CN116400266A - Transformer fault detection method, device and medium based on digital twin model - Google Patents
Transformer fault detection method, device and medium based on digital twin model Download PDFInfo
- Publication number
- CN116400266A CN116400266A CN202310340240.5A CN202310340240A CN116400266A CN 116400266 A CN116400266 A CN 116400266A CN 202310340240 A CN202310340240 A CN 202310340240A CN 116400266 A CN116400266 A CN 116400266A
- Authority
- CN
- China
- Prior art keywords
- transformer
- fault detection
- model
- twin
- neural network
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 245
- 238000003062 neural network model Methods 0.000 claims abstract description 96
- 238000013507 mapping Methods 0.000 claims abstract description 59
- 230000008878 coupling Effects 0.000 claims abstract description 35
- 238000010168 coupling process Methods 0.000 claims abstract description 35
- 238000005859 coupling reaction Methods 0.000 claims abstract description 35
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims description 30
- 239000007787 solid Substances 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 description 6
- 230000035772 mutation Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000007789 gas Substances 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012850 discrimination method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/62—Testing of transformers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Protection Of Transformers (AREA)
Abstract
The invention discloses a transformer fault detection method, device and medium based on a digital twin model, which comprises the following steps: collecting operation data of the entity transformer and establishing a twin database; establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data; updating the operation data of the twin transformer according to the mapping relation; based on the updated operation data and according to a preset fault detection twin model, detecting faults of the entity transformer, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining fault threshold values of the entity transformer. According to the method and the device, the operation data are updated according to the mapping relation, and the fault of the transformer is detected according to the preset fault detection twin model based on the updated operation data, so that the detection precision of the fault of the transformer is improved.
Description
Technical Field
The invention belongs to the technical field of transformer fault detection, and particularly relates to a transformer fault detection method, device and medium based on a digital twin model.
Background
The daily demand of the user for the electricity consumption is larger, so that the user is guaranteed to also put forward higher requirements in the aspect of electricity consumption safety. The transformer is an important pivot for power transmission and transformation of a power system, and once the transformer fails, power failure can occur in a local or even large area, so that the fault detection of the transformer is particularly important. However, in the current method, the fault detection result may be inaccurate by the three-ratio method or the characteristic gas discrimination method.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a transformer fault detection method, device and medium based on a digital twin model, and aims to solve the technical problem of inaccurate transformer fault detection result.
The technical scheme of the invention is as follows:
a transformer fault detection method based on a digital twin model, the detection method comprising:
collecting operation data of the entity transformer and establishing a twin database;
establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data;
updating the operation data of the twin transformer according to the mapping relation;
and detecting the faults of the physical transformer based on the updated operation data and according to a preset fault detection twin model, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining the fault threshold value of the physical transformer.
The accuracy influence shadow is input to the fault detection neural network model as a parameter value, and an input layer of the fault detection neural network model is connected with an output layer of the coupling model.
The step of detecting the fault of the physical transformer based on the updated operation data and according to a preset fault detection twin model comprises the following steps:
analyzing the updated operation data to obtain a damaged value of the entity transformer;
performing iterative computation on the damaged value based on the coupling model;
when the iteration times reach a preset threshold, obtaining a fault threshold of the entity transformer;
generating a fault detection neural network model according to a preset fault detection neural network model, updated operation data and a precision influence factor;
and detecting the faults of the entity transformer according to the fault detection neural network model, the updated operation data and the fault threshold value.
The step of generating the fault detection neural network model according to the preset fault detection neural network model, the updated operation data and the precision influence factor comprises the following steps:
taking the updated partial operation data as a training set, and training a preset fault detection neural network model according to the training set;
determining a precision influence factor cluster according to the number of samples in the training set;
inputting the precision influence factor clusters into a preset algorithm for iterative computation to obtain the fitness value of each precision influence factor in the precision influence factor clusters;
inputting the fitness value into a trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model;
when the fault detection precision reaches the preset precision, obtaining a fault detection neural network model;
when the fault detection precision does not reach the preset precision and the iteration times do not reach the preset times, adjusting each precision influence factor;
and (3) inputting the adjusted precision influence factors into a preset algorithm for iterative calculation, obtaining the fitness value of each adjusted precision influence factor, returning to the step of inputting the fitness value into the trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model.
The step of adjusting each precision influence factor comprises the following steps:
determining a variation factor and a crossover factor based on the training set;
and adjusting each precision influence factor according to the variation factors and the crossing factors.
The step of analyzing the updated operation data to obtain the damaged value of the physical transformer comprises the following steps:
analyzing the electromagnetic distribution of the physical transformer according to the running data after going more;
determining the magnetic leakage area according to the analysis result, and determining the damaged density of the solid transformer according to the magnetic leakage area;
and determining damage values of all parts of the physical transformer according to the damage density.
The step of establishing the mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data comprises the following steps:
storing the operational data to a twin database;
dividing the operation data into available data, data to be analyzed and data to be calculated;
establishing a first mapping relation according to the available data, and establishing a second mapping relation according to the data to be analyzed;
and establishing a third mapping relation through the first mapping relation, the second mapping relation and the data to be calculated.
A digital twin model-based transformer fault detection device, the digital twin model-based transformer fault detection device comprising: the system comprises a data acquisition module, a mapping establishment module, a data updating module and a fault detection module;
the data acquisition module is used for acquiring the operation data of the entity transformer and establishing a twin database;
the mapping establishing module is used for establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data;
the data updating module is used for updating the operation data of the twin transformer according to the mapping relation;
the fault detection module is used for detecting faults of the entity transformer based on updated operation data and according to a preset fault detection twin model, the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of transformer fault detection, and the coupling model is used for determining a fault threshold value of the entity transformer.
A digital twinning model-based transformer fault detection apparatus, the digital twinning model-based transformer fault detection apparatus comprising: the method comprises the steps of a memory, a processor and a digital twin model-based transformer fault detection program which is stored in the memory and can run on the processor, wherein the digital twin model-based transformer fault detection program is executed by the processor to realize the digital twin model-based transformer fault detection method.
A storage medium, on which a digital twin model-based transformer fault detection program is stored, which when executed by a processor implements the steps of the digital twin model-based transformer fault detection method.
The invention has the beneficial effects that:
the invention discloses a transformer fault detection method, device and medium based on a digital twin model, wherein the method comprises the following steps: collecting operation data of the entity transformer and establishing a twin database; establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data; updating the operation data of the twin transformer according to the mapping relation; and detecting the faults of the entity transformer based on the updated operation data and according to a preset fault detection twin model, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining the fault threshold value of the entity transformer. The invention establishes the mapping relation between the entity transformer and the twin transformer based on the twin database, updates the operation data of the twin transformer according to the mapping relation, and detects the faults of the transformer according to the preset fault detection twin model based on the updated operation data, thereby improving the detection precision of the faults of the transformer.
Drawings
FIG. 1 is a schematic diagram of a transformer fault detection device based on a digital twin model of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a first embodiment of a transformer fault detection method based on a digital twin model according to the present invention;
FIG. 3 is a schematic flow chart of a second embodiment of a transformer fault detection method based on a digital twin model according to the present invention;
FIG. 4 is a schematic flow chart of a third embodiment of a transformer fault detection method based on a digital twin model according to the present invention;
fig. 5 is a block diagram of a first embodiment of a transformer fault detection device based on a digital twin model according to the present invention.
Detailed Description
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a transformer fault detection device based on a digital twin model in a hardware operation environment according to an embodiment of the present invention.
As shown in fig. 1, the digital twin model-based transformer fault detection apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display (Display), and the optional user interface 1003 may also include a standard wired interface, a wireless interface, and the wired interface for the user interface 1003 may be a USB interface in the present invention. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
It will be appreciated by those skilled in the art that the structure shown in fig. 1 does not constitute a limitation of a digital twinning model based transformer fault detection apparatus, and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a transformer fault detection program based on a digital twin model may be included in a memory 1005, which is considered to be a type of computer storage medium.
In the transformer fault detection device based on the digital twin model shown in fig. 1, the network interface 1004 is mainly used for connecting a background server, and performing data communication with the background server; the user interface 1003 is mainly used for connecting user equipment; the transformer fault detection device based on the digital twin model calls a transformer fault detection program based on the digital twin model stored in the memory 1005 through the processor 1001, and executes the transformer fault detection method based on the digital twin model provided by the embodiment of the invention.
Based on the hardware structure, the embodiment of the transformer fault detection method based on the digital twin model is provided.
Referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of a transformer fault detection method based on a digital twin model according to the present invention, and the first embodiment of the transformer fault detection method based on the digital twin model according to the present invention is provided.
Step S10: and collecting operation data of the entity transformer and establishing a twin database.
It should be noted that, the execution body of the embodiment may be a computer software service device with functions of data processing, network communication and program running, for example, transformer fault detection based on a digital twin model, or other electronic devices capable of implementing the same or similar functions, which is not limited in this embodiment.
It should be noted that, the operation data of the physical transformer may be collected by the sensor, or may be collected by the transformer operation monitoring system, which is not limited in this embodiment.
It is understood that the operation data of the physical transformer may be a temperature value, a humidity value, a voltage, a current, etc. of the transformer, which is not limited in this embodiment.
It should be noted that, the twin database may update the data of the physical transformer to the twin transformer in real time, so that the operation states of the physical transformer and the twin transformer are synchronized, that is, the twin database stores the real-time operation data of the physical transformer, establishes a mapping relationship between the physical transformer and the twin transformer, and updates the operation data stored by the physical transformer to the twin transformer.
Step S20: and establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data.
It should be noted that, the twin transformers are encapsulated in a preset fault detection twin model, the twin transformers are built by modeling software based on the solid transformers, a transformer box body, an oil storage cabinet, a radiator, a fan and the like are respectively built, the components are combined and assembled to form the twin transformers, the twin transformers are driven by operation data of the solid transformers in the twin database, so that the twin transformers simulate the operation of the solid transformers and gradually tend to the operation state of the solid transformers.
Further, in order to improve the accuracy of detecting the transformer fault, step S20 of this embodiment may include:
storing the operational data to the twin database;
dividing the operation data into available data, data to be analyzed and data to be calculated;
establishing a first mapping relation according to the available data, and establishing a second mapping relation according to the data to be analyzed;
and establishing a third mapping relation through the first mapping relation, the second mapping relation and the data to be calculated.
It should be noted that, for example, the temperature value, the humidity value, and the like can be directly detected by the sensor device or the transformer operation system to obtain the available data, which does not need any calculation; the data to be analyzed is the data which can be obtained only by performing analysis and calculation, for example, the instantaneous power of the transformer, the instantaneous power=the instantaneous voltage and the instantaneous current; the data to be calculated needs to be obtained by analysis and reasoning.
It should be noted that, for example, the load attribute of the transformer needs to directly obtain current and voltage data of the transformer according to the available data, and the current and voltage data are calculated to obtain the instantaneous power of the transformer, and then whether the complexity of the transformer is overloaded is determined according to the instantaneous power of the transformer, and a third mapping relationship is established according to the determination result.
It can be understood that the mapping relation is established according to the physical transformer, namely, the object attribute and the data attribute are established, namely, the attribute of each component of the physical transformer is associated with the corresponding data.
Step S30: and updating the operation data of the twin transformer according to the mapping relation.
It can be understood that the physical transformer may change due to various factors in the operation process, so that the operation data of the physical transformer under various conditions can be synchronized to the twin transformer through the mapping relationship, and the fault of the physical transformer is detected according to the operation data updated by the twin transformer.
Step S40: and detecting the faults of the physical transformer based on the updated operation data and according to a preset fault detection twin model, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining the fault threshold value of the physical transformer.
It should be noted that, the precision influence factor is generally given by experience of the evidence, and the precision and the accuracy cannot meet the requirements, so that the precision of the fault detection result of the transformer is not high at last.
It should be noted that, the coupling model is used to determine a fault threshold of each component of the solid transformer, for example, the coupling model obtains a loss density of each component through electromagnetic analysis, performs coupling iteration according to the loss density to obtain a convective heat transfer coefficient of each component surface of the solid transformer, analyzes according to the convective heat transfer coefficient to obtain a temperature field distribution of the solid transformer, and determines a temperature threshold of each component of the solid transformer according to the temperature field distribution.
Further, in order to improve the fault detection accuracy of the transformer, step S40 of this embodiment may include:
and the precision influence shadow is used as a parameter value to be input into the fault detection neural network model, and an input layer of the fault detection neural network model is connected with an output layer of the coupling model.
The method comprises the steps of obtaining a model of a fault detection neural network, wherein the model of the fault detection neural network comprises an input layer, a mode layer, a summation layer and an output layer, the input layer is the value of an input training sample, the number of neurons in the input layer is the same as the dimension of a sample vector, the mode layer is used for calculating the matching relation between the input characteristic vector of the input layer and a variable in a training set, the summation layer is used for accumulating the probability of the fault to obtain a probability density function of the fault, and the output layer is used for outputting a fault estimated value output by the summation layer and carrying out normalization processing.
It should be noted that, the fault threshold value output by the coupling model is used to calculate at the summing layer in the fault detection neural network model, and determine whether the physical transformer reaches the fault threshold value.
The method comprises the steps of collecting operation data of an entity transformer and establishing a twin database; establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data; updating the operation data of the twin transformer according to the mapping relation; and detecting the faults of the entity transformer based on the updated operation data and according to a preset fault detection twin model, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining the fault threshold value of the entity transformer. In the embodiment, the mapping relation between the entity transformer and the twin transformer is established based on the twin database, the operation data of the twin transformer is updated according to the mapping relation, and the fault of the transformer is detected according to the preset fault detection twin model based on the updated operation data, so that the detection precision of the fault of the transformer is improved.
Referring to fig. 3, fig. 3 is a schematic flow chart of a second embodiment of the method for detecting a transformer fault based on a digital twin model according to the present invention, and the second embodiment of the method for detecting a transformer fault based on a digital twin model according to the present invention is proposed based on the first embodiment shown in fig. 2.
In a second embodiment, the step S40 includes:
step S401: and analyzing the updated operation data to obtain the damage value of the entity transformer.
For example, when the transformer is operated, short-circuit impedance is generated due to magnetic leakage, the updated data is analyzed by finite element software, and the short-circuit impedance calculated by the finite element method is a damaged value.
Further, in order to improve the fault detection accuracy, step S401 of the present embodiment may include:
analyzing the electromagnetic distribution of the physical transformer according to the running data after going more;
determining a magnetic leakage area according to an analysis result, and determining the damaged density of the solid transformer according to the magnetic leakage area;
and determining damage values of all parts of the solid transformer according to the damage density.
The electromagnetic distribution and the damage density of each component of the solid transformer can be determined according to finite element analysis software, and the damage density is multiplied by the oil tank volume to obtain the damage value of each component of the solid transformer.
Step S402: and carrying out iterative calculation on the damage value based on the coupling model.
It can be understood that the damaged value of the physical transformer does not completely affect the operation of the physical transformer, so that the damaged value is iteratively calculated to obtain a threshold value of the physical transformer for failure, and when the damaged value of the physical transformer reaches the threshold value, the physical transformer has failed.
It should be noted that, for example, coupling iteration is performed by using the damaged value as a heat source, so as to obtain a convection heat exchange coefficient on each contact surface of the solid transformer, obtain a temperature field distribution according to the convection heat exchange coefficient, compare the adjacent temperature distribution difference value with a preset difference value, and if the adjacent temperature distribution difference value is greater than the preset difference value, continue the coupling iteration.
Step S403: and when the iteration times reach a preset threshold, obtaining a fault threshold of the entity transformer.
For example, the iteration condition is that the adjacent temperature distribution difference value is smaller than a preset threshold value, when the adjacent temperature distribution difference value is larger than the preset threshold value, the coupling iteration is continued, and when the adjacent temperature distribution difference value is smaller than the preset threshold value, the fault threshold value of the solid transformer is obtained.
It is understood that the preset threshold may be 0.02K, which may be set according to the requirement, which is not limited in this embodiment.
Step S404: and generating a fault detection neural network model according to a preset fault detection neural network model, the updated operation data and the precision influence factor.
It can be understood that the updated operation data and the precision influence factor are used as the input of the preset fault detection neural network model, and the preset fault detection neural network model is trained to obtain the fault detection neural network model.
Step S405: and detecting the faults of the entity transformer according to the fault detection neural network model, the updated operation data and the fault threshold value.
It can be understood that the fault detection neural network model performs fault detection on the physical transformer according to the updated operation data, and indicates that the physical transformer fails when the fault detection value reaches the fault threshold.
The embodiment analyzes the updated operation data to obtain the damaged value of the entity transformer; performing iterative computation on the damage value based on the coupling model; when the iteration times reach a preset threshold, obtaining a fault threshold of the entity transformer; generating a fault detection neural network model according to a preset fault detection neural network model, the updated operation data and the precision influence factor; and detecting the faults of the entity transformer according to the fault detection neural network model, the updated operation data and the fault threshold value. According to the embodiment, through iterative calculation of the damaged value, the fault threshold value of the entity transformer is obtained when the fault threshold value reaches the preset threshold value, the preset fault detection neural network model is trained according to the updated operation data and the precision influence factor to generate the fault detection neural network model, and the entity transformer is subjected to fault detection according to the fault detection neural network model, so that the fault detection precision of the entity transformer is improved.
Referring to fig. 4, fig. 4 is a schematic flow chart of a third embodiment of the method for detecting a transformer fault based on a digital twin model according to the present invention, and the third embodiment of the method for detecting a transformer fault based on a digital twin model according to the present invention is proposed based on the second embodiment shown in fig. 3.
In a third embodiment, the step of generating the fault detection neural network model according to the preset fault detection neural network model, the updated operation data, and the precision influence factor includes:
step S4041: and taking the updated partial operation data as a training set, and training a preset fault detection neural network model according to the training set.
It can be understood that the updated part of operation data is used as a training set, fault calculation is performed on data in the training set, and the preset fault detection neural network is trained according to the calculation result, for example, multiple gases are generated in the oil immersed transformer, and the content of gas components generated by the faults of the transformer is different, so that the percentage of the total content of the fault group is calculated according to the training set, and the preset fault detection neural network model is trained according to the calculated percentage.
Step S4042: and determining a precision influence factor cluster according to the number of samples in the training set.
It will be appreciated that the precision influencing factor cluster includes a plurality of precision influencing factors, and the number of samples in the training set may be set according to the requirement, which is not limited in this embodiment.
Step S4043: and inputting the precision influence factor cluster into a preset algorithm for iterative computation to obtain the fitness value of each precision influence factor in the precision influence factor cluster.
The fitness value of the precision influence factor is calculated according to a preset algorithm, the fitness values of the precision influence factors are compared, and the fitness value of the proper precision influence factor is reserved.
Step S4044: and inputting the fitness value into a trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model.
It can be understood that the fitness value is input into the trained preset fault detection neural network model, then fault detection is performed on the entity transformer, a fault detection result is obtained, if the detection result meets the preset precision, the training of the preset fault detection neural network model is completed, and if the fault detection result does not meet the preset precision, the training needs to be continued.
Step S4045: and when the fault detection precision reaches the preset precision, obtaining a fault detection neural network model.
It can be understood that the iteration is stopped when the fault detection accuracy reaches the preset accuracy, and the fault detection neural network model is obtained.
Step S4046: and when the fault detection precision does not reach the preset precision and the iteration times do not reach the preset times, adjusting the precision influence factors.
The adjustment of each precision influence factor may be a mutation operation or a crossover operation of each precision influence factor.
Further, in order to improve the accuracy of the failure detection neural network model, step S4046 of the present embodiment may include:
determining a variation factor and a crossover factor based on the training set;
and adjusting each precision influence factor according to the variation factor and the crossing factor.
The value of the mutation factor is generally in the range of [0.3 to 0.6 ]]The value of the cross factor is generally between 0.6 and 0.9]Between, according to K (i+1) =x 1 (i)+Z(x 2 (t)-x 3 (t)) performing mutation operation to adjust each precision influence factor, wherein K (i+1) represents mutation operation of the next iteration, and x 1 (i)、x 2 (i) And x 3 (i) The precision influence factor of the current iteration is represented, and Z represents a variation factor; according to
Adjusting each precision influence factor, wherein M (i+1) represents the cross operation of the next iteration, and randl represents [0,1 ]]Random decimal numbers in between, LP represents the cross factor, x (i) representsAny accuracy impact factor of the current iteration.
Step S4047: and inputting the adjusted precision influence factors into the preset algorithm for iterative calculation, obtaining the fitness value of each adjusted precision influence factor, returning to the step of inputting the fitness value into the trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model.
It can be understood that the fitness value of each precision influence factor after adjustment is recalculated, the fitness value is input to a trained preset fault detection neural network model, fault detection is performed on the physical transformer according to the trained preset fault detection neural network model, whether the fault detection precision reaches the preset precision is judged, the fault detection neural network model is obtained after the preset precision is reached, the preset precision is not reached, the iteration times are not reached, and the precision influence factors are adjusted again.
In the embodiment, the updated part of operation data is used as a training set, and a preset fault detection neural network model is trained according to the training set; determining a precision influence factor cluster according to the number of samples in the training set; inputting the precision influence factor cluster into a preset algorithm for iterative computation to obtain the fitness value of each precision influence factor in the precision influence factor cluster; inputting the fitness value into a trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model; when the fault detection precision reaches the preset precision, obtaining a fault detection neural network model; when the fault detection precision does not reach the preset precision and the iteration times do not reach the preset times, adjusting each precision influence factor; and inputting the adjusted precision influence factors into the preset algorithm for iterative calculation, obtaining the fitness value of each adjusted precision influence factor, returning to the step of inputting the fitness value into the trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model. According to the method, the fitness value of the precision influence factor is determined according to a preset algorithm, the fitness value is input into a trained preset fault detection neural network model to perform fault detection on the entity transformer, when the preset precision is not met, the fitness value of the precision influence factor is readjusted, the fitness value is input into the trained preset fault detection neural network model to perform fault detection, and the precision influence factor is continuously screened, so that the precision of the fault detection neural network model is improved.
In addition, referring to fig. 5, an embodiment of the present invention further provides a transformer fault detection device based on a digital twin model, where the transformer fault detection device based on the digital twin model includes: a data acquisition module 10, a mapping establishment module 20, a data update module 30, and a fault detection module 40;
the data acquisition module 10 is used for acquiring the operation data of the entity transformer and establishing a twin database;
the mapping establishing module 20 is configured to establish a mapping relationship between the physical transformer and the twin transformer based on the twin database and according to the operation data;
the data updating module 30 is configured to update operation data of the twin transformer according to the mapping relationship;
the fault detection module 40 is configured to detect a fault of the physical transformer based on the updated operation data and according to a preset fault detection twin model, where the preset fault detection twin model includes a precision influence shadow, a coupling model, and a fault detection neural network model, where the precision influence shadow and the fault detection neural network model are used to determine a precision of fault detection of the transformer, and the coupling model is used to determine a fault threshold of the physical transformer.
The method comprises the steps of collecting operation data of an entity transformer and establishing a twin database; establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data; updating the operation data of the twin transformer according to the mapping relation; and detecting the faults of the entity transformer based on the updated operation data and according to a preset fault detection twin model, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining the fault threshold value of the entity transformer. In the embodiment, the mapping relation between the entity transformer and the twin transformer is established based on the twin database, the operation data of the twin transformer is updated according to the mapping relation, and the fault of the transformer is detected according to the preset fault detection twin model based on the updated operation data, so that the detection precision of the fault of the transformer is improved.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a transformer fault detection program based on a digital twin model, and the transformer fault detection program based on the digital twin model realizes the transformer fault detection method based on the digital twin model when being executed by a processor.
Based on the first embodiment of the transformer fault detection device based on the digital twin model, a second embodiment of the transformer fault detection device based on the digital twin model is provided.
In this embodiment, the fault detection module 40 is configured to analyze the updated operation data to obtain the damaged value of the physical transformer.
Further, the fault detection module 40 is further configured to perform iterative calculation on the damage value based on the coupling model.
Further, the fault detection module 40 is further configured to obtain a fault threshold of the physical transformer when the iteration number reaches a preset threshold.
Further, the fault detection module 40 is further configured to generate a fault detection neural network model according to a preset fault detection neural network model, the updated operation data, and the precision influence factor.
Further, the fault detection module 40 is further configured to detect a fault of the physical transformer according to the fault detection neural network model, the updated operation data and the fault threshold.
Further, the fault detection module 40 is further configured to use the updated part of the operation data as a training set, and train a preset fault detection neural network model according to the training set.
Further, the fault detection module 40 is further configured to determine a precision impact factor cluster according to the number of samples in the training set.
Further, the fault detection module 40 is further configured to input the precision influence factor cluster to a preset algorithm for performing iterative computation, so as to obtain fitness values of precision influence factors in the precision influence factor cluster.
Further, the fault detection module 40 is further configured to input the fitness value to a trained preset fault detection neural network model, and perform fault detection on the physical transformer according to the trained preset fault detection neural network model.
Further, the fault detection module 40 is further configured to obtain a fault detection neural network model when the fault detection precision reaches a preset precision.
Further, the fault detection module 40 is further configured to adjust each precision influence factor when the fault detection precision does not reach the preset precision and the iteration number does not reach the preset number.
Further, the fault detection module 40 is further configured to input each adjusted precision influence factor to the preset algorithm for performing iterative computation, obtain an fitness value of each adjusted precision influence factor, return to input the fitness value to a trained preset fault detection neural network model, and perform fault detection on the physical transformer according to the trained preset fault detection neural network model.
Further, the fault detection module 40 is further configured to determine a mutation factor and a crossover factor based on the training set.
Further, the fault detection module 40 is further configured to adjust each precision influence factor according to the mutation factor and the crossover factor.
Further, the fault detection module 40 is further configured to analyze the electromagnetic distribution of the physical transformer according to the more advanced operation data.
Further, the fault detection module 40 is further configured to determine a magnetic leakage area according to the analysis result, and determine the damaged density distribution of the physical transformer according to the magnetic leakage area.
Further, the fault detection module 40 is further configured to determine a damage value of each component of the physical transformer according to the damage density distribution.
Further, the data acquisition module 10 is further configured to store the operation data to the twin database.
Further, the data acquisition module 10 is further configured to divide the operation data into available data, data to be analyzed and data to be calculated.
Further, the data obtaining module 10 is further configured to establish a first mapping relationship according to the available data, and establish a second mapping relationship according to the data to be analyzed.
Further, the data obtaining module 10 is further configured to establish a third mapping relationship through the first mapping relationship and the second mapping relationship.
Other embodiments or specific implementation manners of the transformer fault detection device based on the digital twin model according to the present invention may refer to the above method embodiments, and will not be described herein.
Claims (10)
1. The transformer fault detection method based on the digital twin model is characterized by comprising the following steps of:
collecting operation data of the entity transformer and establishing a twin database;
establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data;
updating the operation data of the twin transformer according to the mapping relation;
and detecting the faults of the physical transformer based on the updated operation data and according to a preset fault detection twin model, wherein the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of fault detection of the transformer, and the coupling model is used for determining the fault threshold value of the physical transformer.
2. The method for detecting the faults of the transformer based on the digital twin model according to claim 1, wherein the influence shadow of the precision is input into a fault detection neural network model as a parameter value, and an input layer of the fault detection neural network model is connected with an output layer of a coupling model.
3. The method for detecting a fault of a transformer based on a digital twin model according to claim 1, wherein the step of detecting a fault of a physical transformer based on updated operation data and according to a preset fault detection twin model comprises:
analyzing the updated operation data to obtain a damaged value of the entity transformer;
performing iterative computation on the damaged value based on the coupling model;
when the iteration times reach a preset threshold, obtaining a fault threshold of the entity transformer;
generating a fault detection neural network model according to a preset fault detection neural network model, updated operation data and a precision influence factor;
and detecting the faults of the entity transformer according to the fault detection neural network model, the updated operation data and the fault threshold value.
4. A method for detecting a fault in a transformer based on a digital twin model as defined in claim 3, wherein the step of generating the fault detection neural network model based on the preset fault detection neural network model, the updated operation data and the accuracy influencing factor comprises:
taking the updated partial operation data as a training set, and training a preset fault detection neural network model according to the training set;
determining a precision influence factor cluster according to the number of samples in the training set;
inputting the precision influence factor clusters into a preset algorithm for iterative computation to obtain the fitness value of each precision influence factor in the precision influence factor clusters;
inputting the fitness value into a trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model;
when the fault detection precision reaches the preset precision, obtaining a fault detection neural network model;
when the fault detection precision does not reach the preset precision and the iteration times do not reach the preset times, adjusting each precision influence factor;
and (3) inputting the adjusted precision influence factors into a preset algorithm for iterative calculation, obtaining the fitness value of each adjusted precision influence factor, returning to the step of inputting the fitness value into the trained preset fault detection neural network model, and carrying out fault detection on the entity transformer according to the trained preset fault detection neural network model.
5. The method for detecting a transformer fault based on a digital twin model as defined in claim 4, wherein the step of adjusting each precision influence factor comprises:
determining a variation factor and a crossover factor based on the training set;
and adjusting each precision influence factor according to the variation factors and the crossing factors.
6. A method for detecting a transformer fault based on a digital twin model as defined in claim 3, wherein the step of analyzing the updated operation data to obtain a damage value of the physical transformer comprises:
analyzing the electromagnetic distribution of the physical transformer according to the running data after going more;
determining the magnetic leakage area according to the analysis result, and determining the damaged density of the solid transformer according to the magnetic leakage area;
and determining damage values of all parts of the physical transformer according to the damage density.
7. The method for detecting a transformer fault based on a digital twin model according to claim 1, wherein the step of establishing a mapping relationship between the physical transformer and the twin transformer based on the twin database and according to the operation data comprises:
storing the operational data to a twin database;
dividing the operation data into available data, data to be analyzed and data to be calculated;
establishing a first mapping relation according to the available data, and establishing a second mapping relation according to the data to be analyzed;
and establishing a third mapping relation through the first mapping relation, the second mapping relation and the data to be calculated.
8. A digital twin model-based transformer fault detection device, characterized in that the digital twin model-based transformer fault detection device comprises: the system comprises a data acquisition module, a mapping establishment module, a data updating module and a fault detection module;
the data acquisition module is used for acquiring the operation data of the entity transformer and establishing a twin database;
the mapping establishing module is used for establishing a mapping relation between the entity transformer and the twin transformer based on the twin database and according to the operation data;
the data updating module is used for updating the operation data of the twin transformer according to the mapping relation;
the fault detection module is used for detecting faults of the entity transformer based on updated operation data and according to a preset fault detection twin model, the preset fault detection twin model comprises a precision influence shadow, a coupling model and a fault detection neural network model, the precision influence shadow and the fault detection neural network model are used for determining the precision of transformer fault detection, and the coupling model is used for determining a fault threshold value of the entity transformer.
9. A digital twin model-based transformer fault detection apparatus, characterized in that the digital twin model-based transformer fault detection apparatus comprises: a memory, a processor and a digital twinning model based transformer fault detection program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the digital twinning model based transformer fault detection method according to any one of claims 1 to 7.
10. A storage medium, characterized in that the storage medium has stored thereon a digital twin model based transformer fault detection program, which when executed by a processor, implements the steps of the digital twin model based transformer fault detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310340240.5A CN116400266A (en) | 2023-03-31 | 2023-03-31 | Transformer fault detection method, device and medium based on digital twin model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310340240.5A CN116400266A (en) | 2023-03-31 | 2023-03-31 | Transformer fault detection method, device and medium based on digital twin model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116400266A true CN116400266A (en) | 2023-07-07 |
Family
ID=87009775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310340240.5A Pending CN116400266A (en) | 2023-03-31 | 2023-03-31 | Transformer fault detection method, device and medium based on digital twin model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116400266A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117347917A (en) * | 2023-10-09 | 2024-01-05 | 国网黑龙江省电力有限公司大庆供电公司 | Power transmission transformer thermal fault detection system |
-
2023
- 2023-03-31 CN CN202310340240.5A patent/CN116400266A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117347917A (en) * | 2023-10-09 | 2024-01-05 | 国网黑龙江省电力有限公司大庆供电公司 | Power transmission transformer thermal fault detection system |
| CN117347917B (en) * | 2023-10-09 | 2024-05-03 | 国网黑龙江省电力有限公司大庆供电公司 | Power transmission transformer thermal fault detection system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116031888B (en) | Dynamic load prediction-based power flow optimization method, system and storage medium | |
| CN116992399B (en) | Power equipment operation and maintenance assessment method based on power data analysis | |
| CN112990330A (en) | User energy abnormal data detection method and device | |
| CN114386537A (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| CN116400266A (en) | Transformer fault detection method, device and medium based on digital twin model | |
| CN114004162A (en) | Modeling method for smelting load harmonic emission level under multi-working-condition scene | |
| CN115392037A (en) | Equipment fault prediction method, device, equipment and storage medium | |
| CN112381673A (en) | Park electricity utilization information analysis method and device based on digital twin | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| CN101206727B (en) | Data processing apparatus, data processing method | |
| CN114862229A (en) | Power quality evaluation method and device, computer equipment and storage medium | |
| CN107274025B (en) | System and method for realizing intelligent identification and management of power consumption mode | |
| CN113722860A (en) | Transient thermodynamic state online evaluation method, device and medium based on reduced order model | |
| CN112287605A (en) | Flow check method based on graph convolution network acceleration | |
| CN117113086A (en) | Energy storage unit load prediction method, system, electronic equipment and medium | |
| US20220243347A1 (en) | Determination method and determination apparatus for conversion efficiency of hydrogen production by wind-solar hybrid electrolysis of water | |
| CN110991741A (en) | Section constraint probability early warning method and system based on deep learning | |
| CN116307844A (en) | Low-voltage transformer area line loss evaluation analysis method | |
| CN115409317A (en) | Transformer area line loss detection method and device based on feature selection and machine learning | |
| CN114692729A (en) | New energy station bad data identification and correction method based on deep learning | |
| CN113406537A (en) | Quantitative evaluation method for fault degree of power equipment | |
| CN112417794B (en) | Scattering parameter calculation method | |
| CN115498645B (en) | Method and system for reducing dangerous resonance between system impedances | |
| CN113642601B (en) | Medium voltage distribution network transfer operation identification method, device and equipment | |
| CN116595883B (en) | Real-time online system state correction method for numerical reactor |
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 |