CN117872124A - Off-line testing system and method for circuit breaker based on neural network - Google Patents
Off-line testing system and method for circuit breaker based on neural network Download PDFInfo
- Publication number
- CN117872124A CN117872124A CN202410275589.XA CN202410275589A CN117872124A CN 117872124 A CN117872124 A CN 117872124A CN 202410275589 A CN202410275589 A CN 202410275589A CN 117872124 A CN117872124 A CN 117872124A
- Authority
- CN
- China
- Prior art keywords
- state
- circuit breaker
- sample
- test data
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 259
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000003062 neural network model Methods 0.000 claims abstract description 38
- 238000001514 detection method Methods 0.000 claims abstract description 36
- 238000004146 energy storage Methods 0.000 claims description 12
- 239000003990 capacitor Substances 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 14
- 230000035882 stress Effects 0.000 description 8
- 238000009826 distribution Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000011056 performance test Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000004913 activation Effects 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000000243 solution Substances 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 3
- 238000005260 corrosion Methods 0.000 description 3
- 238000009434 installation Methods 0.000 description 3
- 238000009413 insulation Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000013021 overheating Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000010304 firing Methods 0.000 description 1
- 230000009975 flexible effect Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000003892 spreading Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Landscapes
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
One or more embodiments of the present disclosure relate to the field of power electronics, and in particular, to a system and a method for offline testing of a circuit breaker based on a neural network. The off-line testing subsystem is used for conducting off-line short-circuit testing on the sample circuit breaker and the circuit breaker to be tested based on the off-line testing device, wherein the sample circuit breaker is a circuit breaker with a preset state; the data subsystem establishes data connection with the offline test subsystem and detects test data of offline short circuit tests of the sample circuit breaker and the circuit breaker to be tested; the model subsystem reads test data corresponding to the sample circuit breaker, marks the test data according to a preset state of the sample circuit breaker, obtains sample test data, and establishes and trains a state detection neural network model based on the sample test data; and the result subsystem reads the test data corresponding to the circuit breaker to be tested, and obtains the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
Description
Technical Field
One or more embodiments of the present disclosure relate to the field of power electronics, and in particular, to a system and a method for offline testing of a circuit breaker based on a neural network.
Background
A large number of circuit breakers, fuses and other switching devices are installed in transmission and distribution networks. The equipment has the function of timely opening the circuit when the circuit is in fault, short circuit or overload condition so as to protect devices and equipment in the circuit and simultaneously protect a power grid and avoid fault spreading. After installation of these devices, preventive maintenance plans should be formulated accordingly to reduce the risk associated with device failure. Thus, performing off-line testing of circuit breakers is critical to ensuring personnel safety and reducing the risk of equipment failure. The off-line test of the performance of the circuit breaker refers to testing the performance and the breaking function of the circuit breaker to obtain the working performance index of the circuit breaker. The goal of the circuit breaker test is to obtain a trip curve or current time curve (Time current curve, TCC) to determine the life of the circuit breaker in normal operation. The performance evaluation of the circuit breaker is beneficial to minimizing the loss of power failure, downtime and business income, and guaranteeing the safe and reliable operation of the power distribution system.
Offline testing of circuit breakers typically involves two things, namely: (1) Primary test, injecting large current through the breaker to test the current path and function of the breaker; (2) And (3) a secondary injection test, wherein a low current is injected through a tripping mechanism to verify tripping characteristics in the case of overload or fault. In practice, breaker tests often contain more information, given by the breaker test report. The breaker test report gives a table that indicates pass or fail status and test results. If the circuit breaker fails, it will be necessary to provide cause analysis and corrective action suggestions, including: driving and operating the circuit breaker; verifying whether the installation is correct; verifying whether the settings are valid and performing necessary adjustments; whether the circuit breaker is qualified, etc.
It can be seen that the existing circuit breaker test can only give a simple decision as to whether the circuit breaker can operate and the service life (the number of times of breaking), cannot go deep into specific performance indexes, and cannot give effective corrective action suggestions to improve the circuit breaker performance. With industry development, there is a greater need to obtain a comprehensive test of the performance of circuit breakers, requiring accurate performance assessment and advice to assist in the development of the circuit breaker, meeting the production and application requirements.
Disclosure of Invention
One or more embodiments of the present disclosure describe a system and a method for offline testing of a circuit breaker based on a neural network, which can quickly and accurately obtain various states of the circuit breaker by means of a neural network model, and complete the testing of the circuit breaker.
In a first aspect, embodiments of the present disclosure provide a neural network-based offline testing system for a circuit breaker, including:
the off-line testing subsystem is used for conducting off-line short-circuit testing on the sample circuit breaker and the circuit breaker to be tested based on the off-line testing device, wherein the sample circuit breaker is a circuit breaker with a preset state;
the data subsystem establishes data connection with the offline test subsystem and detects test data of offline short circuit tests of the sample circuit breaker and the circuit breaker to be tested;
the model subsystem is used for reading test data corresponding to the sample circuit breaker, marking the test data according to a preset state of the sample circuit breaker, obtaining sample test data, wherein the sample test data comprises single-state sample data and multi-state sample data, the multi-state sample data comprises multi-state test data and associated multi-states, the multi-state test data is obtained by combining each corresponding state with distinguishing test data obtained by comparing reference normal test data, and a state detection neural network model is established and trained based on the sample test data;
and the result subsystem reads the test data corresponding to the circuit breaker to be tested, and obtains the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
Preferably, the off-line testing device includes:
the super capacitor is used for providing power;
the protection switch is connected with the super capacitor and used for protecting a circuit connected with the super capacitor;
the solid-state switch is connected with the protection switch and used for controlling the start and stop of off-line test;
and the acquisition module is used for acquiring test data of the off-line short circuit test.
Preferably, the preset state includes one or more of a preset control system operating condition state, a preset circuit system operating condition state and a preset mechanical system operating condition state.
Preferably, the working condition state of the preset control system is a control normal working condition state, a driving failure state, a signal abnormal state, a feedback failure state or a power undersize state.
Preferably, the preset working condition state of the circuit system is a normal state of the circuit, a short circuit state of the circuit node or an open circuit state of the circuit node.
Preferably, the preset mechanical system working condition state is a mechanical normal working condition state, a cam offset state, a large clearance error state, an energy storage spring failure state, a lock catch failure state, a dead point state or a clamping state.
Preferably, the test data includes at least one of a short-circuit on current, a short-time withstand current, a limit short-circuit breaking current, and an operating short-circuit breaking current.
Preferably, the state detection neural network model is a fully connected neural network, the state detection neural network model comprises an input layer, one or more hidden layers and an output layer, the input of the state detection neural network model is the test data, and the output is the state of the circuit breaker.
In a second aspect, embodiments of the present disclosure provide a method for offline testing of a circuit breaker based on a neural network, including the steps of:
performing off-line short-circuit test on a sample circuit breaker and a circuit breaker to be tested based on a pre-provided off-line test device, wherein the sample circuit breaker is a circuit breaker with a preset state;
detecting test data of the sample circuit breaker and the circuit breaker to be tested for offline short circuit test;
reading test data corresponding to a sample breaker, marking the test data according to a preset state of the sample breaker, and obtaining sample test data, wherein the sample test data comprises single-state sample data and multi-state sample data, the multi-state sample data comprises multi-state test data and associated multi-states, and the multi-state test data is obtained by comparing each corresponding state with differential test data obtained by referring to normal test data;
establishing and training a state detection neural network model based on the sample test data;
and reading the test data corresponding to the circuit breaker to be tested, and obtaining the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
Preferably, the test data comprises current waveform data,
marking the test data according to the preset state of the sample breaker, and the method for obtaining the sample test data comprises the following steps:
taking the test data marked by the preset state as single-state sample data;
randomly selecting a plurality of single-state sample data from the sample data, wherein the preset states of the single-state sample data marks are different;
comparing the current waveform data of the plurality of single-state sample data with preset normal waveform data respectively to obtain differential current waveform data;
synthesizing multi-state current waveform data according to the differential current waveform data, and obtaining multi-state test data based on the multi-state current waveform data;
combining the preset states of the single-state sample data to obtain a combined state;
and marking the multi-state test data by using the combined state to obtain multi-state sample data.
The technical scheme provided by some embodiments of the present specification has the following beneficial effects:
in one or more embodiments of the present disclosure, the offline testing system for a circuit breaker based on a neural network provided herein adopts a neural network model, and according to test data obtained by testing, multiple states of the circuit breaker can be obtained directly through the neural network model, so that the working performance and states of the circuit breaker to be tested can be rapidly mastered, and the efficiency and reliability of analysis after offline testing of the circuit breaker can be improved. The off-line testing system of the breaker based on the neural network provided by the specification can still obtain more accurate test result judgment by means of the neural network model under the condition of various complicated breaker states, and avoids errors and misjudgment of the test result of manual analysis and judgment. The state information of the circuit breaker can be directly obtained by means of the neural network model according to the test data, and the reason of the fault of the circuit breaker can be directly given by the state information, so that corrective measures and suggestions can be conveniently obtained.
Other features and advantages of one or more embodiments of the present disclosure will be further disclosed in the following detailed description, the accompanying drawings.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present description, the drawings that are required in the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present description, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of an offline testing system of a circuit breaker according to an embodiment of the present disclosure.
Fig. 2 is a schematic diagram of a circuit breaker.
Fig. 3 is a schematic diagram of the internal structure of the circuit breaker.
Fig. 4 is a schematic diagram of an offline testing device according to an embodiment of the present disclosure.
Fig. 5 is a schematic diagram of a neural network model according to an embodiment of the present disclosure.
Fig. 6 is a flowchart of a method for offline testing of a circuit breaker according to an embodiment of the present disclosure.
Fig. 7 is a flowchart of a method for obtaining sample test data according to an embodiment of the present disclosure.
FIG. 8 is a graph showing stress test and sample current data provided in the examples of the present disclosure.
Fig. 9 is a schematic diagram of fault identification current data provided in an embodiment of the present disclosure.
Detailed Description
The technical solutions of the embodiments of the present specification are explained and illustrated below with reference to the drawings of the embodiments of the present specification, but the following embodiments are only preferred embodiments of the present specification, and not all the embodiments. Based on the examples in the implementation manner, those skilled in the art may obtain other examples without making any creative effort, which fall within the protection scope of the present specification.
The terms first, second, third and the like in the description and in the claims and in the above drawings are used for distinguishing between different objects and not necessarily for describing a particular sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
In the following description, directional or positional relationships such as the terms "inner", "outer", "upper", "lower", "left", "right", etc., are presented merely to facilitate describing the embodiments and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operate in a particular orientation, and therefore should not be construed as limiting the description.
The data related to the application are information and data authorized by a user or fully authorized by all parties, and the collection of the related data complies with related laws and regulations and standards of related countries and regions.
Application scenario introduction
The one or more off-line breaker testing systems based on the neural network are applied to detection of the breaker and obtain a state detection result of the breaker. Circuit breakers play a vital role in the power grid system and are one of the core devices for the operation and protection of the power system. The main functions of the power system include normal on-off control, overload protection, short-circuit protection, ground fault protection and the like of the circuit, and the safe and stable operation of the power system is ensured. Referring to fig. 1, a sample breaker and a breaker to be tested are connected to an offline testing subsystem 100 disclosed in the present specification, and test data is obtained after an offline short-circuit test is performed by the offline testing subsystem 100. The training of the neural network model is completed by the data subsystem 200 and the model subsystem 300 according to the test data corresponding to the sample breaker, and the trained neural network model is loaded into the result subsystem 400. And the result subsystem 400 reads in the test data corresponding to the circuit breaker to be tested and gives out the result to obtain the state of the circuit breaker to be tested.
Referring to fig. 2 and 3, the basic structure of the circuit breaker includes a housing 10, a three-phase stationary contact terminal 21, a three-phase moving contact terminal 22, an opening and closing indicator 11, a secondary circuit 12, a counter 13, a lock catch 14, a link mechanism 15, an energy storage motor 16, a spring 17, and the like. The most basic function of a circuit breaker is to perform the on and off operations of a circuit. In the daily operation of the power distribution network, the circuit breaker can accurately complete the switching task of the power line or the electrical equipment according to the scheduling instruction or the control signal of the automatic equipment, and the distribution and the adjustment of the power load are realized so as to meet the power supply requirements under various working conditions. The circuit breaker also has perfect protection function. When the current in the distribution network exceeds the set rated value, the circuit breaker can rapidly cut off the circuit, so that safety accidents such as heating and melting of wires and even firing caused by overload are prevented, and power equipment is protected from being damaged. Meanwhile, when a short-circuit fault occurs, the circuit breaker can break the circuit in a very short time (usually at the level of a few milliseconds), so that the rise of short-circuit current is effectively limited, the expansion of the fault is prevented, and the safety and the stability of a power grid are ensured. And simultaneously, the life and property safety of the user can be ensured. By means of the earth fault protection function of the circuit breaker, when a single-phase earth fault occurs in the power system, the circuit breaker timely cuts off a fault line by detecting zero-sequence current or zero-sequence voltage change, and equipment of a non-fault part is prevented from being damaged. And the continuous and reliable power supply of the power distribution network is maintained.
In addition, the modern intelligent circuit breaker can be matched with a relay protection device and an automatic control system to realize real-time monitoring and intelligent judgment of the running state of the power grid, quickly locate and isolate various complex faults, and improve the self-healing capacity and the running efficiency of a power system.
It follows that the circuit breaker is a "guard" in the grid system, which is both an important switching device for power transmission and distribution and a first line of defense for safe operation of the power system. Through accurate control and real-time monitoring of electrical parameters such as current, voltage, the circuit breaker has prevented the emergence of all kinds of electric power accidents effectively, has ensured the safe and stable operation of electric power system, has provided firm energy guarantee for social production and people's life. Therefore, the circuit breaker performance detection and the operational reliability evaluation have important significance. The strict and comprehensive detection of the circuit breaker is an indispensable link before delivery, after installation and during operation. The main detection items and purposes of the circuit breaker will be briefly described.
And (3) mechanical property detection:
the operation mechanism inspection comprises the inspection of an energy storage mechanism, a closing/opening coil and a connecting rod transmission system, so that flexible action without jamming is ensured, and the energy storage motor 16 or the spring 17 can provide enough energy to enable the circuit breaker to complete normal opening and closing operation.
And (3) testing contact pressure, namely measuring the contact pressure of the circuit breaker under different working states, and ensuring that enough contact pressure exists between the contacts to maintain good electric contact and prevent overheating.
The mechanical life test simulates the actual service environment of the circuit breaker, and the durability and the stability of the mechanical structure of the circuit breaker are verified through multiple cyclic operations (such as electric operation and manual operation).
And (3) electrical performance detection:
and (3) carrying out a load current test to confirm that the current carried by the circuit breaker does not exceed a rated value under a normal load condition, and monitoring whether the temperature rise meets the standard requirement.
And a short-circuit breaking capability test is used for verifying whether the circuit breaker can rapidly and reliably cut off large current when a short circuit occurs, and the test comprises the tests of parameters such as transient recovery voltage, breaking time, arc gap recovery and the like.
And (3) overload protection characteristic test, namely adjusting the setting value of the overload tripping device, and testing whether the overload tripping device can accurately trip under specified current.
Insulation performance tests, including power frequency withstand voltage tests, dielectric loss tangent (tan delta) tests, leakage current tests, and the like, ensure that the circuit breaker has a good insulation level.
And (3) detecting special environmental adaptability:
and the high-low temperature test is used for testing the working performance of the circuit breaker under the extreme temperature condition, so that the circuit breaker can stably operate in a wide environment temperature range.
And (5) a damp-heat test, and evaluating the corrosion resistance and electrical performance retention of the circuit breaker in a high-humidity environment.
And an electromagnetic compatibility (EMC) test is carried out to test the resistance of the circuit breaker to external electromagnetic interference and the electromagnetic disturbance degree generated by the circuit breaker.
Other important detection items:
vacuum degree detection, for a vacuum circuit breaker, is to detect the vacuum degree of an internal vacuum arc extinguishing chamber, and ensure good insulation and arc extinguishing capability.
And (3) an aging test, wherein the aging process is accelerated to predict the performance change of the circuit breaker after long-term use.
Corrosion protection detection, the corrosion resistance of circuit breaker materials, especially equipment exposed to outdoor or harsh environmental conditions.
The explosion-proof performance test is suitable for the circuit breaker used in a specific dangerous area, and the capability of the circuit breaker in an explosive gas environment, which cannot become an ignition source, needs to be verified.
On-site operation state inspection:
appearance inspection, checking whether the breaker body and accessories have obvious damage, rust or other abnormal phenomena.
And checking connection points, namely checking whether the connection part of the circuit breaker and the bus or outgoing cable has the problems of overheating, oxidization or loosening and the like.
And (5) checking the indication instrument and the signal to confirm whether the position indication, the fault signal and the telemetry data of the circuit breaker are accurate and effective.
Through the series of strict detection projects, the design and manufacturing quality, the electrical and mechanical properties and the actual running efficiency of the circuit breaker can be comprehensively evaluated, and the circuit breaker can safely, reliably and efficiently fulfill the functions of the circuit breaker in various complicated power system environments.
Specifically, the embodiment of the present specification first provides a circuit breaker offline testing system based on a neural network, including:
the off-line test subsystem 100 is used for performing off-line short-circuit test on a sample circuit breaker and a circuit breaker to be tested based on an off-line test device, wherein the sample circuit breaker is a circuit breaker with a preset state;
the data subsystem 200 establishes data connection with the offline testing subsystem 100, and detects test data of the sample circuit breaker and the circuit breaker to be tested for offline short circuit test;
the model subsystem 300 reads test data corresponding to the sample circuit breaker, marks the test data according to a preset state of the sample circuit breaker, obtains sample test data, wherein the sample test data comprises single-state sample data and multi-state sample data, the multi-state sample data comprises multi-state test data and associated multi-states, the multi-state test data is obtained by combining each corresponding state with distinguishing test data obtained by comparing reference normal test data, and establishes and trains a state detection neural network model based on the sample test data;
and the result subsystem 400 reads the test data corresponding to the circuit breaker to be tested, and obtains the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
Although there have been many studies in the art on the test items and purposes of circuit breakers, as well as the method of testing. However, the test results are still in a rough way at present, and mainly depend on manual research and reading. The test result is manually read, so that the problem of low efficiency and the problem of low accuracy and easy misjudgment of the research and judgment are solved. How to improve the efficiency and accuracy of testing and judging the circuit breaker results becomes a research subject in the field. According to one or more embodiments disclosed in the specification, a circuit breaker off-line testing system and a circuit breaker off-line testing method based on a neural network are provided, a neural network model is used for rapidly and accurately judging a testing result of the short circuit test, and a solution is provided for improving the judging efficiency and accuracy of the testing result of the circuit breaker. The short circuit test of the circuit breaker is a key detection means of the circuit breaker performance, and is mainly used for verifying whether the circuit breaker can rapidly and accurately cut off short circuit current under extreme conditions, so as to ensure safe and stable operation of a power system. In the test process, a preset high rated short-circuit current is introduced into the circuit breaker by simulating possible short-circuit fault conditions of an actual power grid, and the action time, breaking capacity and mechanical and electrical integrity after operation of the circuit breaker are checked. The testing steps comprise: the circuit breaker is connected into a special short circuit test loop; setting corresponding short-circuit current values according to product standards and service environment requirements; starting a short circuit generating device to simulate a short circuit fault; and recording and analyzing data such as action response time, contact separation state and damage condition of the circuit breaker after the equipment bears short-circuit current so as to evaluate whether the short-circuit protection performance of the circuit breaker meets the standard. One or more embodiments disclosed herein also record the current time profile of the three phases during the short circuit test as an integral part of the test results.
The state referred to in the specification refers to the performance, service life, working condition and fault state of the circuit breaker. In this specification, the state is used to characterize the quality, electrical and mechanical properties, and the actual operating performance of the circuit breaker. The state is represented by the code, illustratively, code 000011000 represents a normal state, code 001100110 represents the deformation of the charge spring 17, code 001010010 represents the deformation of the charge spring 17 and the rejection of the secondary circuit 12.
Neural network models are known in the art and have the characteristics of high accuracy and high efficiency. The advantage and disadvantage of the implementation result of this embodiment is that the selection and setting of the sample breaker are more affected. As a recommended embodiment, after the sample breaker is connected to the offline testing subsystem 100, each electrical or mechanical performance index is set according to a step length, and a plurality of short-circuit tests are performed, so as to obtain test data corresponding to a plurality of states of each electrical or mechanical performance index. Or, the device to be tested is opened and closed for a plurality of times until the opening and closing times reach the preset upper limit or various working conditions and faults occur, a judgment standard is manually given and set, and whether the opening and closing are continued is judged according to the working conditions and faults. So that sample test data with sufficient status markers can be generated. For example, when testing the performance of the energy storage spring 17, the deformation of the pure spring 17 is increased step by step in a mechanical manner according to the step length, a short circuit test is performed once under each deformation, a plurality of test data are obtained, and corresponding state mark test data are used to obtain sample test data. In the marking, the state uses a normal state and two states in which the energy storage spring 17 is deformed, as two tags, each deformation amount is marked. On the other hand, a more preferred embodiment is to provide more states, such as, for example, using a normal state, a small deformation of the energy storage spring 17, a middle deformation of the energy storage spring 17, a large deformation of the energy storage spring 17, and marking different deformation amounts of the energy storage spring 17. The specific division mode is carried out by adopting a mode disclosed in the field.
Referring to fig. 4, the off-line testing device includes:
the super capacitor 101 is used for providing power;
the protection switch 102 is connected with the super capacitor 101 and is used for protecting a circuit connected with the super capacitor;
the solid-state switch 103 is connected with the protection switch 102 and used for controlling the start and stop of the off-line test;
and the acquisition module 104 is used for acquiring test data of the offline short circuit test.
The super capacitor 101 can realize the power supply needed by the short circuit test under the condition that the commercial power is used as the charging power supply. In the testing process, the off-line mode, namely the mode of separating from the power grid is adopted, so that impact and influence on the power grid can be avoided. The protection switch 102 can cut off the super capacitor 101 under the condition of fault current, so as to realize safety guarantee. The solid state switch 103 is used to control the start and stop of an off-line test, which is referred to as an off-line short circuit test. The collection module 104 collects test data, and when in implementation, a corresponding collection device is configured according to the technical scheme disclosed in the art according to the selected test data.
As a recommended embodiment, the preset state includes one or more of a preset control system operating condition state, a preset circuit system operating condition state, and a preset mechanical system operating condition state.
As a recommended implementation mode, the working condition state of the preset control system is a control normal working condition state, a driving failure state, a signal abnormal state, a feedback failure state or a power undersize state.
As a recommended implementation mode, the working condition state of the preset circuit system is a normal state of a circuit, a short circuit state of a circuit node or an open circuit state of the circuit node.
As a recommended embodiment, the preset mechanical system working condition state is a mechanical normal working condition state, a cam offset state, a large clearance error state, a failure state of the energy storage spring 17, a failure state of the lock catch 14, a dead point state or a jam state.
The working conditions of the preset control system, the working conditions of the preset circuit system and the working conditions of the preset mechanical system can be mutually combined and arranged to obtain detection and identification of richer states.
As a recommended embodiment, the test data includes at least one of a short-circuit on current, a short-time withstand current, a limit short-circuit breaking current, and an operating short-circuit breaking current. The corresponding selection detection device can realize data acquisition, and then the data subsystem 200 collects and stores the data and generates sample test data.
As a recommended embodiment, referring to fig. 5, the state detection neural network model is a fully connected neural network, the state detection neural network model includes an input layer, one or more hidden layers and an output layer, the input of the state detection neural network model is the test data, and the output is the state of the circuit breaker.
Considering a basic neural network model with two learnable parameters, first, M linear combinations of input variables x1, …, xD are established, in the form of
Where j=1, …, M and superscript (1) indicates that the corresponding parameter is in the first layer of the neural network model. In the formula (1), the parameters are as followsLet the weight, parameter->Is described as offset, and the number ∈ ->Then the pre-activation is the pre-activation. Each of which is>By means of a differentiable nonlinear activation function>To convert into
Within the neural network model, these activation functions are called hidden units. The variables are linearly combined and provided with
Where k=1, …, K, and K is the total number of outputs. This variant corresponds to the second layer of the neural network model, and is alsoIs offset. Final result，/>Activating the function by means of a suitable output unit>A series of outputs converted into a network +.>. The test data is used as input, the state of the breaker sample is coded and then used as output, and the neural network can be applied to a breaker performance test platform.
The nonlinear differentiable function employed in the neural network is a logical sigmoid function of the form
Depending on the effect of the application, the tanh function shown below may be used
Or (b)
。
In a second aspect, an embodiment of the present disclosure provides a method for offline testing a circuit breaker based on a neural network, please refer to fig. 6, including the steps of:
step 102) performing off-line short-circuit test on a sample circuit breaker and a circuit breaker to be tested based on a pre-provided off-line test device, wherein the sample circuit breaker is a circuit breaker with a preset state;
step 104) detecting test data of the sample circuit breaker and the circuit breaker to be tested for offline short circuit test;
step 106) reading test data corresponding to the sample circuit breaker, marking the test data according to a preset state of the sample circuit breaker, and obtaining sample test data, wherein the sample test data comprises single-state sample data and multi-state sample data, the multi-state sample data comprises multi-state test data and associated multi-state, and the multi-state test data is obtained by combining each corresponding state with distinguishing test data obtained by comparing with reference normal test data;
step 108) building and training a state detection neural network model based on the sample test data;
step 110) reading the test data corresponding to the circuit breaker to be tested, and obtaining the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
Referring to fig. 7, the test data includes current waveform data, the test data is marked according to a preset state of the sample breaker, and the method for obtaining the sample test data includes:
step 202) taking the test data marked by the preset state as single-state sample data;
step 204) randomly selecting a plurality of single-state sample data from the sample data, wherein the preset states of the single-state sample data marks are different;
step 206) comparing the current waveform data of the plurality of single-state sample data with preset normal waveform data respectively to obtain differential current waveform data;
step 208) synthesizing multi-state current waveform data according to the differential current waveform data, and obtaining multi-state test data based on the multi-state current waveform data;
step 210) combining the preset states of the single-state sample data to obtain a combined state;
step 212) marking the multi-state test data with the combined state to obtain multi-state sample data.
Example 1
The performance test system for the circuit breaker is used for testing the performance of the circuit breaker, and is required to test whether the mechanical system of the circuit breaker works normally for 10000 times without generating stress relaxation, and if the stress relaxation occurs, the corresponding working times are given.
First, a sample breaker with normal operation of a mechanical system and stress relaxation is prepared. Sample current data is then measured by the offline test subsystem 100, as shown in fig. 8. Sample current data is classified and labeled, where "normal operation" and "stress relaxation" are encoded as 01 and 10, respectively, and all of the above data is stored in the data subsystem 200. Next, the sample current data shown in fig. 8 is used as an input of the neural network, and the data label is used as an output of the neural network, so that training of the neural network is performed. After training is completed, neural network parameters are imported into the results subsystem 400.
And finally, testing the device to be tested and outputting the result. The upper limit of the opening and closing times is set to 10000 times, the off-line testing subsystem 100 continuously performs the opening and closing operation of the circuit breaker, and current data is collected in real time. The result subsystem 400 takes the acquired data as the input of the neural network model, so as to obtain the output of the neural network model, namely the code of 01 or 10, and the device to be tested is judged to be in a normal working state or a stress relaxation state according to the code. And ending the test when the upper limit of the times is reached or the stress relaxation state is generated, and obtaining a result to finish the stress relaxation test of the mechanical system of the device.
Example two
The circuit breaker performance test platform is applied to test the circuit breaker, wherein the neural network is trained (training is performed based on test data of the sample circuit breaker with the locking device disabled). Test current data of a device to be tested after being opened and closed 10000 times are obtained through the result subsystem 400, as shown in fig. 9, the test current data is passed through the result subsystem 400, and the obtained result is that the circuit breaker is separated after being combined, and the circuit breaker belongs to a malfunction fault because of a failure of a locking device. Therefore, the breaker off-line testing system provided by the specification can be used for identifying breaker faults.
The above-described embodiments are merely preferred embodiments of the present disclosure, and do not limit the scope of the disclosure, and various modifications and improvements made by those skilled in the art to the technical solutions of the disclosure should fall within the protection scope defined by the claims of the disclosure without departing from the design spirit of the disclosure.
Claims (10)
1. The off-line testing system of the circuit breaker based on the neural network is characterized by comprising the following components:
the off-line testing subsystem is used for conducting off-line short-circuit testing on the sample circuit breaker and the circuit breaker to be tested based on the off-line testing device, wherein the sample circuit breaker is a circuit breaker with a preset state;
the data subsystem establishes data connection with the offline test subsystem and detects test data of offline short circuit tests of the sample circuit breaker and the circuit breaker to be tested;
the model subsystem is used for reading test data corresponding to the sample circuit breaker, marking the test data according to a preset state of the sample circuit breaker, obtaining sample test data, wherein the sample test data comprises single-state sample data and multi-state sample data, the multi-state sample data comprises multi-state test data and associated multi-states, the multi-state test data is obtained by combining each corresponding state with distinguishing test data obtained by comparing reference normal test data, and a state detection neural network model is established and trained based on the sample test data;
and the result subsystem reads the test data corresponding to the circuit breaker to be tested, and obtains the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
2. The neural network-based circuit breaker off-line testing system of claim 1, wherein,
the off-line testing device comprises:
the super capacitor is used for providing power;
the protection switch is connected with the super capacitor and used for protecting a circuit connected with the super capacitor;
the solid-state switch is connected with the protection switch and used for controlling the start and stop of off-line test;
and the acquisition module is used for acquiring test data of the off-line short circuit test.
3. The neural network-based circuit breaker off-line testing system according to claim 1 or 2, wherein,
the preset state includes one or more of a preset control system operating condition state, a preset circuit system operating condition state, and a preset mechanical system operating condition state.
4. The neural network-based circuit breaker off-line testing system of claim 3, wherein,
the working condition state of the preset control system is a control normal working condition state, a driving failure state, a signal abnormal state, a feedback failure state or a power undersize state.
5. The neural network-based circuit breaker off-line testing system of claim 3, wherein,
the working condition state of the preset circuit system is a circuit normal state, a circuit node short circuit state or a circuit node open circuit state.
6. The neural network-based circuit breaker off-line testing system of claim 3, wherein,
the working condition of the preset mechanical system is a normal working condition, a cam offset condition, a large clearance error condition, an energy storage spring failure condition, a lock catch failure condition, a dead point condition or a jamming condition of the machine.
7. The neural network-based circuit breaker off-line testing system according to claim 1 or 2, wherein,
the test data includes at least one of a short-circuit on current, a short-time withstand current, a limit short-circuit breaking current, and an operating short-circuit breaking current.
8. The neural network-based circuit breaker off-line testing system according to claim 1 or 2, wherein,
the state detection neural network model is a fully-connected neural network and comprises an input layer, one or more hidden layers and an output layer, wherein the input of the state detection neural network model is the test data, and the output is the state of the circuit breaker.
9. The off-line testing method of the circuit breaker based on the neural network is characterized in that,
the method comprises the following steps:
performing off-line short-circuit test on a sample circuit breaker and a circuit breaker to be tested based on a pre-provided off-line test device, wherein the sample circuit breaker is a circuit breaker with a preset state;
detecting test data of the sample circuit breaker and the circuit breaker to be tested for offline short circuit test;
reading test data corresponding to a sample breaker, marking the test data according to a preset state of the sample breaker, and obtaining sample test data, wherein the sample test data comprises single-state sample data and multi-state sample data, the multi-state sample data comprises multi-state test data and associated multi-states, and the multi-state test data is obtained by comparing each corresponding state with differential test data obtained by referring to normal test data;
establishing and training a state detection neural network model based on the sample test data;
and reading the test data corresponding to the circuit breaker to be tested, and obtaining the state of the circuit breaker to be tested based on the test data and the state detection neural network model.
10. The neural network-based circuit breaker off-line testing method of claim 9, wherein,
the test data includes current waveform data,
marking the test data according to the preset state of the sample breaker, and the method for obtaining the sample test data comprises the following steps:
taking the test data marked by the preset state as single-state sample data;
randomly selecting a plurality of single-state sample data from the sample data, wherein the preset states of the single-state sample data marks are different;
comparing the current waveform data of the plurality of single-state sample data with preset normal waveform data respectively to obtain differential current waveform data;
synthesizing multi-state current waveform data according to the differential current waveform data, and obtaining multi-state test data based on the multi-state current waveform data;
combining the preset states of the single-state sample data to obtain a combined state;
and marking the multi-state test data by using the combined state to obtain multi-state sample data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410275589.XA CN117872124A (en) | 2024-03-12 | 2024-03-12 | Off-line testing system and method for circuit breaker based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410275589.XA CN117872124A (en) | 2024-03-12 | 2024-03-12 | Off-line testing system and method for circuit breaker based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117872124A true CN117872124A (en) | 2024-04-12 |
Family
ID=90597185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410275589.XA Pending CN117872124A (en) | 2024-03-12 | 2024-03-12 | Off-line testing system and method for circuit breaker based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117872124A (en) |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108828441A (en) * | 2018-06-12 | 2018-11-16 | 江苏镇安电力设备有限公司 | Fault Diagnosis for HV Circuit Breakers method |
| CN108918111A (en) * | 2018-05-16 | 2018-11-30 | 国网山东省电力公司莱芜供电公司 | Breaker mechanical method for diagnosing faults based on the classification of k- neighbour's Bayes's multi-tag |
| CN109270442A (en) * | 2018-08-21 | 2019-01-25 | 西安工程大学 | High-voltage circuitbreaker fault detection method based on DBN-GA neural network |
| CN110703075A (en) * | 2019-09-10 | 2020-01-17 | 国网浙江省电力有限公司嘉兴供电公司 | High-voltage circuit breaker quick overhaul method based on big data technology |
| CN110716133A (en) * | 2019-09-06 | 2020-01-21 | 国网浙江省电力有限公司嘉兴供电公司 | High-voltage circuit breaker fault studying and judging method based on Internet of things and big data technology |
| CN110929763A (en) * | 2019-10-30 | 2020-03-27 | 西安交通大学 | Multi-source data fusion-based mechanical fault diagnosis method for medium-voltage vacuum circuit breaker |
| CN111289888A (en) * | 2019-12-11 | 2020-06-16 | 嘉兴恒创电力集团有限公司博创物资分公司 | High-voltage circuit breaker state detection and fault diagnosis method based on big data technology |
| CN112698194A (en) * | 2020-12-10 | 2021-04-23 | 云南电网有限责任公司保山供电局 | Comprehensive evaluation method and system for state of circuit breaker operating mechanism |
| CN112729381A (en) * | 2020-12-11 | 2021-04-30 | 广州致新电力科技有限公司 | Fault diagnosis method of high-voltage circuit breaker based on neural network |
| CN112748330A (en) * | 2020-12-24 | 2021-05-04 | 国网江苏省电力有限公司电力科学研究院 | High-voltage circuit breaker operating mechanism fault diagnosis method and device based on distributed integrated neural network |
| CN113191192A (en) * | 2021-03-25 | 2021-07-30 | 云南电网有限责任公司玉溪供电局 | Breaker fault detection method based on wavelet analysis and fuzzy neural network algorithm |
| CN113552448A (en) * | 2021-08-05 | 2021-10-26 | 华能沾化热电有限公司 | Breaker state detection method |
| CN115327369A (en) * | 2022-09-13 | 2022-11-11 | 国网安徽省电力有限公司宿州供电公司 | On-line monitoring and analyzing system for opening and closing current of switch cabinet circuit breaker |
| CN115456106A (en) * | 2022-09-27 | 2022-12-09 | 通号（长沙）轨道交通控制技术有限公司 | High-voltage circuit breaker fault diagnosis model optimization method |
| CN115575811A (en) * | 2022-10-28 | 2023-01-06 | 国网陕西省电力有限公司电力科学研究院 | Method, system, storage medium and equipment for predicting and evaluating electric service life of circuit breaker |
| CN115728628A (en) * | 2022-10-19 | 2023-03-03 | 国网河北省电力有限公司超高压分公司 | On-line monitoring and early warning method and device for circuit breaker control loop fault |
| CN116243155A (en) * | 2023-02-09 | 2023-06-09 | 国网河北省电力有限公司电力科学研究院 | Breaker fault diagnosis method, device, equipment and storage medium |
| CN116524200A (en) * | 2023-05-05 | 2023-08-01 | 东南大学 | High-voltage circuit breaker fault diagnosis method based on image recognition |
| CN117590223A (en) * | 2024-01-18 | 2024-02-23 | 南京飞腾电子科技有限公司 | Online monitoring system and method for circuit breaker |
-
2024
- 2024-03-12 CN CN202410275589.XA patent/CN117872124A/en active Pending
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108918111A (en) * | 2018-05-16 | 2018-11-30 | 国网山东省电力公司莱芜供电公司 | Breaker mechanical method for diagnosing faults based on the classification of k- neighbour's Bayes's multi-tag |
| CN108828441A (en) * | 2018-06-12 | 2018-11-16 | 江苏镇安电力设备有限公司 | Fault Diagnosis for HV Circuit Breakers method |
| CN109270442A (en) * | 2018-08-21 | 2019-01-25 | 西安工程大学 | High-voltage circuitbreaker fault detection method based on DBN-GA neural network |
| CN110716133A (en) * | 2019-09-06 | 2020-01-21 | 国网浙江省电力有限公司嘉兴供电公司 | High-voltage circuit breaker fault studying and judging method based on Internet of things and big data technology |
| CN110703075A (en) * | 2019-09-10 | 2020-01-17 | 国网浙江省电力有限公司嘉兴供电公司 | High-voltage circuit breaker quick overhaul method based on big data technology |
| CN110929763A (en) * | 2019-10-30 | 2020-03-27 | 西安交通大学 | Multi-source data fusion-based mechanical fault diagnosis method for medium-voltage vacuum circuit breaker |
| CN111289888A (en) * | 2019-12-11 | 2020-06-16 | 嘉兴恒创电力集团有限公司博创物资分公司 | High-voltage circuit breaker state detection and fault diagnosis method based on big data technology |
| CN112698194A (en) * | 2020-12-10 | 2021-04-23 | 云南电网有限责任公司保山供电局 | Comprehensive evaluation method and system for state of circuit breaker operating mechanism |
| CN112729381A (en) * | 2020-12-11 | 2021-04-30 | 广州致新电力科技有限公司 | Fault diagnosis method of high-voltage circuit breaker based on neural network |
| CN112748330A (en) * | 2020-12-24 | 2021-05-04 | 国网江苏省电力有限公司电力科学研究院 | High-voltage circuit breaker operating mechanism fault diagnosis method and device based on distributed integrated neural network |
| CN113191192A (en) * | 2021-03-25 | 2021-07-30 | 云南电网有限责任公司玉溪供电局 | Breaker fault detection method based on wavelet analysis and fuzzy neural network algorithm |
| CN113552448A (en) * | 2021-08-05 | 2021-10-26 | 华能沾化热电有限公司 | Breaker state detection method |
| CN115327369A (en) * | 2022-09-13 | 2022-11-11 | 国网安徽省电力有限公司宿州供电公司 | On-line monitoring and analyzing system for opening and closing current of switch cabinet circuit breaker |
| CN115456106A (en) * | 2022-09-27 | 2022-12-09 | 通号（长沙）轨道交通控制技术有限公司 | High-voltage circuit breaker fault diagnosis model optimization method |
| CN115728628A (en) * | 2022-10-19 | 2023-03-03 | 国网河北省电力有限公司超高压分公司 | On-line monitoring and early warning method and device for circuit breaker control loop fault |
| CN115575811A (en) * | 2022-10-28 | 2023-01-06 | 国网陕西省电力有限公司电力科学研究院 | Method, system, storage medium and equipment for predicting and evaluating electric service life of circuit breaker |
| CN116243155A (en) * | 2023-02-09 | 2023-06-09 | 国网河北省电力有限公司电力科学研究院 | Breaker fault diagnosis method, device, equipment and storage medium |
| CN116524200A (en) * | 2023-05-05 | 2023-08-01 | 东南大学 | High-voltage circuit breaker fault diagnosis method based on image recognition |
| CN117590223A (en) * | 2024-01-18 | 2024-02-23 | 南京飞腾电子科技有限公司 | Online monitoring system and method for circuit breaker |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104956186B (en) | The method of electric power distribution system and assessment energy meter including function of measuring | |
| US10910812B2 (en) | Method for testing open phase detection system | |
| CN103513139A (en) | Power transformer fault intelligent diagnosis technology, method and device | |
| Idrees et al. | Fuzzy logic based calculation and analysis of health index for power transformer installed in grid stations | |
| CN103163453A (en) | Nuclear power station relay detection method | |
| CN104898052B (en) | The check system and method for switch mechanical characteristic testing device based on IEC61850 | |
| WO2013131392A1 (en) | Dc power supply dynamic simulation system and dc power supply test system | |
| CN114415625B (en) | Intelligent inspection system and method for mine explosion-proof electrical appliance safety parameters | |
| CN111786462B (en) | Power grid information verification system and method based on multidimensional panoramic data | |
| Zhou et al. | A multi-evidence fusion based integrated method for health assessment of medium voltage switchgears in power grid | |
| Albinali et al. | A centralized substation protection scheme that detects hidden failures | |
| CN117872124A (en) | Off-line testing system and method for circuit breaker based on neural network | |
| CN112150022A (en) | Traction transformer state monitoring and management system | |
| CN115932564A (en) | Fault evaluation method, device and equipment for high-voltage circuit breaker and readable storage medium | |
| CN216351043U (en) | Test circuit of battery pack and battery pack test control system | |
| Pickett et al. | Reducing outages through improved protection, monitoring, diagnostics, and autorestoration in transmission substations—(69 kV and above) | |
| Apostolov et al. | Maintenance testing of multifunctional distance protection ieds | |
| Matavalam et al. | Reliability assessment of industrial circuit breakers with design enhancements | |
| CN206074791U (en) | A kind of faulty action preventing circuit and the normal station low-voltage protection device using the circuit | |
| Kwong et al. | Remarkable life cycle management by effective condition monitoring and assessment system of power transformer in CLP power system | |
| CN220137355U (en) | Position feedback loop broken line detection device | |
| Weerasekera et al. | Condition Assessment of HV Circuit Breaker Contacts Using Dynamic Resistance Measurements | |
| Rebolini et al. | Terna's approach for on-line monitoring system: Intelligent management of assets in a large scale infrastructures | |
| Xu et al. | Research and Application of Operation Characteristic Model and Operation and Maintenance Strategy of High Voltage Switchgear | |
| CN114447873A (en) | Breaker action evaluation method and system based on substation automation information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |