CN116591911A - Intelligent detection operation and maintenance system and method facing to offshore wind turbine generator set - Google Patents
Intelligent detection operation and maintenance system and method facing to offshore wind turbine generator set Download PDFInfo
- Publication number
- CN116591911A CN116591911A CN202310589105.4A CN202310589105A CN116591911A CN 116591911 A CN116591911 A CN 116591911A CN 202310589105 A CN202310589105 A CN 202310589105A CN 116591911 A CN116591911 A CN 116591911A
- Authority
- CN
- China
- Prior art keywords
- generator set
- data
- fault
- module
- offshore wind
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 124
- 238000001514 detection method Methods 0.000 title claims abstract description 75
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 41
- 208000024891 symptom Diseases 0.000 claims abstract description 39
- 238000004891 communication Methods 0.000 claims abstract description 23
- 238000012544 monitoring process Methods 0.000 claims abstract description 22
- 238000010276 construction Methods 0.000 claims description 23
- 230000015654 memory Effects 0.000 claims description 19
- 238000013145 classification model Methods 0.000 claims description 16
- 238000004140 cleaning Methods 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 8
- 238000013075 data extraction Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 4
- 238000010248 power generation Methods 0.000 abstract description 14
- 230000002265 prevention Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 52
- 238000013527 convolutional neural network Methods 0.000 description 10
- 238000012549 training Methods 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 6
- 210000004027 cell Anatomy 0.000 description 6
- 238000003745 diagnosis Methods 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000012795 verification Methods 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 5
- 101100455978 Arabidopsis thaliana MAM1 gene Proteins 0.000 description 4
- 241001123248 Arma Species 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 238000010200 validation analysis Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000002243 precursor Substances 0.000 description 1
- 239000003643 water by type Substances 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- Bioinformatics & Computational Biology (AREA)
- Quality & Reliability (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Artificial Intelligence (AREA)
- Sustainable Development (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Wind Motors (AREA)
Abstract
The invention provides an intelligent detection operation and maintenance system and method facing to an offshore wind turbine generator set, wherein the intelligent detection operation and maintenance system facing to the offshore wind turbine generator set comprises the following components: the system comprises an edge acquisition module, a data communication module, a database module, a data processing module, a fault detection module, a state monitoring module and a maintenance strategy module. The system generates the fault type of the fan generator set and the fault symptom of the fan generator set through the fault detection module, determines a fan maintenance strategy based on the fault type of the fan generator set and the fault symptom of the fan generator set through the offshore fan generator set fault description record database and real-time state data of key parts, achieves early discovery, early prevention and early treatment of the offshore wind generator set fault, reduces the operation and maintenance cost of taking solving measures after the fault occurs, guides the operation and maintenance schedule of the fan generator set more scientifically, reduces the power generation loss, and protects the safety of offshore wind power operation and maintenance personnel.
Description
Technical Field
The invention relates to the technical field of offshore wind turbine generator set detection, in particular to an intelligent detection operation and maintenance system and method facing to an offshore wind turbine generator set.
Background
The coastline of China is long, the offshore wind resources are rich, the offshore wind power has the characteristics of small occupied area, large development scale, high power generation utilization hours and the like, the land wind power faces dilemma, the national policy is good, and the development and construction of the offshore wind power of China are in a good condition. Meanwhile, related problems of operation and maintenance of the offshore wind turbine are also widely focused. Offshore wind turbines have a higher failure rate than onshore because they face a more hostile environment, more difficult maintenance, etc. With the development of offshore wind power, offshore wind farm construction has to be transferred to more distant places and deeper waters. As a result of this change, the operating costs will increase, while facing longer transportation distances, worse climatic conditions and more serious logistical challenges.
The main drawbacks of the current technology are: in offshore wind power operation and maintenance, a method for diagnosing and monitoring faults of a fan generator set is basically discussed, and belongs to 'post-fault management', early warning before faults and early intervention on precursors of the faults are not discussed, and operation and maintenance cost is high. And in offshore wind power, it is relatively difficult to apply in industry by means of a fault diagnosis system of vibration signals only or by means of video detection only. For wind power generation systems, vibration-based fault diagnosis methods require the installation of a considerable number of sensors, and video-based fault diagnosis methods require the assembly of a large number of video detection devices and the establishment of a complete video detection system.
Disclosure of Invention
Therefore, the technical scheme of the invention mainly solves the defects that the prior art lacks management before the failure of the offshore wind turbine generator set and the cost of the failure diagnosis method based on vibration or video is higher, thereby providing an intelligent detection operation and maintenance system and method facing the offshore wind turbine generator set.
In a first aspect, an embodiment of the present invention provides an intelligent detection operation and maintenance system facing an offshore wind turbine generator set, including: the system comprises an edge acquisition module, a data communication module, a database module, a data processing module, a fault detection module, a state monitoring module and a maintenance strategy module;
the edge acquisition module is used for acquiring real-time data information of the offshore wind turbine and sending the real-time data information of the offshore wind turbine to the data communication module;
the data communication module is connected with the edge acquisition module and the database module and is used for sending real-time data information of the offshore wind turbine to the database module;
the database module is used for constructing a basic database of the wind turbine generator set based on real-time data information of the offshore wind turbine generator set; the basic database of the fan generator set comprises an original database of the offshore fan generator set, a fault description record database of the offshore fan generator set and a wind farm environment database;
The data processing module is used for processing the data in the original database of the offshore wind turbine generator set, generating the original data of the wind turbine generator set and sending the original data of the wind turbine generator set to the fault detection module;
the fault detection module is used for acquiring fan generator set data of different fault types, constructing a basic model tree based on the fan generator set data of the different fault types, inputting the fan generator set raw data into the basic model tree, and generating fan generator set fault types and fan generator set fault symptoms;
the state monitoring module is used for monitoring the real-time state of each key part of the offshore wind turbine generator and generating real-time state data of the key part;
the maintenance strategy module is connected with the fault detection module and the state monitoring module and is used for determining a fan maintenance strategy through the offshore wind turbine generator set fault description record database and the real-time state data of the key part based on the fan generator set fault type and the fan generator set fault symptom.
According to the intelligent detection operation and maintenance system for the offshore wind turbine generator set, provided by the embodiment of the invention, the fault detection module is used for generating the fault type of the offshore wind turbine generator set and the fault sign of the offshore wind turbine generator set, and the fan maintenance strategy is determined based on the fault type of the offshore wind turbine generator set and the fault sign of the offshore wind turbine generator set through the offshore wind turbine generator set fault description record database and the real-time state data of the key parts, so that early discovery, early prevention and early treatment of the offshore wind turbine generator set fault are realized, the operation and maintenance cost of taking a solving measure after the fault occurs is reduced, the operation and maintenance period of the offshore wind turbine generator set is guided more scientifically, the power generation loss is reduced, and the safety of offshore wind turbine operation and maintenance personnel is protected.
With reference to the first aspect, in one possible implementation manner, the data processing module includes: the device comprises a data extraction unit, a data cleaning and converting unit and a data loading unit;
the data extraction unit is used for extracting the data in the original database of the offshore wind turbine generator set in a full or incremental mode;
the data cleaning and converting unit is used for cleaning signal class data in the original database of the offshore wind turbine generator set and converting image class data in the original database of the offshore wind turbine generator set;
the data loading unit is used for loading the data in the original database of the offshore wind turbine generator set to the fault detection module in a full-load or incremental mode.
With reference to the first aspect, in another possible implementation manner, the fault detection module includes: the device comprises a first construction unit, a second construction unit, a third construction unit and a first generation unit;
the first construction unit is used for carrying out Fourier transform on the fan generator set data when the fan generator set data is of a signal type, and constructing a first classification model based on the fan generator set data after the Fourier transform;
The second construction unit is used for constructing a second classification model based on the fan generator set data by utilizing meta-learning when the fan generator set data is of an image type;
the third building unit is connected with the first building unit and the second building unit and is used for building the basic model tree based on the first classification model and the second classification model;
the first generating unit is connected with the third constructing unit and is used for inputting the original data of the fan generator set into the basic model tree to generate the fan generator set fault type and the fan generator set fault symptom.
With reference to the first aspect, in another possible implementation manner, the maintenance policy module includes: the device comprises an association unit, a second generation unit, a hierarchical ordering unit and a determination unit;
the association unit is connected with the fault detection module and the database module and is used for associating the fault type of the fan generator set and the fault symptom of the fan generator set with the offshore fan generator set fault description record database to generate a fault processing mode;
the second generating unit is used for generating a fault prediction occurrence time by utilizing an autoregressive moving average model based on the fan generator set fault type and the fan generator set fault symptom;
The grading and sequencing unit is connected with the association unit and the fault detection module and is used for acquiring a wind power plant operation and maintenance standard, grading and sequencing the fault type of the fan generator set and the fault processing mode based on the real-time state data of the key part and the wind power plant operation and maintenance standard, and generating a fault grade;
the determining unit is connected with the fault detection module, the second generating unit and the hierarchical ordering unit and is used for determining the fan maintenance strategy based on the fan generator set fault symptom, the fault prediction occurrence time and the fault grade.
With reference to the first aspect, in another possible implementation manner, the method further includes: a weather module;
the weather module is used for acquiring real-time weather forecast of a target area and determining weather forecast data based on the real-time weather forecast of the target area and weather data of the target area in the wind farm environment database.
With reference to the first aspect, in another possible implementation manner, the method further includes: an intelligent scheduling module;
the intelligent scheduling module is connected with the maintenance strategy module and the weather module and is used for generating overhaul scheduling data according to the fan maintenance strategy, the operation and maintenance ship information, the operation and maintenance personnel information and the weather prediction data in the wind farm environment database.
With reference to the first aspect, in another possible implementation manner, the method further includes: the system economic operation evaluation module;
the system economic operation evaluation module is used for acquiring the generated energy loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, evaluating the operation of the offshore wind farm based on the generated energy loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, and generating an economic operation evaluation result of the offshore wind farm.
In a second aspect, the embodiment of the invention further provides an intelligent detection operation and maintenance method facing to the offshore wind turbine generator set, which comprises the following steps:
the edge acquisition module acquires real-time data information of the offshore wind turbine generator;
the data communication module sends the real-time data information of the offshore wind turbine to the database module;
the database module constructs a basic database of the wind turbine generator set based on real-time data information of the offshore wind turbine generator set; the basic database of the fan generator set comprises an original database of the offshore fan generator set, a fault description record database of the offshore fan generator set and a wind farm environment database;
the data processing module processes the data in the original database of the offshore wind turbine generator set to generate the original data of the wind turbine generator set;
The method comprises the steps that a fault detection module obtains fan generator set data of different fault types, a basic model tree is built based on the fan generator set data of the different fault types, the fan generator set raw data are input into the basic model tree, and the fan generator set fault types and fan generator set fault symptoms are generated;
the state monitoring module monitors the real-time state of each key part of the offshore wind turbine generator to generate real-time state data of the key part;
and the maintenance strategy module determines a fan maintenance strategy based on the fan generator set fault type and the fan generator set fault symptom through the offshore fan generator set fault description record database and the real-time state data of the key part.
In a third aspect, an embodiment of the present invention further discloses an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the steps of an intelligent detection operation and maintenance system for an offshore wind turbine generator set according to the second aspect.
In a fourth aspect, the present invention further discloses a computer readable storage medium, on which a computer program is stored, the computer program implementing the steps of an intelligent detection operation and maintenance system facing an offshore wind turbine generator set according to the second aspect when being executed by a processor.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of an intelligent detection operation and maintenance system facing an offshore wind turbine generator set provided by an embodiment of the invention;
FIG. 2 is a flow chart of an intelligent detection operation and maintenance method facing to an offshore wind turbine generator set provided by an embodiment of the invention;
fig. 3 is a schematic diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In addition, unless explicitly stated or limited otherwise, the terms "mounted," "connected," "coupled," and "connected" are to be construed broadly, and may be, for example, fixedly connected, mechanically connected, or electrically connected; or can be directly connected, or can be indirectly connected through an intermediate medium, or can be communication between the two elements, or can be wireless connection or wired connection. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
The embodiment of the invention provides an intelligent detection operation and maintenance system facing an offshore wind turbine generator set, which is shown in fig. 1 and comprises the following components: the system comprises an edge acquisition module 1, a data communication module 2, a database module 3, a data processing module 4, a fault detection module 5, a state monitoring module 6 and a maintenance strategy module 7;
the edge acquisition module 1 is used for acquiring real-time data information of the offshore wind turbine and sending the real-time data information of the offshore wind turbine to the data communication module 2.
Specifically, the equipment state monitoring sensor and the camera acquire real-time data information such as vibration, temperature, engine unit audio and video, engine unit image data and the like of the generator set, so that the acquisition of the real-time data information of the offshore wind turbine is completed.
Further, each fan generator set is provided with an edge center acquisition module for acquiring data information of each fan generator set.
The data communication module 2 is connected with the edge acquisition module 1 and the database module 3 and is used for sending real-time data information of the offshore wind turbine to the database module 3.
Specifically, real-time data information of the offshore wind turbine generator set collected by the edge center collection module of each fan generator set is sent to the database module 3 by comprehensively utilizing wired communication modes such as submarine cables and wireless communication modes such as satellite communication modes.
The database module 3 is used for constructing a basic database of the wind turbine generator set based on real-time data information of the offshore wind turbine generator set; the basic database of the fan generator set comprises an original database of the offshore fan generator set, a fault description record database of the offshore fan generator set and a wind farm environment database.
Specifically, a basic database of the fan generator sets of three offshore wind farms is constructed from the following three dimensions to completely describe the faults of the fan generator sets of the offshore wind farms; the method comprises the following steps of firstly, an original database of an offshore wind turbine generator set: comprehensively using a relational database and a non-relational database to store the original data of the multi-source heterogeneous offshore wind turbine generator sets, such as the signals, the audio-visual image data and the like of the wind turbine generator sets, acquired by each edge center acquisition module; secondly, a fault description record database of the offshore wind turbine generator set: describing faults of the offshore wind turbine generator set in terms of time, components, fault forms, fault diagnosis technology, fault processing schemes, equipment data before and after the faults occur and the like, and constructing a fault description record database by combining information of the offshore wind turbine generator set (such as information of other offshore wind turbine generators with the same type, including past fault information); and the wind farm environment database is used for storing wind farm history and current environment data, including local weather data, wind farm environment data, operation and maintenance ship information, operation and maintenance personnel information and other related data.
The data processing module 4 is configured to process data in an offshore wind turbine generator set raw database, generate wind turbine generator set raw data, and send the wind turbine generator set raw data to the fault detection module 5.
Specifically, the data of the original database of the offshore wind turbine generator set in the database is processed by adopting E (extract), T (transform) and L (load).
The fault detection module 5 is used for acquiring the fan generator set data of different fault types, constructing a basic model tree based on the fan generator set data of different fault types, inputting the fan generator set raw data into the basic model tree, and generating the fan generator set fault types and the fan generator set fault symptoms.
Specifically, a plurality of methods are adopted to build a basic model of Meta-Learner LSTM (time-loop neural network based on a primitive Learner) for different data types and fan part fault combined primitive learning.
The state monitoring module 6 is used for monitoring the real-time state of each key part of the offshore wind turbine generator and generating real-time state data of the key part.
Specifically, the state monitoring module 6 monitors the real-time state of each key part of the offshore wind turbine generator set, and updates the self-defined time state of the non-key part.
The maintenance strategy module 7 is connected with the fault detection module 5 and the state monitoring module 6 and is used for determining a fan maintenance strategy through the offshore wind turbine generator set fault description record database and the real-time state data of the key parts based on the fan generator set fault type and the fan generator set fault symptom.
According to the intelligent detection operation and maintenance system facing the offshore wind turbine generator set, the fault detection module is used for generating the fault type of the offshore wind turbine generator set and the fault sign of the offshore wind turbine generator set, and the fan maintenance strategy is determined based on the fault type of the offshore wind turbine generator set and the fault sign of the offshore wind turbine generator set through the offshore wind turbine generator set fault description record database and the real-time state data of the key parts, so that early discovery, early prevention and early treatment of the offshore wind turbine generator set fault are realized, the operation and maintenance cost of taking solving measures after the fault occurs is reduced, the operation and maintenance period of the offshore wind turbine generator set is guided more scientifically, the power generation loss is reduced, and the safety of offshore wind power operation and maintenance personnel is protected.
As an alternative embodiment of the present invention, the data processing module 4 includes: a data extraction unit 8, a data cleansing conversion unit 9, and a data loading unit 10;
And the data extraction unit 8 is used for extracting the data in the original database of the offshore wind turbine generator set in a full or incremental mode.
Specifically, if a fault detection model is not constructed before, after the real-time data information of the offshore wind turbine is collected, all data of the database module 3 are extracted in full quantity and put into a cache area (the main purpose is to collect multi-source data); if the related data of part of newly built offshore wind farm fan generator sets are collected later, in order to adjust the framework model in a targeted manner, the related data of the newly built offshore wind farm fan generator sets are extracted in an increment mode and put into a cache area; the data in the database module 3 may also be extracted in full quantities if the model is subsequently adapted to the newly gathered data on a large scale.
The data cleaning and converting unit 9 is used for cleaning and processing the signal class data in the original database of the offshore wind turbine generator set and converting the image class data in the original database of the offshore wind turbine generator set.
Specifically, fourier transformation is performed on signal data, and then incomplete data, repeated data and error data are screened (partial data is screened, or partial fields are screened, a part of useful data is extracted), cleaning (missing value filling, default value setting, enumeration mapping and the like), merging (merging of a plurality of attributes together), splitting, standardization, replacement, verification (time rule, business rule, custom rule), association (other data or mathematics are associated, and data integrity is guaranteed), sorting, calculation and other processes are performed; processing image class data: image labeling, image importing, denoising, image enhancement, conversion of a color image into a gray scale image, conversion of the gray scale image into a binary image, edge detection/segmentation, histogram matching and contour matching.
The data loading unit 10 is used for loading the data in the original database of the offshore wind turbine generator set to the fault detection module 5 in a full-load or incremental mode.
Specifically, according to the actual situation, the full or incremental loading data is determined.
As an alternative embodiment of the present invention, the fault detection module 5 includes: a first construction unit 11, a second construction unit 12, a third construction unit 13, and a first generation unit 14;
a first construction unit 11, configured to perform fourier transform on the fan-generator set data when the fan-generator set data is of a signal type, and construct a first classification model based on the fan-generator set data after fourier transform;
specifically, the signal class data is processed by fourier transform, and then the signal class data is classified by using a Meta-Learner LSTM (Long Short-Term Memory network) as a Meta-Learner and using the LSTM as a base Learner, so that faults and fault symptoms of the signal class data are obtained.
Further, the Meta-Learner LSTM is a two-layer LSTM network, the first layer is a normal LSTM model, the second layer is an LSTM model with approximate random gradient, all loss function values and loss function gradient values are preprocessed and input into the first layer LSTM model, parameters such as learning rate and forgetting gate are calculated, and the loss function gradient values are input into the second layer LSTM for parameter updating.
Further, a task set is established based on the fan generator set data after Fourier transformation, the task set is divided into a training set and a verification set, and the task set is divided intoThe d-th task in (2) randomly extracting T batches of data in the training set, denoted (X) 1 ，Y 1 )，...，(X T ，Y T ) The method comprises the steps of carrying out a first treatment on the surface of the For data of the t-th batch (X t ，Y t ) Calculating a loss function value and a loss function gradient value of a learner (learner), wherein the loss function value L of the learner t The calculation formula of (2) is as follows:
L t ＝L[M(X t ；θ t-1 )，Y t ] (1)
wherein θ t-1 Representing the parameters of the meta learner M after t-1 updates, L representing the loss function.
Further, the gradient value of the loss function of the learner is as followsRepresenting the gradient operator.
Further, the loss function value and the loss function gradient value of the learner are input into the meta-learner, and the cell state is updated using the following formula:
wherein c t Representing updated cell state Θ d-1 Representing the parameters after the d-1 th update of the meta learner R.
Further, the updated parameter value is equal to the updated cell state, i.e., θ t ＝c t 。
Further, after processing all T batches of training data in the d-th task, the loss function value L on the validation set is calculated using the validation set (X, Y) of the d-th task test ：
L test ＝L[M(X t ；θ t )，Y] (3)
Further, the loss function gradient value is (/>Representing gradient operators); further utilize the gradient value of the loss function and the loss function value L test The meta-learner parameter is updated.
Further, when the gradient value of the loss function and the loss function value corresponding to the verification set are smaller (or meet a preset condition), the Meta-Learner LSTM model is better in adaptability to the task, otherwise, the model is poor in adaptability, and the identification requirement of the data cannot be met.
Further, the base learner LSTM may provide the loss function value and the loss function gradient value on each batch of data to the meta learner before inputting the first layer LSTM, where the loss function value and the loss function gradient value on each batch of data may be calculated by LSTM or may be calculated by CNN.
Further, in the iterative process, the meta learner continuously provides updated parameters for the base learner, and for LSTM, the feature dimension of the input data is the latitude of the hidden layer in the LSTM and the layer number of the circulating neural network; for CNN, learning rate, optimizer, iteration times, batch size, activation function, hidden neuron layer number, weight initialization, dropout method (discarding method), regularization and quasi-normalization are adopted; the element learner returns the updated parameters to the base learner after one iteration, and the base learner calculates the loss function value and the loss function gradient through the updated parameters and returns the loss function value and the loss function gradient to the element learner.
Further, in the Meta-Learner LSTM, the Meta-Learner provides the modified LSTM update parameters to the base Learner, the parameters of the Meta-Learner are not initial values of the parameters in the base Learner, the parameters of the Meta-Learner are updated by using SGD, and the calculation of higher derivatives of the loss function does not occur; also, in the Meta-Learner LSTM, parameter updates of both the base Learner and the Meta Learner are performed in the Meta Learner.
And a second construction unit 12, configured to construct a second classification model based on the fan generator set data by meta-learning when the fan generator set data is of an image type.
Specifically, the image class data is classified by using Meta-Learner LSTM, using LSTM as a primitive Learner, and using CNN (Convolutional Neural Networks, convolutional neural network) as a basic Learner, so as to obtain the fault and fault symptoms of the image class data.
The third construction unit 13 is connected to the first construction unit 11 and the second construction unit 12 for constructing a base model tree based on the first classification model and the second classification model.
The first generating unit 14 is connected with the third constructing unit 13, and is used for inputting the original data of the fan generator set into the basic model tree, and generating the fan generator set fault type and the fan generator set fault symptom.
In the alternative embodiment, the method of adopting the Meta-Learner LSTM focuses on solving the problem of small samples, mainly aiming at newly built offshore wind farms, newly commissioned offshore fans and newly replaced generator sets, and solves the problem that the historical data are less and the machine learning model cannot be directly used in 1-2 years of new commission; after the historical data is accumulated, training can be continued on the basic model number by adopting a mode of loading the data in an increment mode, and meanwhile, the Meta-Learner LSTM is also suitable for large samples, namely a common offshore wind field which is put into operation for a period of time, so that the application scene of the offshore wind field is expanded.
As an alternative embodiment of the present invention, the maintenance policy module 7 includes: an association unit 15, a second generation unit 16, a hierarchical ordering unit 17, and a determination unit 18;
the association unit 15 is connected with the fault detection module 5 and the database module 3, and is used for associating the fan generator set fault type and the fan generator set fault symptom with the offshore fan generator set fault description record database to generate a fault processing mode.
Specifically, an APRIORI algorithm (association rule mining algorithm) is adopted to mine fault rules, and then the fault types of the fan generator sets, the fault symptoms of the fan generator sets and the offshore fan generator set fault description record database are associated by the rules to generate a fault processing mode.
The second generating unit 16 is configured to generate a failure prediction occurrence time by using an autoregressive moving average model based on the failure type of the fan generator set and the failure symptom of the fan generator set.
Specifically, in the case of a small sample, a model of an ARMA model (Autoregressive moving average model ) is constructed by using meta learning, and the expected occurrence time of the failure is obtained.
Further, training for the ARMA model using MAML (meta learning); for example, k=10 of MAML (in 10-shot regression tasks, 10 input/output pairs are provided for each task to learn), resulting in ARIMA autoregressive model (since it depends on only its past history value without depending on other interpretation variables, the value at the current time point is equal to the regression of the value at the past several time points) order p (the past p history values relied on), moving average model (the value at the current time point is equal to the regression of the prediction error at the past several time points; prediction error=model prediction value-true value) order q (the past q history prediction error values relied on).
The grading and sequencing unit 17 is connected with the association unit 15 and the fault detection module 5 and is used for acquiring the wind power plant operation and maintenance standard, grading and sequencing the fault type and the fault processing mode of the fan generator set based on the real-time state data of the key part and the wind power plant operation and maintenance standard, and generating a fault grade.
Specifically, according to the operation and maintenance standards issued by the state and related to wind power plants and past production experience, the key parts in the state detection module are real-time state data, and the fault types and the fault processing modes of the fan generator set are ranked in a grading manner according to important condition grades.
The determining unit 18 is connected to the fault detection module 5, the second generating unit 16 and the hierarchical ordering unit 17 for determining a fan maintenance strategy based on the fan generator set fault sign, the fault prediction occurrence time and the fault level.
In the above alternative embodiment, an original database of faults of the offshore wind turbine generator is constructed, a method for learning how to construct a fault model tree of the offshore wind turbine generator is learned under the conditions of small samples and multiple models, a fault rule base of the offshore wind turbine generator is constructed, fault occurrence symptoms are obtained, specific wind field environmental factors are comprehensively considered for scheduling, and the characteristic learning capability of the unknown offshore wind turbine generator is strong.
As an alternative embodiment of the present invention, further comprising: a weather module 19;
the weather module 19 is configured to obtain a real-time weather forecast of the target area, and determine weather prediction data based on the real-time weather forecast of the target area and weather data of the target area in the wind farm environment database.
Specifically, weather conditions are monitored and predicted by utilizing weather data of a target area in a wind farm environment database and combining weather forecast of a weather bureau of the target area.
As an alternative embodiment of the present invention, further comprising: an intelligent scheduling module 20;
the intelligent scheduling module 20 is connected with the maintenance strategy module 7 and the weather module 19 and is used for generating overhaul scheduling data according to a fan maintenance strategy, operation and maintenance ship information, operation and maintenance personnel information and weather prediction data in a wind farm environment database.
Specifically, according to the fan maintenance strategy, the operation and maintenance ship information, the operation and maintenance personnel information and the weather prediction data obtained in the weather module in the database module 3 automatically generate the overhaul schedule by applying the thought of assigning problems in operation and research.
As an alternative embodiment of the present invention, further comprising: a system economic operation evaluation module 21;
the system economic operation evaluation module 21 is configured to obtain a power generation amount loss of the offshore wind farm and an operation and maintenance cost of the offshore wind farm, evaluate operation of the offshore wind farm based on the power generation amount loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, and generate an economic operation evaluation result of the offshore wind farm.
Specifically, the power generation amount loss=the estimated power generation amount-the actual power generation amount; operation and maintenance cost = consumable and spare parts + operation maintenance and periodic overhaul + technical retrofit.
The following describes, by way of a specific embodiment, the operation of an intelligent detection operation and maintenance system for an offshore wind turbine generator set.
Example 1:
acquiring data which can be acquired by all fan generator sets of the offshore wind farm by utilizing an edge data acquisition module;
the collected data is sent to a database module and a state monitoring module through a data communication module;
extracting the collected data from the database module in an incremental form to a data processing module, and preprocessing the data in the data processing module;
the data obtained after processing is loaded in full (incremental loading for the database) to a fault detection module; calling an existing basic model tree of the fault detection module, and modeling all fan generator sets of the newly built offshore wind farm to obtain faults and fault symptoms of the corresponding fan generator sets;
inputting the faults and the fault symptoms obtained in the previous step into a maintenance strategy module to obtain a maintenance strategy of a corresponding fan;
and inputting the maintenance strategy into an intelligent scheduling module to obtain the intelligent scheduling of the newly-built offshore wind farm relative to the fan generator set.
And generating an economic operation report of the newly-built offshore wind farm by taking the year as a unit, and judging whether the economic operation and the operation maintenance are lost.
The embodiment of the invention also discloses an intelligent detection operation and maintenance method facing the offshore wind turbine generator set, which comprises the following steps as shown in fig. 2:
s201, an edge acquisition module acquires real-time data information of the offshore wind turbine.
Specifically, the equipment state monitoring sensor and the camera acquire real-time data information such as vibration, temperature, engine unit audio and video, engine unit image data and the like of the generator set, so that the acquisition of the real-time data information of the offshore wind turbine is completed.
Further, each fan generator set is provided with an edge center acquisition module for acquiring data information of each fan generator set.
S202, the data communication module sends real-time data information of the offshore wind turbine to the database module.
Specifically, real-time data information of the offshore wind turbine generator set collected by the edge center collection module of each fan generator set is sent to the database module by comprehensively utilizing wired communication modes such as submarine cables and wireless communication modes such as satellite communication.
S203, a database module constructs a basic database of the wind turbine generator set based on real-time data information of the offshore wind turbine generator set; the basic database of the fan generator set comprises an original database of the offshore fan generator set, a fault description record database of the offshore fan generator set and a wind farm environment database.
Specifically, a basic database of the fan generator sets of three offshore wind farms is constructed from the following three dimensions to completely describe the faults of the fan generator sets of the offshore wind farms; the method comprises the following steps of firstly, an original database of an offshore wind turbine generator set: comprehensively using a relational database and a non-relational database to store the original data of the multi-source heterogeneous offshore wind turbine generator sets, such as the signals, the audio-visual image data and the like of the wind turbine generator sets, acquired by each edge center acquisition module; secondly, a fault description record database of the offshore wind turbine generator set: describing faults of the offshore wind turbine generator set in terms of time, components, fault forms, fault diagnosis technology, fault processing schemes, equipment data before and after the faults occur and the like, and constructing a fault description record database by combining information of the offshore wind turbine generator set (such as information of other offshore wind turbine generators with the same type, including past fault information); and the wind farm environment database is used for storing wind farm history and current environment data, including local weather data, wind farm environment data, operation and maintenance ship information, operation and maintenance personnel information and other related data.
S204, the data processing module processes the data in the original database of the offshore wind turbine generator set to generate the original data of the wind turbine generator set.
Specifically, the data of the original database of the offshore wind turbine generator set in the database is processed by adopting E (extract), T (transform) and L (load).
S205, the fault detection module acquires fan generator set data of different fault types, builds a basic model tree based on the fan generator set data of the different fault types, inputs the fan generator set raw data into the basic model tree, and generates fan generator set fault types and fan generator set fault symptoms.
Specifically, a plurality of methods are adopted to build a basic model of Meta-Learner LSTM (time-loop neural network based on a primitive Learner) for different data types and fan part fault combined primitive learning.
S206, the state monitoring module monitors the real-time state of each key part of the offshore wind turbine generator to generate real-time state data of the key part.
Specifically, the state monitoring module monitors the states of all key parts of the offshore wind turbine generator set in real time, and updates the self-defined time state of non-key parts.
S207, a maintenance strategy module determines a fan maintenance strategy through the offshore wind turbine generator set fault description record database and the real-time state data of the key parts based on the fan generator set fault type and the fan generator set fault symptom.
According to the intelligent detection operation and maintenance method facing the offshore wind turbine generator set, the fault detection module is used for generating the fault type of the offshore wind turbine generator set and the fault sign of the offshore wind turbine generator set, and the fan maintenance strategy is determined based on the fault type of the offshore wind turbine generator set and the fault sign of the offshore wind turbine generator set through the offshore wind turbine generator set fault description record database and the real-time state data of key parts, so that early discovery, early prevention and early management of the offshore wind turbine generator set fault are realized, the operation and maintenance cost of the solution after the fault occurs is reduced, the operation and maintenance period of the offshore wind turbine generator set is guided more scientifically, the power generation loss is reduced, and the safety of offshore wind power operation and maintenance personnel is protected.
As an optional embodiment of the present invention, S204, that is, the data processing module processes the data in the original database of the offshore wind turbine generator set to generate the original data of the wind turbine generator set, includes:
the data extraction unit extracts data in the original database of the offshore wind turbine generator set in a full or incremental mode.
Specifically, if a fault detection model is not constructed before, after the real-time data information of the offshore wind turbine is collected, all data of the database module are extracted in full quantity and put into a cache area (the main purpose is to collect multi-source data); if the related data of part of newly built offshore wind farm fan generator sets are collected later, in order to adjust the framework model in a targeted manner, the related data of the newly built offshore wind farm fan generator sets are extracted in an increment mode and put into a cache area; the data in the database module may also be extracted in full quantities if the model is subsequently adapted in large scale to the newly collected data.
The data cleaning and converting unit cleans signal class data in the original database of the offshore wind turbine generator set, and converts image class data in the original database of the offshore wind turbine generator set.
Specifically, fourier transformation is performed on signal data, and then incomplete data, repeated data and error data are screened (partial data is screened, or partial fields are screened, a part of useful data is extracted), cleaning (missing value filling, default value setting, enumeration mapping and the like), merging (merging of a plurality of attributes together), splitting, standardization, replacement, verification (time rule, business rule, custom rule), association (other data or mathematics are associated, and data integrity is guaranteed), sorting, calculation and other processes are performed; processing image class data: image labeling, image importing, denoising, image enhancement, conversion of a color image into a gray scale image, conversion of the gray scale image into a binary image, edge detection/segmentation, histogram matching and contour matching.
The data loading unit loads the data in the original database of the offshore wind turbine generator set to the fault detection module in a full-load or incremental mode.
Specifically, according to the actual situation, the full or incremental loading data is determined.
As an optional implementation manner of the present invention, S204, namely, the fault detection module obtains fan generator set data of different fault types, constructs a basic model tree based on the fan generator set data of different fault types, inputs original data of the fan generator set into the basic model tree, and generates a fan generator set fault type and a fan generator set fault sign, including:
when the fan generator set data is of a signal type, the first construction unit carries out Fourier transform on the fan generator set data and constructs a first classification model based on the fan generator set data after the Fourier transform.
Specifically, the signal class data is processed by fourier transform, and then the signal class data is classified by using a Meta-Learner LSTM (Long Short-Term Memory network) as a Meta-Learner and using the LSTM as a base Learner, so that faults and fault symptoms of the signal class data are obtained.
Further, the Meta-Learner LSTM is a two-layer LSTM network, the first layer is a normal LSTM model, the second layer is an LSTM model with approximate random gradient, all loss function values and loss function gradient values are preprocessed and input into the first layer LSTM model, parameters such as learning rate and forgetting gate are calculated, and the loss function gradient values are input into the second layer LSTM for parameter updating.
Further, a task set is established based on the fan generator set data after Fourier transformation, the task set is divided into a training set and a verification set, and for the (d) task in the task set, T batches of data are randomly extracted in the training set and marked as (X) 1 ，Y 1 )，...，(X T ，Y T ) The method comprises the steps of carrying out a first treatment on the surface of the For data of the t-th batch (X t ，Y t ) Calculating a loss function value and a loss function gradient value of a learner (learner), wherein the loss function value L of the learner t The calculation formula of (2) is as follows:
L t ＝L[M(X t ；θ t-1 )，Y t ] (1)
wherein θ t-1 Representing the parameters of the meta learner M after t-1 updates, L representing the loss function.
Further, the gradient value of the loss function of the learner is as followsRepresenting the gradient operator.
Further, the loss function value and the loss function gradient value of the learner are input into the meta-learner, and the cell state is updated using the following formula:
wherein c t Representing updated cell state Θ d-1 Representing the parameters after the d-1 th update of the meta learner R.
Further, the updated parameter value is equal to the updated cell state, i.e., θ t ＝c t 。
Further, after processing all T batches of training data in the d-th task, the loss function value L on the validation set is calculated using the validation set (X, Y) of the d-th task test ：
L test ＝L[M(X t ；θ t )，Y] (3)
Further, the loss function gradient value is (/>Representing gradient operators); further utilize the gradient value of the loss function and the loss function value L test The meta-learner parameter is updated.
Further, when the gradient value of the loss function and the loss function value corresponding to the verification set are smaller (or meet a preset condition), the Meta-Learner LSTM model is better in adaptability to the task, otherwise, the model is poor in adaptability, and the identification requirement of the data cannot be met.
Further, the base learner LSTM may provide the loss function value and the loss function gradient value on each batch of data to the meta learner before inputting the first layer LSTM, where the loss function value and the loss function gradient value on each batch of data may be calculated by LSTM or may be calculated by CNN.
Further, in the iterative process, the meta learner continuously provides updated parameters for the base learner, and for LSTM, the feature dimension of the input data is the latitude of the hidden layer in the LSTM and the layer number of the circulating neural network; for CNN, learning rate, optimizer, iteration times, batch size, activation function, hidden neuron layer number, weight initialization, dropout method (discarding method), regularization and quasi-normalization are adopted; the element learner returns the updated parameters to the base learner after one iteration, and the base learner calculates the loss function value and the loss function gradient through the updated parameters and returns the loss function value and the loss function gradient to the element learner.
Further, in the Meta-Learner LSTM, the Meta-Learner provides the modified LSTM update parameters to the base Learner, the parameters of the Meta-Learner are not initial values of the parameters in the base Learner, the parameters of the Meta-Learner are updated by using SGD, and the calculation of higher derivatives of the loss function does not occur; also, in the Meta-Learner LSTM, parameter updates of both the base Learner and the Meta Learner are performed in the Meta Learner.
When the fan generator set data is of an image type, the second construction unit constructs a second classification model based on the fan generator set data by utilizing meta-learning.
Specifically, the image class data is classified by using Meta-Learner LSTM, using LSTM as a primitive Learner, and using CNN (Convolutional Neural Networks, convolutional neural network) as a basic Learner, so as to obtain the fault and fault symptoms of the image class data.
The third construction unit constructs a base model tree based on the first classification model and the second classification model.
The first generation unit inputs the original data of the fan generator set into the basic model tree, and generates the fan generator set fault type and the fan generator set fault symptom.
As an optional embodiment of the present invention, the step S206, that is, the maintenance policy module determines, based on the fan generator set fault type and the fan generator set fault symptom, a fan maintenance policy through the offshore fan generator set fault description record database and the real-time state data of the key part, including:
The association unit associates the fan generator set fault type and the fan generator set fault symptom with the offshore fan generator set fault description record database to generate a fault processing mode.
Specifically, an APRIORI algorithm (association rule mining algorithm) is adopted to mine fault rules, and then the fault types of the fan generator sets, the fault symptoms of the fan generator sets and the offshore fan generator set fault description record database are associated by the rules to generate a fault processing mode.
The second generating unit generates a failure prediction occurrence time by using an autoregressive moving average model based on the failure type of the fan generator set and the failure symptom of the fan generator set.
Specifically, in the case of a small sample, a model of an ARMA model (Autoregressive moving average model ) is constructed by using meta learning, and the expected occurrence time of the failure is obtained.
Further, training for the ARMA model using MAML (meta learning); for example, k=10 of MAML (in 10-shot regression tasks, 10 input/output pairs are provided for each task to learn), resulting in ARIMA autoregressive model (since it depends on only its past history value without depending on other interpretation variables, the value at the current time point is equal to the regression of the value at the past several time points) order p (the past p history values relied on), moving average model (the value at the current time point is equal to the regression of the prediction error at the past several time points; prediction error=model prediction value-true value) order q (the past q history prediction error values relied on).
The grading and sequencing unit acquires the wind power plant operation and maintenance standard, and performs grading and sequencing on the fault type and the fault processing mode of the fan generator set based on the real-time state data of the key part and the wind power plant operation and maintenance standard to generate a fault grade.
Specifically, according to the operation and maintenance standards issued by the state and related to wind power plants and past production experience, the key parts in the state detection module are real-time state data, and the fault types and the fault processing modes of the fan generator set are ranked in a grading manner according to important condition grades.
The determining unit determines a fan maintenance strategy based on fan generator set fault symptoms, a fault prediction occurrence time and a fault level.
As an alternative embodiment of the present invention, further comprising:
the weather module acquires real-time weather forecast of the target area, and determines weather forecast data based on the real-time weather forecast of the target area and weather data of the target area in the wind farm environment database.
Specifically, weather conditions are monitored and predicted by utilizing weather data of a target area in a wind farm environment database and combining weather forecast of a weather bureau of the target area.
As an alternative embodiment of the present invention, further comprising:
and the intelligent scheduling module generates overhaul scheduling data according to the fan maintenance strategy, the operation and maintenance ship information, the operation and maintenance personnel information and the weather prediction data in the wind farm environment database.
Specifically, according to a fan maintenance strategy, the operation and maintenance ship information, the operation and maintenance personnel information and the weather prediction data obtained from the weather module in the database module automatically generate the overhaul schedule by applying the thought of assigning problems in operation and research.
As an alternative embodiment of the present invention, further comprising:
the system economic operation evaluation module acquires the generating capacity loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, evaluates the operation of the offshore wind farm based on the generating capacity loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, and generates an economic operation evaluation result of the offshore wind farm.
Specifically, the power generation amount loss=the estimated power generation amount-the actual power generation amount; operation and maintenance cost = consumable and spare parts + operation maintenance and periodic overhaul + technical retrofit.
In addition, an electronic device is provided in an embodiment of the present invention, as shown in fig. 3, where the electronic device may include a processor 110 and a memory 120, where the processor 110 and the memory 120 may be connected by a bus or other manner, and in fig. 3, the connection is exemplified by a bus. In addition, the electronic device further includes at least one interface 130, where the at least one interface 130 may be a communication interface or other interfaces, and the embodiment is not limited thereto.
The processor 110 may be a central processing unit (Central Processing Unit, CPU). The processor 110 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSPs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), field programmable gate arrays (Field-Programmable Gate Array, FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above.
The memory 120 is used as a non-transitory computer readable storage medium for storing non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the video compositing method according to the embodiments of the present invention. The processor 110 executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory 120, i.e., implementing an intelligent detection operation and maintenance system facing the offshore wind turbine generator set in the above-described method embodiment.
Memory 120 may include a storage program area that may store an operating system, at least one application program required for functionality, and a storage data area; the storage data area may store data created by the processor 110, etc. In addition, memory 120 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 120 may optionally include memory located remotely from processor 110, which may be connected to processor 110 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.
In addition, at least one interface 130 is used for communication of the electronic device with external devices, such as with a server or the like. Optionally, at least one interface 130 may also be used to connect peripheral input, output devices, such as a keyboard, display screen, etc.
The one or more modules are stored in the memory 120 and when executed by the processor 110 perform an intelligent detection operation and maintenance method for an offshore wind turbine generator set in the embodiment shown in fig. 2.
The specific details of the electronic device may be understood correspondingly with respect to the corresponding related descriptions and effects in the embodiment shown in fig. 1, which are not repeated herein.
It will be appreciated by those skilled in the art that implementing all or part of the above-described embodiment method may be implemented by a computer program to instruct related hardware, where the program may be stored in a computer readable storage medium, and the program may include the above-described embodiment method when executed. The storage medium may be a magnetic Disk, an optical disc, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a Flash Memory (Flash Memory), a Hard Disk (HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.
Claims (10)
1. An intelligent detection operation and maintenance system facing to an offshore wind turbine generator set, which is characterized by comprising: the system comprises an edge acquisition module, a data communication module, a database module, a data processing module, a fault detection module, a state monitoring module and a maintenance strategy module;
the edge acquisition module is used for acquiring real-time data information of the offshore wind turbine and sending the real-time data information of the offshore wind turbine to the data communication module;
the data communication module is connected with the edge acquisition module and the database module and is used for sending real-time data information of the offshore wind turbine to the database module;
the database module is used for constructing a basic database of the wind turbine generator set based on real-time data information of the offshore wind turbine generator set; the basic database of the fan generator set comprises an original database of the offshore fan generator set, a fault description record database of the offshore fan generator set and a wind farm environment database;
The data processing module is used for processing the data in the original database of the offshore wind turbine generator set, generating the original data of the wind turbine generator set and sending the original data of the wind turbine generator set to the fault detection module;
the fault detection module is used for acquiring fan generator set data of different fault types, constructing a basic model tree based on the fan generator set data of the different fault types, inputting the fan generator set raw data into the basic model tree, and generating fan generator set fault types and fan generator set fault symptoms;
the state monitoring module is used for monitoring the real-time state of each key part of the offshore wind turbine generator and generating real-time state data of the key part;
the maintenance strategy module is connected with the fault detection module and the state monitoring module and is used for determining a fan maintenance strategy through the offshore wind turbine generator set fault description record database and the real-time state data of the key part based on the fan generator set fault type and the fan generator set fault symptom.
2. An intelligent detection operation and maintenance system facing an offshore wind turbine generator set according to claim 1, wherein the data processing module comprises: the device comprises a data extraction unit, a data cleaning and converting unit and a data loading unit;
The data extraction unit is used for extracting the data in the original database of the offshore wind turbine generator set in a full or incremental mode;
the data cleaning and converting unit is used for cleaning signal class data in the original database of the offshore wind turbine generator set and converting image class data in the original database of the offshore wind turbine generator set;
the data loading unit is used for loading the data in the original database of the offshore wind turbine generator set to the fault detection module in a full-load or incremental mode.
3. The intelligent detection operation and maintenance system facing an offshore wind turbine generator set of claim 1, wherein the fault detection module comprises: the device comprises a first construction unit, a second construction unit, a third construction unit and a first generation unit;
the first construction unit is used for carrying out Fourier transform on the fan generator set data when the fan generator set data is of a signal type, and constructing a first classification model based on the fan generator set data after the Fourier transform;
the second construction unit is used for constructing a second classification model based on the fan generator set data by utilizing meta-learning when the fan generator set data is of an image type;
The third building unit is connected with the first building unit and the second building unit and is used for building the basic model tree based on the first classification model and the second classification model;
the first generating unit is connected with the third constructing unit and is used for inputting the original data of the fan generator set into the basic model tree to generate the fan generator set fault type and the fan generator set fault symptom.
4. The intelligent detection operation and maintenance system facing an offshore wind turbine generator set according to claim 1, wherein the maintenance strategy module comprises: the device comprises an association unit, a second generation unit, a hierarchical ordering unit and a determination unit;
the association unit is connected with the fault detection module and the database module and is used for associating the fault type of the fan generator set and the fault symptom of the fan generator set with the offshore fan generator set fault description record database to generate a fault processing mode;
the second generating unit is used for generating a fault prediction occurrence time by utilizing an autoregressive moving average model based on the fan generator set fault type and the fan generator set fault symptom;
The grading and sequencing unit is connected with the association unit and the fault detection module and is used for acquiring a wind power plant operation and maintenance standard, grading and sequencing the fault type of the fan generator set and the fault processing mode based on the real-time state data of the key part and the wind power plant operation and maintenance standard, and generating a fault grade;
the determining unit is connected with the fault detection module, the second generating unit and the hierarchical ordering unit and is used for determining the fan maintenance strategy based on the fan generator set fault symptom, the fault prediction occurrence time and the fault grade.
5. The intelligent detection operation and maintenance system facing an offshore wind turbine generator set of claim 1, further comprising: a weather module;
the weather module is used for acquiring real-time weather forecast of a target area and determining weather forecast data based on the real-time weather forecast of the target area and weather data of the target area in the wind farm environment database.
6. The intelligent detection operation and maintenance system facing an offshore wind turbine generator set of claim 5, further comprising: an intelligent scheduling module;
The intelligent scheduling module is connected with the maintenance strategy module and the weather module and is used for generating overhaul scheduling data according to the fan maintenance strategy, the operation and maintenance ship information, the operation and maintenance personnel information and the weather prediction data in the wind farm environment database.
7. The intelligent detection operation and maintenance system facing an offshore wind turbine generator set of claim 1, further comprising: the system economic operation evaluation module;
the system economic operation evaluation module is used for acquiring the generated energy loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, evaluating the operation of the offshore wind farm based on the generated energy loss of the offshore wind farm and the operation and maintenance cost of the offshore wind farm, and generating an economic operation evaluation result of the offshore wind farm.
8. An intelligent detection operation and maintenance method facing to an offshore wind turbine generator set is characterized by comprising the following steps of:
the edge acquisition module acquires real-time data information of the offshore wind turbine generator;
the data communication module sends the real-time data information of the offshore wind turbine to the database module;
the database module constructs a basic database of the wind turbine generator set based on real-time data information of the offshore wind turbine generator set; the basic database of the fan generator set comprises an original database of the offshore fan generator set, a fault description record database of the offshore fan generator set and a wind farm environment database;
The data processing module processes the data in the original database of the offshore wind turbine generator set to generate the original data of the wind turbine generator set;
the method comprises the steps that a fault detection module obtains fan generator set data of different fault types, a basic model tree is built based on the fan generator set data of the different fault types, the fan generator set raw data are input into the basic model tree, and the fan generator set fault types and fan generator set fault symptoms are generated;
the state monitoring module monitors the real-time state of each key part of the offshore wind turbine generator to generate real-time state data of the key part;
and the maintenance strategy module determines a fan maintenance strategy based on the fan generator set fault type and the fan generator set fault symptom through the offshore fan generator set fault description record database and the real-time state data of the key part.
9. An electronic device comprising a processor and a memory, the memory coupled to the processor;
the memory has stored thereon computer readable program instructions which, when executed by the processor, implement the method of claim 8.
10. A computer readable storage medium, on which a computer program is stored which, when executed by a processor, implements the method of any of claims 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310589105.4A CN116591911A (en) | 2023-05-19 | 2023-05-19 | Intelligent detection operation and maintenance system and method facing to offshore wind turbine generator set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310589105.4A CN116591911A (en) | 2023-05-19 | 2023-05-19 | Intelligent detection operation and maintenance system and method facing to offshore wind turbine generator set |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116591911A true CN116591911A (en) | 2023-08-15 |
Family
ID=87604248
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310589105.4A Pending CN116591911A (en) | 2023-05-19 | 2023-05-19 | Intelligent detection operation and maintenance system and method facing to offshore wind turbine generator set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116591911A (en) |
-
2023
- 2023-05-19 CN CN202310589105.4A patent/CN116591911A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Adedipe et al. | Bayesian network modelling for the wind energy industry: An overview | |
| Liu et al. | A small-sample wind turbine fault detection method with synthetic fault data using generative adversarial nets | |
| Wang et al. | Copula‐based model for wind turbine power curve outlier rejection | |
| Dinwoodie et al. | Analysis of offshore wind turbine operation and maintenance using a novel time domain meteo-ocean modeling approach | |
| US11604461B2 (en) | Method and apparatus for monitoring the condition of subsystems within a renewable generation plant or microgrid | |
| US20220292666A1 (en) | Systems and methods for detecting wind turbine operation anomaly using deep learning | |
| CN115796059B (en) | Electrical equipment service life prediction method and system based on deep learning | |
| CN111878322B (en) | Wind power generator device | |
| CN115036922B (en) | Distributed photovoltaic power generation electric quantity prediction method and system | |
| CN116307641B (en) | Digital power plant-oriented resource collaborative scheduling management method and system | |
| CN116383773A (en) | Data center energy efficiency anomaly detection method, system and medium based on self-adaptive prediction interval | |
| Vera-Tudela et al. | On the selection of input variables for a wind turbine load monitoring system | |
| CN112802011A (en) | Fan blade defect detection method based on VGG-BLS | |
| CN110188939B (en) | Wind power prediction method, system, equipment and storage medium of wind power plant | |
| CN116821610A (en) | Method for optimizing wind power generation efficiency by utilizing big data | |
| CN116591911A (en) | Intelligent detection operation and maintenance system and method facing to offshore wind turbine generator set | |
| CN113723670B (en) | Photovoltaic power generation power short-term prediction method with variable time window | |
| CN111523557A (en) | Wind power intelligent fault diagnosis method based on big data | |
| CN115659778A (en) | On-line construction method and on-line updating equipment of wind turbine generator bearing fault diagnosis model | |
| Pahmi et al. | Artificial neural network based forecasting of power under real time monitoring environment | |
| CN112699598A (en) | Intelligent diagnosis method and device for abnormal oil temperature of gear box | |
| Zhang | Comparison of data-driven and model-based methodologies of wind turbine fault detection with SCADA data | |
| Nielsen et al. | Quantifying the value of SHM for wind turbine blades | |
| CN116821730B (en) | Fan fault detection method, control device and storage medium | |
| CN117688464B (en) | Hidden danger analysis method and system based on multi-source sensor data |
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 |