A switchgear defect detection method and system based on multi-target recognition

Abstract

This invention proposes a method and system for defect detection in switchgear based on multi-target identification, belonging to the field of equipment inspection and data identification technology. The detection system includes a switchgear status signal acquisition unit, a switchgear status signal reporting unit, a time interval judgment unit, a signal sequence intersection calculation unit, a signal sequence union calculation unit, a target element determination unit, and a defect result output unit. The detection method involves calculating the union and intersection sequences of multiple abnormal signal sequences to determine at least two elements as target identification elements, thereby determining the defect type of the switchgear. This invention simultaneously considers the intersection and union of multiple abnormal signal sequence sets to determine multi-target identification elements, thus performing defect detection in switchgear. It also considers historical false negative probabilities, improving the timeliness, accuracy, and completeness of switchgear defect detection.

Description

Technical Field

[0001] The present invention belongs to the technical field of equipment detection and data identification, and in particular, relates to a switch equipment defect detection method, system and computer-readable storage medium based on multi-target identification. Background Art

[0002] The power distribution system is the last relay system that distributes electric energy to end users in the traditional power system classification. The distribution system is composed of a variety of distribution equipment (or components) and distribution facilities. Switchgear is widely used in power distribution systems such as substations, and its safe and reliable operation is crucial.

[0003] High-voltage disconnector equipment is one of the main equipment in substations and plays an important role in the operation of power grids. However, outdoor high-voltage disconnectors are operated outdoors for a long time and are affected by atmospheric erosion. In addition, there are certain defects in their own technology, process, assembly, and materials. Many outdoor disconnectors are prone to overheating of the conductive circuit, rust, fracture of the insulating porcelain column, and transmission jamming and failure.

[0004] GIS switchgear with fully enclosed structure is widely used in substations. Compared with open power transmission and transformation equipment, GIS equipment has a series of advantages such as small footprint and space, fast installation, less interference from the external environment, safe and reliable operation, environmental protection, less maintenance workload, and long maintenance cycle. With the increase of equipment service time and the continuous commissioning of new GIS equipment, the number of GIS equipment failures has gradually increased. Tracking and testing the insulation performance of GIS equipment in operation is an important means of conducting GIS status maintenance risk assessment.

[0005] Due to the popularity of smart grids, various working conditions of distribution network equipment are acquired and analyzed through a variety of combined sensors, and potential risks of equipment are diagnosed. Based on the equipment defect portrait level and combined with the current equipment repair and maintenance plan, with the assistance of expert systems, specific and effective solutions are given for different equipment defect problems, making it possible to ensure safer and more stable operation of the equipment.

[0006] However, the data in the intelligent distribution system is huge in scale and complex in type. The sharp increase in intelligent distribution network equipment and the amount of data generated by equipment measurement points have brought opportunities for power equipment defect prediction, power capacity scheduling, etc., but also created challenges. Summary of the invention

[0007] In order to solve the above technical problems, the present invention proposes a switch device defect detection method, system and computer-readable storage medium based on multi-target recognition.

[0008] In a first aspect of the present invention, a switch device defect detection method based on multi-target recognition is proposed. The method can be executed by an electronic device, and the method comprises the following steps:

[0009] S1: Acquire the three most recently reported abnormal signal sequences of the switch device, wherein the three most recently reported abnormal signal sequences are marked as the first abnormal signal sequence S1, the second abnormal signal sequence S2 and the third abnormal signal sequence S3 in order of time;

[0010] S2: Calculate the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0011] S3: If T21≤St, T32≤St and T32≤T21, proceed to step S7; otherwise, proceed to step S4;

[0012] S4: Calculate the union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3, and proceed to step S5; St is the preset minimum reporting time interval of the switch signal;

[0013] S5: searching the preset database for the historical underreporting probability of each element in the union sequence S123, arranging the elements in the union sequence S123 in descending order according to the historical underreporting probability, and proceeding to step S6;

[0014] S6: Determine at least two elements from the union sequence S123 as target identification elements, and proceed to step S9;

[0015] S7: Calculate an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0016] S8: Determine at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, and proceed to step S9;

[0017] S9: Determine the defect type of the switch device based on the target identification element.

[0018] The signal sequence reported by the switch device includes at least two of cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, switch information value and appearance image feature value, and the appearance image feature value includes edge contour feature value.

[0019] The cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value are all preset with their corresponding standard reference values;

[0020] The switch device reports the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value according to the preset minimum reporting time interval St;

[0021] Alternatively, when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switching device reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value.

[0022] The step S8 determines at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, specifically including:

[0023] If both the intersection sequence S12 and the intersection sequence S23 are not empty, at least one target identification element is determined from the intersection sequence S12, and at least one target identification element is determined from the intersection sequence S23 as the target identification element.

[0024] The step S8 determines at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, specifically including:

[0025] If at least one of the intersection sequence S12 and the intersection sequence S23 is not empty, at least two elements are determined from the intersection sequence that is not empty as target identification elements.

[0026] The defect types of the switchgear determined in step S9 include: porcelain bottle breakage, operation failure, conductive circuit overheating, exterior rust, and component deformation and displacement.

[0027] In a second aspect of the present invention, a switchgear defect detection system based on multi-target recognition is proposed, the detection system comprising a switchgear state signal acquisition unit, a switchgear state signal reporting unit, a time interval determination unit, a signal sequence intersection calculation unit, a signal sequence union calculation unit, a target element determination unit, and a defect result output unit;

[0028] The switchgear status signal acquisition unit is used to acquire a switchgear signal sequence according to a preset period, wherein the switchgear signal sequence includes a cable temperature value, a gas leakage value, a power loss value, a partial discharge signal value, a displacement signal value, and a switch information value;

[0029] The switchgear status signal reporting unit reports the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value according to a preset minimum reporting time interval St; or, when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switchgear reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value;

[0030] The time interval judgment unit obtains the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3 recently reported by the switch device, and calculates the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0031] The signal sequence union calculation unit calculates a union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3;

[0032] The signal sequence intersection calculation unit calculates an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0033] A target element determination unit determines a target identification element from the union sequence or the intersection sequence;

[0034] The defect result output unit outputs a defect type of the switch device based on the target identification element.

[0035] The defect type of the switch device output by the defect result output unit includes one of the following results or any combination thereof: porcelain bottle breakage, operation failure, conductive circuit overheating, appearance corrosion, and component deformation and displacement.

[0036] After obtaining the difference T21 and the difference T32, the time interval determination unit performs the following determination:

[0037] If T21≤St, T32≤St and T32≤T21, then activating the signal sequence intersection calculation unit;

[0038] Otherwise, the signal sequence union calculation unit is activated.

[0039] The switch equipment includes a distribution network switch equipment, and the distribution network switch equipment includes an outdoor isolating switch equipment and a GIS switch equipment with a fully enclosed structure.

[0040] In a third aspect of the present invention, a computer-readable storage medium is provided for storing computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the switch device defect detection method based on multi-target recognition as described in the first aspect.

[0041] In a fourth aspect of the present invention, a computer device is further proposed. When a computer instruction is executed on the computer device, the electronic device executes the aforementioned switch device defect detection method based on multi-target recognition.

[0042] Specifically, the computer device can be various electronic devices, such as a portable terminal device including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the switch device defect detection method based on multi-target recognition described in the first aspect are implemented.

[0043] The present invention simultaneously considers the intersection and union of multiple abnormal signal sequence sets to determine multi-target identification elements, thereby performing defect detection of switch equipment, while considering the historical underreporting probability, thereby improving the timeliness, accuracy and completeness of switch equipment defect detection.

[0044] Further advantages of the present invention will be further reflected in detail in the specific embodiments section in conjunction with the drawings of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. 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 paying creative work.

[0046] Figure 1 This is a main flow chart of a switchgear defect detection method based on multi-target recognition according to an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of the abnormal signal sequence reported by the switch device at three consecutive reporting moments.

[0048] Figure 3 This is a schematic diagram of the functional unit composition of a switchgear defect detection system based on multi-target recognition according to an embodiment of the present invention.

[0049] Figure 4 Schematic diagram of the principle correspondence between the switchgear signal sequence and the potential defect type DETAILED DESCRIPTION

[0050] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments that are the same as the present application. Instead, they are only examples of devices and methods that are the same as some aspects of the present application as detailed in the attached claims.

[0051] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0052] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0053] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0054] Figure 1 It is a main flow chart of a switchgear defect detection method based on multi-target recognition according to an embodiment of the present invention.

[0055] Figure 1 The execution subject of the method may be a physical machine, a virtual machine, or a cluster combination of physical machines and / or virtual machines.

[0056] Preferably, Figure 1 The method is jointly executed by a local remote terminal unit of the switch device and a remote data processing platform.

[0057] The method includes steps S1-S9, and each step is specifically performed as follows: Figure 1 Step numbers S1-S9 omitted):

[0058] S1: Acquire the three most recently reported abnormal signal sequences of the switch device, wherein the three most recently reported abnormal signal sequences are marked as the first abnormal signal sequence S1, the second abnormal signal sequence S2 and the third abnormal signal sequence S3 in order of time;

[0059] S2: Calculate the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0060] S3: If T21≤St, T32≤St and T32≤T21, proceed to step S7; otherwise, proceed to step S4;

[0061] S4: Calculate the union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3, and proceed to step S5; St is the preset minimum reporting time interval of the switch signal;

[0062] S5: searching the preset database for the historical underreporting probability of each element in the union sequence S123, arranging the elements in the union sequence S123 in descending order according to the historical underreporting probability, and proceeding to step S6;

[0063] S6: Determine at least two elements from the union sequence S123 as target identification elements, and proceed to step S9;

[0064] S7: Calculate an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0065] S8: Determine at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, and proceed to step S9;

[0066] S9: Determine the defect type of the switch device based on the target identification element.

[0067] Next, combine Figure 2 , further Figure 1 Each step of the embodiment is described in detail.

[0068] The signal sequence reported by the switch device includes at least two of cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, switch information value and appearance image feature value, and the appearance image feature value includes edge contour feature value.

[0069] In the smart grid system, a variety of combined sensors can be configured locally on the switchgear to obtain a variety of corresponding status parameters, including cable temperature values, gas leakage values, power loss values, partial discharge signal values, displacement signal values, switch information values, and appearance image feature values.

[0070] Preferably, the multiple combination sensors may be remote terminal units (RTUs).

[0071] Remote terminal unit (RTU) is an electronic device installed at the local site, used to monitor and measure switchgear installed at the local site ("remote" here refers to the data processing platform). RTU converts the measured status or signal into a data format that can be sent on the communication media. It also converts the data sent from the central computer (the aforementioned remote data processing platform) into commands to achieve functional control of the equipment.

[0072] RTU (Remote Terminal Unit) is a remote measurement and control unit device responsible for monitoring and controlling field signals and industrial equipment. Compared with the commonly used programmable controller PLC, RTU usually has excellent communication capabilities and larger storage capacity, is suitable for harsher temperature and humidity environments, and provides more computing functions. It is precisely because of the perfect functions of RTU that RTU products have been widely used in SCADA systems.

[0073] SCADA systems can be designed to meet various applications (water, electricity, gas, alarm, communication, security, etc.) and meet the design specifications and operating concepts required by customers. SCADA systems can be as simple as a switch connected to the remote end through a pair of wires, or as complex as a computer network consisting of many wireless remote terminal units (RTUs) and communicating with powerful microcomputers installed in the central control room. The remote terminal units of SCADA systems can be implemented with a variety of different hardware and software. This depends on the nature of the controlled site, the complexity of the system, the requirements for data communication, real-time alarm reporting, analog signal measurement accuracy, status monitoring, equipment regulation control and switch control.

[0074] Under normal circumstances, the switch device obtains its own multiple status signals through the RTU according to the first preset period and then reports them.

[0075] For example, if the minimum detection period of the RTU is configured to be 60 seconds, the RTU will detect the status signal of the switch device every 60 seconds. If everything is normal, the RTU will continue to wait for the next detection until the reporting time point is reached. Assuming that the pre-configured minimum reporting interval is 120 seconds, the switch device will report the status signal only when the interval from the last reporting is 120 seconds.

[0076] For example, assuming that the last reporting time node is the 0th second, at the 10th second, the RTU detects the status signal sequence 1 of the switch device once and finds that everything is normal, but the minimum reporting time interval has not been reached. Therefore, until the 70th second, the RTU detects the status signal sequence 2 of the switch device again and finds that everything is normal. At this time, the minimum reporting time interval has not been reached; until the 120th second, the switch device reports the most recent status signal sequence 2; then at 130 seconds, the RTU detects the status signal sequence 3 of the switch device again, and at 190 seconds, the RTU detects the status signal sequence 4 of the switch device again, until the 240th second, the switch device reports the most recent status signal sequence 4, and so on.

[0077] However, if the RTU detects an abnormal signal, it means that the current switchgear may be in an abnormal state. If the abnormal signal is still reported after the minimum reporting time interval, it may cause a delay in defect detection. At this time, it is easy to think of reporting the abnormal signal immediately;

[0078] However, since the status signals of the switchgear are diverse and the RTU sensor may have a single abnormality, if only a certain signal abnormality is reported directly, it may lead to frequent data reporting and frequent "false" false positive signals.

[0079] To this end, this application improves this as follows:

[0080] The cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value are all preset with their corresponding standard reference values;

[0081] The switch device reports the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value according to the preset minimum reporting time interval St;

[0082] Alternatively, when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switching device reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value.

[0083] Corresponding to the above embodiment, the specific implementation improvements are as follows:

[0084] Assuming that the last reporting time node is the 0th second, at the 10th second, the RTU detects the status signal sequence 1 of the switch device and finds that everything is normal, but the minimum reporting time interval has not been reached. Therefore, until the 70th second, the RTU detects the status signal sequence 2 of the switch device again and finds that at least two status signals are abnormal. At this time, the switch device immediately reports the status signal sequence 2, that is, it does not need to wait until the 120th second, but reports directly at the 70th second when the abnormality is found to avoid delay.

[0085] However, the results of only considering a single anomaly are not accurate enough, especially when the data in the smart distribution system is huge in scale and complex in type, and the number of smart distribution network devices increases dramatically.

[0086] To this end, step S1 of the technical solution of this application first clarifies:

[0087] S1: Acquire the three most recently reported abnormal signal sequences of the switch device, wherein the three most recently reported abnormal signal sequences are marked as the first abnormal signal sequence S1, the second abnormal signal sequence S2 and the third abnormal signal sequence S3 in order of time;

[0088] like Figure 2 As shown:

[0089] S1 = {cable temperature value, gas leakage value, power loss value, partial discharge signal value};

[0090] S2 = {cable temperature value, gas leakage value, power loss value, displacement signal value};

[0091] S3 = {cable temperature value, gas leakage value, displacement signal value, switch information value};

[0092] Next, step S2: calculating the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0093] Step S3: If T21≤St, T32≤St and T32≤T21, proceed to step S7; otherwise, proceed to step S4; St is the preset minimum reporting time interval of the switch signal;

[0094] In step S3, if T21≤St, T32≤St and T32≤T21, it means that the RTU of the switchgear detects multiple abnormal signals within the predetermined minimum reporting time interval (in advance within the minimum reporting time interval), and the frequency of the detected abnormal signals is getting higher and higher. At this time, it is necessary to immediately perform defect detection warning, that is, enter step S7:

[0095] S7: Calculate an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0096] Referring to the above example, intersection sequence S12 = {cable temperature value, gas leakage value, power loss value}; intersection sequence S23 = {cable temperature value, gas leakage value};

[0097] S8: Determine at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, and proceed to step S9.

[0098] The step S8 determines at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, specifically including:

[0099] If both the intersection sequence S12 and the intersection sequence S23 are not empty, at least one target identification element is determined from the intersection sequence S12, and at least one target identification element is determined from the intersection sequence S23 as the target identification element.

[0100] Preferably, intersection sequence S12 = {cable temperature value, gas leakage value, power loss value}; intersection sequence S23 = {cable temperature value, gas leakage value}, then the target identification element can be {cable temperature value, gas leakage value} or {cable temperature value, gas leakage value, power loss value} or {gas leakage value, power loss value}.

[0101] Of course, in step S3, if the condition of "T21≤St, T32≤St and T32≤T21" is not satisfied, then the process proceeds to step S3 to calculate the union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2 and the third abnormal signal sequence S3, and then proceeds to step S5;

[0102] S5: searching the preset database for the historical underreporting probability of each element in the union sequence S123, arranging the elements in the union sequence S123 in descending order according to the historical underreporting probability, and proceeding to step S6;

[0103] S6: Determine at least two elements from the union sequence S123 as target identification elements, and proceed to step S9.

[0104] Referring to the above example, the union sequence S123 = {cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, knife switch information value}. Assuming that the union sequence after being arranged in descending order according to the historical underreporting probability is {gas leakage value, knife switch information value, power loss value, partial discharge signal value, displacement signal value, cable temperature value}, then {gas leakage value, knife switch information value, power loss value} can be selected as the target identification element and proceed to step S9.

[0105] S9: Determine the defect type of the switch device based on the target identification element.

[0106] The defect types of the switchgear determined in step S9 include: porcelain bottle breakage, operation failure, conductive circuit overheating, exterior rust, and component deformation and displacement.

[0107] Specifically, according to the correspondence between the target identification element and the defect type, the defect type of the switch device can be determined.

[0108] The specific principles include:

[0109] Gas leakage value: detect the concentration of sulfur hexafluoride gas. Sulfur hexafluoride gas is an important monitoring indicator of the ring network cabinet. Especially when the temperature rises, sulfur hexafluoride will accelerate the decomposition. The sulfur hexafluoride gas monitoring system can display the leakage value through the indicator chip on the single-chip microcomputer. When it exceeds the warning value, an audible and visual alarm will be issued;

[0110] Cable temperature value: The temperature index of cable electrical equipment is another very important index of the power grid. The temperature information is obtained through wireless equipment technology and sent to the designated temperature analysis and monitoring equipment for expert judgment; then, the received cable temperature information is judged by the acquired computer and other advanced equipment. If it exceeds the preset value, an alarm will be issued and data prediction analysis will be formed. In addition, the working current of the high-voltage disconnector can generally only reach about 60% of its rated current. Once the current exceeds 70%, the conductive circuit may overheat.

[0111] Power loss value, partial discharge signal value: Monitor the power loss of the local power consumption through the feeder terminal unit (FTU), and obtain the data of the power modem for different time spans to conduct effective analysis to obtain power loss analysis; cable partial discharge failure can cause the insulation layer to be broken down.

[0112] Displacement signal value: If the same device sends a displacement signal exceeding the threshold within one day, it is considered a frequent signal; if there are many devices with frequent signals, the components may be deformed or shifted.

[0113] Knife switch information value: The knife switch information in the signal data is parsed through string segmentation, regular matching, etc. If the knife switch information is incomplete, it may indicate that the operation is malfunctioning.

[0114] Based on the above method embodiment, Figure 3 A schematic diagram of the functional unit composition of a switchgear defect detection system based on multi-target recognition according to an embodiment of the present invention is given.

[0115] Figure 3 A switch device defect detection system based on multi-target recognition is shown, and the detection system includes a switch device status signal acquisition unit, a switch device status signal reporting unit, a time interval judgment unit, a signal sequence intersection calculation unit, a signal sequence union calculation unit, a target element determination unit and a defect result output unit.

[0116] The switchgear status signal acquisition unit is used to acquire a switchgear signal sequence according to a preset period, wherein the switchgear signal sequence includes a cable temperature value, a gas leakage value, a power loss value, a partial discharge signal value, a displacement signal value, and a switch information value;

[0117] The switchgear status signal reporting unit reports the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value according to a preset minimum reporting time interval St; or, when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switchgear reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value;

[0118] The time interval judgment unit obtains the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3 recently reported by the switch device, and calculates the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0119] The signal sequence union calculation unit calculates a union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3;

[0120] The signal sequence intersection calculation unit calculates an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3;

[0121] A target element determination unit determines a target identification element from the union sequence or the intersection sequence;

[0122] The defect result output unit outputs a defect type of the switch device based on the target identification element.

[0123] The defect type of the switch device output by the defect result output unit includes one of the following results or any combination thereof: porcelain bottle breakage, operation failure, conductive circuit overheating, appearance corrosion, and component deformation and displacement.

[0124] After obtaining the difference T21 and the difference T32, the time interval determination unit performs the following determination:

[0125] If T21≤St, T32≤St and T32≤T21, then activating the signal sequence intersection calculation unit;

[0126] Otherwise, the signal sequence union calculation unit is activated.

[0127] The switch equipment includes a distribution network switch equipment, and the distribution network switch equipment includes an outdoor isolating switch equipment and a GIS switch equipment with a fully enclosed structure.

[0128] In the above embodiment, each time the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value, or when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switch device reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value, that is, the signal sequence reported each time must include the appearance image feature value.

[0129] This is because porcelain bottle breakage failures occur more frequently on 220KV disconnector equipment, and the detection and repair of porcelain bottle breakage failures are very important for the safe operation of the power system.

[0130] The most significant feature of a broken porcelain bottle fault is the straight line fracture of the edge in the appearance feature. Considering this, the appearance image feature value includes the edge contour feature value, and each time it is reported as a signal sequence, it can first predict whether it is a broken porcelain bottle fault.

[0131] Figure 4 It is a schematic diagram of the principle correspondence between the switchgear signal sequence and the potential defect type. Figure 4 The correspondence between the given target identification elements and the defect type can determine the defect type of the switchgear.

[0132] The specific principles include:

[0133] Gas leakage value: detect the concentration of sulfur hexafluoride gas. Sulfur hexafluoride gas is an important monitoring indicator of the ring network cabinet. Especially when the temperature rises, sulfur hexafluoride will accelerate the decomposition. The sulfur hexafluoride gas monitoring system can display the leakage value through the indicator chip on the single-chip microcomputer. When it exceeds the warning value, an audible and visual alarm will be issued;

[0134] Cable temperature value: The temperature index of cable electrical equipment is another very important index of the power grid. The temperature information is obtained through wireless equipment technology and sent to the designated temperature analysis and monitoring equipment for expert judgment; then, the received cable temperature information is judged by the acquired computer and other advanced equipment. If it exceeds the preset value, an alarm will be issued and data prediction analysis will be formed. In addition, the working current of the high-voltage disconnector can generally only reach about 60% of its rated current. Once the current exceeds 70%, the conductive circuit may overheat.

[0135] Power loss value, partial discharge signal value: Monitor the power loss of the local power consumption through the feeder terminal unit (FTU), and obtain the data of the power modem for different time spans to conduct effective analysis to obtain power loss analysis; cable partial discharge failure can cause the insulation layer to be broken down.

[0136] Displacement signal value: If the same device sends a displacement signal exceeding the threshold within one day, it is considered a frequent signal; if there are many devices with frequent signals, the components may be deformed or shifted.

[0137] Knife switch information value: The knife switch information in the signal data is parsed through string segmentation, regular matching, etc. If the knife switch information is incomplete, it may indicate that the operation is malfunctioning.

[0138] The present invention simultaneously considers the intersection and union of multiple abnormal signal sequence sets to determine multi-target identification elements, thereby performing defect detection of switch equipment, while considering the historical underreporting probability, thereby improving the timeliness, accuracy and completeness of switch equipment defect detection.

[0139] The present invention provides multiple embodiments, each of which can constitute an independent technical solution and may contribute to the prior art and solve corresponding technical problems. However, it should be pointed out that different embodiments can be combined with each other without violating logic; at the same time, each embodiment can solve at least one technical problem, but it is not required that each individual embodiment solves multiple or all technical problems.

[0140] It should also be noted that in the specific implementation of this application, if user-related data is involved, when the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0141] The foregoing has shown and described the method embodiments and system of the present invention, but it is understandable to those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

[0142] In the above description, after the embodiment in the form of method flow is elaborated in detail, the corresponding embodiment in the form of system function architecture does not need to be repeated again, but those skilled in the art can directly and unambiguously determine that the embodiment in the form of system function architecture also corresponds to all the contents of the embodiment in the form of method flow. The same is true for the embodiment in the form of computer program code pseudo language.

Claims

1. A switchgear defect detection method based on multi-target recognition, characterized in that: The method comprises the following steps: S1: Acquire the three most recently reported abnormal signal sequences of the switch device, wherein the three most recently reported abnormal signal sequences are marked as the first abnormal signal sequence S1, the second abnormal signal sequence S2 and the third abnormal signal sequence S3 in order of time; S2: Calculate the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3; S3: If T21≤St, T32≤St and T32≤T21, proceed to step S7; otherwise, proceed to step S4; S4: Calculate the union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3, and proceed to step S5; St is the preset minimum reporting time interval of the abnormal signal of the switch device; S5: Obtain the historical underreporting probability of each element in the union sequence S123 from the preset database, arrange the elements in the union sequence S123 in descending order according to the historical underreporting probability, and proceed to step S6; S6: Determine at least two elements from the union sequence S123 as target identification elements, and proceed to step S9; S7: Calculate an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3; S8: Determine at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements; S9: Determine the defect type of the switch device based on the target identification element.

2. A switchgear defect detection method based on multi-target recognition according to claim 1, characterized in that: The signal sequence reported by the switch device includes at least two of cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, switch information value and appearance image feature value, and the appearance image feature value includes edge contour feature value.

3. A switchgear defect detection method based on multi-target recognition as claimed in claim 2, characterized in that: The cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value are all preset with their corresponding standard reference values; The switch device reports the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value according to the preset minimum reporting time interval St; Alternatively, when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switching device reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value.

4. A switchgear defect detection method based on multi-target recognition according to claim 1, characterized in that: The step S8 determines at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, specifically including: If both the intersection sequence S12 and the intersection sequence S23 are not empty, at least one target identification element is determined from the intersection sequence S12, and at least one target identification element is determined from the intersection sequence S23 as the target identification element.

5. The switchgear defect detection method based on multi-target recognition according to claim 1, characterized in that: The step S8 determines at least two elements from the intersection sequence S12 and the intersection sequence S23 as target recognition elements, specifically including: If at least one of the intersection sequence S12 and the intersection sequence S23 is not empty, at least two elements are determined from the intersection sequence that is not empty as target identification elements.

6. A switchgear defect detection method based on multi-target recognition according to claim 1, characterized in that: The defect types of the switchgear determined in step S9 include: porcelain bottle breakage, operation failure, conductive circuit overheating, exterior rust, and component deformation and displacement.

7. A switchgear defect detection system based on multi-target recognition, the detection system comprising a switchgear state signal acquisition unit, a switchgear state signal reporting unit, a time interval determination unit, a signal sequence intersection calculation unit, a signal sequence union calculation unit, a target element determination unit and a defect result output unit, Features: The switchgear status signal acquisition unit is used to acquire a switchgear signal sequence according to a preset period, wherein the switchgear signal sequence includes a cable temperature value, a gas leakage value, a power loss value, a partial discharge signal value, a displacement signal value, and a switch information value; The switchgear status signal reporting unit reports the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value according to a preset minimum reporting time interval St; or, when at least two of the cable temperature value, gas leakage value, power loss value, partial discharge signal value, displacement signal value, and switch information value exceed their respective corresponding standard reference values, the switchgear reports in real time at least two signal sequences that exceed their respective corresponding standard reference values, and simultaneously obtains and reports the appearance image feature value; The time interval judgment unit obtains the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3 recently reported by the switch device, and calculates the difference T21 between the reporting times t1 and t2 corresponding to the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and the difference T32 between the reporting times t2 and t3 corresponding to the second abnormal signal sequence S2 and the third abnormal signal sequence S3; The signal sequence union calculation unit calculates a union sequence S123 of the first abnormal signal sequence S1, the second abnormal signal sequence S2, and the third abnormal signal sequence S3; The signal sequence intersection calculation unit calculates an intersection sequence S12 of the first abnormal signal sequence S1 and the second abnormal signal sequence S2, and an intersection sequence S23 of the second abnormal signal sequence S2 and the third abnormal signal sequence S3; A target element determination unit determines a target identification element from the union sequence or the intersection sequence; A defect result output unit outputs a defect type of the switch device based on the target identification element; After obtaining the difference T21 and the difference T32, the time interval determination unit performs the following determination: If T21≤St, T32≤St and T32≤T21, then activating the signal sequence intersection calculation unit; Otherwise, the signal sequence union calculation unit is activated.

8. A switchgear defect detection system based on multi-target recognition according to claim 7, characterized in that: The defect type of the switch device output by the defect result output unit includes one of the following results or any combination thereof: porcelain bottle breakage, operation failure, conductive circuit overheating, appearance corrosion, and component deformation and displacement.

9. A switchgear defect detection system based on multi-target recognition according to claim 7, characterized in that: The switch equipment includes a distribution network switch equipment, and the distribution network switch equipment includes an outdoor isolating switch equipment and a GIS switch equipment with a fully enclosed structure.

Images