CN114510517A - Data processing method and system for health management of large-scale rotating unit - Google Patents
Data processing method and system for health management of large-scale rotating unit Download PDFInfo
- Publication number
- CN114510517A CN114510517A CN202210188364.1A CN202210188364A CN114510517A CN 114510517 A CN114510517 A CN 114510517A CN 202210188364 A CN202210188364 A CN 202210188364A CN 114510517 A CN114510517 A CN 114510517A
- Authority
- CN
- China
- Prior art keywords
- data
- dimension
- running state
- state data
- module
- 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
- 230000036541 health Effects 0.000 title claims abstract description 57
- 238000003672 processing method Methods 0.000 title claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 78
- 238000012545 processing Methods 0.000 claims abstract description 70
- 238000013135 deep learning Methods 0.000 claims abstract description 59
- 238000004364 calculation method Methods 0.000 claims abstract description 41
- 238000004140 cleaning Methods 0.000 claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000007619 statistical method Methods 0.000 claims abstract description 29
- 230000008569 process Effects 0.000 claims abstract description 22
- 230000002776 aggregation Effects 0.000 claims description 18
- 238000004220 aggregation Methods 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 17
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 238000011835 investigation Methods 0.000 claims description 6
- 238000012800 visualization Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000008878 coupling Effects 0.000 abstract description 6
- 238000010168 coupling process Methods 0.000 abstract description 6
- 238000005859 coupling reaction Methods 0.000 abstract description 6
- 238000012423 maintenance Methods 0.000 description 18
- 230000003190 augmentative effect Effects 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24553—Query execution of query operations
- G06F16/24554—Unary operations; Data partitioning operations
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24553—Query execution of query operations
- G06F16/24554—Unary operations; Data partitioning operations
- G06F16/24556—Aggregation; Duplicate elimination
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Fuzzy Systems (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a data processing method and a system for health management of a large-scale rotating unit, wherein the system comprises a source data layer, a data detail layer, a data dimension increasing layer and an algorithm application display layer, which are respectively used for acquiring running state data of the rotating unit in the running process; carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data list; carrying out dimension expansion processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching; and respectively matching the running state data of each dimensionality in the full detail data wide table with the corresponding deep learning algorithm model to perform data calculation and aggregate statistical analysis, and visually displaying the data calculation and aggregate statistical analysis results. By using the method and the system, the data coupling in the deep learning algorithm is reduced, the reusability of intermediate data is increased, and the labor and time cost of data processing in the health management process is reduced.
Description
Technical Field
The invention relates to the technical field of equipment information monitoring and big data, in particular to a data processing method and a data processing system for health management of a large-scale rotating unit.
Background
With the increasing requirements on the safety and reliability of mechanical equipment, the importance of health management and intelligent operation and maintenance technology is continuously improved, and the development of the health management intelligent operation and maintenance technology for large-scale rotating units is more and more rapid. The intelligent operation and maintenance of the health management of the large-scale rotating unit is realized by collecting the data of the operation process of the rotating unit and utilizing a deep learning method to carry out fault prediction, service life prediction and maintenance decision on mechanical equipment. A traditional deep learning mode usually forms a large amount of intermediate data through a complex data processing flow, and health management intelligent operation and maintenance data have the characteristic of strong coupling, so that data reuse is difficult to be carried out on various algorithm models, and a large amount of manpower and time cost waste are caused. Therefore, it is necessary to establish a data warehouse for health management of a large rotating machine group and construct health management data with high reusability.
A traditional data warehouse adopts dimensional modeling and is generally divided into a data preparation layer, a data detail layer, a data summarization layer, a data mart layer and a data application layer. In the traditional data warehouse field modeling, data layering is usually carried out by taking a business process as a guide, statistical analysis is carried out facing business requirements, and data required by the intelligent operation and maintenance of the health management of the large-scale rotating unit mainly face different requirements of learning models with different depths, so that the traditional data warehouse modeling mode is difficult to apply in the intelligent operation and maintenance of the health management of the large-scale rotating unit, and a data warehouse facing the intelligent operation and maintenance of the health management of the large-scale rotating unit needs to be established aiming at the characteristics of the intelligent operation and maintenance of the health management of the large-scale rotating unit.
Disclosure of Invention
Aiming at the defects of the prior art, the problems to be solved by the invention are as follows: a data processing method and a data processing system facing intelligent operation and maintenance of health management of a large-scale rotating unit are provided, so that the problems of strong data coupling, low data reuse rate and high waste of labor and time cost in the existing intelligent operation and maintenance process of the health management of the rotating unit are solved.
In order to solve the technical problems, the invention adopts the following technical scheme:
a data processing method for health management of a large rotating unit comprises the following steps:
step 1: acquiring running state data of a rotating unit in a running process;
step 2: carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data list;
and step 3: carrying out dimension expansion processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching;
and 4, step 4: and respectively matching the running state data of each dimensionality in the full detail data wide table with the corresponding deep learning algorithm model to perform data calculation and aggregate statistical analysis, and visually displaying the data calculation and aggregate statistical analysis results.
In the data processing method for health management of a large-scale rotating machine set, preferably, the step 2 specifically includes:
performing data slicing processing on the obtained running state data according to different data dimensions to obtain data details of the running state data in different data dimensions;
carrying out data cleaning processing on the running state data of each dimension to remove invalid data in the running state data;
abnormal data investigation is carried out on the running state data of each dimension, and the investigated abnormal data is repaired;
and sorting the operation state data of each dimension after cleaning and repairing to obtain a data detail table.
In the data processing method for health management of a large-scale rotating machine set, preferably, the step 3 specifically includes:
performing time-frequency analysis and parameter index extraction on the cleaned and repaired running state data, adding data dimensionality obtained by the time-frequency analysis and data dimensionality obtained by the parameter index extraction into the data detail table, and performing dimensionality extension processing to obtain a dimensionality data wide table;
and matching other equipment associated with the operation state data, and adding other equipment parameter information associated with the operation state data into the dimension-increased data width table as a new data dimension to obtain a full detailed data width table.
In the data processing method for health management of a large-scale rotating machine set, preferably, the step 4 specifically includes:
reading the running state data of each dimension from the full detail data wide table, and respectively searching a matched deep learning algorithm model from a deep learning algorithm library aiming at the running state data of each dimension, so as to call the matched deep learning algorithm model to respectively perform data calculation on the running state data of each dimension;
respectively carrying out aggregate statistical analysis on the data calculation results of the running state data of each dimension;
and visually displaying the data calculation result and the aggregation statistical analysis result of the running state data.
Correspondingly, the invention provides a data processing system for health management of a large-scale rotating unit, which comprises a source data layer, a data detail layer, a data dimension increasing layer and an algorithm application display layer;
the source data layer is used for storing the acquired running state data of the rotating unit in the running process;
the data detail layer is used for carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data detail table;
the data dimension increasing layer is used for performing dimension expanding processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching;
the algorithm application display layer is used for matching the running state data of each dimensionality in the full-detail data wide table with the corresponding deep learning algorithm model respectively to perform data calculation and aggregate statistical analysis, and performing visual display on data calculation and aggregate statistical analysis results.
In the data processing system for health management of the large-scale rotating machine set, preferably, the data detail layer comprises a data slicing module, a data cleaning module, a data repairing module and a data sorting module;
the data slicing module is used for carrying out data slicing processing on the obtained running state data according to different data dimensions to obtain data details of the running state data in different data dimensions;
the data cleaning module is used for cleaning the data of the running state data of each dimension and removing invalid data in the running state data;
the data restoration module is used for performing abnormal data investigation on the running state data of each dimension and restoring the investigated abnormal data;
the data sorting module is used for sorting the operation state data of each dimensionality after cleaning and repairing to obtain a data detail table.
In the data processing system for health management of the large-scale rotating machine set, the data dimension increasing layer preferably comprises a signal processing module, an index extracting module and a dimension expanding module;
the signal processing module is used for performing time-frequency analysis on the cleaned and repaired running state data;
the index extraction module is used for extracting parameter indexes of the cleaned and repaired running state data;
the dimension expansion module is used for adding the data dimension obtained by time-frequency analysis and the data dimension obtained by parameter index extraction into the data detail table, and performing dimension expansion processing to obtain a dimension-increased data wide table; and matching other devices associated with the operation state data, and adding parameter information of other devices associated with the operation state data into the dimension-increased data width table as a new data dimension to obtain a full detailed data width table.
In the data processing system for health management of the large-scale rotating machine set, the algorithm application display layer preferably comprises a deep learning algorithm library module, an aggregation statistical module and a visualization module;
the deep learning algorithm library module is provided with a deep learning algorithm library, and a plurality of deep learning algorithm models matched with running state data of different dimensions of the rotating unit are stored in the deep learning algorithm library; the deep learning algorithm library module is used for reading the running state data of each dimension from the full detailed data wide table, and searching a matched deep learning algorithm model from the deep learning algorithm library aiming at the running state data of each dimension, so that the matched deep learning algorithm model is called to perform data calculation on the running state data of each dimension;
the aggregation statistical module is used for respectively carrying out aggregation statistical analysis on the data calculation results of the running state data of each dimension;
the visualization module is used for visually displaying the data calculation result and the aggregation statistical analysis result of the running state data.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention relates to a data processing method and a system for health management of a large-scale rotating unit, wherein the system comprises a source data layer, a data detail layer, a data dimension increasing layer and an algorithm application display layer, which are respectively used for acquiring running state data of the rotating unit in the running process; carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data list; carrying out dimension expansion processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching; respectively matching the running state data of each dimension in the full detail data wide table with a corresponding deep learning algorithm model for data calculation and aggregate statistical analysis, and visually displaying the data calculation and aggregate statistical analysis results; therefore, the data processing system for the health management of the large-scale rotating unit can store a large amount of operating data of the rotating unit into the data source according to different themes, is convenient for calling the data with different themes to carry out the health management analysis, and solves the problems of multiple data theme types and complex data classification in the operating process of the large-scale rotating unit.
2. According to the data processing system for health management of the large-scale rotating unit, the source data of the health management of the large-scale rotating unit is subjected to data layering processing through system layering, so that data of each layer can be reused for multiple times, and each system layering can also run in parallel respectively, so that the problem of strong data coupling in the intelligent operation and maintenance process of the health management of the large-scale rotating unit is solved, and the time and labor cost for processing the source data are reduced.
3. In the data processing method and the data processing system for the health management of the large-scale rotating machine set, data required by a deep learning algorithm model are matched according to multiple dimensions, so that data required by the algorithm are obtained for calculation, a traditional data warehouse usually only carries out data layering and aggregation statistics according to a business process instead of carrying out data layering according to a specific deep learning model, and therefore the method solves the problem that the data are matched with the deep learning algorithm for calculation in the intelligent operation and maintenance process of the health management of the large-scale rotating machine set, and is more suitable for the intelligent operation and maintenance of the health management of the rotating machine set.
Drawings
For purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made in detail to the present invention as illustrated in the accompanying drawings, in which:
FIG. 1 is a flow chart of a data processing method for health management of a large-scale rotating unit.
FIG. 2 is a structural diagram of a data processing system for health management of large rotating machines according to the present invention.
FIG. 3 is a diagram of an embodiment of a data processing system for health management of a large rotating group according to the present invention.
Detailed Description
The following detailed description of embodiments of the invention is provided in connection with the accompanying drawings.
The method for intelligently processing the operation and maintenance data for health management of the large-scale rotating unit solves the problems of strong data coupling, low data reuse rate and high time and labor consumption of different-depth learning model algorithm scheduling in the health management process of the rotating unit.
As shown in fig. 1, the data processing method for health management of a large rotating machine set according to the present invention includes the following steps:
step 1: and acquiring running state data of the rotating unit in the running process.
And 2, step: and carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data list.
And step 3: and carrying out dimension expansion processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching.
And 4, step 4: and respectively matching the running state data of each dimensionality in the full detail data wide table with the corresponding deep learning algorithm model to perform data calculation and aggregate statistical analysis, and visually displaying the data calculation and aggregate statistical analysis results.
Correspondingly, as shown in fig. 2, the invention also provides a data processing system for health management of a large-scale rotating unit, which comprises a source data layer, a data detail layer, a data dimension increasing layer and an algorithm application display layer;
the source data layer is used for storing the acquired running state data of the rotating unit in the running process;
the data detail layer is used for carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data detail table;
the data dimension increasing layer is used for performing dimension expanding processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching;
the algorithm application display layer is used for matching the running state data of each dimensionality in the full detail data wide table with the corresponding deep learning algorithm model respectively to perform data calculation and aggregate statistical analysis, and performing visual display on data calculation and aggregate statistical analysis results.
Firstly, importing data collected during the working operation of a large-scale rotating unit into a source data layer to obtain an original data fact table. The source Data refers to equipment running state Data acquired by the rotating unit in different Acquisition modes in the running process, the Data source of the large rotating unit is divided into multiple subjects, such as vibration CMS (Condition Monitoring System) vibration Data, SCADA (Supervisory Control And Data Acquisition) Monitoring Data And the like, different subject source Data have different Data processing modes in the follow-up process, And the different subject source Data correspond to different deep learning algorithms. As shown in fig. 3, the present embodiment is described with CMS vibration data as a subject object.
The source data layer is used for storing source data, for example, CMS vibration data is common data in the operation process of a rotating unit, the vibration data comprises various dimensions, such as wind field names, fan numbers, gear box numbers, measuring points, measuring point types, sampling time and the like, and the CMS vibration data fact table is imported into the source data layer according to multi-dimensional statistics.
The data detail layer is mainly used for preprocessing data, and comprises data slicing, data cleaning, data repairing and the like, wherein the corresponding functional module comprises a data slicing module, a data cleaning module and a data repairing module. The CMS data has the characteristics of high acquisition frequency and large data volume, so that a data slicing module needs to be called at a data detail layer to slice the CMS data, and the sliced CMS data can be better applied to deep learning.
Specifically, the data detail layer comprises a data slicing module, a data cleaning module, a data repairing module and a data sorting module;
the data slicing module is used for carrying out data slicing processing on the obtained running state data according to different data dimensions to obtain data details of the running state data in different data dimensions;
the data cleaning module is used for cleaning the data of the running state data of each dimension and removing invalid data in the running state data;
the data restoration module is used for performing abnormal data investigation on the running state data of each dimension and restoring the investigated abnormal data;
the data sorting module is used for sorting the operation state data of each dimensionality after cleaning and repairing to obtain a data detail table.
Different modules in the data detail layer can be respectively called according to different subject data characteristics and different data cleaning and repairing requirements.
Therefore, step 2 of the data processing method for health management of the large-scale rotating machine set specifically comprises the following steps: performing data slicing processing on the obtained running state data according to different data dimensions to obtain data details of the running state data in different data dimensions; carrying out data cleaning processing on the running state data of each dimension to remove invalid data in the running state data; abnormal data investigation is carried out on the running state data of each dimension, and the investigated abnormal data is repaired; and sorting the operation state data of each dimension after cleaning and repairing to obtain a data detail table.
The data dimension increasing layer is mainly used for further processing the data of the data detail layer according to the requirements of a deep learning algorithm, and dimension expansion is carried out in the modes of signal processing, index extraction, dimension table matching and the like. For example, because deep learning model algorithms have different dimensional requirements for CMS vibration signals, it is necessary to perform various processing on CMS data to meet the requirements of different depth learning algorithms.
Specifically, the data dimension increasing layer comprises a signal processing module, an index extracting module and a dimension expanding module;
the signal processing module is used for performing time-frequency analysis on the cleaned and repaired running state data;
the index extraction module is used for extracting parameter indexes of the cleaned and repaired running state data;
the dimension expansion module is used for adding the data dimension obtained by time-frequency analysis and the data dimension obtained by parameter index extraction into the data detail table, and performing dimension expansion processing to obtain an increased dimension data wide table; and matching other devices associated with the operation state data, and adding parameter information of other devices associated with the operation state data into the dimension-increased data width table as a new data dimension to obtain a full detailed data width table.
Therefore, step 3 of the data processing method for health management of the large-scale rotating machine set specifically comprises the following steps: performing time-frequency analysis and parameter index extraction on the cleaned and repaired running state data, adding data dimensionality obtained by the time-frequency analysis and data dimensionality obtained by the parameter index extraction into the data detail table, and performing dimensionality extension processing to obtain a dimensionality data wide table; and matching other equipment associated with the operation state data, and adding other equipment parameter information associated with the operation state data into the dimension-increased data width table as a new data dimension to obtain a full detailed data width table.
For example, for CMS vibration data, a signal processing module is first required to be called to perform time-frequency analysis processing on the CMS vibration data, including wavelet packet transformation, envelope and other time-frequency analysis methods, different signal time-frequency analysis processing methods represent one dimension of a fact table, and signal data subjected to signal time-frequency analysis processing is added to the fact table of a data detail layer according to different dimensions to form a data dimension increasing table; then, parameter index extraction is carried out on the dimension increasing fact table subjected to signal processing, an index extraction module index extraction method is called to carry out parameter index extraction processing on the data dimension increasing table to obtain data comprising kurtosis, inclination, mean value and the like, different indexes represent a dimension, and the index data extracted through indexes are added into the data dimension increasing table to continue extending the dimension of the fact table; finally, dimension expansion is carried out on the augmented data wide table obtained by sorting the fact table after dimension increase to obtain a full detailed data wide table with complete information, the dimension expansion is carried out on the augmented data wide table by calling a dimension table in a dimension expansion module, other equipment associated with the operation state data is matched, other equipment parameter information associated with the operation state data is extracted, the CMS vibration signal is closely connected with the gear box, therefore, the gear box dimension table can be matched with the fact table of the CMS operation state data, the parameter information of the gear box such as the name of the gear box and other mapping information is used as a new data dimension and is added into the augmented data wide table to form the full detailed data wide table.
The algorithm application display layer is mainly used for matching with the full-detail fact broad table by using a deep learning algorithm in the algorithm module, namely each algorithm needs to indicate data dimension information needing to be matched, namely the algorithm dimension table needs to comprise dimensions such as a wind field number, a fan number, measuring point information, sensor codes, sampling frequency, slice length, a signal processing mode, needed indexes and the like, and meanwhile, the address where the algorithm is located needs to be used for calling.
Specifically, the algorithm application display layer comprises a deep learning algorithm library module, an aggregation statistical module and a visualization module;
the deep learning algorithm library module is provided with a deep learning algorithm library, and a plurality of deep learning algorithm models matched with running state data of different dimensions of the rotating unit are stored in the deep learning algorithm library; the deep learning algorithm library module is used for reading the running state data of each dimension from the full detailed data wide table, and searching a matched deep learning algorithm model from the deep learning algorithm library aiming at the running state data of each dimension, so that the matched deep learning algorithm model is called to perform data calculation on the running state data of each dimension;
the aggregation statistical module is used for respectively carrying out aggregation statistical analysis on the data calculation results of the running state data of each dimension;
and the visualization module is used for visually displaying the data calculation result and the aggregation statistical analysis result of the running state data.
Therefore, step 4 of the data processing method for health management of the large-scale rotating machine set specifically comprises the following steps: reading the running state data of each dimension from the full detail data wide table, and respectively searching a matched deep learning algorithm model from a deep learning algorithm library aiming at the running state data of each dimension, so as to call the matched deep learning algorithm model to respectively perform data calculation on the running state data of each dimension; performing aggregation statistical analysis on the data calculation results of the running state data of each dimension respectively; and visually displaying the data calculation result and the aggregation statistical analysis result of the running state data.
In specific implementation, an algorithm dimension table in a deep learning algorithm library module needs to be manually maintained, corresponding data can be obtained when different depth learning algorithm models in the deep learning algorithm library module are matched with data dimensions in a full data detail width table one by one, the accuracy of multiple times of calculation of test models is often calculated to obtain an average value when deep learning test data are calculated and applied, and calculation results of multiple methods can be compared, so that the calculation results of deep learning need to be aggregated, summarized and counted, and are visually displayed.
In summary, the data processing system for health management of the large-scale rotating unit enables source data of the health management of the large-scale rotating unit to be processed in a data layering manner through system layering, enables data of each layer to be multiplexed for multiple times, and enables the system layering to operate in parallel respectively, so that the problem of strong data coupling in the intelligent operation and maintenance process of the health management of the large-scale rotating unit is solved, and time and labor cost for processing the source data are reduced; meanwhile, in the method and the system, data required by the deep learning algorithm model are matched according to multiple dimensions, so that data required by the algorithm are obtained for calculation, and a traditional data warehouse usually only carries out data layering and aggregation statistics according to a service flow instead of carrying out data layering according to a specific deep learning model, so that the method solves and systematically solves the problem that the data is matched with the deep learning algorithm for calculation in the intelligent operation and maintenance process of health management of the large-scale rotating unit, and is more suitable for the intelligent operation and maintenance of the health management of the rotating unit.
Finally, it is noted that the above-mentioned embodiments illustrate rather than limit the invention, and that, while the invention has been described with reference to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A data processing method for health management of a large-scale rotating unit is characterized by comprising the following steps:
step 1: acquiring running state data of a rotating unit in a running process;
step 2: carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data list;
and step 3: carrying out dimension expansion processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching;
and 4, step 4: and respectively matching the running state data of each dimensionality in the full detail data wide table with the corresponding deep learning algorithm model to perform data calculation and aggregate statistical analysis, and visually displaying the data calculation and aggregate statistical analysis results.
2. The data processing method for health management of the large-scale rotating unit according to claim 1, wherein the step 2 specifically comprises:
performing data slicing processing on the obtained running state data according to different data dimensions to obtain data details of the running state data in different data dimensions;
carrying out data cleaning processing on the running state data of each dimension to remove invalid data in the running state data;
abnormal data investigation is carried out on the running state data of each dimension, and the investigated abnormal data is repaired;
and sorting the operation state data of each dimension after cleaning and repairing to obtain a data detail table.
3. The data processing method for health management of the large-scale rotating unit according to claim 1, wherein the step 3 specifically comprises:
performing time-frequency analysis and parameter index extraction on the cleaned and repaired running state data, adding data dimensionality obtained by the time-frequency analysis and data dimensionality obtained by the parameter index extraction into the data detail table, and performing dimensionality extension processing to obtain a dimensionality data wide table;
and matching other equipment associated with the operation state data, and adding other equipment parameter information associated with the operation state data into the dimension-increased data width table as a new data dimension to obtain a full detailed data width table.
4. The data processing method for health management of the large-scale rotating machine set according to claim 1, wherein the step 4 is specifically:
reading the running state data of each dimension from the full detail data wide table, and respectively searching a matched deep learning algorithm model from a deep learning algorithm library aiming at the running state data of each dimension, so as to call the matched deep learning algorithm model to respectively perform data calculation on the running state data of each dimension;
respectively carrying out aggregate statistical analysis on the data calculation results of the running state data of each dimension;
and visually displaying the data calculation result and the aggregation statistical analysis result of the running state data.
5. A data processing system for health management of a large-scale rotating unit is characterized by comprising a source data layer, a data detail layer, a data dimension increasing layer and an algorithm application display layer;
the source data layer is used for storing the acquired running state data of the rotating unit in the running process;
the data detail layer is used for carrying out data cleaning and repairing treatment on the obtained running state data to obtain a data detail table;
the data dimension increasing layer is used for performing dimension expanding processing and dimension information matching processing on the cleaned and repaired running state data to obtain a full detailed data wide table after dimension increasing and information matching;
the algorithm application display layer is used for matching the running state data of each dimensionality in the full-detail data wide table with the corresponding deep learning algorithm model respectively to perform data calculation and aggregate statistical analysis, and performing visual display on data calculation and aggregate statistical analysis results.
6. The data processing system for health management of large-scale rotating units according to claim 5, wherein the data detail layer comprises a data slicing module, a data cleaning module, a data repairing module and a data sorting module;
the data slicing module is used for carrying out data slicing processing on the obtained running state data according to different data dimensions to obtain data details of the running state data in different data dimensions;
the data cleaning module is used for cleaning the data of the running state data of each dimension and removing invalid data in the running state data;
the data restoration module is used for performing abnormal data investigation on the running state data of each dimension and restoring the investigated abnormal data;
the data sorting module is used for sorting the operation state data of each dimensionality after cleaning and repairing to obtain a data detail table.
7. The data processing system for health management of the large-scale rotating units according to claim 5, wherein the data dimension increasing layer comprises a signal processing module, an index extraction module and a dimension expansion module;
the signal processing module is used for performing time-frequency analysis on the cleaned and repaired running state data;
the index extraction module is used for extracting parameter indexes of the cleaned and repaired running state data;
the dimension expansion module is used for adding the data dimension obtained by time-frequency analysis and the data dimension obtained by parameter index extraction into the data detail table, and performing dimension expansion processing to obtain a dimension-increased data wide table; and matching other devices associated with the operation state data, and adding parameter information of other devices associated with the operation state data into the dimension-increased data width table as a new data dimension to obtain a full detailed data width table.
8. The data processing system for health management of large rotating units according to claim 5, wherein the algorithm application display layer comprises a deep learning algorithm library module, an aggregation statistical module and a visualization module;
the deep learning algorithm library module is provided with a deep learning algorithm library, and a plurality of deep learning algorithm models matched with running state data of different dimensions of the rotating unit are stored in the deep learning algorithm library; the deep learning algorithm library module is used for reading the running state data of each dimension from the full detailed data wide table, and searching a matched deep learning algorithm model from the deep learning algorithm library aiming at the running state data of each dimension, so that the matched deep learning algorithm model is called to perform data calculation on the running state data of each dimension;
the aggregation statistical module is used for respectively carrying out aggregation statistical analysis on the data calculation results of the running state data of each dimension;
the visualization module is used for visually displaying the data calculation result and the aggregation statistical analysis result of the running state data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210188364.1A CN114510517A (en) | 2022-02-28 | 2022-02-28 | Data processing method and system for health management of large-scale rotating unit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210188364.1A CN114510517A (en) | 2022-02-28 | 2022-02-28 | Data processing method and system for health management of large-scale rotating unit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114510517A true CN114510517A (en) | 2022-05-17 |
Family
ID=81553350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210188364.1A Pending CN114510517A (en) | 2022-02-28 | 2022-02-28 | Data processing method and system for health management of large-scale rotating unit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114510517A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117112039A (en) * | 2023-08-24 | 2023-11-24 | 中邮通建设咨询有限公司 | Transmission optimization system and operation method of data center |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111506660A (en) * | 2020-04-21 | 2020-08-07 | 瑞纳智能设备股份有限公司 | Heat supply network real-time data warehouse system |
| CN112765232A (en) * | 2021-01-05 | 2021-05-07 | 广东核电合营有限公司 | Data management method and system for nuclear power plant major equipment |
| CN113482863A (en) * | 2021-07-16 | 2021-10-08 | 南京安维士传动技术股份有限公司 | State evaluation and health management system of wind generating set |
| CN113988421A (en) * | 2021-10-29 | 2022-01-28 | 北京中恒博瑞数字电力科技有限公司 | Wind turbine generator power intelligent prediction method based on deep learning |
| CN114077635A (en) * | 2021-11-22 | 2022-02-22 | 南京轨道交通系统工程有限公司 | Urban rail intelligent operation and maintenance platform based on big data analysis and deep learning |
-
2022
- 2022-02-28 CN CN202210188364.1A patent/CN114510517A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111506660A (en) * | 2020-04-21 | 2020-08-07 | 瑞纳智能设备股份有限公司 | Heat supply network real-time data warehouse system |
| CN112765232A (en) * | 2021-01-05 | 2021-05-07 | 广东核电合营有限公司 | Data management method and system for nuclear power plant major equipment |
| CN113482863A (en) * | 2021-07-16 | 2021-10-08 | 南京安维士传动技术股份有限公司 | State evaluation and health management system of wind generating set |
| CN113988421A (en) * | 2021-10-29 | 2022-01-28 | 北京中恒博瑞数字电力科技有限公司 | Wind turbine generator power intelligent prediction method based on deep learning |
| CN114077635A (en) * | 2021-11-22 | 2022-02-22 | 南京轨道交通系统工程有限公司 | Urban rail intelligent operation and maintenance platform based on big data analysis and deep learning |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117112039A (en) * | 2023-08-24 | 2023-11-24 | 中邮通建设咨询有限公司 | Transmission optimization system and operation method of data center |
| CN117112039B (en) * | 2023-08-24 | 2024-04-26 | 中邮通建设咨询有限公司 | Transmission optimization system and operation method of data center |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840666B (en) | Model construction method and system for predicting future wind turbine generator failure | |
| CN111060337A (en) | Running equipment real-time fault diagnosis method based on expert system | |
| CN112486136B (en) | Fault early warning system and method | |
| CN110618384A (en) | Motor performance test platform | |
| CN112883075B (en) | Landslide universal type ground surface displacement monitoring data missing and outlier processing method | |
| CN114510517A (en) | Data processing method and system for health management of large-scale rotating unit | |
| CN113551765A (en) | Sound spectrum analysis and diagnosis method for equipment fault | |
| CN111993157A (en) | Machining equipment production state monitoring system and method based on electric signals | |
| CN115099284A (en) | Hot rolling mill vibration data processing method, system, terminal and storage medium | |
| CN117078232B (en) | Processing equipment fault prevention system and method based on big data | |
| Cao et al. | Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator | |
| CN117520816A (en) | Vibration fault characteristic signal extraction method combining EMD and tensor decomposition | |
| CN112598142A (en) | Wind turbine generator overhaul work quality examination assisting method and system | |
| CN112524077A (en) | Method, device and system for detecting fan fault | |
| CN117151214A (en) | Method, device, equipment and medium for analyzing operation and maintenance of airplane windshield | |
| CN117076510A (en) | Automatic inspection method for intelligent manufacturing | |
| CN111488968A (en) | Method and system for extracting comprehensive energy metering data features | |
| CN115750341A (en) | Roots compressor fault diagnosis method and system | |
| CN114004996A (en) | Abnormal sound detection method, abnormal sound detection device, electronic equipment and medium | |
| CN113268552A (en) | Generator equipment hidden danger early warning method based on locality sensitive hashing | |
| CN114579827B (en) | Method, device and equipment for processing data performance curve of industrial equipment | |
| CN118690066A (en) | Comprehensive visualization system for bridge health monitoring data | |
| CN113393143B (en) | Fin machining state monitoring method based on information fusion | |
| CN116150636B (en) | Fault monitoring method and system for transmission module | |
| CN118535618A (en) | Digital report feature analysis method and system based on data mining |
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 |