CN115392118A - Debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch - Google Patents
Debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch Download PDFInfo
- Publication number
- CN115392118A CN115392118A CN202210987816.2A CN202210987816A CN115392118A CN 115392118 A CN115392118 A CN 115392118A CN 202210987816 A CN202210987816 A CN 202210987816A CN 115392118 A CN115392118 A CN 115392118A
- Authority
- CN
- China
- Prior art keywords
- model
- simulation
- module
- voltage isolating
- isolating switch
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
Abstract
The invention relates to the technical field of intelligent diagnosis of high-voltage isolating switches, in particular to a debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switches. The system comprises a joint simulation unit, a simulation unit and a simulation unit, wherein the joint simulation unit is used for acquiring a large amount of simulation data through mechatronic joint simulation; a database management unit for storing data of all full life cycles involved in the management system; and the deep training unit is used for acquiring the intelligent diagnosis model of the mechanical state of the high-voltage isolating switch through training the deep neural network model. According to the design, a deep neural network model is sequentially trained progressively by using a large number of simulation samples and limited real samples, so that the parameters of a fault diagnosis model are optimized, and the problem that the precision of an intelligent diagnosis algorithm is low due to too few training samples of a high-voltage isolating switch is effectively solved; the efficiency and the precision of the high-voltage isolating switch mechanical state diagnosis technology realized by intelligent algorithms of technologies such as a deep neural network and the like can be improved.
Description
Technical Field
The invention relates to the technical field of intelligent diagnosis of high-voltage isolating switches, in particular to a debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switches.
Background
Mechanical faults of the high-voltage isolating switch occur frequently, and the operation safety of a power system is seriously threatened. The high-voltage isolating switch belongs to high-voltage switch equipment which is used in a large amount in a power system, and if the high-voltage isolating switch breaks down, the safety of maintainers and the normal operation of a power grid are seriously damaged; therefore, the normal operation of the high-voltage isolating switch plays a vital role in the safety of maintainers and the normal operation of a power grid. With the increasing demand for land utilization in China, the GIS type isolating switch is increasingly applied to the safety guarantee link in the field of high-voltage power transmission with the advantages of small occupied area, high integration degree and the like.
The intelligent diagnosis technology which is carried out through the motor driving power in the switching-on and switching-off process of the isolating switch is one of the online diagnosis technologies of the mechanical state of the high-voltage isolating switch with clear prospect. However, in order to achieve higher precision, an intelligent algorithm related to the intelligent diagnosis technology often needs a large number of training samples, tens of thousands of training samples are difficult to obtain simply through industrial sites and actual measurement, and the obtained samples are difficult to clean, screen and the like, so that the diagnosis efficiency and accuracy of the intelligent diagnosis technology are low, and further the industry cannot recognize that the mechanical state result of the high-voltage isolation switch is realized through the intelligent algorithm of the technology such as the deep neural network.
In view of this, in order to improve the efficiency and the accuracy of the technology for diagnosing the mechanical state of the high-voltage isolating switch by using an intelligent algorithm of technologies such as a deep neural network, a debugging system for intelligently diagnosing the mechanical fault of the high-voltage isolating switch is provided.
Disclosure of Invention
The invention aims to provide a debugging system for intelligently diagnosing mechanical faults of a high-voltage isolating switch, so as to solve the problems in the background technology.
In order to solve the technical problems, the invention provides a debugging system for intelligently diagnosing mechanical faults of a high-voltage isolating switch, which comprises
The combined simulation unit is used for modeling the physical and asynchronous motors of the high-voltage isolating switch respectively, then further performing electromechanical integrated combined simulation on the two models, and acquiring a large amount of simulation data through multi-state repeated simulation;
the system comprises a database management unit, a data processing unit and a data processing unit, wherein the database management unit is used for storing all data of the full life cycle related to a management system, and comprises basic data imported for modeling of the combined simulation unit, simulation sample data obtained from the combined simulation unit, subsequent model training supported by combining with field measured data, and massive historical operating data of the high-voltage isolating switch;
and the deep training unit is used for acquiring an intelligent diagnosis model of the mechanical state of the high-voltage isolating switch by training a deep neural network model by adopting simulation sample data and field measured data in the database management unit, so that intelligent diagnosis and debugging of mechanical faults of the high-voltage isolating switch are realized.
As a further improvement of the technical scheme, the joint simulation unit comprises a physical model building module, an asynchronous motor simulation module and an electromechanical integrated debugging module; the physical model building module and the asynchronous motor simulation module run in parallel, and the signal output ends of the physical model building module and the asynchronous motor simulation module are simultaneously connected with the signal input end of the mechatronic debugging module;
wherein:
the physical model building module also comprises engineering drawing software which is used for building a physical model of the high-voltage isolating switch according to the fine proportion of 1;
the asynchronous motor simulation module also comprises electric power simulation software which is used for building an asynchronous motor model in the electric power simulation software according to the fine proportion of 1;
the electromechanical integration debugging module also comprises a joint simulation module, the joint simulation module is composed of joint simulation interfaces of two types of simulation software, namely engineering drawing software and electric power simulation software, the electromechanical integration debugging module is used for constructing an electromechanical integration joint simulation model formed by combining the high-voltage isolating switch physical model and the asynchronous motor model, and the electromechanical integration joint simulation model is debugged by adopting the joint simulation module.
The high-voltage isolating switch physical model and the asynchronous motor model are in butt joint combination through a combined simulation interface of two kinds of simulation software.
As a further improvement of the technical solution, the database management unit comprises a basic physical parameter module, an analog simulation sample module and an experimental actual measurement sample module; the basic physical parameter module, the simulation sample module and the experimental actual measurement sample module coexist in parallel;
wherein:
the physical parameters stored and managed in the basic physical parameter module are obtained by actual measurement of a real object system of the high-voltage isolating switch or are preset and used for importing the engineering drawing software and the electric power simulation software to build and simulate a physical model or an asynchronous motor model;
the sample data stored and managed in the simulation sample module is a sample result of a power-time signal of the driving motor, which is obtained by setting various mechanical states in the mechatronic combined simulation model established in the mechatronic debugging module and repeatedly simulating;
the sample data stored and managed in the experimental actual measurement sample module is a limited sample of the power-time signal of the driving motor in different mechanical states of the high-voltage isolating switch obtained through experiments or field actual measurement.
As a further improvement of the technical solution, in the process of obtaining simulation sample data by the mechatronic debugging module, the mechanical state of the high-voltage disconnecting switch is simulated by setting constraint conditions, wherein the mechanical state that can be simulated includes but is not limited to switching-off, switching-on, switching-off (operation, standby, maintenance) and various switching-on and switching-off states.
As a further improvement of the technical scheme, the specific method for acquiring the limited sample of the power-time signal of the driving motor in different mechanical states of the high-voltage isolating switch by the experimental actual measurement sample module in an experimental or field actual measurement mode comprises the following steps:
actually measuring a power-time curve of a driving motor in the switching-on and switching-off process of the high-voltage isolating switch by building a high-voltage isolating switch experimental platform or in a transformer substation; sampling the mechanical state of the high-voltage isolating switch in the maintenance process to obtain small sample data of a power-time curve of a driving motor based on the actual opening and closing process of the high-voltage isolating switch; and the mechanical state based on the actual high-voltage isolating switch corresponds to the mechanical state simulated in the mechatronic debugging module.
As a further improvement of the technical solution, the deep training unit includes a deep neural network model, a trained multilayer model and an intelligent diagnosis model; the signal output end of the deep neural network model is connected with the signal input end of the trained multilayer model, and the signal output end of the trained multilayer model is connected with the signal input end of the intelligent diagnosis model;
wherein:
the deep neural network model is a multilayer neural network model established on the historical operating data of massive high-voltage isolating switches and used as a training basis for machine learning;
the trained multilayer model is a new model which is output after the deep neural network model is trained by utilizing a large amount of sample data stored and managed in the simulation sample module and is used for intermediate transition;
the intelligent diagnosis model is a high-precision intelligent diagnosis model of the mechanical state of the high-voltage isolating switch, which is obtained by training the last layers of the trained multilayer model through sample data stored and managed in the experimental actual measurement sample module and is used for performing subsequent intelligent diagnosis operation of the mechanical state of the high-voltage isolating switch.
As a further improvement of the technical solution, when the intelligent diagnosis model is obtained by training the last several layers of the trained multilayer model through the sample data stored and managed in the experimental actual measurement sample module, the number of the trained model layers is determined according to the precision requirement of the intelligent diagnosis model for the mechanical state of the high-voltage disconnecting switch with high precision.
The invention also aims to provide a debugging method for intelligent diagnosis of mechanical faults of the high-voltage isolating switch, which is operated based on the debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switch and comprises the following steps:
step 1, importing preset physical parameter data from a basic physical parameter module in the database management unit into the joint simulation unit;
step 2, constructing a physical model of the high-voltage isolating switch in the physical model construction module through engineering drawing software; building an asynchronous motor model in the asynchronous motor simulation module through electric power simulation software;
step 3, constructing a mechatronic joint simulation model formed by combining the high-voltage isolating switch physical model and the asynchronous motor model in the mechatronic debugging module, and debugging the mechatronic joint simulation model by adopting the joint simulation modules of the engineering drawing software and the electric power simulation software;
step 4, setting various mechanical states in the mechanical-electrical integration combined simulation model established in the step 3, repeatedly simulating, obtaining a sample result of a power-time signal of the driving motor and storing the sample result into the simulation sample module of the database management unit;
step 5, training the deep neural network model in the deep training unit by using a large number of samples stored in the analog simulation sample module;
step 6, obtaining limited samples of the power-time signals of the driving motor under different mechanical states of the high-voltage isolating switch in an experimental or field actual measurement mode, and storing the limited samples into an experimental actual measurement sample module of the database management unit;
step 7, training the last layers of the trained multilayer model trained in the step 5 through small samples stored in the experimental actual measurement sample module, so as to obtain the high-precision intelligent diagnosis model of the mechanical state of the high-voltage isolating switch;
and S8, importing the real-time operation data of the high-voltage isolating switch into the intelligent diagnosis model, so that the intelligent diagnosis of the mechanical fault of the high-voltage isolating switch can be realized.
The invention also provides a system operation platform device, which comprises a processor, a memory and a computer program stored in the memory and operated on the processor, wherein the processor is used for implementing the debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switch when executing the computer program.
The fourth objective of the present invention is to provide a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the above debugging system for intelligent diagnosis of mechanical fault of a high-voltage disconnecting switch.
Compared with the prior art, the invention has the beneficial effects that:
1. in the debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switch, firstly, a high-voltage isolating switch physical model and an asynchronous motor model corresponding to a real high-voltage isolating switch physical system are built according to a high-precision proportion of 1;
2. in the debugging system for the intelligent diagnosis of the mechanical fault of the high-voltage isolating switch, the intelligent diagnosis of the mechanical fault of the high-voltage isolating switch is carried out by adopting a high-precision intelligent diagnosis model obtained through training, the efficiency and the precision of the technology for realizing the diagnosis of the mechanical state of the high-voltage isolating switch by an intelligent algorithm of technologies such as a deep neural network and the like can be improved, and therefore the intelligent diagnosis result of the mechanical fault state of the high-voltage isolating switch can be more easily recognized by the industry.
Drawings
FIG. 1 is a block diagram of an exemplary overall system architecture of the present invention;
FIG. 2 is a block flow diagram of an exemplary method of the present invention;
FIG. 3 is a block diagram of an exemplary electronic computer platform assembly in accordance with the present invention.
The various reference numbers in the figures mean:
100. a joint simulation unit; 101. a physical model building module; 1011. engineering drawing software; 102. an asynchronous motor simulation module; 1021. power simulation software; 103. a mechatronic debugging module; 1031. a joint simulation module;
200. a database management unit; 201. a basic physical parameter module; 202. an analog simulation sample module; 203. an experimental actual measurement sample module;
300. a deep training unit; 301. a deep neural network model; 302. training the multi-layer model; 303. an intelligent diagnosis model.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
As shown in fig. 1-3, the present embodiment provides a debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch, which comprises
The combined simulation unit 100 is used for modeling physical and asynchronous motors of the high-voltage isolating switch respectively, then performing electromechanical integrated combined simulation on the two models, and acquiring a large amount of simulation data through multi-state repeated simulation;
the database management unit 200 is used for storing all data of the full life cycle related in the management system, and comprises basic data imported for modeling of the combined simulation unit 100, simulation sample data obtained from the combined simulation unit 100, subsequent model training supported by combining with field measured data, and massive historical operating data of the high-voltage isolating switch;
and the deep training unit 300, wherein the deep training unit 300 is used for acquiring an intelligent diagnosis model of the mechanical state of the high-voltage isolating switch by training the deep neural network model by adopting simulation sample data and field measured data in the database management unit 200, so that intelligent diagnosis and debugging of mechanical faults of the high-voltage isolating switch are realized.
In this embodiment, the joint simulation unit 100 includes a physical model building module 101, an asynchronous motor simulation module 102, and an electromechanical integration debugging module 103; the physical model building module 101 and the asynchronous motor simulation module 102 run in parallel, and the signal output ends of the physical model building module 101 and the asynchronous motor simulation module 102 are connected with the signal input end of the electromechanical integrated debugging module 103;
wherein:
the physical model building module 101 further comprises engineering drawing software 1011, which is used for building a high-voltage isolating switch physical model in the engineering drawing software 1011 according to the refinement proportion of 1;
the asynchronous motor simulation module 102 further comprises power simulation software 1021, which is used for building an asynchronous motor model in the power simulation software 1021 according to a refinement ratio of 1;
the mechatronic debugging module 103 further comprises a joint simulation module 1031, the joint simulation module 1031 is composed of joint simulation interfaces of two kinds of simulation software, namely engineering drawing software 1011 and electric power simulation software 1021, the mechatronic debugging module 103 is used for constructing a mechatronic joint simulation model formed by combining a physical model of the high-voltage isolating switch and an asynchronous motor model, and debugging the mechatronic joint simulation model by adopting the joint simulation module 1031.
Further, in the co-simulation unit 100:
because the output power of the high-voltage isolating switch driving motor directly depends on the torque acted on the main shaft of the driving motor, and the torque acted on the main shaft of the motor is directly related to the mechanical state of the high-voltage isolating switch, a mechanical model of the high-voltage isolating switch is required to be constructed firstly when the power of the driving motor changes through simulation analysis of the high-voltage isolating switch in each mechanical state; when a high-voltage isolating switch model is constructed through commercial modeling software, the highly relevant parts of the mechanical analysis need to be subjected to refined modeling so as to improve the precision, and meanwhile, the irrelevant parts of the mechanical analysis are simplified or omitted so as to reduce the running time;
meanwhile, the power of the driving motor in the switching-on and switching-off process of the high-voltage isolating switch is related to the mechanical structure connected to the main shaft of the driving motor and also related to the motor parameters of the driving motor, so that the actual situation can be better restored by establishing a refined driving motor model of the high-voltage isolating switch in commercial software.
The high-voltage isolating switch physical model is built through the modeling function of the commercial engineering drawing software 1011, and the asynchronous motor model is built through the modeling function of the commercial power simulation software 1021, and the operations are performed according to the refinement ratio of 1.
Furthermore, the built physical model of the high-voltage isolating switch and the asynchronous motor model are combined through a combined simulation interface of two kinds of simulation software to form an electromechanical integrated simulation model of the high-voltage isolating switch.
In this embodiment, the database management unit 200 includes a basic physical parameter module 201, a simulation sample module 202, and an experimental actual measurement sample module 203; the basic physical parameter module 201, the simulation sample module 202 and the experimental actual measurement sample module 203 coexist in parallel;
wherein:
the basic physical parameter module 201 stores the managed physical parameters, which are obtained by actually measuring a real object system of the high-voltage isolating switch or are set by design in advance, and is used for importing engineering drawing software 1011 and electric power simulation software 1021 to build and simulate a physical model or an asynchronous motor model;
the sample data stored and managed in the simulation sample module 202 is a sample result of a power-time signal of the driving motor, which is obtained by setting various mechanical states in the mechatronic combined simulation model established in the mechatronic debugging module 103 and repeatedly performing simulation;
the sample data stored and managed in the experimental actual measurement sample module 203 is a limited sample of the power-time signal of the driving motor in different mechanical states of the high-voltage isolating switch, which is obtained through an experiment or a field actual measurement mode.
Further, in the process of obtaining simulation sample data through the mechatronic debugging module 103, the mechanical state of the high-voltage disconnecting switch is simulated through setting of constraint conditions, wherein the mechanical state capable of being simulated includes but is not limited to switching-off, switching-on, switching-off (operation, standby and maintenance) and various switching-on and switching-off states.
Further, the specific method for acquiring the limited sample of the power-time signal of the driving motor in different mechanical states of the high-voltage isolating switch by the experimental actual measurement sample module 203 through an experimental or field actual measurement mode includes the following steps:
actually measuring a power-time curve of a driving motor in the switching-on and switching-off process of the high-voltage isolating switch by building a high-voltage isolating switch experiment platform or in a transformer substation; sampling the mechanical state of the high-voltage isolating switch in the maintenance process to obtain small sample data of a power-time curve of the driving motor based on the actual opening and closing process of the high-voltage isolating switch;
wherein the mechanical state based on the actual high voltage disconnector corresponds to the simulated mechanical state in the mechatronic commissioning module 103.
In this embodiment, the deep training unit 300 includes a deep neural network model 301, a trained multilayer model 302, and an intelligent diagnosis model 303; the signal output end of the deep neural network model 301 is connected with the signal input end of the trained multilayer model 302, and the signal output end of the trained multilayer model 302 is connected with the signal input end of the intelligent diagnosis model 303;
wherein:
the deep neural network model 301 is a multilayer neural network model established on historical operating data of massive high-voltage isolating switches and used as a training basis for machine learning;
the trained multilayer model 302 is a new model for intermediate transition, which is output after the deep neural network model 301 is trained by using a large amount of sample data stored and managed in the analog simulation sample module 202;
the intelligent diagnosis model 303 is a high-precision intelligent diagnosis model of the mechanical state of the high-voltage isolating switch, which is obtained by training the last layers of the trained multilayer model 302 through sample data stored and managed in the experimental actual measurement sample module 203 and is used for performing subsequent intelligent diagnosis operation of the mechanical state of the high-voltage isolating switch.
The multilayer deep neural network is initially trained by using samples of the power-time curves of the drive motor of the obtained high-voltage isolating switch in each mechanical state, which are stored in the simulation sample module 202, so as to optimize parameters of the multilayer neural network.
Further, when the intelligent diagnosis model 303 is obtained by training the last several layers of the trained multilayer model 302 through the sample data stored and managed in the experimental actual measurement sample module 203, the number of the trained model layers is determined according to the precision requirement of the intelligent diagnosis model of the mechanical state of the high-voltage disconnecting switch with high precision.
As shown in fig. 2, the present embodiment further provides a debugging method for intelligently diagnosing a mechanical fault of a high-voltage isolator, where the method is operated based on the debugging system for intelligently diagnosing a mechanical fault of a high-voltage isolator, and includes the following steps:
step 1, importing preset physical parameter data from a basic physical parameter module 201 in a database management unit 200 into a joint simulation unit 100;
step 2, constructing a physical model of the high-voltage isolating switch in the physical model construction module 101 through engineering drawing software 1011; constructing an asynchronous motor model in the asynchronous motor simulation module 102 through electric power simulation software 1021;
step 3, constructing a mechatronic joint simulation model formed by combining a high-voltage isolating switch physical model and an asynchronous motor model in the mechatronic debugging module 103, and debugging the mechatronic joint simulation model by adopting a joint simulation module 1031 of two kinds of software, namely engineering drawing software 1011 and electric power simulation software 1021;
step 4, setting various mechanical states in the mechanical-electrical integration combined simulation model established in the step 3, repeatedly simulating, obtaining a sample result of the power-time signal of the driving motor and storing the sample result into an analog simulation sample module 202 of the database management unit 200;
step 5, training the deep neural network model 301 in the deep training unit 300 by using a large number of samples stored in the simulation sample module 202;
step 6, obtaining limited samples of the power-time signals of the driving motor in different mechanical states of the high-voltage isolating switch in an experimental or field actual measurement mode, and storing the limited samples into an experimental actual measurement sample module 203 of the database management unit 200;
step 7, training the last layers of the trained multilayer model 302 trained in the step 5 through a small sample stored in the experimental actual measurement sample module 203, so as to obtain a high-precision intelligent diagnosis model 303 of the mechanical state of the high-voltage isolating switch;
and S8, the real-time operation data of the high-voltage isolating switch is introduced into the intelligent diagnosis model 303, so that the intelligent diagnosis of the mechanical fault of the high-voltage isolating switch can be realized.
As shown in fig. 3, the present embodiment further provides a system operation platform device, which includes a processor, a memory, and a computer program stored in the memory and executed on the processor.
The processor comprises one or more than one processing core, the processor is connected with the memory through the bus, the memory is used for storing program instructions, and the debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switch is realized when the processor executes the program instructions in the memory.
Alternatively, the memory may be implemented by any type or combination of volatile and non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
In addition, the invention also provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the debugging system for intelligent diagnosis of mechanical faults of the high-voltage isolating switch is realized.
Optionally, the present invention also provides a computer program product containing instructions, which when run on a computer, causes the computer to execute the above aspects of the debugging system for intelligent diagnosis of mechanical faults of a high-voltage disconnector.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, and the program may be stored in a computer-readable storage medium, where the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk.
The foregoing shows and describes the general principles, principal features, and advantages of the invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention, and that various changes and modifications may be made without departing from the spirit and scope of the present invention, which fall within the scope of the claimed invention. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. A debug system for high voltage isolator mechanical failure intelligent diagnosis, its characterized in that: comprises that
The combined simulation unit (100) is used for modeling physical and asynchronous motors of the high-voltage isolating switch respectively, then performing electromechanical integrated combined simulation on the two models, and acquiring a large amount of simulation data through multi-state repeated simulation;
the system comprises a database management unit (200), a database management unit (200) and a high-voltage isolation switch, wherein the database management unit (200) is used for storing all data of a full life cycle related to a management system, and comprises basic data imported for modeling of the joint simulation unit (100), simulation sample data obtained from the joint simulation unit (100) and massive historical operating data of the high-voltage isolation switch, and the subsequent model training is supported by combining with field measured data;
the deep training unit (300) is used for acquiring an intelligent diagnosis model of the mechanical state of the high-voltage isolating switch by training a deep neural network model by adopting simulation sample data and field measured data in the database management unit (200), so that intelligent diagnosis and debugging of mechanical faults of the high-voltage isolating switch are realized.
2. The debugging system for intelligent diagnosis of mechanical faults of high-voltage isolating switches according to claim 1, wherein: the combined simulation unit (100) comprises a physical model building module (101), an asynchronous motor simulation module (102) and an electromechanical integrated debugging module (103); the physical model building module (101) and the asynchronous motor simulation module (102) run in parallel, and the signal output ends of the physical model building module (101) and the asynchronous motor simulation module (102) are connected with the signal input end of the electromechanical integration debugging module (103) at the same time;
wherein:
the physical model building module (101) further comprises engineering drawing software (1011) for building a high-voltage isolating switch physical model according to the refinement proportion of 1;
the asynchronous motor simulation module (102) further comprises electric power simulation software (1021), and the electric power simulation software (1021) is used for building an asynchronous motor model according to the refinement proportion of 1;
the electromechanical integration debugging module (103) further comprises a joint simulation module (1031), the joint simulation module (1031) is composed of joint simulation interfaces of two kinds of simulation software, namely engineering drawing software (1011) and electric power simulation software (1021), the electromechanical integration debugging module (103) is used for constructing a mechatronic joint simulation model formed by combining the high-voltage isolating switch physical model and the asynchronous motor model, and the joint simulation module (1031) is used for debugging the mechatronic joint simulation model.
3. The debugging system for intelligent diagnosis of mechanical faults of high-voltage isolating switches according to claim 2, characterized in that: the database management unit (200) comprises a basic physical parameter module (201), a simulation sample module (202) and an experimental actual measurement sample module (203); the basic physical parameter module (201), the simulation sample module (202) and the experimental actual measurement sample module (203) coexist in parallel;
wherein:
the physical parameters stored and managed in the basic physical parameter module (201) are obtained by actual measurement of a real object system of the high-voltage isolating switch or are preset and used for importing engineering drawing software (1011) and electric power simulation software (1021) to build and simulate a physical model or an asynchronous motor model;
the sample data stored and managed in the simulation sample module (202) is a sample result of a power-time signal of the driving motor, which is obtained by setting various mechanical states in the mechatronic joint simulation model established in the mechatronic debugging module (103) and repeatedly simulating;
the sample data stored and managed in the experimental actual measurement sample module (203) is a limited sample of the power-time signal of the driving motor in different mechanical states of the high-voltage isolating switch, which is obtained through an experimental or field actual measurement mode.
4. The debugging system for intelligent diagnosis of mechanical faults of high-voltage isolating switches according to claim 3, characterized in that: in the process of acquiring simulation sample data through the electromechanical integration debugging module (103), the mechanical state of the high-voltage isolating switch is simulated through constraint condition setting, wherein the simulated mechanical state comprises but is not limited to switching-off, switching-on, switching-off and various switching-on and switching-off states.
5. The debugging system for intelligent diagnosis of mechanical faults of high-voltage isolating switches according to claim 4, characterized in that: the specific method for acquiring the limited samples of the power-time signals of the driving motor of the high-voltage isolating switch in different mechanical states by the experimental actual measurement sample module (203) through an experimental or field actual measurement mode comprises the following steps:
actually measuring a power-time curve of a driving motor in the switching-on and switching-off process of the high-voltage isolating switch by building a high-voltage isolating switch experiment platform or in a transformer substation; sampling the mechanical state of the high-voltage isolating switch in the maintenance process to obtain small sample data of a power-time curve of the driving motor based on the actual opening and closing process of the high-voltage isolating switch; wherein the mechanical state based on the actual high voltage disconnector corresponds to the simulated mechanical state in the mechatronic commissioning module (103).
6. The debugging system for intelligent diagnosis of mechanical faults of high-voltage isolating switches according to claim 3, characterized in that: the deep training unit (300) comprises a deep neural network model (301), a trained multilayer model (302) and an intelligent diagnosis model (303); the signal output end of the deep neural network model (301) is connected with the signal input end of the trained multilayer model (302), and the signal output end of the trained multilayer model (302) is connected with the signal input end of the intelligent diagnosis model (303);
wherein:
the deep neural network model (301) is a multilayer neural network model established on the historical operating data of massive high-voltage isolating switches and used as a training basis for machine learning;
the trained multilayer model (302) is a new model which is output after the deep neural network model (301) is trained by utilizing a large amount of sample data stored and managed in the simulation sample module (202) and is used for intermediate transition;
the intelligent diagnosis model (303) is a high-precision intelligent diagnosis model of the mechanical state of the high-voltage isolating switch, which is obtained by training the last layers of the trained multilayer model (302) through the sample data stored and managed in the experimental actual measurement sample module (203), and is used for performing subsequent intelligent diagnosis operation of the mechanical state of the high-voltage isolating switch.
7. The debugging system for intelligent diagnosis of mechanical faults of high-voltage isolating switches according to claim 6, wherein: when the intelligent diagnosis model (303) is obtained by training the last layers of the trained multilayer model (302) through the sample data stored and managed in the experimental actual measurement sample module (203), the number of the trained model layers is determined according to the precision requirement of the intelligent diagnosis model for the mechanical state of the high-voltage disconnecting switch with high precision.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210987816.2A CN115392118A (en) | 2022-08-17 | 2022-08-17 | Debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210987816.2A CN115392118A (en) | 2022-08-17 | 2022-08-17 | Debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115392118A true CN115392118A (en) | 2022-11-25 |
Family
ID=84120604
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210987816.2A Pending CN115392118A (en) | 2022-08-17 | 2022-08-17 | Debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115392118A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116184182A (en) * | 2022-12-15 | 2023-05-30 | 国网安徽省电力有限公司电力科学研究院 | GIS isolating switch mechanical state identification method based on curve similarity |
-
2022
- 2022-08-17 CN CN202210987816.2A patent/CN115392118A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116184182A (en) * | 2022-12-15 | 2023-05-30 | 国网安徽省电力有限公司电力科学研究院 | GIS isolating switch mechanical state identification method based on curve similarity |
| CN116184182B (en) * | 2022-12-15 | 2024-04-12 | 国网安徽省电力有限公司电力科学研究院 | GIS isolating switch mechanical state identification method based on curve similarity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106295956B (en) | Reliability assessment method considering repairable multi-state complex characteristic of power supply system of nuclear power plant | |
| CN102411106B (en) | Fault monitoring method and device for power transformer | |
| CN108229049A (en) | The method that multistate system mission reliability modeling is carried out based on performance model | |
| CN103744424A (en) | Field device logic control test method | |
| CN109936113B (en) | Protection action intelligent diagnosis method and system based on random forest algorithm | |
| CN115392118A (en) | Debugging system for intelligent diagnosis of mechanical fault of high-voltage isolating switch | |
| CN105787161A (en) | Hierarchical modeling method for electric power simulation system | |
| Georgilakis et al. | A systematic stochastic petri net based methodology for transformer fault diagnosis and repair actions | |
| CN110611312B (en) | On-line simulation system and method for panoramic data of substation bay level | |
| CN113887606A (en) | Electronic equipment control system fault diagnosis method based on fault tree establishment | |
| CN104698935B (en) | A kind of method for optimizing protection supervisory equipment exploitation | |
| CN106802599A (en) | A kind of diagnosing fault of power transformer system based on expert database | |
| CN115248994A (en) | Method and system for predicting and managing faults of electromechanical static pressure servo mechanism | |
| CN110361609A (en) | Extra-high voltage equipment monitors system and method | |
| Niu et al. | The study of fault diagnosis the high-voltage circuit breaker based on neural network and expert system | |
| CN112345841A (en) | Kalman filtering fault detection method applied to coastal city alternating current charging pile | |
| CN216998283U (en) | Virtual-real combined debugging system for large coke oven mechanical equipment | |
| CN113801666B (en) | Virtual-real combined large-scale coke oven mechanical equipment debugging system and debugging method | |
| Niang et al. | Formal Verification for Validation of PSEEL's PLC Program | |
| NL2032669B1 (en) | Intelligent method for diagnosing mechanical fault of high-voltage isolating switch based on transfer learning | |
| Feng et al. | Reliability evaluation for distribution system based on probabilistic model checking | |
| CN113033821A (en) | Load identification machine learning system applied to algorithm core unit | |
| Hutcheson et al. | Function-based design of a spacecraft power system diagnostics testbed | |
| Manivannan et al. | Towards a realtime knowledge-based simulation system for diagnosing machine tool failure | |
| Han et al. | Implementation and verification of prognostics and health management system using a configurable system of systems architecture |
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 |