CN116578869A - Fault diagnosis method, fault diagnosis device and electronic device for power system - Google Patents

Fault diagnosis method, fault diagnosis device and electronic device for power system Download PDF

Info

Publication number
CN116578869A
CN116578869A CN202310556638.2A CN202310556638A CN116578869A CN 116578869 A CN116578869 A CN 116578869A CN 202310556638 A CN202310556638 A CN 202310556638A CN 116578869 A CN116578869 A CN 116578869A
Authority
CN
China
Prior art keywords
power system
fault diagnosis
fault
target
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310556638.2A
Other languages
Chinese (zh)
Inventor
郭海平
郭琦
卢远宏
郭天宇
张�杰
黄立滨
涂亮
胡云
洪泽祺
刘宇嫣
苏明章
伍文聪
陈智豪
胡玉峰
赵艳军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China South Power Grid International Co ltd
China Southern Power Grid Co Ltd
Original Assignee
China South Power Grid International Co ltd
China Southern Power Grid Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China South Power Grid International Co ltd, China Southern Power Grid Co Ltd filed Critical China South Power Grid International Co ltd
Priority to CN202310556638.2A priority Critical patent/CN116578869A/en
Publication of CN116578869A publication Critical patent/CN116578869A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/241Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/243Classification techniques relating to the number of classes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/12Computing arrangements based on biological models using genetic models
    • G06N3/126Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00Computing arrangements based on specific mathematical models
    • G06N7/01Probabilistic graphical models, e.g. probabilistic networks
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04SSYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • Y04S10/52Outage or fault management, e.g. fault detection or location

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Genetics & Genomics (AREA)
  • Physiology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The application provides a fault diagnosis method, a fault diagnosis device and an electronic device of a power system. The method comprises the following steps: performing data preprocessing and feature extraction on the received original data set to obtain a target data set; determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on a target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the method comprises the steps of genetic algorithm and Bayesian optimization algorithm, and presetting a fault diagnosis model for carrying out fault diagnosis on a target power system; the power system data of the target power system received in real time is input into the target fault diagnosis model to obtain the fault diagnosis result of the target power system, and the problem that in the prior art, the accuracy and the efficiency are low due to the fact that the fault diagnosis is carried out on the power system by adopting manual experience and offline analysis is solved.

Description

Fault diagnosis method, fault diagnosis device and electronic device for power system
Technical Field
The present application relates to the field of power systems, and more particularly, to a fault diagnosis method, a fault diagnosis apparatus, a computer-readable storage medium, and an electronic apparatus for a power system.
Background
The power system is an infrastructure of modern industry and civilians, and its safe, stable and efficient operation is of paramount importance. However, during operation of the power system, equipment faults and power system anomalies are difficult to avoid.
In order to find and solve the problems in time, fault diagnosis of a power system becomes an important research field. The traditional fault diagnosis method generally depends on manual experience and offline analysis, but has the problems of lower accuracy and lower efficiency when the traditional fault diagnosis method performs fault diagnosis on the power system.
Therefore, a method for diagnosing faults of an electric power system with high accuracy and high efficiency is needed.
Disclosure of Invention
The application mainly aims to provide a fault diagnosis method, a fault diagnosis device, a computer readable storage medium and an electronic device for a power system, which at least solve the problem of low accuracy and efficiency caused by adopting manual experience and offline analysis to perform fault diagnosis on the power system in the prior art.
In order to achieve the above object, according to one aspect of the present application, there is provided a fault diagnosis method of an electric power system, comprising: the method comprises the steps of processing received original data sets, namely performing data preprocessing and feature extraction to obtain target data sets, wherein the original data sets comprise power system data of a target power system and power system data of other power systems, and the power system data comprise operation data, equipment state information and fault alarm data of the corresponding power systems; training, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system; and a fault diagnosis step of inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
Optionally, performing data preprocessing and feature extraction on the received original data set to obtain a target data set, including: based on a data preprocessing method, carrying out data preprocessing on the original data set to obtain a preprocessed original data set, wherein the data preprocessing method comprises the steps of removing repeated data, correcting error data, filling a missing value, and segmenting words and extracting word stems of text data; extracting characteristics of the text data in the preprocessed original data set to obtain text characteristics, extracting characteristics of the numerical data in the preprocessed original data set to obtain first numerical characteristics, and extracting characteristics of the category data in the preprocessed original data set to obtain category characteristics; converting the text feature and the category feature based on a target conversion method to obtain a second numerical feature, wherein the target conversion method comprises one of the following steps: word embedding and vectorization; the target dataset is composed of the first numerical feature and the second numerical feature.
Optionally, inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system, including: performing anomaly detection and filtering processing on the power system data of the target power system received in real time to obtain processed power system data; and inputting the processed power system data into the target fault diagnosis model to obtain the fault diagnosis result of the target power system, wherein the fault diagnosis result comprises whether the target power system has a fault or not, and the fault diagnosis result also comprises a fault type under the condition that the target power system has the fault.
Optionally, after inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system, the fault diagnosis method further includes: determining fault occurrence time of the target power system according to fault types and fault recording data in the fault diagnosis result, wherein the fault recording data comprises current waveforms, voltage waveforms and equipment state information of the target power system; and analyzing equipment, topological structures and equipment association relations of the target power system according to the fault occurrence time, so as to locate faults of the target power system.
Optionally, after inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system, the fault diagnosis method further includes: converting the fault diagnosis result based on a natural language generation technology to obtain a natural language text corresponding to the fault diagnosis result, and extracting keywords from the natural language text to obtain a fault type, a name of fault equipment, a fault position and a fault reason in the fault diagnosis result; determining a fault processing scheme corresponding to the fault type in the fault diagnosis result according to the fault type, the name of the fault equipment, the fault position, the fault reason and the knowledge base in the fault diagnosis result; and filling the fault processing scheme into a fault processing strategy template to obtain the fault processing strategy.
Optionally, the preset fault diagnosis model is obtained by model fusion of a plurality of different machine learning models.
Optionally, the fault diagnosis method further includes: and adding the power system data of the target power system received in real time into the original data set, and executing the processing step, the training step and the fault diagnosis step at least once according to the data.
According to another aspect of the present application, there is provided a fault diagnosis apparatus of an electric power system, including: the processing unit is used for executing the processing steps, carrying out data preprocessing and feature extraction on the received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system; the training unit is used for executing the training step, determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system; and the execution unit is used for executing a fault diagnosis step, and inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
According to still another aspect of the present application, there is provided a computer-readable storage medium including a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform any one of the fault diagnosis methods of the power system.
According to still another aspect of the present application, there is provided an electronic apparatus including a memory in which a computer program is stored, and a processor configured to execute any one of the fault diagnosis methods of the power system by the computer program.
By applying the technical scheme, firstly, a target data set for training a preset fault diagnosis model is constructed, namely, the original data set comprising power system data of a target power system and power system data of other power systems is subjected to data preprocessing and feature extraction to obtain the target data set; then, determining a model structure and super parameters of a preset fault diagnosis model based on a super parameter optimization algorithm, and training the preset fault diagnosis model by adopting a target data set so as to obtain a target fault diagnosis model; and finally, processing the power system data of the target power system acquired in real time by using the target fault diagnosis model to obtain a fault diagnosis result of the target power system. Compared with the prior art that the method relies on manual experience and offline analysis, the method and the device adopt the target fault diagnosis model obtained after training, process the power system data of the target power system acquired in real time to obtain the fault diagnosis result of the target power system, ensure that the obtained fault diagnosis result is accurate, ensure higher efficiency of fault diagnosis of the power system, and further solve the problem that in the prior art, the accuracy and the efficiency are lower due to the fact that the manual experience and the offline analysis are adopted to perform fault diagnosis on the power system.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:
fig. 1 illustrates a hardware block diagram of a mobile terminal performing a fault diagnosis method of a power system according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a fault diagnosis method of an electric power system according to an embodiment of the present application;
fig. 3 shows a schematic structural diagram of a fault diagnosis apparatus for an electric power system according to an embodiment of the present application.
Wherein the above figures include the following reference numerals:
102. a processor; 104. a memory; 106. a transmission device; 108. and an input/output device.
Detailed Description
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the application herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
As described in the background art, in the prior art, the fault diagnosis is performed on the power system by using manual experience and offline analysis, which results in lower accuracy and efficiency, and in order to solve the above problems, the embodiments of the present application provide a fault diagnosis method, a fault diagnosis device, a computer readable storage medium and an electronic device for the power system.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.
The method embodiments provided in the embodiments of the present application may be performed in a mobile terminal, a computer terminal or similar computing device. Taking the mobile terminal as an example, fig. 1 is a block diagram of a hardware structure of the mobile terminal of a fault diagnosis method of a power system according to an embodiment of the present application. As shown in fig. 1, a mobile terminal may include one or more (only one is shown in fig. 1) processors 102 (the processor 102 may include, but is not limited to, a microprocessor MCU or a processing device such as a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 for communication functions and an input-output device 108. It will be appreciated by those skilled in the art that the structure shown in fig. 1 is merely illustrative and not limiting of the structure of the mobile terminal described above. For example, the mobile terminal may also include more or fewer components than shown in fig. 1, or have a different configuration than shown in fig. 1.
The memory 104 may be used to store a computer program, for example, a software program of application software and a module, such as a computer program corresponding to a fault diagnosis method of a power system in an embodiment of the present application, and the processor 102 executes the computer program stored in the memory 104 to perform various functional applications and data processing, that is, to implement the above-described method. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located relative to the processor 102, which may be connected to the mobile terminal via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, simply referred to as NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is configured to communicate with the internet wirelessly.
In the present embodiment, a fault diagnosis method of a power system operating on a mobile terminal, a computer terminal, or a similar computing device is provided, it is to be noted that the steps shown in the flowcharts of the drawings may be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be executed in an order different from that here.
Fig. 2 is a flowchart of a fault diagnosis method of a power system according to an embodiment of the present application. As shown in fig. 2, the fault diagnosis method includes the steps of:
step S201, a processing step, in which data preprocessing and feature extraction are carried out on a received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system;
in the processing step, the original data set is subjected to data preprocessing and feature extraction, so that the follow-up training of a preset fault diagnosis model based on the target data set is further ensured, and the obtained target fault diagnosis model is accurate.
In addition, in the practical application process, similar fault characteristics and rules may exist between different power systems, so that the prior knowledge can be provided for the preset fault diagnosis model of the target power system by utilizing the power system data of other power systems by means of transfer learning and field self-adaptive technology. That is, the power system data of the target power system and the power system data of other power systems can be adopted to train the preset fault diagnosis model, so that the requirement of training data can be effectively reduced, and the generalization performance of the target fault diagnosis model can be improved.
In the above processing steps, the power system data of the target power system and the power system data of other power systems may be collected through various approaches such as a corresponding SCADA system (data acquisition and monitoring control system, supervisory Control and Data Acquisition, abbreviated as SCADA) and/or a smart grid monitoring device. While the collected power system data may include structured data (e.g., numeric data) and unstructured data (e.g., textual data).
Step S202, a training step, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
In the training step, the model structure and the super parameters of the preset fault diagnosis model can be automatically searched through a super parameter optimization algorithm (namely a target algorithm, such as a Bayesian optimization algorithm or a genetic algorithm, and the like), so that the performance of the target fault diagnosis model obtained through subsequent training in practical application is improved. In addition, in the model training process of the preset fault diagnosis model by adopting the target data set, the parameters of the preset fault diagnosis model can be dynamically adjusted by utilizing the self-adaptive adjustment strategy, so that the preset fault diagnosis model can be better adapted to the characteristics of training data.
And step S203, a fault diagnosis step, namely inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
Through the embodiment, firstly, a target data set for training a preset fault diagnosis model is constructed, namely, data preprocessing and feature extraction are carried out on an original data set comprising power system data of a target power system and power system data of other power systems to obtain the target data set; then, determining a model structure and super parameters of a preset fault diagnosis model based on a super parameter optimization algorithm, and training the preset fault diagnosis model by adopting a target data set so as to obtain a target fault diagnosis model; and finally, processing the power system data of the target power system acquired in real time by using the target fault diagnosis model to obtain a fault diagnosis result of the target power system. Compared with the prior art that the method relies on manual experience and offline analysis, the method and the device adopt the target fault diagnosis model obtained after training, process the power system data of the target power system acquired in real time to obtain the fault diagnosis result of the target power system, ensure that the obtained fault diagnosis result is accurate, ensure higher efficiency of fault diagnosis of the power system, and further solve the problem that in the prior art, the accuracy and the efficiency are lower due to the fact that the manual experience and the offline analysis are adopted to perform fault diagnosis on the power system.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
In a specific implementation process, the step S201 may be implemented through step S2011, step S2012, step S2013, and step S2014. In step S2011, based on the data preprocessing method, the data preprocessing is performed on the original data set to obtain the preprocessed original data set, so that the quality and reliability of the obtained preprocessed original data set are ensured to be better. The data preprocessing method comprises the steps of removing repeated data, correcting error data, filling a missing value, and segmenting and extracting word stems of text data. Step S2012, performing feature extraction on the text data in the preprocessed original data set to obtain text features, performing feature extraction on the numerical data in the preprocessed original data set to obtain first numerical features, and performing feature extraction on the category data in the preprocessed original data set to obtain category features; step S2013, converting the text feature and the category feature based on a target conversion method to obtain a second numerical feature, where the target conversion method includes one of the following steps: word embedding and vectorization; in step S2014, the first numerical feature and the second numerical feature form the target data set, so that the obtained target data set is further ensured to be more accurate, and the digitized target data set is easier to preset for reading of the fault diagnosis model. And training a preset fault diagnosis model by adopting a target data set later, so that the obtained target fault diagnosis model is further ensured to be accurate and reliable.
In the above embodiment, the data preprocessing method is not limited to removing duplicate data, correcting error data, filling missing values, and word segmentation and stem extraction of text data, but may be any other possible data preprocessing and data cleaning method in the prior art.
In the actual application process, the preprocessed original data set may include text-type data, i.e., text data for short, numeric-type data, i.e., numeric data for short, and category-type data, i.e., category data for short. Wherein keywords and phrases are extracted from text data (e.g., fault descriptions, repair records, etc.) as text features. As the numerical characteristics, characteristics (e.g., mean, variance, peak, etc.) are extracted and counted from numerical data (e.g., data of current, voltage, temperature, etc.). Category data (e.g., device type, fault type, etc.) is encoded as category characteristics.
Since the category features and the text features are not numerical features, in order to facilitate the reading of a preset fault diagnosis model, the extracted category features and text features can be converted into numerical features by word embedding and/or vectorization and other methods. In a specific embodiment, word2Vec or BERT Word embedding models may be used to convert category features and text features to numeric features.
In order to avoid interference to the target fault diagnosis model caused by the power system data of the target power system received in real time, and further ensure that the accuracy of the obtained fault diagnosis result is high, the above step S203 of the present application may be implemented through step S2031 and step S2032. Step S2031, performing anomaly detection and filtering processing on the power system data of the target power system received in real time, to obtain processed power system data; step S2032, inputting the processed power system data into the target fault diagnosis model to obtain the fault diagnosis result of the target power system, where the fault diagnosis result includes whether the target power system has a fault, and where the fault diagnosis result includes a fault type when the target power system has a fault.
In the above embodiment, the power system data of the target power system received in real time is processed by using the target fault diagnosis model, and before the fault diagnosis result is obtained, the abnormality detection and filtering processing are performed on the received power system data of the target power system, so that the abnormality data (such as noise, interference, false alarm, etc.) in the power system data of the target power system received in real time can be identified. By eliminating abnormal data in the power system data of the target power system received in real time, the target fault diagnosis model can be prevented from being interfered, and the accuracy of the target fault diagnosis model can be improved.
In the actual application process, the fault diagnosis result is not limited to include whether the target power system has a fault or not, and in the case that the target power system has a fault, the fault diagnosis result also includes a fault type, and the fault diagnosis result may also include a device name related to the fault.
In order to locate the fault of the target power system more accurately, in some embodiments, the fault diagnosis method of the present application further includes step S204 and step S205. Step S204, after inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system, determining a fault occurrence time of the target power system according to a fault type and fault recording data in the fault diagnosis result, where the fault recording data includes a current waveform, a voltage waveform and equipment state information of the target power system; and step S205, analyzing the equipment, the topological structure and the equipment association relation of the target power system according to the fault occurrence time, so as to locate the fault of the target power system.
In a specific embodiment of the present application, when the fault is located on the target power system based on the fault diagnosis result, analysis can be performed by combining with actual fault recording data of the target power system. The method comprises the following specific steps:
step S1: and determining the type of the fault, equipment related to the fault and the possible occurrence position of the fault according to the fault diagnosis result.
Step S2: and acquiring actual fault recording data from a protection device, an SCADA system, intelligent power grid equipment and the like of the target power system. The fault log data typically includes current waveforms, voltage waveforms, and device status information for the target power system.
Step S3: and determining the fault occurrence time according to the fault diagnosis result and the fault recording data. Wherein the fault occurrence time can be used to analyze data of a plurality of devices simultaneously, thereby precisely locating the location of the fault occurrence of the target power system.
Step S4: and analyzing waveform characteristics in the fault record data by combining the fault type and equipment information of the target power system. For example, for short-circuit faults, abrupt changes in current waveform and decreases in fault voltage may be of concern; for overload faults, the case where the current of longer duration exceeds the nominal value may be of concern.
Step S5: according to the topological structure of the target power system, the connection relation between the devices is analyzed, so that the fault locating range is reduced, and the locating accuracy is improved.
Step S6: according to the running state, the historical fault record, the maintenance information and the like among the devices, the association relation of the devices is analyzed, and further determination of the fault source and the influence range is facilitated.
Step S7: and combining the analysis to accurately locate the fault position. Fault location algorithms (e.g., traveling wave location, power flow analysis, etc.) may be used to assist in the determination.
Step S8: after the fault location is determined, the cause of the fault can be further analyzed, helping to optimize the maintenance scheme and providing support for the prevention of similar faults.
In some specific implementations, the fault diagnosis method of the present application further includes step S206, step S207, and step S208. Step S206, after inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system, converting the fault diagnosis result based on a natural language generation technology to obtain a natural language text corresponding to the fault diagnosis result, and extracting keywords from the natural language text to obtain a fault type, a name of a fault device, a location of a fault and a fault cause in the fault diagnosis result; step S207, determining a fault processing scheme corresponding to the fault type in the fault diagnosis result according to the fault type, the name of the fault equipment, the fault position, the fault reason and the knowledge base in the fault diagnosis result; step S208, filling the fault processing scheme into a fault processing strategy template to obtain the fault processing strategy, so that the generated fault processing strategy is ensured to be accurate, the readability of the generated fault processing strategy is ensured to be high, and the stability and the reliability of the target power system are further ensured to be high.
Using Natural Language Generation (NLG) techniques, fault handling strategies generated from fault diagnosis results may also include emergency handling measures (e.g., isolating faulty equipment, adjusting operational strategies, etc.), repair plans (e.g., replacing damaged components, adjusting equipment parameters, etc.), and equipment optimization suggestions (e.g., upgrading equipment, improving operational strategies, etc.). These suggestions may provide references to field engineers. Meanwhile, as the running state of the power system changes, the fault processing strategy can be updated in real time so as to ensure an optimal processing scheme.
In a specific embodiment of the present application, a Natural Language Generation (NLG) technique can convert structured data (numeric data in a fault diagnosis result) into a natural language text, and provide an understandable suggestion for power system fault processing, so that a more accurate fault processing strategy can be generated according to the fault diagnosis result, and the specific steps are as follows:
step S1: and converting the fault diagnosis result based on a natural language generation technology to obtain a natural language text corresponding to the fault diagnosis result.
Step S2: keyword extraction is performed on the natural language text, such as fault type, name of fault device, location of fault, and cause of fault.
Step S3: relevant knowledge and experience of fault processing of the power system are collected, and a knowledge base is established, wherein the knowledge base can comprise the mapping relation between fault types and processing methods, equipment characteristics, operation rules and the like.
Step S4: and analyzing a fault processing scheme suitable for the current fault according to the fault type, the name of fault equipment, the fault position, the fault reason and the knowledge base in the extracted fault diagnosis result. The fault handling scheme may include emergency handling measures, equipment repair schemes, equipment optimization suggestions, and the like.
Step S5: according to the fault handling scheme, a text template (i.e., a fault handling policy template) for generating a fault handling policy is designed. The fault handling policy template should include padding slots to insert the actual fault information during the generation process.
Step S6: and filling the fault diagnosis result and the fault processing scheme into a fault processing strategy template, and generating a natural language text by using an NLG technology. This may use existing NLG models, e.g., GPT-4.
Step S7: and evaluating whether the generated fault processing strategy is accurate, easy to understand and targeted. The quality of advice may be confirmed by subjective assessment or discussion with professionals.
Step S8: the knowledge base is continuously updated according to the actual fault handling process and feedback information so as to generate more accurate fault handling suggestions.
Step S9: the generated fault handling policy may require real-time updates based on real-time changes in the operating state of the target power system. And when the system state changes, repeating the steps to generate a new fault processing strategy.
Through the steps, a Natural Language Generation (NLG) technology is utilized, and a relatively accurate fault processing strategy can be generated according to a fault diagnosis result. In addition, as fault handling practices accumulate, the knowledge base will continue to be refined and the generated recommendations will be more accurate and efficient.
In order to further improve the robustness and accuracy of the target fault diagnosis model obtained through training, an integrated learning and model fusion technology can be adopted, and a plurality of different machine learning technologies are combined to obtain the preset fault diagnosis model. For example, an ensemble learning method such as Bagging, boosting, stacking may be used, so in an embodiment of the present application, the preset fault diagnosis model is obtained by performing model fusion by using a plurality of different machine learning models.
In a specific embodiment of the present application, the machine learning may further use a supervised learning model such as a support vector machine, a decision tree, and a neural network to classify the fault; an unsupervised learning model such as a clustering algorithm and a dimension reduction technology can be used for finding potential failure modes. Meanwhile, in the process of training the preset fault diagnosis model, historical fault data can be utilized, and parameters of the preset fault diagnosis model can be optimized through methods such as cross verification and grid search.
Of course, in the actual application process, the fault diagnosis results of a plurality of different machine learning can be integrated, so that more accurate fault diagnosis results can be obtained.
For the real-time change of the running state of the power system, the incremental learning and online learning technology can be adopted to update the target fault diagnosis model in real time. As new fault data and equipment status information is collected, the parameters and structure of the target fault diagnostic model may be continually adjusted to better accommodate the actual operating conditions of the power system. In a specific embodiment of the present application, the fault diagnosis method further includes adding the power system data of the target power system received in real time to the raw data set, and performing the processing step, the training step, and the fault diagnosis step at least once in this way.
In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the fault diagnosis method of the power system of the present application will be described in detail with reference to specific embodiments.
The embodiment relates to a specific fault diagnosis method of a power system, which comprises the following steps:
In the event of a fault in the target power system, the monitoring system collects operational data, equipment status information and fault alarm records of the target power system.
Step S1: and performing anomaly detection and filtering processing on the operation data, the equipment state information and the fault alarm record of the target power system.
Step S2: features are extracted by Natural Language Processing (NLP) and converted into numerical features.
Step S3: and (3) inputting the numerical characteristics obtained in the step (S2) into a target fault diagnosis model. And processing the numerical characteristics in the step S2 by adopting a target fault diagnosis model to obtain a fault diagnosis result.
Step S4: and according to the fault diagnosis result and the topological structure of the target power system, performing fault determination on the target power system, namely determining a certain device in a certain specific area where the fault occurs. Further analysis of historical fault data and status information of the device, finds that the fault may be caused by equipment aging and improper operating parameters.
Step S5: through Natural Language Generation (NLG) technology, a targeted fault handling strategy is provided for field engineers. Such as isolating faulty equipment, replacing damaged components, adjusting equipment operating parameters, and the like. The field engineer performs the process according to the fault handling strategy.
The embodiment of the application also provides a fault diagnosis device of the power system, and the fault diagnosis device of the power system can be used for executing the fault diagnosis method for the power system. The device is used for realizing the above embodiments and preferred embodiments, and is not described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
The following describes a fault diagnosis apparatus for an electric power system according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a fault diagnosis apparatus of a power system according to an embodiment of the present application. As shown in fig. 3, the fault diagnosis apparatus includes:
the processing unit 10 is configured to perform a processing step, perform data preprocessing and feature extraction on a received original data set to obtain a target data set, where the original data set includes power system data of a target power system and power system data of other power systems, and the power system data includes operation data, equipment state information and fault alarm data of the corresponding power system;
In the processing step, the original data set is subjected to data preprocessing and feature extraction, so that the follow-up training of a preset fault diagnosis model based on the target data set is further ensured, and the obtained target fault diagnosis model is accurate.
In addition, in the practical application process, similar fault characteristics and rules may exist between different power systems, so that the prior knowledge can be provided for the preset fault diagnosis model of the target power system by utilizing the power system data of other power systems by means of transfer learning and field self-adaptive technology. That is, the power system data of the target power system and the power system data of other power systems can be adopted to train the preset fault diagnosis model, so that the requirement of training data can be effectively reduced, and the generalization performance of the target fault diagnosis model can be improved.
In the above processing steps, the power system data of the target power system and the power system data of other power systems may be collected through various approaches such as a corresponding SCADA system (data acquisition and monitoring control system, supervisory Control and Data Acquisition, abbreviated as SCADA) and/or a smart grid monitoring device. While the collected power system data may include structured data (e.g., numeric data) and unstructured data (e.g., textual data).
A training unit 20, configured to perform a training step, determine a model structure and a hyper parameter of a preset fault diagnosis model based on a target algorithm, and train the preset fault diagnosis model based on the target data set to obtain a target fault diagnosis model, where the target algorithm includes one of the following: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
in the training step, the model structure and the super parameters of the preset fault diagnosis model can be automatically searched through a super parameter optimization algorithm (namely a target algorithm, such as a Bayesian optimization algorithm or a genetic algorithm, and the like), so that the performance of the target fault diagnosis model obtained through subsequent training in practical application is improved. In addition, in the model training process of the preset fault diagnosis model by adopting the target data set, the parameters of the preset fault diagnosis model can be dynamically adjusted by utilizing the self-adaptive adjustment strategy, so that the preset fault diagnosis model can be better adapted to the characteristics of training data.
And an execution unit 30, configured to execute a fault diagnosis step, and input the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
Through the embodiment, the processing unit is configured to construct a target data set for training a preset fault diagnosis model, that is, perform data preprocessing and feature extraction on an original data set including power system data of a target power system and power system data of other power systems to obtain the target data set; the training unit is used for determining a model structure and super parameters of a preset fault diagnosis model based on a super parameter optimization algorithm, and training the preset fault diagnosis model by adopting a target data set so as to obtain a target fault diagnosis model; the execution unit is used for processing the power system data of the target power system acquired in real time by using the target fault diagnosis model to obtain a fault diagnosis result of the target power system. Compared with the prior art that the method relies on manual experience and offline analysis, the method and the device adopt the target fault diagnosis model obtained after training, process the power system data of the target power system acquired in real time to obtain the fault diagnosis result of the target power system, ensure that the obtained fault diagnosis result is accurate, ensure higher efficiency of fault diagnosis of the power system, and further solve the problem that in the prior art, the accuracy and the efficiency are lower due to the fact that the manual experience and the offline analysis are adopted to perform fault diagnosis on the power system.
In a specific implementation process, the processing unit comprises a first processing module, a feature extraction module, a conversion module and a combination module, wherein the first processing module is used for preprocessing the data of the original data set based on a data preprocessing method to obtain the preprocessed original data set, and therefore the quality and the reliability of the obtained preprocessed original data set are guaranteed to be good. The data preprocessing method comprises the steps of removing repeated data, correcting error data, filling a missing value, and segmenting and extracting word stems of text data. The feature extraction module is used for carrying out feature extraction on the text data in the preprocessed original data set to obtain text features, carrying out feature extraction on the numerical data in the preprocessed original data set to obtain first numerical features, and carrying out feature extraction on the category data in the preprocessed original data set to obtain category features; the conversion module is configured to convert the text feature and the class feature based on a target conversion method to obtain a second numerical feature, where the target conversion method includes one of the following steps: word embedding and vectorization; the combination module is used for forming the target data set by the first numerical value characteristic and the second numerical value characteristic, so that the obtained target data set is more accurate, and the digitized target data set is easier to preset and read the fault diagnosis model. And training a preset fault diagnosis model by adopting a target data set later, so that the obtained target fault diagnosis model is further ensured to be accurate and reliable.
In the above embodiment, the data preprocessing method is not limited to removing duplicate data, correcting error data, filling missing values, and word segmentation and stem extraction of text data, but may be any other possible data preprocessing and data cleaning method in the prior art.
In the actual application process, the preprocessed original data set may include text-type data, i.e., text data for short, numeric-type data, i.e., numeric data for short, and category-type data, i.e., category data for short. Wherein keywords and phrases are extracted from text data (e.g., fault descriptions, repair records, etc.) as text features. As the numerical characteristics, characteristics (e.g., mean, variance, peak, etc.) are extracted and counted from numerical data (e.g., data of current, voltage, temperature, etc.). Category data (e.g., device type, fault type, etc.) is encoded as category characteristics.
Since the category features and the text features are not numerical features, in order to facilitate the reading of a preset fault diagnosis model, the extracted category features and text features can be converted into numerical features by word embedding and/or vectorization and other methods. In a specific embodiment, word2Vec or BERT Word embedding models may be used to convert category features and text features to numeric features.
In order to avoid the power system data of the target power system received in real time, generate interference to the target fault diagnosis model, and further ensure that the accuracy of the obtained fault diagnosis result is higher, the execution unit of the application further comprises a second processing module and an execution module, wherein the second processing module is used for performing anomaly detection and filtering processing on the power system data of the target power system received in real time to obtain the processed power system data; the execution module is configured to input the processed power system data into the target fault diagnosis model to obtain the fault diagnosis result of the target power system, where the fault diagnosis result includes whether the target power system has a fault, and where the fault diagnosis result includes a fault type when the target power system has a fault.
In the above embodiment, the power system data of the target power system received in real time is processed by using the target fault diagnosis model, and before the fault diagnosis result is obtained, the abnormality detection and filtering processing are performed on the received power system data of the target power system, so that the abnormality data (such as noise, interference, false alarm, etc.) in the power system data of the target power system received in real time can be identified. By eliminating abnormal data in the power system data of the target power system received in real time, the target fault diagnosis model can be prevented from being interfered, and the accuracy of the target fault diagnosis model can be improved.
In the actual application process, the fault diagnosis result is not limited to include whether the target power system has a fault or not, and in the case that the target power system has a fault, the fault diagnosis result also includes a fault type, and the fault diagnosis result may also include a device name related to the fault.
In order to locate a fault of a target power system more accurately, in some embodiments, the fault diagnosis apparatus of the present application further includes a first determining unit and an analyzing unit. The first determining unit is configured to determine, after inputting power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system, a fault occurrence time of the target power system according to a fault type and fault recording data in the fault diagnosis result, where the fault recording data includes a current waveform, a voltage waveform, and equipment state information of the target power system; the analysis unit is used for analyzing the equipment, the topological structure and the equipment association relation of the target power system according to the fault occurrence time, so as to locate the fault of the target power system.
In a specific embodiment of the present application, when the fault is located on the target power system based on the fault diagnosis result, analysis can be performed by combining with actual fault recording data of the target power system. The method comprises the following specific steps:
step S1: and determining the type of the fault, equipment related to the fault and the possible occurrence position of the fault according to the fault diagnosis result.
Step S2: and acquiring actual fault recording data from a protection device, an SCADA system, intelligent power grid equipment and the like of the target power system. The fault log data typically includes current waveforms, voltage waveforms, and device status information for the target power system.
Step S3: and determining the fault occurrence time according to the fault diagnosis result and the fault recording data. Wherein the fault occurrence time can be used to analyze data of a plurality of devices simultaneously, thereby precisely locating the location of the fault occurrence of the target power system.
Step S4: and analyzing waveform characteristics in the fault record data by combining the fault type and equipment information of the target power system. For example, for short-circuit faults, abrupt changes in current waveform and decreases in fault voltage may be of concern; for overload faults, the case where the current of longer duration exceeds the nominal value may be of concern.
Step S5: according to the topological structure of the target power system, the connection relation between the devices is analyzed, so that the fault locating range is reduced, and the locating accuracy is improved.
Step S6: according to the running state, the historical fault record, the maintenance information and the like among the devices, the association relation of the devices is analyzed, and further determination of the fault source and the influence range is facilitated.
Step S7: and combining the analysis to accurately locate the fault position. Fault location algorithms (e.g., traveling wave location, power flow analysis, etc.) may be used to assist in the determination.
Step S8: after the fault location is determined, the cause of the fault can be further analyzed, helping to optimize the maintenance scheme and providing support for the prevention of similar faults.
In some specific implementations, the fault diagnosis apparatus of the present application further includes an extraction unit, a second determination unit, and a filling unit. The extraction unit is used for inputting the power system data of the target power system received in real time into the target fault diagnosis model, converting the fault diagnosis result based on a natural language generation technology after obtaining the fault diagnosis result of the target power system, obtaining a natural language text corresponding to the fault diagnosis result, and extracting keywords from the natural language text to obtain the fault type, the name of fault equipment, the position of fault and the fault reason in the fault diagnosis result; the second determining unit is configured to determine a fault handling scheme corresponding to the fault type in the fault diagnosis result according to the fault type, the name of the fault device, the location of the fault, the fault cause and the knowledge base in the fault diagnosis result; the filling unit is used for filling the fault processing scheme into the fault processing strategy template to obtain the fault processing strategy, so that the generated fault processing strategy is ensured to be accurate, the readability of the generated fault processing strategy is ensured to be high, and the stability and the reliability of the target power system are further ensured to be high.
Using Natural Language Generation (NLG) techniques, fault handling strategies generated from fault diagnosis results may also include emergency handling measures (e.g., isolating faulty equipment, adjusting operational strategies, etc.), repair plans (e.g., replacing damaged components, adjusting equipment parameters, etc.), and equipment optimization suggestions (e.g., upgrading equipment, improving operational strategies, etc.). These suggestions may provide references to field engineers. Meanwhile, as the running state of the power system changes, the fault processing strategy can be updated in real time so as to ensure an optimal processing scheme.
In a specific embodiment of the present application, a Natural Language Generation (NLG) technique can convert structured data (numeric data in a fault diagnosis result) into a natural language text, and provide an understandable suggestion for power system fault processing, so that a more accurate fault processing strategy can be generated according to the fault diagnosis result, and the specific steps are as follows:
step S1: and converting the fault diagnosis result based on a natural language generation technology to obtain a natural language text corresponding to the fault diagnosis result.
Step S2: keyword extraction is performed on the natural language text, such as fault type, name of fault device, location of fault, and cause of fault.
Step S3: relevant knowledge and experience of fault processing of the power system are collected, and a knowledge base is established, wherein the knowledge base can comprise the mapping relation between fault types and processing methods, equipment characteristics, operation rules and the like.
Step S4: and analyzing a fault processing scheme suitable for the current fault according to the fault type, the name of fault equipment, the fault position, the fault reason and the knowledge base in the extracted fault diagnosis result. The fault handling scheme may include emergency handling measures, equipment repair schemes, equipment optimization suggestions, and the like.
Step S5: according to the fault handling scheme, a text template (i.e., a fault handling policy template) for generating a fault handling policy is designed. The fault handling policy template should include padding slots to insert the actual fault information during the generation process.
Step S6: and filling the fault diagnosis result and the fault processing scheme into a fault processing strategy template, and generating a natural language text by using an NLG technology. This may use existing NLG models, e.g., GPT-4.
Step S7: and evaluating whether the generated fault processing strategy is accurate, easy to understand and targeted. The quality of advice may be confirmed by subjective assessment or discussion with professionals.
Step S8: the knowledge base is continuously updated according to the actual fault handling process and feedback information so as to generate more accurate fault handling suggestions.
Step S9: the generated fault handling policy may require real-time updates based on real-time changes in the operating state of the target power system. And when the system state changes, repeating the steps to generate a new fault processing strategy.
Through the steps, a Natural Language Generation (NLG) technology is utilized, and a relatively accurate fault processing strategy can be generated according to a fault diagnosis result. In addition, as fault handling practices accumulate, the knowledge base will continue to be refined and the generated recommendations will be more accurate and efficient.
In order to further improve the robustness and accuracy of the target fault diagnosis model obtained through training, an integrated learning and model fusion technology can be adopted, and a plurality of different machine learning technologies are combined to obtain the preset fault diagnosis model. For example, an ensemble learning method such as Bagging, boosting, stacking may be used, so in an embodiment of the present application, the preset fault diagnosis model is obtained by performing model fusion by using a plurality of different machine learning models.
In a specific embodiment of the present application, the machine learning may further use a supervised learning model such as a support vector machine, a decision tree, and a neural network to classify the fault; an unsupervised learning model such as a clustering algorithm and a dimension reduction technology can be used for finding potential failure modes. Meanwhile, in the process of training the preset fault diagnosis model, historical fault data can be utilized, and parameters of the preset fault diagnosis model can be optimized through methods such as cross verification and grid search.
Of course, in the actual application process, the fault diagnosis results of a plurality of different machine learning can be integrated, so that more accurate fault diagnosis results can be obtained.
For the real-time change of the running state of the power system, the incremental learning and online learning technology can be adopted to update the target fault diagnosis model in real time. As new fault data and equipment status information is collected, the parameters and structure of the target fault diagnostic model may be continually adjusted to better accommodate the actual operating conditions of the power system. In a specific embodiment of the present application, the fault diagnosis apparatus further includes a receiving unit configured to add the power system data of the target power system received in real time to the raw data set, and perform the processing step, the training step, and the fault diagnosis step at least once in this manner.
The fault diagnosis device of the power system comprises a processor and a memory, wherein the processing unit, the training unit, the execution unit and the like are all stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions. The modules are all located in the same processor; alternatively, the above modules may be located in different processors in any combination.
The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one kernel, and the problem of low accuracy and efficiency caused by fault diagnosis of the power system by adopting manual experience and offline analysis in the prior art is solved by adjusting kernel parameters.
The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
The embodiment of the invention provides a computer readable storage medium, which comprises a stored program, wherein the program is used for controlling equipment where the computer readable storage medium is located to execute the fault diagnosis method of the power system.
Specifically, the fault diagnosis method of the electric power system includes:
step S201, a processing step, in which data preprocessing and feature extraction are carried out on a received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system;
Step S202, a training step, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
and step S203, a fault diagnosis step, namely inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
An embodiment of the present invention provides an electronic device including a memory in which a computer program is stored, and a processor configured to execute the fault diagnosis method of the power system by the computer program.
Specifically, the fault diagnosis method of the electric power system includes:
step S201, a processing step, in which data preprocessing and feature extraction are carried out on a received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system;
Step S202, a training step, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
and step S203, a fault diagnosis step, namely inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program stored in the memory and capable of running on the processor, wherein the processor realizes at least the following steps when executing the program:
step S201, a processing step, in which data preprocessing and feature extraction are carried out on a received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system;
Step S202, a training step, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
and step S203, a fault diagnosis step, namely inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
The device herein may be a server, PC, PAD, cell phone, etc.
The application also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with at least the following method steps:
step S201, a processing step, in which data preprocessing and feature extraction are carried out on a received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system;
Step S202, a training step, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
and step S203, a fault diagnosis step, namely inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may be implemented in program code executable by computing devices, so that they may be stored in a storage device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
It will be appreciated by those skilled in the art that 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 flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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 one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. 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 storage media for a computer 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, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
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 one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:
1) In the fault diagnosis method, firstly, a target data set for training a preset fault diagnosis model is constructed, namely, the original data set comprising power system data of a target power system and power system data of other power systems is subjected to data preprocessing and feature extraction to obtain the target data set; then, determining a model structure and super parameters of a preset fault diagnosis model based on a super parameter optimization algorithm, and training the preset fault diagnosis model by adopting a target data set so as to obtain a target fault diagnosis model; and finally, processing the power system data of the target power system acquired in real time by using the target fault diagnosis model to obtain a fault diagnosis result of the target power system. Compared with the prior art that the method relies on manual experience and offline analysis, the method and the device adopt the target fault diagnosis model obtained after training, process the power system data of the target power system acquired in real time to obtain the fault diagnosis result of the target power system, ensure that the obtained fault diagnosis result is accurate, ensure higher efficiency of fault diagnosis of the power system, and further solve the problem that in the prior art, the accuracy and the efficiency are lower due to the fact that the manual experience and the offline analysis are adopted to perform fault diagnosis on the power system.
2) In the fault diagnosis device, the processing unit is used for constructing a target data set for training a preset fault diagnosis model, namely, carrying out data preprocessing and feature extraction on an original data set comprising power system data of a target power system and power system data of other power systems to obtain the target data set; the training unit is used for determining a model structure and super parameters of a preset fault diagnosis model based on a super parameter optimization algorithm, and training the preset fault diagnosis model by adopting a target data set so as to obtain a target fault diagnosis model; the execution unit is used for processing the power system data of the target power system acquired in real time by using the target fault diagnosis model to obtain a fault diagnosis result of the target power system. Compared with the prior art that the method relies on manual experience and offline analysis, the method and the device adopt the target fault diagnosis model obtained after training, process the power system data of the target power system acquired in real time to obtain the fault diagnosis result of the target power system, ensure that the obtained fault diagnosis result is accurate, ensure higher efficiency of fault diagnosis of the power system, and further solve the problem that in the prior art, the accuracy and the efficiency are lower due to the fact that the manual experience and the offline analysis are adopted to perform fault diagnosis on the power system.
The above description is only of the preferred embodiments of the present application and is not intended to limit the present application, but various modifications and variations can be made to the present application by 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 protection scope of the present application.

Claims (10)

1. A fault diagnosis method of an electric power system, characterized by comprising:
the method comprises the steps of processing received original data sets, namely performing data preprocessing and feature extraction to obtain target data sets, wherein the original data sets comprise power system data of a target power system and power system data of other power systems, and the power system data comprise operation data, equipment state information and fault alarm data of the corresponding power systems;
training, namely determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
And a fault diagnosis step of inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
2. The method of claim 1, wherein the performing data preprocessing and feature extraction on the received raw data set to obtain the target data set includes:
based on a data preprocessing method, carrying out data preprocessing on the original data set to obtain a preprocessed original data set, wherein the data preprocessing method comprises the steps of removing repeated data, correcting error data, filling a missing value, and segmenting words and extracting word stems of text data;
extracting characteristics of the text data in the preprocessed original data set to obtain text characteristics, extracting characteristics of the numerical data in the preprocessed original data set to obtain first numerical characteristics, and extracting characteristics of the category data in the preprocessed original data set to obtain category characteristics;
converting the text feature and the category feature based on a target conversion method to obtain a second numerical feature, wherein the target conversion method comprises one of the following steps: word embedding and vectorization;
The target dataset is composed of the first numerical feature and the second numerical feature.
3. The fault diagnosis method according to claim 1, wherein inputting the power system data of the target power system received in real time into the target fault diagnosis model, to obtain a fault diagnosis result of the target power system, comprises:
performing anomaly detection and filtering processing on the power system data of the target power system received in real time to obtain processed power system data;
and inputting the processed power system data into the target fault diagnosis model to obtain the fault diagnosis result of the target power system, wherein the fault diagnosis result comprises whether the target power system has a fault or not, and the fault diagnosis result also comprises a fault type under the condition that the target power system has the fault.
4. The fault diagnosis method according to any one of claims 1 to 3, characterized in that, after inputting power system data of the target power system received in real time into the target fault diagnosis model, the fault diagnosis method further comprises:
Determining fault occurrence time of the target power system according to fault types and fault recording data in the fault diagnosis result, wherein the fault recording data comprises current waveforms, voltage waveforms and equipment state information of the target power system;
and analyzing equipment, topological structures and equipment association relations of the target power system according to the fault occurrence time, so as to locate faults of the target power system.
5. The fault diagnosis method according to any one of claims 1 to 3, characterized in that, after inputting power system data of the target power system received in real time into the target fault diagnosis model, the fault diagnosis method further comprises:
converting the fault diagnosis result based on a natural language generation technology to obtain a natural language text corresponding to the fault diagnosis result, and extracting keywords from the natural language text to obtain a fault type, a name of fault equipment, a fault position and a fault reason in the fault diagnosis result;
determining a fault processing scheme corresponding to the fault type in the fault diagnosis result according to the fault type, the name of the fault equipment, the fault position, the fault reason and the knowledge base in the fault diagnosis result;
And filling the fault processing scheme into a fault processing strategy template to obtain the fault processing strategy.
6. A fault diagnosis method according to any one of claims 1 to 3, wherein the predetermined fault diagnosis model is obtained by model fusion of a plurality of different machine learning models.
7. A fault diagnosis method according to any one of claims 1 to 3, characterized in that the fault diagnosis method further comprises:
and adding the power system data of the target power system received in real time into the original data set, and executing the processing step, the training step and the fault diagnosis step at least once according to the data.
8. A fault diagnosis apparatus for an electric power system, comprising:
the processing unit is used for executing the processing steps, carrying out data preprocessing and feature extraction on the received original data set to obtain a target data set, wherein the original data set comprises power system data of a target power system and power system data of other power systems, and the power system data comprises operation data, equipment state information and fault alarm data of the corresponding power system;
The training unit is used for executing the training step, determining a model structure and super parameters of a preset fault diagnosis model based on a target algorithm, and training the preset fault diagnosis model based on the target data set to obtain the target fault diagnosis model, wherein the target algorithm comprises one of the following steps: the preset fault diagnosis model is used for carrying out fault diagnosis on the target power system;
and the execution unit is used for executing a fault diagnosis step, and inputting the power system data of the target power system received in real time into the target fault diagnosis model to obtain a fault diagnosis result of the target power system.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform the fault diagnosis method of the electric power system according to any one of claims 1 to 7.
10. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to execute the fault diagnosis method of the power system according to any one of claims 1 to 7 by means of the computer program.
CN202310556638.2A 2023-05-16 2023-05-16 Fault diagnosis method, fault diagnosis device and electronic device for power system Pending CN116578869A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310556638.2A CN116578869A (en) 2023-05-16 2023-05-16 Fault diagnosis method, fault diagnosis device and electronic device for power system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310556638.2A CN116578869A (en) 2023-05-16 2023-05-16 Fault diagnosis method, fault diagnosis device and electronic device for power system

Publications (1)

Publication Number Publication Date
CN116578869A true CN116578869A (en) 2023-08-11

Family

ID=87535431

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310556638.2A Pending CN116578869A (en) 2023-05-16 2023-05-16 Fault diagnosis method, fault diagnosis device and electronic device for power system

Country Status (1)

Country Link
CN (1) CN116578869A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117272152A (en) * 2023-11-16 2023-12-22 晶科储能科技有限公司 Energy storage system fault diagnosis method, system, electronic equipment and storage medium
CN117668751A (en) * 2023-11-30 2024-03-08 广东一业建设股份有限公司 High-low voltage power system fault diagnosis method and device

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117272152A (en) * 2023-11-16 2023-12-22 晶科储能科技有限公司 Energy storage system fault diagnosis method, system, electronic equipment and storage medium
CN117272152B (en) * 2023-11-16 2024-08-06 晶科储能科技有限公司 Energy storage system fault diagnosis method, system, electronic equipment and storage medium
CN117668751A (en) * 2023-11-30 2024-03-08 广东一业建设股份有限公司 High-low voltage power system fault diagnosis method and device
CN117668751B (en) * 2023-11-30 2024-04-26 广东一业建设股份有限公司 High-low voltage power system fault diagnosis method and device

Similar Documents

Publication Publication Date Title
CN116578869A (en) Fault diagnosis method, fault diagnosis device and electronic device for power system
CN111435366A (en) Equipment fault diagnosis method and device and electronic equipment
CN109143094B (en) Abnormal data detection method and device for power battery
CN117408162B (en) Power grid fault control method based on digital twin
CN113570200A (en) Power grid operation state monitoring method and system based on multidimensional information
CN110187275B (en) Method and system for detecting health state of permanent magnet motor
CN116739829B (en) Big data-based power data analysis method, system and medium
CN113742993A (en) Method, device, equipment and storage medium for predicting life loss of dry-type transformer
CN116304928A (en) Power supply equipment fault prediction method, device, equipment and storage medium
CN117394529A (en) SCADA-based auxiliary decision method and system for main distribution network loop-closing reverse power supply control conditions
CN118311429B (en) Generator fault detection method and device based on power plant
CN115081673B (en) Abnormality prediction method and device for oil and gas pipeline, electronic equipment and medium
CN117172753A (en) Intelligent hydropower station inspection system and method based on mobile internet
CN112183610A (en) Transformer fault prediction method and transformer fault prediction device
CN113758652B (en) Oil leakage detection method and device for converter transformer, computer equipment and storage medium
CN118395358A (en) Intelligent anti-misoperation topology analysis method for transformer substation
CN113850305A (en) Generator fault diagnosis and health state assessment method and electronic equipment
CN117254446A (en) Method and device for determining abnormal source node of power system
CN113570473B (en) Equipment fault monitoring method, device, computer equipment and storage medium
CN114330569A (en) Method and equipment for detecting fan unit component fault and storage medium
CN116318386A (en) Failure prediction method of optical module, system and storage medium thereof
CN113807462A (en) AI-based network equipment fault reason positioning method and system
CN112688836A (en) Energy routing equipment online dynamic sensing method based on deep self-coding network
CN112700016A (en) Predictive maintenance method and apparatus
CN118536009B (en) Power data model construction method and system based on generation type artificial intelligence

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