Extra-high voltage transformer multi-parameter state evaluation method, system, equipment and medium
By constructing a UHV transformer health status evaluation model based on domain knowledge causal graphs, the uncertainty problem of equipment status evaluation is solved, and the accuracy and reliability of the evaluation are improved.
Patent Information
- Application Number
- CN202411840883.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-13
- Publication Date
- 2025-04-08
AI Technical Summary
The status evaluation of ultra-high voltage transformer has strong internal state unknown and uncertainty, which leads to poor analysis accuracy and makes it difficult to effectively evaluate the health status of the equipment.
A multi-parameter state evaluation method is adopted to obtain monitoring parameters, and a health status evaluation model based on the causal graph of domain knowledge is constructed, data cleaning, dimensionality reduction and causal intervention analysis are carried out, and the causal intervention distribution is calculated to evaluate the health status of the transformer.
It improves the reliability and accuracy of UHV transformer status evaluation, can more accurately evaluate the health status of the equipment, and reduces the uncertainty of fault prediction.
Smart Images
Figure CN119782986A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of transformer status assessment, and relates to a method, system, equipment and medium for multi-parameter status assessment of an ultra-high voltage transformer. Background Art
[0002] The reliable operation of UHV transformers is the basis for the safety and stability of UHV transmission lines. As the equipment is put into operation for a longer period of time, the defect rate of transformer equipment is on the rise, and the safe operation of equipment still faces severe challenges. In terms of equipment failures, such as winding short circuits, inter-turn / inter-pancake short circuits, etc., this not only causes the UHV transmission line to stop operating, but also causes serious economic losses and adverse social impacts.
[0003] In recent years, with the development of online monitoring technology and computer technology, artificial intelligence technology has effectively improved the quality of power transformer status evaluation. However, due to the complex internal structure of UHV transformers, short operation time, sparse data measurement points, and few fault samples, the status evaluation of UHV transformers still faces the problem of poor analysis accuracy due to the unknown and strong uncertainty of the internal state. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method, system, device and medium for multi-parameter status assessment of ultra-high voltage transformers. The method, system, device and medium can accurately assess the health status of ultra-high voltage transformers.
[0005] In order to achieve the above object, the present invention adopts the following technical scheme:
[0006] In one aspect of the present invention, the multi-parameter state assessment method for a UHV transformer of the present invention comprises:
[0007] Obtain the monitoring parameters of the current UHV transformer;
[0008] The monitoring parameters of the current UHV transformer are input into a UHV transformer health status assessment model to obtain the health status of the UHV transformer, wherein the UHV transformer health status assessment model is constructed based on a domain knowledge causal graph of the UHV transformer.
[0009] The multi-parameter state assessment method for ultra-high voltage transformers of the present invention is further improved in that:
[0010] Furthermore, before inputting the monitoring parameters of the UHV transformer into the UHV transformer health status assessment model, the method further includes:
[0011] Obtain the domain knowledge causal graph of UHV transformers;
[0012] According to the domain knowledge causal graph, obtain the N key entities of monitoring parameters of the UHV transformer [X1, X2, ..., X N ];
[0013] For the N monitoring parameter key entities [X1, X2, ..., X N ] to clean and reduce the data dimension, and obtain the variable set [Z1, Z2, ..., Z M ], where a variable after dimension reduction corresponds to several key entities of monitoring parameters;
[0014] The variable set [Z1, Z2, ..., Z M ] is the cause variable Z, and the key entity of the running state in the domain knowledge causal graph is the result variable Y. For the N monitoring parameter key entities [X1, X2, ..., X N ] Perform multi-parameter intervention operations and calculate the causal intervention distribution P(Y=y|do(Z=z)) of the result variable Y=y when the multi-parameter intervention operation makes the cause variable Z=z;
[0015] According to the N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct a health status assessment model for ultra-high voltage transformers.
[0016] Furthermore, the process of obtaining the domain knowledge causal graph of the UHV transformer is as follows:
[0017] Get the text data of UHV transformer;
[0018] Extracting entities and relationships from the text data of the UHV transformer to obtain key entities of operating status, key entities of monitoring parameters, and causal relationships between the key entities of operating status and the key entities of monitoring parameters;
[0019] The key entities of the operating status and the key entities of the monitoring parameters are used as nodes V in the graph structure representation, and the causal relationship between the key entities of the operating status and the key entities of the monitoring parameters is used as the edge E in the graph structure representation to construct a domain knowledge causal graph.
[0020] Further, the N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct the transformer health status assessment model as follows:
[0021] The variable set [Z1, Z2, ..., Z M ] is the first-level indicator, with the N monitoring parameter key entities [X1, X2, ..., X N ] is taken as the secondary indicator, and the causal intervention distribution P(Y=y|do(Z=z)) is taken as the indicator weight to construct the UHV transformer health status assessment model.
[0022] In a second aspect of the present invention, the multi-parameter status assessment system for ultra-high voltage transformers of the present invention comprises:
[0023] The first acquisition module is used to obtain the monitoring parameters of the current UHV transformer;
[0024] An evaluation module is used to input the monitoring parameters of the current UHV transformer into a UHV transformer health status evaluation model to obtain the health status of the UHV transformer, wherein the UHV transformer health status evaluation model is constructed based on a domain knowledge causal graph of the UHV transformer.
[0025] The UHV transformer multi-parameter state assessment system of the present invention is further improved in that:
[0026] Furthermore, it also includes:
[0027] The second acquisition module is used to obtain the domain knowledge causal graph of the UHV transformer;
[0028] The third acquisition module is used to obtain N key entities of monitoring parameters of the UHV transformer [X1, X2, ..., X N ];
[0029] A preprocessing module is used to process the N monitoring parameter key entities [X1, X2, ..., X N ] to clean and reduce the data dimension, and obtain the variable set [Z1, Z2, ..., Z M ], where a variable after dimension reduction corresponds to several key entities of monitoring parameters;
[0030] The calculation module is used to calculate the variable set [Z1, Z2, ..., Z M ] is the cause variable Z, and the key entity of the running state in the domain knowledge causal graph is the result variable Y. For the N monitoring parameter key entities [X1, X2, ..., X N ] Perform multi-parameter intervention operations and calculate the causal intervention distribution P(Y=y|do(Z=z)) of the result variable Y=y when the multi-parameter intervention operation makes the cause variable Z=z;
[0031] The first building module is used to monitor the N key entities of the parameters [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct a health status assessment model for ultra-high voltage transformers.
[0032] Furthermore, the second acquisition module includes:
[0033] A fourth acquisition module is used to acquire text data of the UHV transformer;
[0034] An extraction module is used to extract entities and relationships from the text data of the UHV transformer to obtain key entities of operating status, key entities of monitoring parameters, and causal relationships between the key entities of operating status and the key entities of monitoring parameters;
[0035] The second construction module is used to construct a domain knowledge causal graph using the key entities of the operating status and the key entities of the monitoring parameters as nodes V in the graph structure representation, and using the causal relationship between the key entities of the operating status and the key entities of the monitoring parameters as edges E in the graph structure representation.
[0036] Further, the N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct the transformer health status assessment model as follows:
[0037] The variable set [Z1, Z2, ..., Z M ] is the first-level indicator, with the N monitoring parameter key entities [X1, X2, ..., X N ] is taken as the secondary indicator, and the causal intervention distribution P(Y=y|do(Z=z)) is taken as the indicator weight to construct the UHV transformer health status assessment model.
[0038] In a third aspect of the present invention, the computer device of the present invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the multi-parameter state assessment method of the ultra-high voltage transformer when executing the computer program.
[0039] In a fourth aspect of the present invention, the computer-readable storage medium of the present invention stores a computer program, and when the computer program is executed by a processor, the steps of the multi-parameter state assessment method of the ultra-high voltage transformer are implemented.
[0040] The present invention has the following beneficial effects:
[0041] During specific operation, the method, system, device and medium for multi-parameter status assessment of ultra-high voltage transformers described in the present invention input the monitoring parameters of the current ultra-high voltage transformer into the ultra-high voltage transformer health status assessment model to obtain the health status of the ultra-high voltage transformer, wherein the ultra-high voltage transformer health status assessment model is constructed based on the domain knowledge causal graph of the ultra-high voltage transformer to explore the causal relationship between the monitoring parameters and the operating status of the ultra-high voltage transformer, thereby improving the reliability and accuracy of the ultra-high voltage transformer status assessment. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:
[0043] Figure 1 A flow chart of constructing a UHV transformer health status assessment model in the present invention;
[0044] Figure 2 It is a schematic diagram of the cause-effect diagram of the UHV transformer operation status;
[0045] Figure 3 It is a system structure diagram of the present invention. DETAILED DESCRIPTION
[0046] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only an embodiment of a part of the present invention, not all embodiments, and is not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concepts disclosed in the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.
[0047] The accompanying drawings show schematic diagrams of structures according to embodiments disclosed in the present invention. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.
[0048] Embodiment 1
[0049] refer to Figure 1 The multi-parameter state assessment method for ultra-high voltage transformers of the present invention comprises the following steps:
[0050] 1) Obtain monitoring parameters of UHV transformer;
[0051] 2) Inputting the monitoring parameters of the UHV transformer into the UHV transformer health status assessment model to assess the health status of the UHV transformer.
[0052] The process of step 2) is:
[0053] 21) Construct a UHV transformer health status assessment model;
[0054] The process of step 21) is:
[0055] 211) Embedding domain experience knowledge;
[0056] Entity extraction and relationship extraction are performed on text data such as transformer case reports, maintenance guidelines, and test reports. The key entities of operating status and key entities of monitoring parameters, as well as the causal relationship between the two types of entities, are mined. The key entities of operating status and key entities of monitoring parameters are used as nodes V in the graph structure representation, and the causal relationship is used as the edge E in the graph structure representation to construct a domain knowledge causal graph G = (V, E).
[0057] 212) Status evaluation causal discovery;
[0058] For the online monitoring data of UHV transformers, according to the entity V in the domain knowledge causal graph G = (V, E), N key entities of monitoring parameters [X1, X2, ..., X N ] to form the original data set.
[0059] 213) Data cleaning;
[0060] For monitoring parameter X i , fill the data items containing NaN or missing data according to the mean method.
[0061] The data is normalized to 0-1. Scaling to the 0-1 interval, according to the formula Get new data value
[0062] Categorize new data values into continuous values Convert to discrete value
[0063] 214) Data dimensionality reduction;
[0064] Since there are many monitoring points for UHV transformers and the amount of monitoring data is large, methods such as principal component analysis (PCA) or linear discriminant analysis (LDA) are used to reduce the dimension of discrete data and convert N key entities of monitoring parameters [X1, X2, ..., X N ] is reduced to M [Z1,Z2,...,Z M ], using the dimension reduction variables as intermediate latent variables to reduce the computational resource consumption of subsequent steps, where each dimension reduction parameter Z i Contains any number of monitoring parameters X i , all dimension reduction parameters [Z1,Z2,...,Z M ] contains all monitoring parameters [X1,X2,...,X N ].
[0065] 215)Causal discovery;
[0066] refer to Figure 2 , let the variable set after dimension reduction [Z1,Z2,...,Z M ] is the cause variable, and the set of key entities of the running state [Y1,Y2,...,Y N ] is the result variable, and the function f = {f:Z→Y} represents the causal relationship between the cause variable and the result variable. i With the outcome variable Y i The relationship between them is given to the cause variable Z i , reconstruct the causal graph G dr =(V dr ,E dr ).
[0067] Using the Bayesian causal learning method, calculate the cause variable Z i With the outcome variable Y i The function between them is used to obtain the monitoring parameter X i The observed distribution of the running state P(Y|Z) is the domain knowledge causal graph G dr =(V dr ,E dr ) dr Assignment, at the same time, for the result variable Y i , the probability P(Y i |Z), that is, the probability that the UHV transformer operating state is the outcome variable.
[0068] At the same time, the monitoring parameters [X1,X2,...,X N ] as the cause variable, the variable set after dimension reduction [Z1, Z2, ..., Z M] as the result variable, and calculate the function between the cause variable and the result variable to complete the causal diagram.
[0069] 22) Causal intervention analysis;
[0070] Perform intervention operations on all monitored parameters. Let do(Z=z) represent the execution of multi-parameter intervention operations on variable Z to fix its value at z. Calculate the causal intervention distribution P(Y=y|do(Z=z)), that is, the probability that the result variable Y=y when the cause variable Z=z is made through intervention.
[0071] Since the causal graph G dr =(V dr ,E dr ) has a many-to-many relationship with the cause variables and the result variables, that is, one cause variable may lead to multiple result variables, and one result variable may be caused by multiple cause variables. Based on the cause variables, the intervention effects of the cause variables on different result variables are comprehensively calculated.
[0072] 23) Using the dimension reduction parameters [Z1, Z2, ..., Z M ] is listed as the first-level indicator, with the corresponding monitoring parameters [X1, X2, ..., X N ] is taken as the secondary indicator, and the causal intervention distribution is used as the indicator weight to construct the UHV transformer health status assessment model.
[0073] 24) Inputting the monitoring parameters of the UHV transformer into a UHV transformer health status assessment model to assess the health status of the UHV transformer.
[0074] Embodiment 2
[0075] The specific process of this embodiment is as follows:
[0076] 1) Embedding domain experience knowledge;
[0077] Entity extraction and relationship extraction are performed on text data such as transformer case reports, maintenance guidelines, and test reports to mine the key entities of operating status and the key entities of monitoring parameters, as well as the causal relationship between the key entities of operating status and the key entities of monitoring parameters. The key entities of operating status include "normal", "attention" and "abnormal", etc., and the key entities of monitoring parameters include "partial discharge", "vibration", "temperature", "current", "acetylene" and "hydrogen", etc., among which each key entity of monitoring parameters is causally connected to each key entity of operating status, and the key entity of monitoring parameters points to the key entity of operating status.
[0078] The key entities of the operating status and the key entities of the monitoring parameters are taken as nodes V in the graph structure representation, and the causal relationship is taken as the edge E in the graph structure representation to construct a domain knowledge causal graph G = (V, E).
[0079] 2) state evaluation causal discovery;
[0080] For the online monitoring data of UHV transformers collected within a certain period of time, according to the entity node V in the domain knowledge causal graph G = (V, E), the N key entities of monitoring parameters [X1, X2, ..., X N ] to form the original data set. Suppose the original data set is a continuous value with N columns and K rows, indicating that there are K data values for each type of monitoring parameter.
[0081] 21) Data cleaning;
[0082] For each column of monitoring parameter data X i ={x1,...,x K}, fill the data items containing NaN or missing data according to the mean method.
[0083] The data is normalized to 0-1. Scaling to the 0-1 interval, according to the formula Calculate new data values When the i-th column data X i ={x1,...,x K The maximum value of} is 1069.86 and the minimum value is 0.099. The new data value after normalization of the original data value 242.055 is 0.226178.
[0084] Categorize new data values into continuous values Convert to discrete value Specifically, the quartile method is used to project the continuous data values into four equal-proportional intervals, that is, the value between the interval [0, 0.25) is defined as 1, the value between the interval [0.25, 0.5) is defined as 2, the value between the interval [0.5, 0.75) is defined as 3, and the value between the interval [0.75, 1] is defined as 4. At this point, all continuous data are converted into 4 types of discrete data.
[0085] 22) Data dimensionality reduction;
[0086] Since there are many monitoring points for UHV transformers and the amount of monitoring data is large, methods such as principal component analysis (PCA) or linear discriminant analysis (LDA) are used to reduce the dimension of discrete data and convert N key entities of monitoring parameters [X1, X2, ..., X N ] is reduced to M [Z1,Z2,...,Z M ], and use the dimension reduction variables as intermediate hidden variables to reduce the computing resource consumption of subsequent steps. Among them, each dimension reduction parameter Z i Contains any number of monitoring parameters Xi , all dimension reduction parameters [Z1,Z2,...,Z M ] contains all monitoring parameters [X1,X2,...,X N ].
[0087] 23) Cause and effect discovery;
[0088] Suppose the variable set after dimension reduction [Z1, Z2, ..., Z M ] is the cause variable, and the key entity set of the running state [Y1,Y2,...,Y N ] is the result variable, and the function f = {f:Z→Y} represents the causal relationship between the cause variable and the result variable. i With the outcome variable Y i The relationship between them is given to the cause variable Z i , reconstruct the causal graph G dr =(V dr ,E dr ).
[0089] The Bayesian causal learning method is used to calculate the function between the cause variable and the result variable, and the observed distribution P(Y|Z) of the monitoring parameters and the operating status is obtained, which is the domain knowledge causal graph G dr =(V dr ,E dr ) dr Assignment, at the same time, for the result variable Y i , calculate the probability P(Y i |Z), that is, the probability that the UHV transformer operating state is the outcome variable.
[0090] Connect the result variable Y1 with the cause variables Z1, Z2, and Z3, and the calculation formula for the observed distribution P(Y1|Z) is
[0091] Similarly, the monitoring parameters [X1,X2,...,X N ] as the cause variable, the variable set after dimension reduction [Z1, Z2, ..., Z M ] as the result variable, calculate the function between the two, and improve the cause-effect diagram of the UHV transformer operating status.
[0092] 3) causal intervention analysis;
[0093] Perform intervention operations on all monitored parameters, and let do(X=x) represent the intervention operation on variable X to fix its value to x. Calculate the causal intervention distribution P(Y=y|do(X=x)), that is, the probability that the result variable Y=y when the cause variable X=x is intervened.
[0094] Since the causal graph G dr =(V dr ,E dr ) has a many-to-many relationship with the cause variables and the result variables, that is, one cause variable may lead to multiple result variables, and one result variable may be caused by multiple cause variables. Based on the cause variables, the intervention effects of the cause variables on different result variables are comprehensively calculated.
[0095] For example, the cause variable "temperature" has four discrete values: "1", "2", "3", and "4". The intervention operation is performed to keep the value of "temperature" at "4". For the result variables "normal", "caution", and "abnormal", the intervention distributions P(Y = "normal" | do("temperature" = 4)), P(Y = "caution" | do("temperature" = 4)), and P(Y = "abnormal" | do("temperature" = 4)) are calculated respectively. Among them, the greater the difference between the intervention distribution P(Y = "normal" | do("temperature" = 4)) and the observation distribution (conditional probability) P(Y = "normal" | "temperature" = 4), the greater the intervention effect of the cause variable "temperature" on the result variable.
[0096] 4) Construct an UHV transformer operation status assessment model;
[0097] Based on the UHV transformer operating status causal diagram and causal intervention distribution, an UHV transformer operating status evaluation system is constructed.
[0098] Specifically, the dimension reduction parameters [Z1, Z2, ..., Z M ] as the first-level indicator, and the corresponding monitoring parameters [X1, X2, ..., X N ] is used as the secondary indicator, the causal intervention distribution P(Y|do(X)) is used as the indicator weight α(Y|X), and a transformer health status assessment model is established based on this.
[0099] For example, the calculation formula of the score f(Y1) of the operating state Y1 is derived from the dimension reduction parameters connected to it, the monitoring parameters and their related weights, that is,
[0100] The scores of different operating states are calculated, and the operating state with the highest score is taken as the final evaluation result. At the same time, all operating states and corresponding scores are output as a reference for interpretable results.
[0101] Embodiment 3
[0102] refer to Figure 3 The multi-parameter status assessment system for ultra-high voltage transformers of the present invention comprises:
[0103] The first acquisition module is used to obtain the monitoring parameters of the current UHV transformer;
[0104] An evaluation module is used to input the monitoring parameters of the current UHV transformer into a UHV transformer health status evaluation model to obtain the health status of the UHV transformer, wherein the UHV transformer health status evaluation model is constructed based on a domain knowledge causal graph of the UHV transformer.
[0105] As an embodiment of the present invention, it also includes:
[0106] The second acquisition module is used to obtain the domain knowledge causal graph of the UHV transformer;
[0107] The third acquisition module is used to obtain N key entities of monitoring parameters of the UHV transformer [X1, X2, ..., X N ];
[0108] A preprocessing module is used to process the N monitoring parameter key entities [X1, X2, ..., X N ] to clean and reduce the data dimension, and obtain the variable set [Z1, Z2, ..., Z M ], where a variable after dimension reduction corresponds to several key entities of monitoring parameters;
[0109] The calculation module is used to calculate the variable set [Z1, Z2, ..., Z M ] is the cause variable Z, and the key entity of the running state in the domain knowledge causal graph is the result variable Y. For the N monitoring parameter key entities [X1, X2, ..., X N ] Perform multi-parameter intervention operations and calculate the causal intervention distribution P(Y=y|do(Z=z)) of the result variable Y=y when the multi-parameter intervention operation makes the cause variable Z=z;
[0110] The first building module is used to monitor the N key entities of the parameters [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct a health status assessment model for ultra-high voltage transformers.
[0111] As an implementation manner of the present invention, the second acquisition module includes:
[0112] A fourth acquisition module is used to acquire text data of the UHV transformer;
[0113] An extraction module is used to extract entities and relationships from the text data of the UHV transformer to obtain key entities of operating status, key entities of monitoring parameters, and causal relationships between the key entities of operating status and the key entities of monitoring parameters;
[0114] The second construction module is used to construct a domain knowledge causal graph using the key entities of the operating status and the key entities of the monitoring parameters as nodes V in the graph structure representation, and using the causal relationship between the key entities of the operating status and the key entities of the monitoring parameters as edges E in the graph structure representation.
[0115] As an implementation mode of the present invention, the N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct the transformer health status assessment model as follows:
[0116] The variable set [Z1, Z2, ..., Z M ] is the first-level indicator, with the N monitoring parameter key entities [X1, X2, ..., X N ] is taken as the secondary indicator, and the causal intervention distribution P(Y=y|do(Z=z)) is taken as the indicator weight to construct the UHV transformer health status assessment model.
[0117] The division of modules in the embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.
[0118] Embodiment 4
[0119] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of implementing the method for multi-parameter status assessment of ultra-high voltage transformers are implemented, for example, including: obtaining the monitoring parameters of the current ultra-high voltage transformer; inputting the monitoring parameters of the current ultra-high voltage transformer into the health status assessment model of the ultra-high voltage transformer to obtain the health status of the ultra-high voltage transformer, wherein the health status assessment model of the ultra-high voltage transformer is constructed based on the domain knowledge causal graph of the ultra-high voltage transformer. The memory may include a memory, such as a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk memory, etc. The processor, the network interface, and the memory are interconnected through an internal bus, and the internal bus may be an industrial standard architecture bus, a peripheral component interconnection standard bus, an extended industrial standard architecture bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program may include a program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.
[0120] Embodiment 5
[0121] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of implementing the method for multi-parameter status assessment of ultra-high voltage transformers are implemented, for example, including: obtaining the monitoring parameters of the current ultra-high voltage transformer; inputting the monitoring parameters of the current ultra-high voltage transformer into the ultra-high voltage transformer health status assessment model to obtain the health status of the ultra-high voltage transformer, wherein the ultra-high voltage transformer health status assessment model is constructed based on the domain knowledge causal graph of the ultra-high voltage transformer. Specifically, the computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.
[0122] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0123] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0124] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A multi-parameter state assessment method for ultra-high voltage transformers, characterized in that: include: Obtain the monitoring parameters of the current UHV transformer; The monitoring parameters of the current UHV transformer are input into a UHV transformer health status assessment model to obtain the health status of the UHV transformer, wherein the UHV transformer health status assessment model is constructed based on a domain knowledge causal graph of the UHV transformer.
2. The method for multi-parameter status assessment of a UHV transformer according to claim 1, characterized in that: Before inputting the monitoring parameters of the UHV transformer into the UHV transformer health status assessment model, the method further includes: Obtain the domain knowledge causal graph of UHV transformers; According to the domain knowledge causal graph, obtain the N key entities of monitoring parameters of the UHV transformer [X1, X2, ..., X N ]; For the N monitoring parameter key entities [X1, X2, ..., X N ] to clean and reduce the data dimension, and obtain the variable set [Z1, Z2, ..., Z M ], where a variable after dimension reduction corresponds to several key entities of monitoring parameters; The variable set [Z1, Z2, ..., Z M ] is the cause variable Z, and the key entity of the running state in the domain knowledge causal graph is the result variable Y. For the N monitoring parameter key entities [X1, X2, …, X N ] Perform multi-parameter intervention operations and calculate the causal intervention distribution P(Y=y|do(Z=z)) of the result variable Y=y when the multi-parameter intervention operation makes the cause variable Z=z; According to the N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct a health status assessment model for ultra-high voltage transformers.
3. The method for multi-parameter status assessment of a UHV transformer according to claim 2, characterized in that: The process of obtaining the domain knowledge causal graph of the UHV transformer is as follows: Get the text data of UHV transformer; Extracting entities and relationships from the text data of the UHV transformer to obtain key entities of operating status, key entities of monitoring parameters, and causal relationships between the key entities of operating status and the key entities of monitoring parameters; The key entities of the operating status and the key entities of the monitoring parameters are used as nodes V in the graph structure representation, and the causal relationship between the key entities of the operating status and the key entities of the monitoring parameters is used as the edge E in the graph structure representation to construct a domain knowledge causal graph.
4. The method for multi-parameter status assessment of a UHV transformer according to claim 2, characterized in that: The N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct the transformer health status assessment model as follows: The variable set [Z1, Z2, ..., Z M ] is the first-level indicator, with the N monitoring parameter key entities [X1, X2, ..., X N ] is taken as the secondary indicator, and the causal intervention distribution P(Y=y|do(Z=z)) is taken as the indicator weight to construct the UHV transformer health status assessment model.
5. A multi-parameter status assessment system for ultra-high voltage transformers, characterized in that: include: The first acquisition module is used to obtain the monitoring parameters of the current UHV transformer; An evaluation module is used to input the monitoring parameters of the current UHV transformer into a UHV transformer health status evaluation model to obtain the health status of the UHV transformer, wherein the UHV transformer health status evaluation model is constructed based on a domain knowledge causal graph of the UHV transformer.
6. The UHV transformer multi-parameter status assessment system according to claim 5, characterized in that: Also includes: The second acquisition module is used to obtain the domain knowledge causal graph of the UHV transformer; The third acquisition module is used to obtain N key entities of monitoring parameters of the UHV transformer [X1, X2, ..., X N ]; A preprocessing module is used to process the N monitoring parameter key entities [X1, X2, ..., X N ] to clean and reduce the data dimension, and obtain the variable set [Z1, Z2, ..., Z M ], where a variable after dimension reduction corresponds to several key entities of monitoring parameters; The calculation module is used to calculate the variable set [Z1, Z2, ..., Z M ] is the cause variable Z, and the key entity of the running state in the domain knowledge causal graph is the result variable Y. For the N monitoring parameter key entities [X1, X2, ..., X N ] Perform multi-parameter intervention operations and calculate the causal intervention distribution P(Y=y|do(Z=z)) of the result variable Y=y when the multi-parameter intervention operation makes the cause variable Z=z; The first building module is used to monitor the N key entities of the parameters [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct a health status assessment model for ultra-high voltage transformers.
7. The UHV transformer multi-parameter status assessment system according to claim 6, characterized in that: The second acquisition module includes: A fourth acquisition module is used to acquire text data of the UHV transformer; An extraction module is used to extract entities and relationships from the text data of the UHV transformer to obtain key entities of operating status, key entities of monitoring parameters, and causal relationships between the key entities of operating status and the key entities of monitoring parameters; The second construction module is used to construct a domain knowledge causal graph using the key entities of the operating status and the key entities of the monitoring parameters as nodes V in the graph structure representation, and using the causal relationship between the key entities of the operating status and the key entities of the monitoring parameters as edges E in the graph structure representation.
8. The UHV transformer multi-parameter status assessment system according to claim 6, characterized in that: The N monitoring parameter key entities [X1, X2, ..., X N ], the variable set after dimension reduction [Z1,Z2,...,Z M ] and the causal intervention distribution P(Y=y|do(Z=z)) to construct the transformer health status assessment model as follows: The variable set [Z1, Z2, ..., Z M ] is the first-level indicator, with the N monitoring parameter key entities [X1, X2, ..., X N ] is taken as the secondary indicator, and the causal intervention distribution P(Y=y|do(Z=z)) is taken as the indicator weight to construct the UHV transformer health status assessment model.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the multi-parameter state assessment method for ultra-high voltage transformers as described in any one of claims 1-4 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the multi-parameter state assessment method for ultra-high voltage transformers as described in any one of claims 1 to 4 are implemented.