding fault.

[0005] Prior art document 2 (CN117521804A) discloses a method for identifying and reasoning hidden faults in relay protection based on a knowledge graph. The method includes the following steps: A. Construct a knowledge graph for identifying relay protection faults; B. Real-time sense the operation information and fault characteristics of line protection, bus protection, transformer protection, bus-tie sectional protection, circuit breaker protection, sampling circuit, input / output circuit, pressure plate, and setting value. Use a document vectorization model to perform distributed vectorization representation on the above relay protection fault entity description information. Based on the knowledge reasoning algorithm of distributed representation, substitute it into the knowledge graph for identifying relay protection faults, and calculate through single-step and multi-step causal relationship reasoning to obtain different types of hidden faults in relay protection.

[0006] The deficiency of the prior art document 1 lies in that it only analyzes the alarm data of the protection device. Only when the existing protection device has triggered an alarm, will it conduct fault analysis and give the fault cause and fault handling method for this alarm. It cannot give early warnings for other hidden abnormalities of the device that have not triggered an alarm. Moreover, there are often multiple alarms simultaneously occurring in the protection device, and these alarms often have a causal relationship logically, that is, multiple alarms caused by the same reason. However, the method in technical document 1 may locate these alarms as multiple faults.

[0007] The prior art document 2 mainly focuses on the power grid itself. According to the current data information, it identifies whether there are faults such as operation errors and circuit breaks in a certain power grid area, all of which are static fault inferences. Summary of the Invention

[0008] To solve the deficiencies in the prior art, the present invention provides a method and system for intelligently detecting the faults and health status of a device, which can quickly locate the faulty board or chip and improve the accuracy rate of the signals sent by the relay protection device.

[0009] The present invention adopts the following technical solutions.

[0010] The first aspect of the present invention provides a method for intelligently detecting the faults and health status of a device, including:

[0011] Collect the current data of each module in the relay protection device. The current data includes the working information, current alarm information, current protection function status, and current protection action information of each board or chip in the relay protection device; according to the current data, historical data of each module, and the fault and health diagnosis logic in the knowledge base, infer in real time the current faults and fault causes of the relay protection device, the current health status, and the reasons affecting the current health status. Among them, the fault and health diagnosis logic is stored in the knowledge base in the form of rules.

[0012] Optionally, the fault and health diagnosis logic of the relay protection device includes at least one of the following: the fault judgment logic of each board or chip, the alarm judgment logic, the protection function judgment logic, the protection function blocking logic, and the protection alarm judgment logic.

[0013] Optionally, according to the current data, historical data of each module, and the fault and health diagnosis logic in the knowledge base, inferring in real time the current faults and fault causes of the relay protection device, the current health status, and the reasons affecting the current health status includes:

[0014] Compare the current data of each module with the data in the normal state, and conduct abnormal analysis of the faults according to the comparison results;

[0015] Store knowledge in the knowledge base in a rule-based manner. The rules are represented in the form of if-T-Then, and the knowledge is used to represent fault and health diagnosis logic. Here, if is the condition, T is the duration for meeting the condition, and Then is the resulting consequence. In the case where the analysis result is abnormal, traverse the knowledge base and extract all the knowledge related to the current data of each module in the knowledge base. If the number of all the knowledge related to the current data of each module is less than the first threshold, check one by one whether the current data meets the knowledge to determine whether it is the cause represented by the knowledge. Otherwise, combine the historical data and convert all the extracted knowledge into a knowledge graph with time edges.

[0016] Traverse all the boundary nodes of the corresponding knowledge graph to infer the current fault and its cause, the current health status and the cause affecting the current health status.

[0017] Optionally, the relay protection device faults include first-class alarms, second-class alarms, and third-class anomalies. The first-class alarms are alarms that directly affect the operation of the relay protection device. When a first-class alarm is triggered, the relay protection device stops all protection functions and only executes partial display and monitoring functions. When a second-class alarm is triggered, only the functions related to the second-class alarm are affected, and the remaining protection functions work normally. When a third-class anomaly is triggered, it does not affect the operation of any protection function and only gives an anomaly prompt.

[0018] Optionally, the health status includes device normal, device faulty, and device sub-healthy. When the relay protection device has no alarms, the device health status is 100%. When any first-class alarm occurs in the relay protection device, the device health status is 0%. Calculate the device sub-healthy x% according to the following formula: x = 100 – K1*N1 – K2*N2; where N1 is the number of times the second-class alarm is triggered, N2 is the number of times the third-class anomaly is triggered, K1 is the second-class alarm health status coefficient, and K2 is the third-class anomaly health status coefficient.

[0019] Optionally, the method further includes: explaining the real-time reasoning process and presenting the reasoning process to the user in a visual manner.

[0020] Optionally, the method further includes: obtaining a fault query instruction from the user; searching in the historical data based on the fault information in the fault query instruction to obtain all the historical data of the fault; analyzing all the historical data of the fault to obtain the historical fault cause of the fault.

[0021] The second aspect of the present invention provides a system for intelligently detecting the faults and health status of a device, and the system includes:

[0022] The intelligent diagnosis module is mounted on the internal network of the relay protection device and is used to collect the current data of each module mounted on the internal network of the relay protection device. The current data includes the working information of each board or chip in the relay protection device, the current alarm information, the current protection function status, and the current protection action information. According to the current data, historical data of each module, and the fault and health diagnosis logic in the knowledge base, it can infer in real time the current faults and their causes, the current health status, and the reasons affecting the current health status of the relay protection device.

[0023] The intelligent diagnosis module includes a data acquisition module, a data storage module, a knowledge base, and a status inference engine. Among them,

[0024] The data acquisition module is used to collect the current data of each module mounted on the internal network of the relay protection device;

[0025] The data storage module is used to store the current data of each module collected by the data acquisition module as historical data;

[0026] The knowledge base is used to store the fault and health diagnosis logic of the relay protection device;

[0027] The status inference engine is used to infer in real time the current faults and their causes, the current health status, and the reasons affecting the current health status of the device according to the current data of each module collected by the data acquisition module, the historical data in the data storage module, and the fault and health diagnosis logic of the relay protection device in the knowledge base.

[0028] Optionally, the intelligent diagnosis module further includes:

[0029] An interpreter, which is used to explain the real-time inference process and display the inference process to the user in a visual way.

[0030] Optionally, each module mounted on the internal network of the relay protection device includes at least one of the following: a power supply module, a communication and management module, a human-machine interface module, a protection CPU module, an intelligent input module, and an intelligent output module. Among them, the power supply module is used to provide power for the relay protection device; the communication and management module is used to undertake the communication between the relay protection device and the outside world, exchange information, and perform fault recording; the human-machine interface module is used to display the current current and voltage, protection function status, and device alarm information; the protection CPU module is used for sampling, protection action principle judgment, accident recording, and software and hardware self-check; the intelligent input module is used to access digital input signals; the intelligent output module is used to output digital output signals.

[0031] Compared with the prior art, the beneficial effects of the present invention at least include:

[0032] Compared with the prior art document 1, the present invention is significantly different in that the present invention monitors and analyzes all the information that can be monitored, such as the current working information, fault information, current alarm information, current protection function status, action information, etc. of each board or chip of the relay protection device, compares the currently collected data with the correct data to find abnormal conditions, report the current device fault status, and infer the cause of the device failure based on the collected data. Through the fault location and cause analysis of the device that has issued an alarm, the issuance of abnormal alarms and the protection function status and protection action status analysis of the device that has not issued an alarm but has an abnormality, such as the location of the cause of the protection function not being put into use, the analysis of the cause of the protection action not being triggered, and the analysis of the cause of the protection action triggering, the faulty board or chip can be quickly located, and the accuracy of the signal issued by the relay protection device can be improved.

[0033] Compared with the prior art document 2, the significant difference of the present invention lies in that it is based on a method that combines traditional rules and knowledge graphs, and introduces the concept of time edges in the knowledge graph. Based on historical data and current data, it identifies the implicit and explicit faults of the relay protection device itself, identifies abnormalities and causes, and judges the health status of the device, and has the logical reasoning function in time sequence; based on this significant difference, the technical effects actually achieved by the present invention include at least: 1) fault location and cause analysis of the device for which an alarm has been issued; 2) issuance of abnormal alarms when the device has not issued an alarm but has abnormalities itself; 3) analysis of protection function status and protection action status, such as locating the cause of the protection function not being put into use, analyzing the cause of the protection action not being triggered, and analyzing the cause of the protection action being triggered. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0035] Figure 1 An application environment diagram of a method for intelligently detecting device failure and health status provided by an embodiment of the present invention;

[0036] Figure 2 A schematic diagram of the structure of a knowledge graph in a method for intelligently detecting device failure and health status provided by an embodiment of the present invention;

[0037] Figure 3 A schematic diagram of the structure of an intelligent diagnosis module in a system for intelligently detecting device failure and health status provided by an embodiment of the present invention;

[0038] Figure 4A schematic structural diagram of a relay protection device provided by an embodiment of the present invention;

[0039] Figure 5 A schematic internal structure diagram of an electronic device provided by an embodiment of the present invention. Specific embodiments

[0040] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only a part of the embodiments of the present invention, rather than all embodiments. Based on the spirit of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0041] The application environment diagram of the embodiments of the present invention is as Figure 1 shown. The relay protection device is connected to a circuit breaker, a transmission line, or a transformer. By monitoring the health status of the relay protection device and using the data collected by the relay protection device, the current faults of the device, the causes of the faults, the current health status, and the reasons affecting the current health status are obtained, thereby reducing misoperation of protection actions, improving the accuracy of signals sent by the relay protection device, and enabling quick location of faulty boards or chips.

[0042] Embodiment 1 of the present invention provides a method for intelligently detecting the faults and health status of a device, including:

[0043] S1. Collect the current data of each module in the relay protection device. The current data includes the working information of each board or chip in the relay protection device, the current alarm information, the current protection function status, and the current protection action information.

[0044] Specifically, the working information includes at least one of the following: working voltage, number of chip reads and writes. The current alarm includes at least one of the following: PT disconnection alarm, CT disconnection alarm, protection CPU plug-in abnormality alarm, and configuration file error alarm. The current protection function status includes at least one of the following: overcurrent protection function is effective, differential protection function is effective, reclosing function is ineffective. The current protection action information includes at least one of the following: overcurrent protection action, action current, and action time.

[0045] S2. According to the current data, historical data of each module, and the fault and health diagnosis logic in the knowledge base, infer in real time the current faults of the relay protection device, the causes of the faults, the current health status, and the reasons affecting the current health status.

[0046] Optionally, the fault and health diagnosis logic of the relay protection device includes at least one of the following: the fault judgment logic of each board or chip, the alarm judgment logic, the protection function judgment logic, the protection function blocking logic, and the protection alarm judgment logic.

[0047] Specifically, the fault judgment logic of each board or chip of the relay protection device includes the following examples:

[0048] For example: IF the working voltage of CPU1 < x V AND

[0049] the working voltage of CPU2 < x V

[0050] THEN the power supply board is faulty

[0051] IF the number of chip reads and writes > the read and write count limit AND

[0052] the ID of the board where the chip is located = y

[0053] THEN the chip on the X board is faulty

[0054] It should be noted that V is the voltage unit, x is the preset threshold, which is obtained according to expert experience. y is the board number.

[0055] Specifically, the alarm judgment logic of the relay protection device includes the following examples:

[0056] For example: IF the current value collected by AD1!= the current value collected by AD2

[0057] THEN the current acquisition is abnormal

[0058] IF the voltage value collected by AD1!= the voltage value collected by AD2

[0059] THEN the voltage acquisition is abnormal

[0060] IF the current acquisition is abnormal OR the voltage acquisition is abnormal

[0061] THEN the sampled value is incorrect

[0062] IF the current acquisition is abnormal OR the voltage acquisition is abnormal OR the sampled value is incorrect

[0063] THEN the device is faulty

[0064] Specifically, the protection function judgment logic of the relay protection device includes the following examples:

[0065] For example: IF ﹥ the change amount starting current setting value IQD OR

[0066] △3i0 ﹥ the change amount starting current setting value IQD

[0067] THEN Protection activation

[0068] IF Protection activation AND

[0069] The current current I > overcurrent operation current setting value Iset AND

[0070] The duration T during which the current current is greater than the overcurrent operation current setting value > overcurrent operation time setting value Tset AND

[0071] Overcurrent protection function status valid = 1

[0072] THEN Overcurrent protection action

[0073] Among them, IQD is the variable activation current setting value, △3i0 is the sudden change of zero-sequence current, is the current change value of a certain phase, and the is calculated according to the following formula:

[0074]

[0075] Among them, refers to three phase types of AB, BC, and CA, K refers to the current sampling moment; Ts is the current value sampling period, K - Ts refers to the sampling moment 1 sampling period before the K moment; K - 2Ts refers to the sampling moment 2 sampling periods before the K moment, represents the current value of a certain phase at the K moment, represents the current value sampled 1 sampling period before the K moment, represents the current value sampled 2 sampling periods before the K moment.

[0076] Specifically, the locking logic of each protection function of the relay protection device includes the following examples:

[0077] For example: IF the overcurrent protection hard pressure plate is put in = 0 OR

[0078] the overcurrent protection soft pressure plate is put in = 0 OR

[0079] the overcurrent protection control word is put in = 0

[0080] THEN the overcurrent protection function is locked

[0081] IF CT disconnection alarm = 1

[0082] THEN the overcurrent protection function is locked

[0083] IF the SV data quality is not valid = 0

[0084] THEN Lock all protections

[0085] IF dual AD inconsistency = 1

[0086] THEN block all protections

[0087] IF the SV maintenance status is inconsistent with the device

[0088] THEN block all protections

[0089] IF the SV sampling is interrupted

[0090] THEN block all protections

[0091] IF the SV sampling is out of step

[0092] THEN block all protections

[0093] Specifically, the protection alarm judgment logics of the relay protection device include the following examples:

[0094] For example: IF zero-sequence current > zero-sequence starting current setting value AND

[0095] the duration for which the zero-sequence current is greater than the zero-sequence starting current setting value > N s

[0096] THEN CT disconnection alarm

[0097] It should be noted that N is a setting value, and its magnitude is obtained based on expert experience.

[0098] Optionally, the faults of the relay protection device include first-class alarms, second-class alarms, and third-class anomalies. The first-class alarms are alarms that directly affect the operation of the relay protection device. When a first-class alarm is triggered, the relay protection device stops all protection functions and only executes some display and monitoring functions; when a second-class alarm is triggered, it only affects the functions related to the second-class alarm, and the remaining protection functions work normally; when a third-class anomaly is triggered, it does not affect the operation of any protection function and only gives an anomaly prompt.

[0099] Optionally, the health status includes device normal, device fault, and device sub-health; when the relay protection device has no alarms, the device health status is 100%; when any type of alarm occurs in the relay protection device, the device health status is 0%;

[0100] Calculate the device sub-health x% according to the following formula: x = 100 – K1*N1 – K2*N2; where, N1 is the number of second-class alarm triggers, N2 is the number of third-class anomaly triggers, K1 is the second-class alarm health status coefficient, and K2 is the third-class anomaly health status coefficient.

[0101] Specifically, K1 takes a value of 20%, and K2 takes a value of 2%.

[0102] S2 includes:

[0103] S2.1. Compare the current data of each module with the data in the normal state; and perform abnormal analysis on the faults according to the comparison results.

[0104] Specifically, performing abnormal analysis on the faults includes fault location and cause analysis for the faults with alarms issued, abnormal alarm issuance for the devices without alarms but with abnormalities themselves, and analysis of the protection function status and protection action status. For example, location of the reason for the protection function not being put into operation, analysis of the reason for the protection action not being triggered, and analysis of the reason for the protection action being triggered, etc.

[0105] S2.2. Store the knowledge in the knowledge base in the form of rules, where the rules are represented in the form of if-T-Then. The knowledge is used to represent the fault and health diagnosis logic. Among them, if is the condition, T is the duration for meeting the condition, and Then is the resulting consequence; in the case where the analysis result is abnormal, traverse the knowledge base and extract all the knowledge related to the current data of each module in the knowledge base. If the number of all the knowledge related to the current data of each module is less than the first threshold, check one by one whether the current data meets the knowledge to determine whether it is the reason represented by the knowledge. Otherwise, combine the historical data to transform the extracted all knowledge into a knowledge graph with time edges.

[0106] Optionally, the first threshold can take a value of 3 or a value of 5.

[0107] Specifically, combining the historical data to transform the extracted all knowledge into a knowledge graph with time edges includes: taking the result stored in Then in the rule as the entity of the knowledge graph, the condition stored in if as the cause, and the time stored in T as the time parameter of the edge. In this way, based on the method combining traditional rules and knowledge graphs, it is possible to identify the hidden and explicit faults of the relay protection device itself and have the logical reasoning ability in the time sequence.

[0108] Specifically, if includes the conditional relationship and causal relationship in the above logic, and takes the conditional relationship and causal relationship in if as the edges. Then includes each module, fault, fault cause, and health status, and takes the result stored in Then as the entity, and the time parameter in T as the time attribute of the edge.

[0109] S2.3. Traverse all the boundary nodes of the corresponding knowledge graph to infer the current fault and its cause, the current health status and the reason affecting the current health status.

[0110] Combined with Figure 2 shown in Figure 2 is a knowledge graph transformed from the blocking of the overcurrent protection function. By traversing all the boundary nodes of the knowledge graph such as CT disconnection and SV sampling interruption, the current fault and its cause, the current health status and the reason affecting the current health status are inferred.

[0111] In this embodiment, not only can the fault location and cause analysis of the already issued alarms be performed, but also the fault location of the abnormal alarms that the device has not issued alarms but itself has abnormalities can be performed. In addition, the protection function status and protection action status analysis, etc. can be performed, so as to quickly locate the faulty board or chip and improve the accuracy of the signals sent by the relay protection device.

[0112] Optionally, the transformed knowledge graph with time edges is input into a pre-trained graph neural network model. Through the node features, edge relationships, and time series information of the knowledge graph, using the prediction results of the nodes and edges output by the graph neural network model, the current fault and its cause, the current health status, and the reasons affecting the current health status are inferred.

[0113] Specifically, the graph neural network model is constructed based on the graph convolutional network combined with the recurrent neural network. The graph neural network model is trained using historical data and labeled fault cases. By optimizing the loss function and model structure, the reliability of the model is improved.

[0114] Specifically, the following loss function is used during the training process:

[0115]

[0116] where n is the number of nodes in the knowledge graph, T is the length of the time series, y i (t) is the true value of node i at time t, is the predicted value of node i at time t, and λ is the time decay factor, which is used to adjust the loss weights at different time points.

[0117] Optionally, the method further includes:

[0118] S3. Explain the real-time inference process and display the inference process to the user in a visual manner.

[0119] Specifically, a 3D model is used to display the knowledge graph and the inference process, and the paths involved in each inference process are explained. For example, using the Gemini Flash Thinking model, the inference process is deeply visualized and displayed to the user.

[0120] Specifically, different colors are used to represent the states of the nodes, different colors and thicknesses are used to represent the connection states of the edges, and time attributes are assigned to the edges. The inference paths are highlighted with arrows and displayed, and a time axis is added below the graph to show the state changes at different time points.

[0121] Optionally, the method further includes:

[0122] S4. Obtain the user's fault query instruction, search in the historical data based on the fault information in the fault query instruction to obtain all the historical data of the fault, and analyze all the historical data of the fault to obtain the historical fault cause.

[0123] Combined with Figure 3 , Figure 4 As shown in

[0124] An intelligent detection device fault and health status system provided by Embodiment 2 of the present invention, the system includes:

[0125] The intelligent diagnosis module is mounted on the internal network of the relay protection device and is used to collect the current data of each module mounted on the internal network of the relay protection device. The current data includes the working information of each board or chip in the relay protection device, the current alarm information, the current protection function status, and the current protection action information; according to the current data, historical data of each module and the fault and health diagnosis logic in the knowledge base, the current fault and fault cause of the relay protection device, the current health status and the reasons affecting the current health status are inferred in real time;

[0126] The data acquisition module is used to collect the current data of each module mounted on the internal network of the relay protection device;

[0127] The data storage module is used to store the current data of each module collected by the data acquisition module as historical data;

[0128] The knowledge base is used to store the fault and health diagnosis logic of the relay protection device;

[0129] The status inference machine is used to infer the current fault and its fault cause, the current health status and the reasons affecting the current health status of the device in real time according to the current data of each module collected by the data acquisition module, the historical data in the data storage module, and the fault and health diagnosis logic of the relay protection device in the knowledge base.

[0130] Optionally, the intelligent diagnosis module further includes: an interpreter, which is used to interpret the real-time inference process and display the inference process to the user in a visual manner.

[0131] Optionally, each module mounted on the intranet in the relay protection device includes at least one of the following: a power module, a communication and management module, a human-machine interface module, a protection CPU module, an intelligent input module, and an intelligent output module; wherein, the power module is used to provide power for the relay protection device, the communication and management module is used to undertake communication between the relay protection device and the outside world, exchange information and perform fault recording, the human-machine interface module is used to display the current current and voltage, protection function status, and device alarm information; the protection CPU module is used for sampling, judging the protection action principle, accident recording, and self-checking of software and hardware; the intelligent input module is used to access digital quantity signals; the intelligent output module is used to output digital quantity signals.

[0132] Optionally, the intranet includes a CAN network and / or an Ethernet network, and the intranet is used to collect data of each module connected to the intranet in real time.

[0133] Combined with Figure 5 As shown, Embodiment 3 of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is loaded into the processor, it implements the method for intelligent detection of the fault and health status of the device in Embodiment 1. The memory includes an internal memory and a non-volatile storage medium. The electronic device further includes a system bus, an input / output interface, and a communication interface. The system bus is used for communication between the processor, the memory, and the input / output interface. The input / output interface is used to connect the system bus and the communication interface. The communication interface is used to communicate with external devices.

[0134] It should be noted that the electronic device can be a computer device or a terminal device.

[0135] Compared with the prior art, the beneficial effects of the present invention at least include:

[0136] Compared with the prior art document 1, the significant difference of the present invention is that the present invention monitors and analyzes all observable information such as the current working information, fault information, current alarm information, current protection function status, and action information of each board or chip of the relay protection device, compares the currently collected data with the correct data to find abnormal situations, reports the current device fault status, and infers the cause of the device fault based on the collected data. Through the fault location and cause analysis of the faults that have been alarmed by the device, the abnormal alarm of the device that has not issued an alarm but has abnormalities itself, and the analysis of the protection function status and protection action status, such as the cause location of the unactivated protection function, the cause analysis of the untriggered protection action, and the cause analysis of the triggered protection action, etc., it is possible to quickly locate the faulty board or chip and improve the accuracy of the signals sent by the relay protection device.

[0137] Compared with the prior art document 2, the significant difference of the present invention lies in that, based on the method combining traditional rule-based and knowledge graph, and introducing the concept of time edge into the knowledge graph, based on historical data and current data, for the relay protection device itself, it identifies the latent and explicit faults of the device itself, abnormal identification and cause reasoning, device health status judgment, and has the logical reasoning function in chronological order; based on this significant difference, the technical effects actually achieved by the present invention at least include: 1) Fault location and cause analysis of the faults for which the device has issued alarms; 2) Sending abnormal alarms for the abnormalities that exist in the device itself but for which the device has not issued alarms; 3) Analysis of the protection function status and protection action status, such as location of the reasons for the protection function not being put into operation, analysis of the reasons for the protection action not being triggered, analysis of the reasons for the protection action being triggered, etc.

[0138] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution is prior or posterior, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0139] The present disclosure may be a system, a method, and / or a computer program product. The computer program product may include a computer-readable storage medium having thereon computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0140] The computer-readable storage medium may be a tangible device that can retain and store instructions used by an instruction execution device. The computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (non-exhaustive list) of the computer-readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanically encoded device, such as a punched card or raised structures in grooves storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage medium used herein is not construed as an instantaneous signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated through a waveguide or other transmission medium (e.g., an optical pulse through an optical fiber cable), or an electrical signal transmitted through a wire.

[0141] The computer-readable program instructions described herein can be downloaded to various computing / processing devices from a computer-readable storage medium or downloaded to an external computer or external storage device through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing / processing device.

[0142] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, etc., and conventional procedural programming languages such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider). In some embodiments, by using the state information of the computer-readable program instructions to customize an electronic circuit, such as a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit can execute the computer-readable program instructions to implement various aspects of the present disclosure.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent substitutions can still be made to the specific embodiments of the present invention, and any modification or equivalent substitution that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A method for intelligently detecting device failure and health status, characterized in that: include: Collecting current data of each module in the relay protection device, the current data including working information of each board or chip in the relay protection device, current alarm information, current protection function status and current protection action information; Based on the current data, historical data and fault and health diagnosis logic in the knowledge base of each module, the current fault and fault cause, current health status and the reasons affecting the current health status of the relay protection device can be inferred in real time; Among them, the fault and health diagnosis logic is stored in the knowledge base in a rule-based manner.

2. The method for intelligently detecting device failure and health status according to claim 1, characterized in that: The fault and health diagnosis logic of the relay protection device includes at least one of the following: each board or chip fault judgment logic, alarm judgment logic, protection function judgment logic, protection function locking logic and protection alarm judgment logic.

3. The method for intelligently detecting device failure and health status according to claim 1, characterized in that: Based on the current data, historical data and fault and health diagnosis logic in the knowledge base of each module, the current fault and fault cause, current health status and reasons affecting the current health status of the relay protection device are inferred in real time, including: Compare the current data of each module with the data in the normal state, and perform abnormal analysis on the fault according to the comparison result; The knowledge is stored in the knowledge base in a rule-based manner. The rules are expressed in the form of if-T-Then. The knowledge is used to characterize the fault and health diagnosis logic, where if is the condition, T is the duration of satisfying the condition, and Then is the result. When the analysis result is abnormal, the knowledge base is traversed to extract all the knowledge related to the current data of each module in the knowledge base. If the number of all the knowledge related to the current data of each module is less than the first threshold, the current data is checked one by one to see if it satisfies the knowledge to determine whether it is the cause of the knowledge representation. Otherwise, all the extracted knowledge is converted into a knowledge graph with time edges in combination with historical data. Traverse all boundary nodes of the corresponding knowledge graph, infer the current fault and fault cause, the current health status and the reasons affecting the current health status.

4. The method for intelligently detecting device failure and health status according to claim 1, characterized in that: The faults of the relay protection device include Class I alarm, Class II alarm and Class III abnormalities. The Class I alarm is an alarm that directly affects the operation of the relay protection device. When the Class I alarm is triggered, the relay protection device stops all protection functions and only performs some display and monitoring functions; when the Class II alarm is triggered, only the functions related to the Class II alarm are affected, and the other protection functions work normally; when the Class III abnormality is triggered, it does not affect the operation of any protection function, and only gives an abnormal prompt.

5. The method for intelligently detecting device failure and health status according to claim 1, characterized in that: The health status includes normal device, device failure and sub-healthy device; when the relay protection device has no alarm, the device health status is 100%; when any type of alarm occurs in the relay protection device, the device health status is 0%; The sub-health x% of the device is calculated according to the following formula: x=100-K1*N1-K2*N2; wherein N1 is the number of Class II alarm triggers, N2 is the number of Class III abnormal triggers, K1 is the Class II alarm health status coefficient, and K2 is the Class III abnormal health status coefficient.

6. The method for intelligently detecting device failure and health status according to claim 1, characterized in that: The method further comprises: The real-time reasoning process is explained and displayed to the user in a visual way.

7. The method for intelligently detecting device failure and health status according to claim 1, characterized in that: The method further comprises: Obtain the user's fault query instructions; Based on the fault information in the fault query instruction, search in the historical data to obtain all historical data of the fault; Analyze all historical data of the fault to obtain the historical fault causes.

8. A system for intelligently detecting device failure and health status using the method for intelligently detecting device failure and health status according to any one of claims 1 to 7, characterized in that: The system comprises: The intelligent diagnosis module is mounted on the intranet of the relay protection device and is used to collect the current data of each module mounted on the intranet in the relay protection device, wherein the current data includes the working information of each board or chip in the relay protection device, the current alarm information, the current protection function status and the current protection action information; based on the current data of each module, the historical data and the fault and health diagnosis logic in the knowledge base, the current fault and fault cause, the current health state and the reasons affecting the current health state of the relay protection device are inferred in real time; The intelligent diagnosis module includes a data acquisition module, a data storage module, a knowledge base and a state inference engine, wherein: Data acquisition module, used to collect current data of each module in the relay protection device mounted on the intranet; A data storage module is used to store the current data of each module collected by the data collection module as historical data; Knowledge base for storing fault and health diagnosis logic of relay protection devices; The state inference engine is used to infer the current fault of the device and its cause, the current health state and the reasons affecting the current health state in real time based on the current data of each module collected by the data acquisition module, the historical data in the data storage module and the fault and health diagnosis logic of the relay protection device in the knowledge base.

9. The system for intelligently detecting device failure and health status according to claim 8, characterized in that: The intelligent diagnosis module also includes: The interpreter is used to interpret the real-time reasoning process and present the reasoning process to the user in a visual way.

10. The system for intelligently detecting device failure and health status according to claim 8, characterized in that: The modules mounted on the intranet in the relay protection device include at least one of the following: a power supply module, a communication and management module, a human-machine interface module, a protection CPU module, an intelligent input module and an intelligent output module; wherein the power supply module is used to provide power for the relay protection device, the communication and management module is used to undertake the communication, information exchange and fault recording between the relay protection device and the outside world, the human-machine interface module is used to display the current current and voltage, protection function status, and device alarm information; the protection CPU module is used for sampling, protection action principle judgment, accident recording and software and hardware self-test; the intelligent input module is used to access switch signals; the intelligent output module is used to output switch signals.

Images