CN110852484A - Fault early warning system and method for wind generating set - Google Patents
Fault early warning system and method for wind generating set Download PDFInfo
- Publication number
- CN110852484A CN110852484A CN201910979917.3A CN201910979917A CN110852484A CN 110852484 A CN110852484 A CN 110852484A CN 201910979917 A CN201910979917 A CN 201910979917A CN 110852484 A CN110852484 A CN 110852484A
- Authority
- CN
- China
- Prior art keywords
- early warning
- module
- fault
- data
- model
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- 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
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- 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
- G06F16/26—Visual data mining; Browsing structured data
-
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses a wind generating set fault early warning system and a method, which overcome the problem that the value of modules cannot be deployed and evaluated according to different scenes in the prior art, and comprise a data warehouse module for storing and arranging information, an early warning model module for calculating a fault characteristic value, an early warning rule module for triggering early warning codes according to a preset rule by using the fault characteristic value, a fault diagnosis module for generating complete early warning information according to an early warning code recognition base, a front-end display module for processing and feeding back faults, and early warning function evaluation for evaluating the quality of an early warning function. The invention is suitable for development and construction of a group big data system and a wind power plant edge system under the conditions of different wind generating set models, operating environments and data acquisition, and evaluates the value of each module in the whole function.
Description
Technical Field
The invention relates to the field of wind power generation, in particular to a fault early warning system of a wind generating set.
Background
In recent years, the wind power accumulated installed capacity is continuously increased at a rate of 20%, the large-scale in-service wind generating set enters a failure high-occurrence period due to the reasons of insufficient design and manufacturing level, bad operating conditions, increased operating years and the like at the initial development stage, the safety accidents of the wind power plant are continuous, and the operation and maintenance cost is high. With the continuous development of new generation information technologies such as cloud computing, big data, internet +, and the like, a new management mode is brought to the operation and maintenance of the wind power plant, namely an active operation and maintenance mode, the health state of the wind power generator set is actively monitored, a maintenance plan is made in advance, major safety accidents are prevented, the reliability of the wind power generator set can be effectively improved, the operation and maintenance cost of the wind power plant is reduced, fault model modules cannot be deployed in different operation scenes aiming at different types of wind power generator sets, and no module effect evaluation module is used for evaluating the value of each module in the whole function.
For example, a "wind turbine generator failure early warning system" disclosed in chinese patent literature, whose publication number CN107038453A includes a real-time monitoring module, an early warning model module, an early warning result feedback module, and a knowledge base module, where the real-time monitoring module reads parameters of a wind turbine in real time and transmits the parameters to the early warning module for analysis; when the early warning standard is met, an alarm mechanism is triggered, information is transmitted to an early warning result feedback module, and a worker records the feedback result into a knowledge base module; and the knowledge base module continuously accumulates the results in the early warning result feedback module, and when the accuracy of the early warning result is lower than 60%, the results in the knowledge base module are fed back to the early warning model module. However, for different types of wind turbines and different operation scenes, the method needs to redeploy a set of fault model modules, and has no module effect evaluation module, so that the value of each module in the overall function is evaluated.
Disclosure of Invention
The invention provides a wind generating set fault early warning system and a method for overcoming the problem that the value of modules cannot be deployed and evaluated according to different scenes in the prior art, and the system and the method are suitable for development and construction of a group big data system and a wind farm edge system under the conditions of different wind generating set models, operating environments and collected data, and evaluate the value of each module in the whole function.
In order to achieve the purpose, the invention adopts the following technical scheme:
a wind generating set fault early warning system comprises:
the data warehouse module is used for storing basic information, configuration information, operation data and the like of the wind power plant and the wind turbine generator, and cleaning and sorting the data;
the early warning model module is used for calculating a fault characteristic value according to information of the wind power plant and the wind turbine generator by using data in the data warehouse;
the early warning rule module is used for triggering early warning codes according to preset rules by utilizing the fault characteristic values;
the fault diagnosis module is associated with the early warning information knowledge base to generate complete early warning information, wherein the complete early warning information comprises fault positions, fault reasons and processing suggestions;
the front-end display module initiates an early warning list to an external production management system, processes and feeds back fault and processing information, operates the early warning list through an application program, and processes and feeds back the fault and the processing information to be stored back to the database module if the early warning list initiates the external production management system;
evaluating the early warning function, namely evaluating the quality of the early warning function through a feedback early warning list;
the data warehouse module, the early warning model module, the early warning rule module and the fault diagnosis module are sequentially connected with the front-end display module and are all connected with early warning function evaluation. In the scheme, the early warning model module extracts a fault characteristic value through algorithms such as statistical analysis, machine learning and deep mining, wherein the fault characteristic value is an index reflecting fault occurrence or fault severity. And the early warning rule module is used for establishing the relation between the fault characteristic value and the fault through a logic relation, and the fault is represented by an early warning code. And the fault diagnosis module is used for establishing the relation between the early warning codes and information such as fault positions, fault severity, fault reasons, processing suggestions and the like through correlation comparison based on a fault knowledge base to generate a quasi early warning list, wherein the quasi early warning list is early warning information which is not formally initiated and comprises unit basic information, early warning codes, fault positions, fault severity, fault reasons and processing suggestions. And the front-end display module judges whether to initiate an early warning list to a wind power plant site or not by checking the content of the quasi early warning list. And the early warning function evaluation module is used for evaluating each module by extracting performance indexes of each link, such as early warning order accuracy rate, quasi-early warning order initiation rate and the like.
Preferably, the wind generating set fault early warning system further comprises a data conversion module, the data warehouse module comprises structured data and unstructured data, the structured data are stored in a database form, the unstructured data are stored in a file form, and the unstructured data are converted into the structured data through the data conversion module.
Preferably, the early warning model module comprises a model data input module, an early warning model program module, a model data output module, a fault early warning system application module and an early warning model management module, the model data input module, the early warning model program module, the model data output module and the fault early warning system application module are sequentially connected, and the early warning model program module and the fault early warning system application module are both connected with the early warning model management module. The data needed by calculation are obtained from the data warehouse through the model data input module, the data service can adopt the data service provided by a big data platform, algorithms such as statistical analysis, machine learning, deep mining and the like are modularized into an early warning model program module, the early warning model program module is called by the model management module, the fault characteristic value is calculated periodically, and the fault characteristic value is transmitted to the fault early warning application module through the model data output module.
A method of a wind generating set fault early warning system comprises the following steps:
s1: calculating a fault characteristic value according to information of the wind power plant and the wind turbine generator;
s2: triggering an early warning code according to a preset rule by using the fault characteristic value;
s3: associating the early warning information knowledge base to generate complete early warning information;
s4: initiating an early warning list to an external production management system, and processing and feeding back fault and processing information;
s5: and the whole fault early warning process is supervised through the fed back early warning list, and the quality of each module is evaluated.
Preferably, the specific process of S1 includes the following steps:
s101: the data warehouse module is used for storing basic information, configuration information and operation data of the wind power plant and the wind turbine generator, cleaning and sorting the data, wherein the data comprises structured data and unstructured data, the structured data is directly used for model fast calculation, and the unstructured data is converted into the structured data through the data conversion module;
s102: the model data input module acquires structural data required by calculation from the data warehouse module;
s103: modeling statistical analysis, machine learning and deep mining algorithms into an early warning model program module;
s104: and the model management module calls an early warning model program module, periodically calculates a fault characteristic value, and transmits the fault characteristic value to the fault early warning application module through the model data output module.
Preferably, the formula for calculating the fault characteristic value in S104 is as follows:
in the formula xrmsIs a valid value, x, of a fault characteristic valueiThe residual value is the difference value between the measured value and the predicted value, and the predicted value is obtained by the ant colony algorithm,And obtaining by a machine learning algorithm or a deep learning algorithm, and outputting the effective value of the fault characteristic value to a fault early warning application module through a model data output module.
Preferably, the specific process of S2 includes the following steps:
s201: a threshold value A and a threshold value B are preset in the early warning rule module,
s202: linking the fault characteristic value with a corresponding fault through a logical relation, wherein the fault is represented by an early warning code;
the logic relation is as follows: if the condition 1 or the condition 2 is met, an early warning code is reported, if the temperature residual value of the rear bearing of the high-speed shaft of the gearbox is greater than a threshold value A, or if the bandwidth filtering vibration value of the high-speed shaft of the gearbox is greater than a threshold value B, the early warning code is reported.
Preferably, the specific process of S3 includes the following steps:
s301: the fault diagnosis module is internally provided with an early warning information knowledge base, and the early warning information knowledge base comprises the classification of each part of the wind generating set, the fault mode of each sub-part, the fault reason and the processing suggestion;
s302: associating each early warning code with corresponding fault position, fault severity, fault reason and processing opinion information through association comparison based on an early warning information knowledge base;
s303: and the fault diagnosis module generates a quasi-early warning list.
Preferably, the S4 specifically includes the following contents:
s401: the front-end display module displays the early warning list in a list form or a pop-up window form, and can display the detailed content of the early warning list;
s402: and judging whether an early warning list is initiated to a wind power plant site, and generating a feedback early warning list by the monitoring parameters of the wind generating set related to the early warning list and the fault characteristic value calculated by the early warning model and feeding the feedback early warning list back to the data warehouse module.
Preferably, the S5 specifically includes the following contents:
s501: setting four grades of excellent, good, medium and poor in the early warning function evaluation module, wherein 80% -100% of the interval corresponds to grade excellent, 60% -80% of the interval corresponds to grade good, and 0% -40% of the interval corresponds to grade poor in the 40% -60% of the interval corresponds to grade, extracting performance indexes of all links, and the completion rate, the accuracy rate and the initiation rate of an early warning order are used for expressing the performance of problem handling;
s502: the accuracy rate of the early warning list is used for evaluating the performance of the front-end display module, and the initiation rate of the quasi early warning list is used for evaluating the performance of the rules of the early warning module;
s503: and judging the levels of the early warning order completion rate, the early warning order accuracy rate and the early warning order initiation rate by reading the intervals of the early warning order completion rate, the early warning order accuracy rate and the early warning order initiation rate.
Therefore, the invention has the following beneficial effects:
according to the fault early warning characteristics of the wind generating set, the fault early warning function is divided into a plurality of links, the boundary is clear, no coupling exists between the links, and the system deployment and optimization are facilitated;
the method is suitable for development and construction of a group big data system and a wind power plant edge system under the conditions of different wind generating set models, operating environments and data acquisition, and effectively realizes an active maintenance strategy of the wind power plant;
the system has an evaluation function, and the value of each module in the whole function can be evaluated by extracting performance indexes of each link, such as the accuracy rate of the early warning list and the initiation rate of the quasi early warning list.
Drawings
FIG. 1 is a schematic flow diagram of the present invention.
Fig. 2 is a block diagram of the structure of the embodiment of the present invention.
Fig. 3 is a block diagram of an early warning model module according to an embodiment of the present invention.
In the figure: 1. the early warning system comprises a data warehouse module 2, an early warning model module 3, an early warning rule module 4, a fault diagnosis module 5, a front end display module 6, an early warning function evaluation module 7, an early warning model program module 8, an early warning model management module 9, a model data input module 10, a model data output module 11, a fault early warning system application module 12 and a data conversion module.
Detailed Description
The invention is further described with reference to the following detailed description and accompanying drawings.
Example (b):
the wind generating set fault early warning system of the present embodiment, as shown in fig. 1 to 3, includes:
and the data warehouse module 1 is connected with the early warning function evaluation module 6 and the early warning model module 2, stores the basic information, the configuration information, the operation data and the like of the wind power plant and the wind turbine generator, and cleans and arranges the data. The data stored in the data warehouse module 1 comprises structured data and unstructured data, and a Hadoop big data platform, an Ariiyun big data platform and other industrial big data platforms are adopted; the structured data is stored in a database form and comprises an Hbase distributed type column storage database in Hadoop, a Maxcomputer offline database in Arizon and the like, and the structured data is directly used for model fast calculation; the unstructured data are stored in a file form and comprise an HDFS distributed file system in Hadoop and an OSS file system in Alice cloud, the unstructured data are converted into structured data through a data conversion module 12, and the data conversion module 12 serves for converting the unstructured data into application service of the structured data.
And the early warning model module 2 is connected with the early warning function evaluation module 6 and calculates a fault characteristic value by using the data in the data warehouse module 1. The early warning model module 2 comprises an early warning model program module 7, an early warning model management module 8, a model data input module 9, a model data output module 10 and a fault early warning system application module 11, wherein the model data input module 9, the early warning model program module 7, the model data output module 10 and the fault early warning system application module 11 are sequentially connected, and the early warning model program module 7 and the fault early warning system application module 11 are both connected with the early warning model management module 8; data required by calculation are obtained from a data warehouse through a model data input module 9, and data services can be provided by a big data platform and comprise MapReduce of Hadoop and PyODPS provided by Ariyun; the algorithm of statistical analysis, machine learning, deep excavation and the like is modularized into an early warning model program module 7, the early warning model program module 7 is called by a model management module 8, the fault characteristic value is calculated periodically, and the fault characteristic value is transmitted to a fault early warning application module 11 through a model data output module 10, such as communication means of kafka or webservice and the like.
And the early warning rule module 3 is connected with the early warning function evaluation 6 and triggers an early warning code according to a preset rule by utilizing the fault characteristic value. Establishing a relation between a fault characteristic value and a fault through a logical relation, wherein the fault is represented by an early warning code, and the logical relation is as follows: if the condition 1 is met and/or the condition 2 is met, the early warning code XX is reported, if: taking data of 2 hours before the current time, wherein the ratio of the residual temperature value of the rear bearing of the high-speed shaft of the gearbox exceeding 2 degrees is more than 50 percent, and the bandwidth filtering vibration value of the high-speed shaft of the gearbox exceeds 2m/s2If the ratio is greater than 10%, an early warning code 31025 is reported, which indicates the abnormal wear failure of the high-speed shaft bearing of the gearbox.
And the fault diagnosis module 4 is connected with the early warning function evaluation 6, associates an early warning information knowledge base according to the early warning codes, and generates complete early warning information comprising fault positions, fault reasons and processing suggestions. Based on a fault knowledge base, establishing a relation between an early warning code and fault position, fault severity, fault reason and processing suggestion information through correlation and comparison, and generating a quasi early warning list, wherein the fault knowledge base comprises the classification of each part of a wind generating set, a fault mode of each part, the fault reason and processing suggestion, and comprises a primary system transmission system of the wind generating set, a secondary system gear box, a tertiary system gear box high-speed shaft and bearing, a quaternary system gear box high-speed shaft rear bearing, the fault mode is bearing pitting, the fault reason is poor lubrication of the gear box, the processing suggestion is to check the lubrication oil level and grease condition of the gear box, and the oil of the gear box needs to be replaced when the grease is abnormal.
And the front-end display module 5 is connected with the early warning function evaluation 6, operates the early warning list through an application program, and processes and feeds back the fault and the processing information to the data warehouse 1 if the early warning list starts an external production management system. Whether an early warning order is initiated to a wind power plant site is judged by checking the content of the accurate early warning order, and in order to improve the judgment accuracy, a corresponding data analysis module needs to be developed for auxiliary analysis so that an engineer can check the monitoring parameters of the wind generating set related to the early warning and the fault characteristic value calculated by the early warning model.
The early warning function evaluation module 6 is used for evaluating each sub-module by extracting the performance indexes of each link, for example, the completion rate of an early warning list is used for evaluating the performance of the processing problem of a field engineer, the accuracy rate of the early warning list is used for evaluating the performance of the front-end display module, and the initiation rate of a quasi early warning list is used for evaluating the performance of an early warning rule.
The embodiment of the invention correspondingly provides a wind generating set fault early warning method, which is applied to the wind generating set fault early warning system and comprises the following steps:
s1: calculating a fault characteristic value according to information of the wind power plant and the wind turbine generator;
s2: triggering an early warning code according to a preset rule by using the fault characteristic value;
s3: associating the early warning information knowledge base to generate complete early warning information;
s4: initiating an early warning list to an external production management system, and processing and feeding back fault and processing information;
s5: and the whole fault early warning process is supervised through the fed back early warning list, and the quality of each module is evaluated.
In the scheme, the data warehouse module 1 stores basic information, configuration information, operation data and the like of a wind power plant and a wind turbine generator, the primary data is utilized, the early warning model module 2 calculates a fault characteristic value, the early warning rule module 3 triggers an early warning code according to a preset rule, the fault diagnosis module 4 associates an early warning information knowledge base according to the early warning code to generate complete early warning information which comprises a fault position, a fault reason and a processing suggestion, and finally, the complete early warning information passes through the front-end display module 5, initiates an early warning list to an external production management system to process and feed back fault and processing information, and the early warning function evaluation module 6 supervises the whole fault early warning process for checking the advantages and disadvantages of each module.
Preferably, the specific process of S1 includes the following steps:
s101: the data warehouse module 1 stores basic information, configuration information and operation data of a wind power plant and a wind turbine generator, and cleans and sorts the data, wherein the data comprises structured data and unstructured data, the structured data is directly used for model fast calculation, and the unstructured data is converted into the structured data through the data conversion module 12;
s102: the model data input module 9 acquires structural data required for calculation from the data warehouse module 1;
s103: the algorithm of statistical analysis, machine learning and deep excavation is modularized into an early warning model program module 7;
s104: the model management module 8 calls the early warning model program module 7, periodically calculates the fault characteristic value, and transmits the fault characteristic value to the fault early warning application module 11 through the model data output module 10.
Preferably, the formula for calculating the fault characteristic value in S104 is as follows:
in the formula xrmsIs a valid value, x, of a fault characteristic valueiThe residual value is a residual value, the residual value is a difference value between an actual measurement value and a predicted value, the predicted value is obtained through an ant colony algorithm, a machine learning algorithm or a deep learning algorithm, and an effective value of a fault characteristic value is output to the fault early warning application module 11 through the model data output module 10.
In the scheme, data required by calculation are acquired from a data warehouse through a model data input service 9, and the data service can adopt data services provided by a big data platform, including MapReduce of Hadoop and PyODPS provided by Ariyun; the algorithm of statistical analysis, machine learning, deep mining and the like is modularized into an early warning model program module 7, the early warning model program module 7 is called by a model management module 8, fault characteristic values are calculated periodically and transmitted to a fault early warning application module 11 through a model data output service 10, such as communication means of kafka or webservice and the like.
Preferably, the specific process of S2 includes the following steps:
s201: a threshold value A and a threshold value B are preset in the early warning rule module 3,
s202: linking the fault characteristic value with a corresponding fault through a logical relation, wherein the fault is represented by an early warning code;
the threshold A is 50%, and the threshold B is 10%;
the logic relation is as follows: if the condition 1 is met or the condition 2 is met, an early warning code is reported, if: taking data of 2 hours before the current time, wherein the ratio of the residual temperature value of the rear bearing of the high-speed shaft of the gearbox exceeding 2 degrees is more than 50 percent, and the bandwidth filtering vibration value of the high-speed shaft of the gearbox exceeds 2m/s2If the ratio is greater than 10%, an early warning code 31025 is reported, which indicates the abnormal wear failure of the high-speed shaft bearing of the gearbox.
Preferably, the specific process of S3 includes the following steps:
s301: an early warning information knowledge base is arranged in the fault diagnosis module 4, and comprises the classification of each part of the wind generating set, the fault mode of each sub-part, the fault reason and the processing suggestion;
s302: associating each early warning code with corresponding fault position, fault severity, fault reason and processing opinion information through association comparison based on an early warning information knowledge base;
s303: the fault diagnosis module 4 generates a quasi-early warning list.
The early warning information knowledge base comprises classification of each part of the wind generating set, fault modes of each sub-part, fault reasons and treatment suggestions, for example, a primary system transmission system, a secondary system gear box, a tertiary system gear box high-speed shaft and a bearing, and a quaternary system gear box high-speed shaft rear bearing of the wind generating set, the fault mode is bearing pitting, the fault reasons are poor lubrication of the gear box, the treatment suggestions are that the lubricating oil level and the grease condition of the gear box are checked, and the gear box oil needs to be replaced if the grease is abnormal.
Preferably, the S4 specifically includes the following contents:
s401: the front-end display module 5 displays the early warning list in a list form or a pop-up window form, and can display the detailed content of the early warning list;
s402: and judging whether an early warning list is initiated to a wind power plant site, and generating a feedback early warning list by the monitoring parameters of the wind generating set related to the early warning list and the fault characteristic value calculated by the early warning model and feeding the feedback early warning list back to the data warehouse module 1.
Preferably, the S5 specifically includes the following contents:
s501: setting four grades of excellent, good, medium and poor in the early warning function evaluation module 6, wherein 80% -100% of intervals correspond to excellent grades, 60% -80% of intervals correspond to good grades, and 0% -40% of intervals correspond to poor grades in 40% -60% of corresponding grades, extracting performance indexes of all links, and the completion rate, the accuracy rate and the initiation rate of an early warning order are used for expressing the performance of problem processing;
s502: the accuracy rate of the early warning list is used for evaluating the performance of the front-end display module 5, and the initiation rate of the quasi early warning list is used for evaluating the performance of the rules of the early warning module 3;
s503: and judging the levels of the early warning order completion rate, the early warning order accuracy rate and the early warning order initiation rate by reading the intervals of the early warning order completion rate, the early warning order accuracy rate and the early warning order initiation rate.
The invention provides a wind generating set fault early warning system and a method, according to the fault early warning characteristics of the wind generating set, the fault early warning function is divided into a plurality of links, the boundary is clear, no coupling exists between the links, and the system deployment and optimization are convenient; the method is suitable for development and construction of a group big data system and a wind power plant edge system under the conditions of different wind generating set models, operating environments and data acquisition, and effectively realizes an active maintenance strategy of the wind power plant; the system has an evaluation function, and the value of each module in the whole function can be evaluated by extracting performance indexes of each link, such as the accuracy rate of the early warning list and the initiation rate of the quasi early warning list.
Claims (10)
1. The utility model provides a wind generating set trouble early warning system which characterized by includes:
the data warehouse module (1) is used for storing basic information, configuration information, operation data and the like of the wind power plant and the wind turbine generator, and cleaning and sorting the data;
the early warning model module (2) is used for calculating a fault characteristic value according to the information of the wind power plant and the wind turbine generator by utilizing the data in the data warehouse module (1);
the early warning rule module (3) triggers an early warning code according to a preset rule by utilizing the fault characteristic value;
the fault diagnosis module (4) is associated with the early warning information knowledge base to generate complete early warning information, wherein the complete early warning information comprises fault positions, fault reasons and processing suggestions;
the front-end display module (5) initiates an early warning list to an external production management system, processes and feeds back fault and processing information, operates the early warning list through an application program, and processes and feeds back the fault and the processing information to be stored back to the database module (1) if the early warning list initiates the external production management system;
the early warning function evaluation (6) evaluates the quality of the early warning function through the fed back early warning list;
the early warning system comprises a data warehouse module (1), an early warning model module (2), an early warning rule module (3), a fault diagnosis module (4) and a front end display module (5), wherein the data warehouse module, the early warning model module, the early warning rule module and the fault diagnosis module are sequentially connected and are all connected with early warning function assessment (6).
2. The wind generating set fault pre-warning system according to claim 1, further comprising a data conversion module (12), wherein the data warehouse module (1) comprises structured data and unstructured data, the structured data is stored in a database form, the unstructured data is stored in a file form, and the unstructured data is converted into the structured data through the data conversion module (12).
3. The wind generating set fault early warning system according to claim 1, wherein the early warning model module (2) comprises a model data input module (9), an early warning model program module (7), a model data output module (10), a fault early warning system application module (11) and an early warning model management module (8), the model data input module (9), the early warning model program module (7), the model data output module (10) and the fault early warning system application module (11) are sequentially connected, and the early warning model program module (7) and the fault early warning system application module (11) are both connected with the early warning model management module (8).
4. A method of a wind park fault pre-warning system according to claim 1, comprising the steps of:
s1: calculating a fault characteristic value according to information of the wind power plant and the wind turbine generator;
s2: triggering an early warning code according to a preset rule by using the fault characteristic value;
s3: associating the early warning information knowledge base to generate complete early warning information;
s4: initiating an early warning list to an external production management system, and processing and feeding back fault and processing information;
s5: and the whole fault early warning process is supervised through the fed back early warning list, and the quality of each module is evaluated.
5. The method of the wind generating set fault pre-warning system according to claim 4, wherein the specific process of S1 includes the following steps:
s101: the data warehouse module (1) stores basic information, configuration information and operation data of a wind power plant and a wind turbine generator, and cleans and sorts the data, wherein the data comprises structured data and unstructured data, the structured data is directly used for model fast calculation, and the unstructured data is converted into the structured data through the data conversion module (12);
s102: the model data input module (9) acquires structural data required by calculation from the data warehouse module (1);
s103: the statistical analysis, machine learning and deep mining algorithm is modularized into an early warning model program module (7);
s104: the model management module (8) calls an early warning model program module (7), periodically calculates a fault characteristic value, and transmits the fault characteristic value to the fault early warning application module (11) through the model data output module (10).
6. The method of the wind generating set fault pre-warning system according to claim 5, wherein the formula for calculating the fault characteristic value in the step S104 is as follows:
in the formula xrmsIs a valid value, x, of a fault characteristic valueiThe residual value is a residual value, the residual value is a difference value between an actual measurement value and a predicted value, the predicted value is obtained through an ant colony algorithm, a machine learning algorithm or a deep learning algorithm, and the effective value of the fault characteristic value is output to a fault early warning application module (11) through a model data output module (10).
7. The method of the wind generating set fault pre-warning system according to claim 4, wherein the specific process of S2 includes the following steps:
s201: a threshold value A and a threshold value B are preset in the early warning rule module (3),
s202: linking the fault characteristic value with a corresponding fault through a logical relation, wherein the fault is represented by an early warning code;
the logic relation is as follows: if the condition 1 or the condition 2 is met, an early warning code is reported, if the temperature residual value of the rear bearing of the high-speed shaft of the gearbox is greater than a threshold value A, or if the bandwidth filtering vibration value of the high-speed shaft of the gearbox is greater than a threshold value B, the early warning code is reported.
8. The method of the wind generating set fault pre-warning system according to claim 4, wherein the specific process of S3 includes the following steps:
s301: an early warning information knowledge base is arranged in the fault diagnosis module (4), and the early warning information knowledge base comprises the classification of each part of the wind generating set, the fault mode of each sub-part, the fault reason and the processing suggestion;
s302: associating each early warning code with corresponding fault position, fault severity, fault reason and processing opinion information through association comparison based on an early warning information knowledge base;
s303: and the fault diagnosis module (4) generates a quasi-early warning list.
9. The method of the wind generating set fault pre-warning system according to claim 4, wherein the S4 specifically includes the following contents:
s401: the front-end display module (5) displays the early warning list in a list form or a pop-up window form, and can display the detailed content of the early warning list;
s402: and judging whether an early warning list is initiated to a wind power plant site, and generating a feedback early warning list by the monitoring parameters of the wind generating set related to the early warning list and the fault characteristic value calculated by the early warning model and feeding the feedback early warning list back to the data warehouse module (1).
10. The method of the wind generating set fault pre-warning system according to claim 4, wherein the S5 specifically includes the following contents:
s501: setting four grades of excellent, good, medium and poor in the early warning function evaluation module (6), wherein 80% -100% of intervals correspond to excellent grades, 60% -80% of intervals correspond to good grades, and 0% -40% of intervals correspond to poor grades in 40% -60% of corresponding grades, extracting performance indexes of all links, and the completion rate, the accuracy rate and the initiation rate of an early warning order are used for evaluating the performance of processing problems;
s502: the accuracy rate of the early warning list is used for expressing the performance of the front-end display module (5), and the initiation rate of the quasi early warning list is used for evaluating the performance of the rules of the early warning module (3);
s503: and judging the levels of the early warning order completion rate, the early warning order accuracy rate and the early warning order initiation rate by reading the intervals of the early warning order completion rate, the early warning order accuracy rate and the early warning order initiation rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910979917.3A CN110852484B (en) | 2019-10-15 | 2019-10-15 | Fault early warning system and method for wind generating set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910979917.3A CN110852484B (en) | 2019-10-15 | 2019-10-15 | Fault early warning system and method for wind generating set |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110852484A true CN110852484A (en) | 2020-02-28 |
| CN110852484B CN110852484B (en) | 2023-03-10 |
Family
ID=69596418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910979917.3A Active CN110852484B (en) | 2019-10-15 | 2019-10-15 | Fault early warning system and method for wind generating set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110852484B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111397930A (en) * | 2020-03-17 | 2020-07-10 | 成都智造天下科技有限公司 | Construction equipment fault detection and risk intelligent early warning system and method |
| CN111526203A (en) * | 2020-04-29 | 2020-08-11 | 华润电力技术研究院有限公司 | Fan fault early warning system and method |
| CN111721527A (en) * | 2020-05-18 | 2020-09-29 | 浙江工业大学 | Wind generating set gearbox fault positioning method based on CMS system big data combined standard deviation and wavelet entropy |
| CN115596620A (en) * | 2022-12-15 | 2023-01-13 | 深圳鹏锐信息技术股份有限公司（Cn） | Wind generating set fault intelligent analysis method and system based on artificial intelligence |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103033359A (en) * | 2012-12-19 | 2013-04-10 | 西安交通大学 | Wind turbine generator main gear fault diagnosis method of multi-feature multilevel comprehensive judgment |
| CN103903408A (en) * | 2014-04-04 | 2014-07-02 | 内蒙古大唐国际新能源有限公司 | Device fault detecting and early warning method and system |
| CN107038453A (en) * | 2017-03-06 | 2017-08-11 | 辽宁大唐国际新能源有限公司 | The fault early warning system and determination methods of a kind of Wind turbines |
| CN108732494A (en) * | 2017-04-21 | 2018-11-02 | 上海电气集团股份有限公司 | A kind of wind-driven generator abnormity diagnosis processing system |
| CN109947088A (en) * | 2019-04-17 | 2019-06-28 | 北京天泽智云科技有限公司 | Equipment fault early-warning system based on model lifecycle management |
| CN110322018A (en) * | 2019-06-19 | 2019-10-11 | 上海电力学院 | A kind of power plant fans fault early warning system based on fuzzy reasoning |
-
2019
- 2019-10-15 CN CN201910979917.3A patent/CN110852484B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103033359A (en) * | 2012-12-19 | 2013-04-10 | 西安交通大学 | Wind turbine generator main gear fault diagnosis method of multi-feature multilevel comprehensive judgment |
| CN103903408A (en) * | 2014-04-04 | 2014-07-02 | 内蒙古大唐国际新能源有限公司 | Device fault detecting and early warning method and system |
| CN107038453A (en) * | 2017-03-06 | 2017-08-11 | 辽宁大唐国际新能源有限公司 | The fault early warning system and determination methods of a kind of Wind turbines |
| CN108732494A (en) * | 2017-04-21 | 2018-11-02 | 上海电气集团股份有限公司 | A kind of wind-driven generator abnormity diagnosis processing system |
| CN109947088A (en) * | 2019-04-17 | 2019-06-28 | 北京天泽智云科技有限公司 | Equipment fault early-warning system based on model lifecycle management |
| CN110322018A (en) * | 2019-06-19 | 2019-10-11 | 上海电力学院 | A kind of power plant fans fault early warning system based on fuzzy reasoning |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111397930A (en) * | 2020-03-17 | 2020-07-10 | 成都智造天下科技有限公司 | Construction equipment fault detection and risk intelligent early warning system and method |
| CN111526203A (en) * | 2020-04-29 | 2020-08-11 | 华润电力技术研究院有限公司 | Fan fault early warning system and method |
| CN111526203B (en) * | 2020-04-29 | 2022-11-01 | 华润电力技术研究院有限公司 | Fan fault early warning system and method |
| CN111721527A (en) * | 2020-05-18 | 2020-09-29 | 浙江工业大学 | Wind generating set gearbox fault positioning method based on CMS system big data combined standard deviation and wavelet entropy |
| CN111721527B (en) * | 2020-05-18 | 2022-04-08 | 浙江工业大学 | Wind generating set gearbox fault positioning method based on CMS system big data combined standard deviation and wavelet entropy |
| CN115596620A (en) * | 2022-12-15 | 2023-01-13 | 深圳鹏锐信息技术股份有限公司（Cn） | Wind generating set fault intelligent analysis method and system based on artificial intelligence |
| CN115596620B (en) * | 2022-12-15 | 2023-03-07 | 深圳鹏锐信息技术股份有限公司 | Wind generating set fault intelligent analysis method and system based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110852484B (en) | 2023-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110852484B (en) | Fault early warning system and method for wind generating set | |
| Li et al. | A real-time inspection and opportunistic maintenance strategies for floating offshore wind turbines | |
| Zaher et al. | A multi-agent fault detection system for wind turbine defect recognition and diagnosis | |
| CN104951851B (en) | A kind of Wind turbines State Forecasting Model method for building up based on gray relative regression support vector machine | |
| CN111597682B (en) | Method for predicting remaining life of bearing of gearbox of wind turbine | |
| CN102510125B (en) | Method and device for monitoring operation conditions of power primary equipment | |
| CN110991666A (en) | Fault detection method, model training method, device, equipment and storage medium | |
| Catelani et al. | Risk assessment of a wind turbine: A new FMECA-based tool with RPN threshold estimation | |
| CN103512651B (en) | The Wind turbines state monitoring method reported to the police based on vibration mutation and device | |
| CN109685371B (en) | Bayesian network-based dynamic probability wind turbine generator running state comprehensive evaluation method | |
| WO2013071931A1 (en) | A system and method for identifying suggestions to remedy wind turbine faults | |
| CN108564313A (en) | The method and device of Wind turbines status monitoring and health evaluating based on fault tree | |
| CN108928744B (en) | Container crane online diagnosis and maintenance system based on big data | |
| CN111651505A (en) | Data-driven equipment operation situation analysis and early warning method and system | |
| CN112390127B (en) | Health degree model-based preventive maintenance strategy generation method for escalator | |
| CN108072524A (en) | A kind of gearbox of wind turbine bearing fault method for early warning | |
| CN103309342A (en) | Safety verification scheme aiming at industrial control system | |
| CN114184956B (en) | Service wind generating set fault prediction method based on big data management | |
| Shi et al. | Study of wind turbine fault diagnosis and early warning based on SCADA data | |
| Sun et al. | Failure analysis of floating offshore wind turbines with correlated failures | |
| Sun et al. | Failure analysis of floating offshore wind turbines considering working condition correlations | |
| CN103994199A (en) | Gear case maintenance time determining method based on state monitoring | |
| CN113743510A (en) | Gear transmission system FMMEA analysis method based on fault tree analysis | |
| CN108491622A (en) | A kind of fault diagnosis method and system of Wind turbines | |
| KR20220089853A (en) | Method for Failure prediction and prognostics and health management of renewable energy generation facilities using machine learning technology |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |