CN113656938A - Monitoring method, control device and storage medium for distribution network transformer - Google Patents
Monitoring method, control device and storage medium for distribution network transformer Download PDFInfo
- Publication number
- CN113656938A CN113656938A CN202110767526.2A CN202110767526A CN113656938A CN 113656938 A CN113656938 A CN 113656938A CN 202110767526 A CN202110767526 A CN 202110767526A CN 113656938 A CN113656938 A CN 113656938A
- Authority
- CN
- China
- Prior art keywords
- transformer
- prediction model
- data
- distribution network
- fault
- 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
- 238000009826 distribution Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012544 monitoring process Methods 0.000 title claims abstract description 28
- 238000003860 storage Methods 0.000 title claims abstract description 20
- 238000003745 diagnosis Methods 0.000 claims abstract description 64
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims description 44
- 238000007726 management method Methods 0.000 claims description 36
- 230000006698 induction Effects 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 9
- 230000015556 catabolic process Effects 0.000 claims description 8
- 238000006731 degradation reaction Methods 0.000 claims description 8
- 230000004927 fusion Effects 0.000 claims description 7
- 238000004088 simulation Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 12
- 230000036541 health Effects 0.000 abstract description 4
- 238000009472 formulation Methods 0.000 abstract description 2
- 239000000203 mixture Substances 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000006978 adaptation Effects 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004587 chromatography analysis Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The embodiment of the invention provides a monitoring method, a control device and a storage medium for a power distribution network transformer, and belongs to the technical field of power distribution networks. The monitoring method of the distribution network transformer comprises the following steps: acquiring characteristic data of the transformer, wherein the characteristic data comprises operation data and/or fault data of the transformer; training a preset fault diagnosis and management prediction model through the acquired characteristic data; and monitoring the transformer in real time through the trained preset fault diagnosis and management prediction model. The embodiment of the invention can be used for predicting and diagnosing two situations based on the fault diagnosis and management prediction model, can evaluate the actual state of the transformer equipment, guides the formulation of maintenance strategies and improves the health level of the equipment; and the fault diagnosis and management prediction model is fused and constructed on the basis of a digital twin technology at the intelligent terminal, so that the high-efficiency control can be realized, the monitoring efficiency can be improved, and the operation cost can be reduced.
Description
Technical Field
The invention relates to the technical field of power distribution networks, in particular to a supervision method, a control device and a storage medium of a power distribution network transformer.
Background
The power distribution network has the characteristics of multiple voltage levels, complex network structure, various equipment types, multiple and wide operation points, relatively poor safety environment and the like, so that the safety risk factors of the power distribution network are relatively more. In recent years, the construction investment of the power distribution network in China is continuously increased, the development of the power distribution network has remarkable effect, and with the increase of the capacity of power equipment and the enlargement of the scale of the power distribution network, power energy is provided for various users, so that higher requirements on the safe operation and the power supply reliability of a power system are provided.
The distribution transformer is a large number of distribution equipment which is widely applied, the working state of the distribution transformer directly affects the electricity quality of users, and the position of the transformer in an electric power system is very important and is one of the most important and expensive electric equipment in the electric power system. Therefore, the improvement of the operation reliability of the transformer, especially the large-scale power transformer, has very important significance for the safe and reliable operation of the whole power grid.
At present, the management of the transformer is mainly based on a temperature monitoring mode, the transformer is controlled within a reasonable working temperature range, and in addition, the periodic electrification detection and fault diagnosis of the transformer comprise a vibration analysis method, an oil chromatography analysis and the like. The disadvantages for the management and maintenance of the transformer are: management and maintenance mainly comprise three methods based on historical statistical probability, sensor data drive and a physical model, but the methods have limitations and defects, for example, the historical statistical probability has certain randomness; the prediction precision is poor for more real-time states based on the sensing data driving model; based on the problems of poor fidelity of the physical model, low effective utilization rate of data and the like. With the mass adoption of integrated, low maintenance equipment, early custom equipment overhauls, test cycles have not been able to accommodate advances in the level of equipment diagnosis and management.
Disclosure of Invention
The embodiment of the invention aims to provide a monitoring method for a distribution network transformer, which can comprehensively monitor the full life cycle of the distribution network transformer.
In order to achieve the above object, an embodiment of the present invention provides a method for supervising a distribution network transformer, where the method for supervising a distribution network transformer includes: acquiring characteristic data of the transformer, wherein the characteristic data comprises operation data and/or fault data of the transformer; training a preset fault diagnosis and management prediction model through the acquired characteristic data; and monitoring the transformer in real time through the trained preset fault diagnosis and management prediction model.
Optionally, the acquiring the characteristic data of the transformer includes: acquiring an induction signal of a transformer; filtering and denoising the induction signal; and extracting the characteristics of the filtered and noise-reduced induction signals.
Optionally, before the training of the preset fault diagnosis and management prediction model, the method for supervising the distribution network transformer further includes: performing virtual space mapping on the characteristic data through a digital twinning technology; and respectively constructing a fault alarm diagnosis model and a performance analysis prediction model.
Optionally, the training of the preset fault diagnosis and management prediction model includes: training the fault alarm diagnosis model and the performance analysis prediction model by taking the characteristic data as input; and fitting the trained fault alarm diagnosis model and the performance analysis prediction model.
Optionally, the fault alarm diagnosis model is constructed by a classification random validation algorithm, and the training of the fault alarm diagnosis model includes: acquiring corresponding state variables aiming at the fault data in the characteristic data, and establishing a random fault sample, wherein the state variables comprise a normal state, an early warning state and an alarm state; and training the fault alarm diagnosis model by taking the random fault sample as input.
Optionally, the performance analysis prediction model is constructed by a data fusion driving algorithm, and the training of the performance analysis prediction model includes: acquiring corresponding simulation induction data showing the performance degradation of the transformer aiming at the operation data in the characteristic data, and establishing an operation state sample; and training the performance analysis prediction model by taking the running state sample as input.
Optionally, the simulation induction data showing the performance degradation of the transformer is a time sequence, and the training of the performance analysis prediction model includes: predicting the output of the next sequence m by taking the first m-1 sequence in the time sequence as regression input; wherein the expected value of error between the predicted output and the acquired signal corresponding to the actual operating data is minimized by a minimum cost function.
Optionally, the minimum cost function is represented by:
wherein, E2]Representing performance index expectation, Y () representing the output predicted by the performance analysis prediction model, S () representing the time series, ξS() Is Gaussian white noise sequence, m represents the number of time sequences, k represents the prediction step length, nyIndicating the order of the output.
Optionally, the expected error value is set based on a lightweight boundary expected value of a delay correction operator Z, where the delay correction operator is represented by the following equation:
where m represents the number of time series, n represents the number of types in the time series, Q () represents the minimum cost function, Y represents the output predicted by the performance analysis prediction model, and S represents the input of the performance analysis prediction model.
An embodiment of the present invention further provides a control device, where the control device includes: the monitoring system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the monitoring method of the distribution network transformer.
The embodiment of the invention also provides a machine-readable storage medium, wherein the machine-readable storage medium stores instructions, and the instructions enable a machine to execute the monitoring method for the distribution network transformer according to any one of the above.
Through the technical scheme, the fault diagnosis and management prediction model can be used for predicting and diagnosing two situations, the actual state of the transformer equipment can be evaluated, the establishment of maintenance strategies can be guided, and the health level of the equipment can be improved; and the fault diagnosis and management prediction model is fused and constructed on the basis of a digital twin technology at the intelligent terminal, so that the high-efficiency control can be realized, the monitoring efficiency can be improved, and the operation cost can be reduced.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
fig. 1 is a schematic flow chart of a monitoring method for a distribution network transformer according to an embodiment of the present invention;
FIG. 2 is a schematic structure of a performance analysis prediction model provided by an embodiment of the present invention;
FIG. 3 is an exemplary diagram provided by an embodiment of the present invention;
FIG. 4 is a supplementary illustration of FIG. 3;
fig. 5 is a schematic structural diagram of an example of an intelligent terminal corresponding to a control device according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
Before describing the embodiments of the present invention, the disadvantages of the prior art and the design idea of the present invention are introduced:
at present, the management of the distribution network transformer is mainly based on a temperature monitoring mode, the transformer is controlled within a reasonable working temperature range, and in addition, the periodic live detection and fault diagnosis of the transformer comprise a vibration analysis method, an oil chromatography analysis and the like. For the on-line monitoring of the transformer, a real-time system-level fusion method for monitoring and managing the transformer is lacked, so that the effective management and prediction of the transformer are realized, abnormal conditions occurring in the operation are timely processed and reported, and a monitoring device based on the method is also lacked.
Furthermore, management and maintenance of the distribution network transformer mainly include three methods based on historical statistical probability, sensing data driving and physical models, but the single methods have limitations and defects, for example, the historical statistical probability has certain randomness; the prediction precision is poor for more real-time states based on the sensing data driving model; based on the problems of poor fidelity of the physical model, low effective utilization rate of data and the like. With the mass adoption of integrated, low maintenance equipment, early custom equipment overhauls, test cycles have not been able to accommodate advances in the level of equipment diagnosis and management.
The appearance of the digital twinning technology provides a good solution for supervision of the distribution network transformer. Based on the characteristics of digital twin virtual-real mapping, the intelligent predictive maintenance of the transformer is realized by utilizing a life cycle high-fidelity model and intelligent sensing data and adopting a model and data fusion strategy.
Fig. 1 is a schematic flow chart of a monitoring method for a distribution network transformer according to an embodiment of the present invention, please refer to fig. 1, where the monitoring method for a distribution network transformer may include the following steps:
step S110: the method comprises the steps of obtaining characteristic data of the transformer, wherein the characteristic data comprise operation data and/or fault data of the transformer.
Characteristic data of the transformer, such as operation data of current, voltage, temperature, vibration, oil gas components and the like, can be acquired through various sensors and acquisition devices arranged in the transformer; the fault history data can also be obtained by a register, for example, configured by the transformer.
Preferably, the acquiring the characteristic data of the transformer includes: acquiring an induction signal of a transformer; filtering and denoising the induction signal; and extracting the characteristics of the filtered and noise-reduced induction signals.
For example, a sensor arranged in a transformer is used for acquiring an induction signal of the transformer for a period of time; normalizing the induction signal, such as filtering and noise reduction; and performing feature extraction on the processed signals to generate a data matrix of the transformer. For example, the characteristic data S ═ of an arbitrary transformer (S)1,S2…Sn) The corresponding feature tag may be L ═ (L)1,l2…ln) The feature tag L indicates the type of feature data, e.g./1Is a current,/2Voltage, etc.
Step S120: and training a preset fault diagnosis and management prediction model through the acquired characteristic data.
Preferably, before the training of the preset fault diagnosis and management prediction model, the method for supervising the distribution network transformer further includes: performing virtual space mapping on the characteristic data through a digital twinning technology; and respectively constructing a fault alarm diagnosis model and a performance analysis prediction model.
Further preferably, the training of the preset fault diagnosis and management prediction model includes: training the fault alarm diagnosis model and the performance analysis prediction model by taking the characteristic data as input; and fitting the trained fault alarm diagnosis model and the performance analysis prediction model.
According to the embodiment of the invention, aiming at different types of characteristic data, a fault alarm diagnosis model and a performance analysis prediction model can be respectively constructed, and fitting is carried out after training to form a convergence model, so that monitoring management can be better carried out on the power distribution network transformer.
1) And constructing a fault alarm diagnosis model.
Preferably, the fault alarm diagnosis model is constructed by a classification random validation algorithm, and the training of the fault alarm diagnosis model includes: acquiring corresponding state variables aiming at the fault data in the characteristic data, and establishing a random fault sample, wherein the state variables comprise a normal state, an early warning state and an alarm state; and training the fault alarm diagnosis model by taking the random fault sample as input.
For an application scenario that transformer fault data can be obtained, is relatively intuitive and needs quick response, a model constructed based on a classification random confirmation algorithm, namely a fault alarm diagnosis model, can be adopted to obtain a quick and accurate fault early warning result, such as transformer temperature alarm and the like. For example, the characteristic data S ═ (S)1,S2…Sn) Middle S1Is a temperature parameter, sets different range values of the temperature parameter, and corresponds to a characteristic label l1The state is one of a normal state, an early warning state and an alarm state.
By way of example, the state variable corresponding to the fault data is obtained based on a digital twin technology, then the fault alarm diagnosis model is trained through random fault samples based on a classification random confirmation algorithm to make a quick decision to complete fault monitoring, meanwhile, the model output result is used as an observed value to correct a simulation theory derivation value of the fault alarm diagnosis model, and other faults are predicted, so that the prediction and diagnosis results are more accurate and comprehensive.
2) And constructing a performance analysis prediction model.
Preferably, the performance analysis prediction model is constructed by a data fusion driving algorithm, and the training of the performance analysis prediction model includes: acquiring corresponding simulation induction data showing the performance degradation of the transformer aiming at the operation data in the characteristic data, and establishing an operation state sample; and training the performance analysis prediction model by taking the running state sample as input.
And for application scenes that the transformer operation data can be obtained and are difficult to distinguish, a performance analysis prediction model is constructed by adopting a data fusion-based driving algorithm. For example, running data such as vibration, oil temperature and oil gas parameters are simulated through a digital twin technology, simulated induction data showing performance degradation of the transformer are obtained, and a performance analysis prediction model is trained by using running state samples formed by the simulated induction data.
Preferably, the trained performance analysis prediction model can be transferred to a practical application environment and compared with practical operation data for correction, so that a prediction diagnosis result can be accurately output.
Further, in the embodiment of the present invention, it is preferable to set the analog sensing data showing the performance degradation of the transformer as a time sequence, which can be understood as a data structure constructed in time sequence for acquiring the operation data. By way of example, S ═ S (S) is specified for the characteristic data1,S2…Sn) Here, the operating data in the feature data is mainly referred to, and the corresponding feature tag may be L ═ L1,l2…ln) Arbitrary feature Sn＝(Sn1,Sn2…Snm) N is the type of the characteristic data type, and m is the serial number of the collected characteristic sequence, i.e. the time sequence. Setting TmFor the corresponding time for collecting the characteristic data, the time step can be in the unit of hour or day, and the characteristic data S and the data structure corresponding to the collecting time T show the performance of the transformerThe degraded analog sensing data can be represented by the following formula:
preferably, the simulation induction data showing the performance degradation of the transformer is a time series, and the training of the performance analysis prediction model includes: predicting the output of the next sequence m by taking the first m-1 sequence in the time sequence as regression input; wherein the expected value of error between the predicted output and the acquired signal corresponding to the actual operating data is minimized by a minimum cost function.
Fig. 2 is a schematic structure of a performance analysis prediction model, please refer to fig. 2, which trains the performance analysis prediction model with a time sequence of formula (1), and model trains the first m-1 sequences to obtain the mth predicted value.
In the embodiment of the invention, the error expected value between the prediction output and the actually acquired signal is minimized by preferably minimizing the cost function, the adaptation of the model is completed, and the synchronization between the prediction output and the actually acquired signal is kept, namely the performance analysis prediction model achieves the expectation of training.
Preferably, in order to achieve a more accurate performance analysis prediction model, the embodiment of the present invention sets a time series comparison model, which can be represented by the following formula:
Y(m)＝f(Y(m-1),Y(m-2),…,Y(m-k),S(m-1),S(m-2),…,S(m
-k))+ξ(k) (2)
where f () is a nonlinear function of the smoothing process, y (m) represents the output of the performance analysis prediction model, ξ (k) represents a gaussian white noise sequence, and k represents the prediction step size.
By simplifying the equation (2), the following equation can be obtained:
A(z-1)Ym+1＝B(z-1)Sm (3)
wherein z is time delay and is less than or equal to m.
Wherein the content of the first and second substances,
nythe order of the output is represented by,
wherein the content of the first and second substances,
nswhich represents the order of the input and,
the embodiment of the present invention preferably sets a minimum cost function to minimize the performance index expectation E [ ], wherein the minimum cost function can be represented by the following formula:
the minimum expected error value can represent the consistency of the output predicted value and the collected input parameters, and the smaller the expected value, the better the consistency. But for lightweight models, the choice of the error expectation determines the convergence speed of the model. Larger error expectations can destabilize the model, and smaller error expectations can slow the prediction process too slowly. Therefore, a lightweight boundary error expected value setting mode based on the time delay correction operator Z is provided by combining the Lyapunov function.
Preferably, the expected error value is set based on a lightweight boundary expected value of a delay correction operator Z, wherein the delay correction operator can be represented by the following formula:
and fitting the trained fault alarm diagnosis model and the trained performance analysis prediction model to obtain a preset fault diagnosis and management prediction model, namely implanting the preset fault diagnosis and management prediction model into an actual application scene to complete model adaptation, such as implanting into an intelligent terminal in a platform area.
Step S130: and monitoring the transformer in real time through the trained preset fault diagnosis and management prediction model.
By way of example, the whole life cycle of the distribution network transformer is supervised in real time through a fault diagnosis and management prediction model. Acquiring current characteristic data of the power distribution network, preprocessing the characteristic data, inputting the processed characteristic data into a fault diagnosis and management prediction model, predicting the fault and performance of the transformer, and making schemes such as health assessment, maintenance measures and the like.
Preferably, for the time series corresponding to the feature data used for training in step S120, a data storage module may be established to store the time series as historical data (in the past); meanwhile, storing the current characteristic data acquired in real time, and comparing and analyzing the current state of the transformer (at present) by combining historical data; and (3) combining a digital twin technology, simulating the state of the equipment through two models, obtaining the performance index variation trend of the equipment, and providing reference (in the future) for the operation and maintenance of the equipment. Based on the above, the characteristic parameters specific to each transformer device, such as the temperature change and vibration of the device under the ambient temperature and the current and voltage load, are described through the historical data. The real-time perception of the state is realized through the current data parameters, the perception effectiveness and the perception dynamics are guaranteed by comparing with historical data, in addition, the historical data are stored and injected into a digital twin model, the prediction of the future state is realized, and the customized maintenance is realized.
It should be noted that the operation data and the fault data in the feature data are acquired simultaneously, so the formula (1) may also include the fault data, which is also acquired and stored according to time sequence.
An embodiment of the present invention further provides a control device, where the control device includes: the monitoring system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the monitoring method of the distribution network transformer in steps S110-S130.
Fig. 3 is an illustration provided by an embodiment of the present invention, and please refer to fig. 3, which illustrates an exemplary embodiment of the present invention by taking the control device as a digital twin system of a distribution transformer as an example.
The digital twin system of the distribution transformer can acquire characteristic data of the distribution transformer through a sensor configured on a distribution transformer body, wherein real-time operation data can be acquired through an operation sensor, and real-time fault data can be acquired through an alarm sensor; the space virtual mapping based on the digital twinning technology can complete the digital reproduction of the transformer in the digital twinning system of the distribution transformer. Simulating the fault according to the fault monitoring and predicting process of the transformer; simulating the performance state according to the performance reduction rule of the transformer; and constructing a fault diagnosis and management prediction model, training the fault diagnosis and management prediction model by taking the acquired characteristic data as a sample, and transferring the trained fault diagnosis and management prediction model to an actual supervision environment of the transformer, such as a transformer area control system, so as to supervise the whole life cycle of the transformer. Meanwhile, the management results of the characteristic data, the fault diagnosis, the performance and the like of the transformer are stored in the database module and used as historical storage to continue training the fault diagnosis and management prediction model in the digital twin system of the distribution transformer, and the fault diagnosis and management prediction model is updated and maintained.
Further, referring to fig. 4, for the obtained feature data, the sensing signal of the transformer body is obtained, the signal is processed, for example, after filtering and denoising, feature extraction is performed on the signal, and feature data as shown in formula (1) can be formed to form a training set. Aiming at the training of the fault diagnosis and management prediction model, the training is centralized, and for the fault data which can be intuitively obtained, a fault alarm diagnosis model can be constructed through a classification random algorithm and trained through the fault data, so that the fault alarm diagnosis model can diagnose the fault alarm of the transformer; for the operation data which cannot be intuitively obtained, a performance analysis prediction model can be constructed through a data fusion driving algorithm and trained through the operation data, so that the performance analysis prediction model can analyze and predict the performance of the transformer; and fitting the trained fault alarm diagnosis model and the performance analysis prediction model to comprehensively monitor the transformer.
In summary, the embodiment of the invention can be used for predicting and diagnosing two situations based on the fault diagnosis and management prediction model, and can evaluate the actual state of the transformer equipment, guide the formulation of the maintenance strategy and improve the health level of the equipment; the fault diagnosis and management prediction model is constructed in a fusion manner on the basis of a digital twin technology at the intelligent terminal, so that high-efficiency control can be realized, the monitoring efficiency can be improved, and the operation cost can be reduced; for the characteristic data, the data prediction and the characteristic label of the time sequence are combined, the output of the next time is predicted according to the regression input of each sampling time, the error between the predicted output and the actually acquired signal is minimized through minimizing a cost function, the adaptation of the model is completed, and the synchronization between the predicted output and the actually acquired signal is kept; in addition, the method for setting the lightweight boundary expected value of the time delay correction operator effectively balances the relation between the lightweight system and the prediction precision, realizes dynamic adjustment and solves the problem of insufficient model training resources.
Fig. 5 is a schematic structural diagram of an example of an intelligent terminal corresponding to a control device according to an embodiment of the present invention, please refer to fig. 5, where the intelligent terminal may include: MCU, algorithm storage module, algorithm module, acquisition unit, power module, input/output module and communication module. The storage module is connected with the MCU and used for storing the characteristic data acquired in real time. The algorithm module can program the monitoring method of the distribution network transformer in the steps S110-S130 and has an upgrading function. And the MCU calls an algorithm module program to analyze and calculate the characteristic data acquired by the acquisition module. The acquisition module can acquire transformer characteristic data including analog quantity, digital quantity and switching value data, such as current and voltage values, temperature values, switching state signals, fault alarm signals and the like. The communication module supports wired and wireless transmission modes, is connected with the MCU and is used for transmitting monitored characteristic data, and can comprise a local communication RS232 interface, an RS485 interface and a GPRS module for remote communication.
Further, the embodiment of the present invention also provides a machine-readable storage medium, where the machine-readable storage medium has instructions stored thereon, and the instructions cause a machine to execute the method for supervising a distribution network transformer described in steps S110 to S130.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (11)
1. A supervision method for a distribution network transformer is characterized by comprising the following steps:
acquiring characteristic data of the transformer, wherein the characteristic data comprises operation data and/or fault data of the transformer;
training a preset fault diagnosis and management prediction model through the acquired characteristic data;
and monitoring the transformer in real time through the trained preset fault diagnosis and management prediction model.
2. The method for supervising a distribution network transformer according to claim 1, wherein the obtaining of the characteristic data of the transformer comprises:
acquiring an induction signal of a transformer;
filtering and denoising the induction signal; and
and performing feature extraction on the filtered and noise-reduced induction signal.
3. The method for supervising a distribution network transformer according to claim 1, wherein before the training of the preset fault diagnosis and management prediction model, the method for supervising a distribution network transformer further comprises:
performing virtual space mapping on the characteristic data through a digital twinning technology;
and respectively constructing a fault alarm diagnosis model and a performance analysis prediction model.
4. The method for supervising a distribution network transformer according to claim 3, wherein the training of the preset fault diagnosis and management prediction model comprises:
training the fault alarm diagnosis model and the performance analysis prediction model by taking the characteristic data as input; and
and fitting the trained fault alarm diagnosis model and the performance analysis prediction model.
5. The supervision method for distribution network transformers according to claim 4, wherein the fault alarm diagnosis model is constructed by a classification random validation algorithm, and the training of the fault alarm diagnosis model comprises:
acquiring corresponding state variables aiming at the fault data in the characteristic data, and establishing a random fault sample, wherein the state variables comprise a normal state, an early warning state and an alarm state;
and training the fault alarm diagnosis model by taking the random fault sample as input.
6. The supervision method for distribution network transformers according to claim 4, wherein the performance analysis prediction model is constructed by a data fusion driving algorithm, and the training of the performance analysis prediction model comprises:
acquiring corresponding simulation induction data showing the performance degradation of the transformer aiming at the operation data in the characteristic data, and establishing an operation state sample;
and training the performance analysis prediction model by taking the running state sample as input.
7. The method of claim 6, wherein the simulated induction data showing transformer performance degradation is a time series, and wherein the training of the performance analysis prediction model comprises:
predicting the output of the next sequence m by taking the first m-1 sequence in the time sequence as regression input;
wherein the expected value of error between the predicted output and the acquired signal corresponding to the actual operating data is minimized by a minimum cost function.
8. The supervision method of distribution network transformers according to claim 7, characterized in that the minimum cost function is represented by the following equation:
wherein, E2]Representing performance index expectation, Y () representing the output predicted by the performance analysis prediction model, S () representing the time series, ξS() Is Gaussian white noise sequence, m represents the number of time sequences, k represents the prediction step length, nyIndicating the order of the output.
9. The method of claim 7, wherein the error expectation is set based on a lightweight boundary expectation of a delay correction operator Z, wherein the delay correction operator is represented by the following equation:
where m represents the number of time series, n represents the number of types in the time series, Q () represents the minimum cost function, Y represents the output predicted by the performance analysis prediction model, and S represents the input of the performance analysis prediction model.
10. A control device, characterized in that the control device comprises: memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the method of supervision of a distribution network transformer according to any of claims 1-9.
11. A machine-readable storage medium having stored thereon instructions for causing a machine to perform the method of supervision of a distribution network transformer according to any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110767526.2A CN113656938A (en) | 2021-07-07 | 2021-07-07 | Monitoring method, control device and storage medium for distribution network transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110767526.2A CN113656938A (en) | 2021-07-07 | 2021-07-07 | Monitoring method, control device and storage medium for distribution network transformer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113656938A true CN113656938A (en) | 2021-11-16 |
Family
ID=78489964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110767526.2A Pending CN113656938A (en) | 2021-07-07 | 2021-07-07 | Monitoring method, control device and storage medium for distribution network transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113656938A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113835060A (en) * | 2021-11-23 | 2021-12-24 | 武汉格蓝若智能技术有限公司 | Power transformer online state monitoring method and system based on digital twinning |
| CN114167204A (en) * | 2021-11-19 | 2022-03-11 | 广东电网有限责任公司电力调度控制中心 | Distribution transformer abnormal operation recognition early warning method |
| CN114252810A (en) * | 2022-01-12 | 2022-03-29 | 中国电力科学研究院有限公司 | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium |
| CN114422884A (en) * | 2021-11-26 | 2022-04-29 | 北京智芯微电子科技有限公司 | Distribution transformer fault sample collection method, device and system |
| CN114626641A (en) * | 2022-05-13 | 2022-06-14 | 山东汇能电气有限公司 | Transformer power failure prediction system based on data processing |
| CN114784701A (en) * | 2022-04-21 | 2022-07-22 | 中国电力科学研究院有限公司 | Power distribution network live working autonomous navigation method, system, equipment and storage medium |
| CN115529331A (en) * | 2022-09-26 | 2022-12-27 | 云南大学 | Power distribution network box type equipment fault monitoring method and system and computer equipment |
| CN115542856A (en) * | 2022-10-08 | 2022-12-30 | 国网山西省电力公司电力科学研究院 | Equipment fault diagnosis early warning system based on digital twin modeling and use method |
| CN116760195A (en) * | 2023-08-18 | 2023-09-15 | 国网浙江省电力有限公司宁波供电公司 | Method, system, equipment and storage medium for monitoring all-element resources of power system |
| CN117375237A (en) * | 2023-10-20 | 2024-01-09 | 浙江日新电气有限公司 | Substation operation and maintenance method and system based on digital twin technology |
| CN117368651A (en) * | 2023-12-07 | 2024-01-09 | 江苏索杰智能科技有限公司 | Comprehensive analysis system and method for faults of power distribution network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109061341A (en) * | 2018-07-10 | 2018-12-21 | 杭州安脉盛智能技术有限公司 | Kalman filtering transformer fault prediction technique and system neural network based |
| US20190293697A1 (en) * | 2018-03-26 | 2019-09-26 | Oracle International Corporation | Estimating the remaining useful life of a power transformer based on real-time sensor data and periodic dissolved gas analyses |
| CN112418451A (en) * | 2020-11-05 | 2021-02-26 | 武汉大学 | Transformer fault diagnosis positioning system based on digital twinning |
| CN112684379A (en) * | 2020-11-25 | 2021-04-20 | 江苏科技大学 | Transformer fault diagnosis system and method based on digital twinning |
| CN112685949A (en) * | 2020-11-25 | 2021-04-20 | 江苏科技大学 | Transformer health prediction method based on digital twinning |
-
2021
- 2021-07-07 CN CN202110767526.2A patent/CN113656938A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190293697A1 (en) * | 2018-03-26 | 2019-09-26 | Oracle International Corporation | Estimating the remaining useful life of a power transformer based on real-time sensor data and periodic dissolved gas analyses |
| CN109061341A (en) * | 2018-07-10 | 2018-12-21 | 杭州安脉盛智能技术有限公司 | Kalman filtering transformer fault prediction technique and system neural network based |
| CN112418451A (en) * | 2020-11-05 | 2021-02-26 | 武汉大学 | Transformer fault diagnosis positioning system based on digital twinning |
| CN112684379A (en) * | 2020-11-25 | 2021-04-20 | 江苏科技大学 | Transformer fault diagnosis system and method based on digital twinning |
| CN112685949A (en) * | 2020-11-25 | 2021-04-20 | 江苏科技大学 | Transformer health prediction method based on digital twinning |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114167204A (en) * | 2021-11-19 | 2022-03-11 | 广东电网有限责任公司电力调度控制中心 | Distribution transformer abnormal operation recognition early warning method |
| CN113835060A (en) * | 2021-11-23 | 2021-12-24 | 武汉格蓝若智能技术有限公司 | Power transformer online state monitoring method and system based on digital twinning |
| CN114422884A (en) * | 2021-11-26 | 2022-04-29 | 北京智芯微电子科技有限公司 | Distribution transformer fault sample collection method, device and system |
| CN114252810B (en) * | 2022-01-12 | 2024-04-30 | 中国电力科学研究院有限公司 | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium |
| CN114252810A (en) * | 2022-01-12 | 2022-03-29 | 中国电力科学研究院有限公司 | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium |
| CN114784701A (en) * | 2022-04-21 | 2022-07-22 | 中国电力科学研究院有限公司 | Power distribution network live working autonomous navigation method, system, equipment and storage medium |
| CN114784701B (en) * | 2022-04-21 | 2023-07-25 | 中国电力科学研究院有限公司 | Autonomous navigation method, system, equipment and storage medium for live working of power distribution network |
| CN114626641A (en) * | 2022-05-13 | 2022-06-14 | 山东汇能电气有限公司 | Transformer power failure prediction system based on data processing |
| CN114626641B (en) * | 2022-05-13 | 2022-08-23 | 山东汇能电气有限公司 | Transformer power failure prediction system based on data processing |
| CN115529331A (en) * | 2022-09-26 | 2022-12-27 | 云南大学 | Power distribution network box type equipment fault monitoring method and system and computer equipment |
| CN115542856A (en) * | 2022-10-08 | 2022-12-30 | 国网山西省电力公司电力科学研究院 | Equipment fault diagnosis early warning system based on digital twin modeling and use method |
| CN116760195B (en) * | 2023-08-18 | 2024-01-02 | 国网浙江省电力有限公司宁波供电公司 | Method, system, equipment and storage medium for monitoring all-element resources of power system |
| CN116760195A (en) * | 2023-08-18 | 2023-09-15 | 国网浙江省电力有限公司宁波供电公司 | Method, system, equipment and storage medium for monitoring all-element resources of power system |
| CN117375237A (en) * | 2023-10-20 | 2024-01-09 | 浙江日新电气有限公司 | Substation operation and maintenance method and system based on digital twin technology |
| CN117368651A (en) * | 2023-12-07 | 2024-01-09 | 江苏索杰智能科技有限公司 | Comprehensive analysis system and method for faults of power distribution network |
| CN117368651B (en) * | 2023-12-07 | 2024-03-08 | 江苏索杰智能科技有限公司 | Comprehensive analysis system and method for faults of power distribution network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113656938A (en) | Monitoring method, control device and storage medium for distribution network transformer | |
| CN112731159B (en) | Method for pre-judging and positioning battery faults of battery compartment of energy storage power station | |
| CN111027165B (en) | Power battery pack management system and method based on digital twinning | |
| CN108375715B (en) | Power distribution network line fault risk day prediction method and system | |
| CN111143447B (en) | Dynamic monitoring early warning decision system and method for weak links of power grid | |
| CN114154429A (en) | Digital twin body construction method and device of energy storage system and storage medium | |
| CN112700032A (en) | Fault prediction system and method for low-voltage direct-current power distribution and utilization system | |
| CN115423009A (en) | Cloud edge coordination-oriented power equipment fault identification method and system | |
| CN115437344A (en) | Digital twin comprehensive energy fault diagnosis system, method, device and medium | |
| CN110766236A (en) | Power equipment state trend prediction method based on statistical analysis and deep learning | |
| CN110716520B (en) | Flight control servo actuator reliability evaluation modeling method based on multi-source information fusion | |
| CN115932592A (en) | Electric vehicle battery fault diagnosis method based on digital twin model | |
| CN117390944A (en) | Substation operation condition simulation system | |
| CN117574264A (en) | Transformer fault diagnosis method and system based on knowledge constraint neural network | |
| CN110956308A (en) | Chemical equipment health state prediction method based on time sequence | |
| CN115640980A (en) | Power grid engineering cost dynamic management system based on target control | |
| CN112100902B (en) | Lithium ion battery life prediction method based on flow data | |
| CN114579659A (en) | System and method for estimating and sorting utilization potential of power battery by gradient utilization | |
| CN111026624B (en) | Fault prediction method and device of power grid information system | |
| CN114169253A (en) | Data flow dynamic prediction method and system based on Flink and LSTM | |
| CN114066163A (en) | Power equipment state analysis method and device based on three-layer architecture mode | |
| CN111367255A (en) | Performance evaluation test system and method for multi-variable control system | |
| CN117422004B (en) | Carbon potential prediction method and system based on neural network | |
| CN117034157B (en) | Hydropower equipment fault identification method and system combining multimodal operation data | |
| Zhu et al. | Prediction Method of Ohmic Resistance and Charge Transfer Resistance for Lithium-Ion Batteries Based on CSA-SVR |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211116 |
|
| RJ01 | Rejection of invention patent application after publication |