CN117829821B - Cloud platform-based composite material equipment maintenance and management method - Google Patents
Cloud platform-based composite material equipment maintenance and management method Download PDFInfo
- Publication number
- CN117829821B CN117829821B CN202410240194.6A CN202410240194A CN117829821B CN 117829821 B CN117829821 B CN 117829821B CN 202410240194 A CN202410240194 A CN 202410240194A CN 117829821 B CN117829821 B CN 117829821B
- Authority
- CN
- China
- Prior art keywords
- data
- feature
- life cycle
- layer
- feature word
- 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.)
- Active
Links
- 239000002131 composite material Substances 0.000 title claims abstract description 33
- 238000012423 maintenance Methods 0.000 title claims abstract description 15
- 238000007726 management method Methods 0.000 title claims abstract description 12
- 239000013598 vector Substances 0.000 claims description 152
- 230000002093 peripheral effect Effects 0.000 claims description 37
- 238000000605 extraction Methods 0.000 claims description 33
- 238000000034 method Methods 0.000 claims description 29
- 238000004891 communication Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 6
- 230000007547 defect Effects 0.000 abstract description 4
- 238000012545 processing Methods 0.000 abstract description 3
- 238000001845 vibrational spectrum Methods 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000006247 magnetic powder Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
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
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a composite material equipment maintenance management method based on a cloud platform, which relates to the technical field of data processing and effectively improves the accuracy of defect detection of composite material equipment. According to the invention, one data in the historical state data is sequentially selected as main data and the other data are auxiliary data through a plurality of pieces of historical state data pre-stored in each working part, a life cycle judging network and a device topology network are established, each sensor acquires real-time state data of the working part at the position of the working part, each topological node in the device topology network is associated with the corresponding life cycle judging network, the real-time state data of each working part is input into the device topology network, the device topology network distributes the real-time state data through the life cycle judging network, an actual life cycle change amount is obtained according to the distribution result, and the current residual life cycle of the working part is updated according to the actual life cycle change amount.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a cloud platform-based composite material equipment maintenance and management method.
Background
Composite equipment maintenance refers to the process of servicing, repairing, and maintaining equipment manufactured using composite materials. The composite material is formed by combining two or more materials with different properties, has the characteristics of excellent mechanical property, chemical property, thermal property and the like, and is generally used in various industries such as aerospace, automobile manufacturing, constructional engineering, energy and the like. In order to ensure proper operation and extend the useful life of composite equipment, it is important to conduct equipment maintenance on a regular basis.
The existing composite material equipment maintenance technology is realized through the existing technologies such as ultrasonic detection, magnetic powder detection, infrared thermal imaging and the like. However, due to the complexity and heterogeneity of composite materials, there may be certain limitations in detecting internal defects of composite materials, which may not be accurately detected by current nondestructive detection methods.
Therefore, how to improve the accuracy of defect detection of composite material equipment is a difficulty of the prior art, and a cloud platform-based composite material equipment maintenance and management method is provided for the purpose.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a composite material equipment maintenance and management method based on a cloud platform.
In order to achieve the above object, the present invention provides the following technical solutions:
a cloud platform-based composite material equipment maintenance management method comprises the following steps:
s1, setting a cloud computing platform, installing various sensor devices on each working part of composite material equipment, and simultaneously obtaining the standard life cycle of each working part through the Internet;
S2, the cloud computing platform sequentially selects one data in the historical state data as main data and the other data as auxiliary data according to a plurality of pieces of historical state data pre-stored by each working part, and a life cycle judging network is established;
S3, the sensor devices of the working parts are in communication connection with the cloud computing platform, the cloud computing platform establishes a device topology network according to a communication connection result, and then the sensors acquire real-time state data of the working parts at the positions of the sensors;
S4, associating each topological node in the equipment topological network with a corresponding life cycle judging network, inputting real-time state data of each working part into the equipment topological network by the cloud computing platform, distributing the real-time state data by the equipment topological network through the life cycle judging network, and obtaining an actual life cycle variation according to a distribution result;
and S5, updating the current residual life cycle of the working part according to the actual life cycle variable quantity.
Further, the life cycle judging network comprises a feature extraction layer, a data association layer and a state evaluation layer;
The feature extraction layer is used for extracting corresponding feature data from each item of state data in the historical state data and establishing corresponding feature word vectors according to the feature data;
The data association layer is provided with a characteristic data frame, the characteristic data frame is provided with a central characteristic data layer and a peripheral characteristic data layer, and the central characteristic data layer and the peripheral characteristic data layer are used for distributing and arranging each characteristic word vector, wherein the peripheral characteristic data layer is provided with three independent areas;
the state evaluation layer is used for establishing a life cycle judgment network according to the arrangement condition of each feature word vector in the feature data frame.
Further, the process of establishing the feature word vector according to the historical state data comprises the following steps:
the cloud computing platform inputs all the historical state data into a life cycle judging network, and the life cycle judging network transfers the historical state data to a feature extraction layer;
The feature extraction layer is provided with a plurality of feature extraction pointers, and then the feature extraction layer sequentially traverses all the feature extraction pointers on each historical state data, and according to the characteristics of the traversing result, each feature extraction pointer obtains X parts of feature data from the historical state data, wherein X is a natural number larger than 0;
And establishing a high-dimensional vector space, mapping all feature data into the high-dimensional vector space by a feature extraction layer, and converting each feature data into a corresponding feature Word vector by a Word2Vec technology.
Further, the process of distributing and arranging the feature word vectors includes:
The feature extraction layer performs normalization operation on each feature word vector, so that each feature word vector is scaled to a unit length, and then the feature extraction layer sends all feature word vectors to the data association layer;
The data association layer classifies each feature word vector according to the state data types, and then sequentially selects all feature word vectors corresponding to the state data of one type to be mapped to the central feature data layer of the feature data frame, and maps the feature word vectors corresponding to the state data of other types to each region in the peripheral feature data layer.
Obtaining hamming distances between each feature word vector in each region and each feature word vector in the central feature data layer;
counting the total number of hamming distances between each feature word vector in the peripheral feature data layer and each feature word vector in the central feature data layer respectively, and marking the total number as a peripheral association value;
Arranging peripheral association values of the feature word vectors in the same region from high to low, and rearranging the feature word vectors in the region according to an arrangement result;
After rearranging the feature word vectors of the peripheral feature data layer, calculating the Hamming distance between the feature word vectors in the central feature data layer and the feature word vectors in the areas respectively, and counting the total Hamming distance number of the feature word vectors in the central feature data layer and marking the total Hamming distance number as an internal association value, and counting the total Hamming distance number of the feature word vectors in the central feature data layer to the areas and marking the total Hamming distance number as an area association value;
Arranging the internal association values from high to low, and further arranging each feature word vector in the central feature data layer from the edge position to the central position according to the arrangement sequence;
Meanwhile, an arrangement trend is set according to the area association value of each feature word vector in the central feature data layer, and the acquisition process of the arrangement trend comprises the following steps: Wherein/> Representing the arrangement trend of the feature word vector in the central feature data layer to the ith region, and F i represents the region association value of the feature word vector in the central feature data layer to the ith region, i=1, 2,3;
And then carrying out translation adjustment on each feature word vector in the arranged central feature data layer according to the arrangement trend.
Further, the process of establishing the life cycle judgment network according to the arrangement condition of the feature word vectors comprises the following steps:
The data association layer sends the four feature data frames after the arrangement is completed to the state evaluation layer, and according to the residual life cycle of the working part corresponding to each historical state data, the life cycle variable quantity between each feature word vector in the peripheral feature data layer and the central feature data layer is established by the machine learning collecting technology, so that a life cycle judgment network is obtained.
Further, the process of collecting the real-time status data includes:
the method comprises the steps that sensor devices of all working parts are in communication connection with a cloud computing platform, meanwhile, the cloud computing platform establishes equipment topology nodes according to the distribution of all the sensor devices, the equipment topology nodes are composed of a plurality of space nodes, the distribution of all the space nodes in an equipment topology network is mapped with the space of all the working parts in composite material equipment in equal proportion, meanwhile, the current residual life cycle of the corresponding working parts is marked on all the space nodes, and numbers are set for all the space nodes;
the cloud computing platform establishes a data transmission channel between each space node in the equipment topology network and a sensor device on a corresponding working part;
when the composite material equipment runs, each sensor device collects real-time state data of the working part where the sensor device is located, and the real-time state data is sent to the corresponding space node through the data transmission channel.
Further, the process of generating the actual life cycle variable according to the real-time status data includes:
The cloud computing platform sets the same number as the space nodes for each life cycle judging network, further judges the space nodes with the same number in the topology network of the network matching device according to the life cycle, and associates the life cycle judging network with the corresponding space nodes according to the matching result;
When the space node receives the real-time state data acquired by the sensor device, acquiring a method for establishing a life cycle judgment network, converting each real-time state data into corresponding real-time feature word vectors, sequentially selecting one real-time feature word vector of the category data to map to a central feature data layer, and mapping the real-time feature word vectors of the other three category data to each region of a peripheral feature data layer;
Matching the real-time feature word vector with the feature word vector in the life cycle judging network, mapping the real-time feature word vector to a corresponding position according to a matching result, and counting hamming distances between the real-time feature word vector in the peripheral feature data layer and each real-time feature word vector in the central feature data layer respectively;
selecting the real-time feature word vector with the largest Hamming distance between the real-time feature word vector and the real-time feature word vector in the central feature data layer as an associated vector, repeating the operation, further counting the life cycle variation of the real-time feature word vector in each region and the associated vector corresponding to the real-time feature word vector, and adding the life cycle variation of the three regions to obtain the total life cycle variation;
And obtaining the total life cycle variable quantity of each kind of data under the condition that the real-time feature word vector of each kind of data is mapped to the central feature data layer, adding the total life cycle variable quantity under the four conditions, and then taking an average value, and further recording the average value as the actual life cycle variable quantity of the corresponding working part.
Further, the process of updating the current remaining life cycle according to the actual life cycle variation comprises:
The cloud computing platform updates the current residual life cycle of the space node in the equipment topology network according to the actual life cycle variable quantity of the working part;
When the life cycle after the current update is less than or equal to 0, replacing corresponding working parts, resetting the current residual life cycle of the corresponding space node in the equipment topology network, and otherwise, not performing any operation.
Compared with the prior art, the invention has the beneficial effects that:
According to the invention, one data in the historical state data is sequentially selected as main data and the other data are auxiliary data through a plurality of pieces of historical state data pre-stored in each working part, a life cycle judging network and a device topology network are established, each topology node in the device topology network is associated with the corresponding life cycle judging network, the real-time state data of each working part are input into the device topology network, the actual life cycle variation is obtained, the current residual life cycle of the working part is updated according to the actual life cycle variation, and the accuracy of defect detection of the composite material device is effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments described in the present application.
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, based on the examples herein, which are within the scope of the invention as defined by the claims, will be within the scope of the invention as defined by the claims.
Example 1
As shown in fig. 1, a method for maintaining and managing composite material equipment based on a cloud platform comprises the following steps:
s1, setting a cloud computing platform, installing various sensor devices on each working part of composite material equipment, and simultaneously obtaining the standard life cycle of each working part through the Internet;
S2, the cloud computing platform sequentially selects one data in the historical state data as main data and the other data as auxiliary data according to a plurality of pieces of historical state data pre-stored by each working part, and a life cycle judging network is established;
S3, the sensor devices of the working parts are in communication connection with the cloud computing platform, the cloud computing platform establishes a device topology network according to a communication connection result, and then the sensors acquire real-time state data of the working parts at the positions of the sensors;
S4, associating each topological node in the equipment topological network with a corresponding life cycle judging network, inputting real-time state data of each working part into the equipment topological network by the cloud computing platform, distributing the real-time state data by the equipment topological network through the life cycle judging network, and obtaining an actual life cycle variation according to a distribution result;
and S5, updating the current residual life cycle of the working part according to the actual life cycle variable quantity.
Example 2
This embodiment is a further limitation of embodiment 1, said step S1 being implemented by:
Setting various sensor devices for each working part on the composite acquisition material equipment, wherein the types of the sensor devices comprise a temperature sensor, a zero-frequency acceleration sensor, a pressure sensor, a laser camera and the like;
the method comprises the steps that a cloud computing platform is arranged, and further the cloud computing platform obtains lives of all working parts through the Internet, and is used for receiving real-time state data collected by all sensor devices, establishing a device topology network and establishing a life cycle judging network according to historical state data.
Example 3
This embodiment is a further limitation of embodiment 1, said step S2 being implemented by:
The cloud computing platform pre-stores K historical state data of each working part, wherein the historical state data comprises a temperature change curve, a pressure change curve, laser image data, a surface vibration spectrum and a residual life cycle composition of the corresponding working part at the time of generating the historical state data of the working part in unit time;
the surface vibration spectrum represents vibration conditions of m positions of the surface of the corresponding working part in unit time, m data acquisition points are arranged on the working part through a zero-frequency acceleration sensor, tiny vibration occurs to the data acquisition points when the working part runs, the zero-frequency acceleration sensor generates corresponding electric signal output according to the vibration conditions of the data acquisition points, and further generates the surface vibration spectrum of the data acquisition points of the corresponding positions according to the electric signal, wherein K and m are natural numbers larger than 0.
Further, the cloud computing platform establishes a life cycle judging network, wherein the life cycle judging network comprises a feature extraction layer, a data association layer and a state evaluation layer;
The feature extraction layer is used for extracting corresponding feature data from each item of state data in the historical state data and establishing corresponding feature word vectors according to the feature data;
The data association layer is provided with a characteristic data frame, and the characteristic data frame is provided with a central characteristic data layer and a peripheral characteristic data layer, and is used for distributing and arranging each characteristic word vector;
It should be noted that, the peripheral feature data layer is provided with three independent areas;
The state evaluation layer is used for establishing a life cycle judgment network according to the arrangement condition of each characteristic word vector in the characteristic data frame;
the cloud computing platform inputs all the historical state data into a life cycle judging network, and the life cycle judging network transfers the historical state data to a feature extraction layer;
The feature extraction layer is provided with a plurality of feature extraction pointers, and then the feature extraction layer sequentially traverses all the feature extraction pointers on each historical state data, and according to the characteristics of the traversing result, each feature extraction pointer obtains X parts of feature data from the historical state data, wherein X is a natural number larger than 0;
the feature data includes data segments of various status data under normal conditions and abnormal conditions, for example, for a surface vibration spectrum, a feature extraction pointer is used as feature data under abnormal conditions from a data segment having a difference between the surface vibration spectrum and the data segment under normal conditions;
establishing a high-dimensional vector space, mapping all feature data into the high-dimensional vector space by a feature extraction layer, and converting each feature data into a corresponding feature Word vector by a Word2Vec technology;
It should be noted that the Word2Vec technique is a technique for representing words as vectors, and by mapping words to vector representations in a high-dimensional space, semantically similar words have similar vector representations.
Further, the feature extraction layer performs normalization operation on each feature word vector, so that each feature word vector is scaled to a unit length, and then the feature extraction layer sends all feature word vectors to the data association layer;
The data association layer classifies each feature word vector according to the state data types, and then sequentially selects all feature word vectors corresponding to the state data of one type to be mapped to the central feature data layer of the feature data frame, and maps the feature word vectors corresponding to the state data of other types to each region in the peripheral feature data layer.
Calculating hamming distances h between each feature word vector in each region and each feature word vector in the central feature data layer;
Counting the total number of hamming distances between each feature word vector in the peripheral feature data layer and each feature word vector in the central feature data layer, and marking as a peripheral association value H, wherein the calculation formula of the peripheral association value is as follows: Where Num represents the total number of hamming distances, h p represents the value of the p-th hamming distance;
Arranging peripheral association values of the feature word vectors in the same region from high to low, and rearranging the feature word vectors in the region according to an arrangement result;
After rearranging the feature word vectors of the peripheral feature data layer, calculating the Hamming distance between the feature word vectors in the central feature data layer and the feature word vectors in the areas respectively, and counting the total Hamming distance number of the feature word vectors in the central feature data layer and marking the total Hamming distance number as an internal association value, and counting the total Hamming distance number of the feature word vectors in the central feature data layer to the areas and marking the total Hamming distance number as an area association value;
Arranging the internal association values from high to low, and further arranging each feature word vector in the central feature data layer from the edge position to the central position according to the arrangement sequence;
Meanwhile, an arrangement trend is set according to the area association value of each feature word vector in the central feature data layer, and the acquisition process of the arrangement trend comprises the following steps: Wherein/> Representing the arrangement trend of the feature word vector in the central feature data layer to the ith region, and F i represents the region association value of the feature word vector in the central feature data layer to the ith region, i=1, 2,3;
And then carrying out translation adjustment on each feature word vector in the arranged central feature data layer according to the arrangement trend.
Further, the data association layer sends the four feature data frames after the arrangement is completed to the state evaluation layer, and according to the residual life cycle of the working part corresponding to each historical state data, the life cycle variable quantity between each feature word vector in the peripheral feature data layer and the central feature data layer is established by the machine learning collecting technology, so that a life cycle judgment network is obtained;
For example, the lifecycle variation between the feature word vector of one peripheral feature data layer and one feature word vector in the central feature data layer is 20 minutes, and the lifecycle variation between the feature word vector of the peripheral feature data layer and another feature word vector in the central feature data layer is 18 minutes, which means that the lifecycle variation between each feature word vector in the peripheral feature data layer and the associated one feature word vector in the central feature data layer is different.
Example 4
This embodiment is a further limitation of embodiment 1, said step S3 being implemented by:
The sensor devices of the working parts are in communication connection with the cloud computing platform, and the cloud computing platform sets numbers for the sensors according to the communication connection result, wherein the numbers can be H 1,1、H1,2、……、Ha,j, H a,j represents the number of the jth sensor device of the a-th working part, a is a natural number larger than 0, and j=1, 2,3 and 4;
Meanwhile, the cloud computing platform establishes equipment topology nodes according to the distribution of each sensor device, wherein the equipment topology nodes consist of a space nodes, the distribution of each space node in an equipment topology network is mapped with the space of each working part in the composite material equipment in equal proportion, each space node is marked with the current residual life cycle of the corresponding working part, and a number H 1、H2、……、Ha is set for each space node;
the cloud computing platform establishes a data transmission channel between each space node in the equipment topology network and a sensor device on a corresponding working part;
and each sensor device acquires real-time state data of the working part where the sensor device is positioned when the composite material equipment runs, and the real-time state data is sent to the corresponding space node through the data transmission channel.
Example 5
This embodiment is a further limitation of embodiment 1, said step S4 being implemented by:
the cloud computing platform sets the same number H 1、H2、……、Ha as each space node for each life cycle judging network, further judges the space nodes with the same number in the topology network of the network matching device according to the life cycle, and associates the life cycle judging network with the corresponding space nodes according to the matching result;
When the space node receives the real-time state data acquired by the sensor device, acquiring a method for establishing a life cycle judgment network, converting each real-time state data into corresponding real-time feature word vectors, sequentially selecting one real-time feature word vector of the category data to map to a central feature data layer, and mapping the real-time feature word vectors of the other three category data to each region of a peripheral feature data layer;
Because the data sources and the data processing modes of the real-time state data and the historical state data are the same, the real-time characteristic word vector generated according to the real-time state data and the characteristic word vector generated according to the historical state data have the tendency;
Matching the real-time feature word vector with the feature word vector in the life cycle judging network, mapping the real-time feature word vector to a corresponding position according to a matching result, and counting hamming distances between the real-time feature word vector in the peripheral feature data layer and each real-time feature word vector in the central feature data layer respectively;
selecting the real-time feature word vector with the largest Hamming distance between the real-time feature word vector and the real-time feature word vector in the central feature data layer as an associated vector, repeating the operation, further counting the life cycle variation of the real-time feature word vector in each region and the associated vector corresponding to the real-time feature word vector, and adding the life cycle variation of the three regions to obtain the total life cycle variation;
And obtaining the total life cycle variable quantity of each kind of data under the condition that the real-time feature word vector of each kind of data is mapped to the central feature data layer, adding the total life cycle variable quantity under the four conditions, and then taking an average value, and further recording the average value as the actual life cycle variable quantity of the corresponding working part.
Example 6
This embodiment is a further limitation of embodiment 1, said step S5 being implemented by:
the cloud computing platform updates the current residual life cycle of the space nodes in the equipment topological network according to the actual life cycle variable quantity of the working parts, wherein the updating formula is as follows: t Updating =T Residual of -T variation of ; wherein T Updating 、T Residual of 、T variation of represents the updated life cycle, the current remaining life cycle and the actual life cycle change rate, respectively;
When the life cycle after the current update is less than or equal to 0, replacing corresponding working parts, resetting the current residual life cycle of the corresponding space node in the equipment topology network, and otherwise, not performing any operation.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (5)
1. The composite material equipment maintenance management method based on the cloud platform is characterized by comprising the following steps of:
s1, setting a cloud computing platform, installing various sensor devices on each working part of composite material equipment, and simultaneously obtaining the standard life cycle of each working part through the Internet;
S2, the cloud computing platform sequentially selects one data in the historical state data as main data and the other data as auxiliary data according to a plurality of pieces of historical state data pre-stored by each working part, and a life cycle judging network is established;
the life cycle judging network comprises a feature extraction layer, a data association layer and a state evaluation layer;
The feature extraction layer is used for extracting corresponding feature data from each item of state data in the historical state data and establishing corresponding feature word vectors according to the feature data;
The data association layer is provided with a characteristic data frame, the characteristic data frame is provided with a central characteristic data layer and a peripheral characteristic data layer, and the central characteristic data layer and the peripheral characteristic data layer are used for distributing and arranging each characteristic word vector, wherein the peripheral characteristic data layer is provided with three independent areas;
The state evaluation layer is used for establishing a life cycle judgment network according to the arrangement condition of each characteristic word vector in the characteristic data frame;
the process of establishing the feature word vector according to the historical state data comprises the following steps:
the cloud computing platform inputs all the historical state data into a life cycle judging network, and the life cycle judging network transfers the historical state data to a feature extraction layer;
The feature extraction layer is provided with a plurality of feature extraction pointers, all feature extraction pointers are further called to sequentially traverse each historical state data, and each feature extraction pointer obtains X parts of feature data from the historical state data according to the characteristics of the traversing result, wherein X is a natural number larger than 0;
establishing a high-dimensional vector space, mapping all feature data into the high-dimensional vector space by a feature extraction layer, and converting each feature data into a corresponding feature Word vector by a Word2Vec technology;
S3, the sensor devices of the working parts are in communication connection with the cloud computing platform, the cloud computing platform establishes a device topology network according to a communication connection result, and then the sensors acquire real-time state data of the working parts at the positions of the sensors;
S4, associating each topological node in the equipment topological network with a corresponding life cycle judging network, inputting real-time state data of each working part into the equipment topological network by the cloud computing platform, distributing the real-time state data by the equipment topological network through the life cycle judging network, and obtaining an actual life cycle variation according to a distribution result;
the process of generating the actual life cycle variable according to the real-time state data comprises the following steps:
The cloud computing platform sets the same number as the space nodes for each life cycle judging network, further judges the space nodes with the same number in the topology network of the network matching device according to the life cycle, and associates the life cycle judging network with the corresponding space nodes according to the matching result;
When the space node receives the real-time state data acquired by the sensor device, acquiring a method for establishing a life cycle judgment network, converting each real-time state data into corresponding real-time feature word vectors, sequentially selecting one real-time feature word vector of the category data to map to a central feature data layer, and mapping the real-time feature word vectors of the other three category data to each region of a peripheral feature data layer;
Matching the real-time feature word vector with the feature word vector in the life cycle judging network, mapping the real-time feature word vector to a corresponding position according to a matching result, and counting hamming distances between the real-time feature word vector in the peripheral feature data layer and each real-time feature word vector in the central feature data layer respectively;
The method comprises the steps of selecting real-time feature word vectors in a peripheral feature data layer, marking the real-time feature word vector with the largest Hamming distance between the real-time feature word vectors in a central feature data layer as an associated vector, further counting the life cycle variation of the real-time feature word vector in each region and the associated vector corresponding to the real-time feature word vector, and adding the life cycle variation of the three regions to obtain a total life cycle variation;
Acquiring total life cycle variable quantities of various types of data under the condition that the real-time feature word vectors of the various types of data are mapped to the central feature data layer, adding the total life cycle variable quantities under the four conditions, and then taking an average value, and further recording the average value as an actual life cycle variable quantity of a corresponding working part;
and S5, updating the current residual life cycle of the working part according to the actual life cycle variable quantity.
2. The cloud platform-based composite equipment maintenance management method of claim 1, wherein the process of distributing and arranging the feature word vectors comprises:
The feature extraction layer performs normalization operation on each feature word vector, so that each feature word vector is scaled to a unit length, and then the feature extraction layer sends all feature word vectors to the data association layer;
The data association layer classifies each feature word vector according to the state data types, and then sequentially selects all feature word vectors corresponding to the state data of one type to be mapped to a central feature data layer of a feature data frame, and maps feature word vectors corresponding to the state data of other types to each region in a peripheral feature data layer;
Obtaining hamming distances between each feature word vector in each region and each feature word vector in the central feature data layer;
counting the total number of hamming distances between each feature word vector in the peripheral feature data layer and each feature word vector in the central feature data layer respectively, and marking the total number as a peripheral association value;
Arranging peripheral association values of the feature word vectors in the same region from high to low, and rearranging the feature word vectors in the region according to an arrangement result;
After rearranging the feature word vectors of the peripheral feature data layer, calculating the Hamming distance between the feature word vectors in the central feature data layer and the feature word vectors in the areas respectively, and counting the total Hamming distance number of the feature word vectors in the central feature data layer and marking the total Hamming distance number as an internal association value, and counting the total Hamming distance number of the feature word vectors in the central feature data layer to the areas and marking the total Hamming distance number as an area association value;
Arranging the internal association values from high to low, and further arranging each feature word vector in the central feature data layer from the edge position to the central position according to the arrangement sequence;
Meanwhile, an arrangement trend is set according to the area association value of each feature word vector in the central feature data layer, and the acquisition process of the arrangement trend comprises the following steps: Wherein/> Representing the arrangement trend of the feature word vector in the central feature data layer to the ith region, and F i represents the region association value of the feature word vector in the central feature data layer to the ith region, i=1, 2,3;
And then carrying out translation adjustment on each feature word vector in the arranged central feature data layer according to the arrangement trend.
3. The cloud platform-based composite equipment maintenance management method according to claim 2, wherein the process of establishing the life cycle judgment network according to the arrangement condition of the feature word vectors comprises the steps of:
The data association layer sends the four feature data frames after the arrangement is completed to the state evaluation layer, and according to the residual life cycle of the working part corresponding to each historical state data, the life cycle variable quantity between each feature word vector in the peripheral feature data layer and the central feature data layer is established by the machine learning collecting technology, so that a life cycle judgment network is obtained.
4. The cloud platform-based composite equipment maintenance management method according to claim 3, wherein the process of collecting the real-time status data comprises the following steps:
the method comprises the steps that sensor devices of all working parts are in communication connection with a cloud computing platform, meanwhile, the cloud computing platform establishes equipment topology nodes according to the distribution of all the sensor devices, the equipment topology nodes are composed of a plurality of space nodes, the distribution of all the space nodes in an equipment topology network is mapped with the space of all the working parts in composite material equipment in equal proportion, meanwhile, the current residual life cycle of the corresponding working parts is marked on all the space nodes, and numbers are set for all the space nodes;
the cloud computing platform establishes a data transmission channel between each space node in the equipment topology network and a sensor device on a corresponding working part;
when the composite material equipment runs, each sensor device collects real-time state data of the working part where the sensor device is located, and the real-time state data is sent to the corresponding space node through the data transmission channel.
5. The cloud platform based composite equipment maintenance management method of claim 4, wherein updating the current remaining lifecycle according to the actual lifecycle variation comprises:
The cloud computing platform updates the current residual life cycle of the space node in the equipment topology network according to the actual life cycle variable quantity of the working part;
When the life cycle after the current update is less than or equal to 0, replacing corresponding working parts, resetting the current residual life cycle of the corresponding space node in the equipment topology network, and otherwise, not performing any operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410240194.6A CN117829821B (en) | 2024-03-04 | 2024-03-04 | Cloud platform-based composite material equipment maintenance and management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410240194.6A CN117829821B (en) | 2024-03-04 | 2024-03-04 | Cloud platform-based composite material equipment maintenance and management method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117829821A CN117829821A (en) | 2024-04-05 |
| CN117829821B true CN117829821B (en) | 2024-05-31 |
Family
ID=90515744
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410240194.6A Active CN117829821B (en) | 2024-03-04 | 2024-03-04 | Cloud platform-based composite material equipment maintenance and management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117829821B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102590652A (en) * | 2012-01-10 | 2012-07-18 | 山东电力研究院 | Electric-information-based equipment performance evaluation system and method |
| CN114386774A (en) * | 2021-12-21 | 2022-04-22 | 中国中煤能源集团有限公司 | CPIM-based three-dimensional visual full-life-cycle management platform for coal preparation plant |
| CN115601007A (en) * | 2022-09-28 | 2023-01-13 | 北京九天翱翔科技有限公司（Cn） | Diagnosis reasoning and maintenance decision method for complete life cycle of complex electromechanical system |
| CN116934064A (en) * | 2023-06-16 | 2023-10-24 | 国网浙江省电力有限公司象山县供电公司 | Equipment node data monitoring method based on power topology system |
| CN117493498A (en) * | 2024-01-02 | 2024-02-02 | 陕西博古德智能科技有限公司 | Electric power data mining and analysis system based on industrial Internet |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008007160A2 (en) * | 2006-07-11 | 2008-01-17 | Abb Research Ltd. | A life cycle management system for intelligent electronic devices |
-
2024
- 2024-03-04 CN CN202410240194.6A patent/CN117829821B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102590652A (en) * | 2012-01-10 | 2012-07-18 | 山东电力研究院 | Electric-information-based equipment performance evaluation system and method |
| CN114386774A (en) * | 2021-12-21 | 2022-04-22 | 中国中煤能源集团有限公司 | CPIM-based three-dimensional visual full-life-cycle management platform for coal preparation plant |
| CN115601007A (en) * | 2022-09-28 | 2023-01-13 | 北京九天翱翔科技有限公司（Cn） | Diagnosis reasoning and maintenance decision method for complete life cycle of complex electromechanical system |
| CN116934064A (en) * | 2023-06-16 | 2023-10-24 | 国网浙江省电力有限公司象山县供电公司 | Equipment node data monitoring method based on power topology system |
| CN117493498A (en) * | 2024-01-02 | 2024-02-02 | 陕西博古德智能科技有限公司 | Electric power data mining and analysis system based on industrial Internet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117829821A (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111950627B (en) | Multi-source information fusion method and application thereof | |
| CN112765768B (en) | Discrete workshop digital traceability method based on Internet of things | |
| CN109086470A (en) | A kind of method for diagnosing faults based on fuzzy preference relation and D-S evidence theory | |
| CN110596506A (en) | Converter fault diagnosis method based on time convolution network | |
| CN110889091A (en) | Machine tool thermal error prediction method and system based on temperature sensitive interval segmentation | |
| CN111652835A (en) | Method for detecting insulator loss of power transmission line based on deep learning and clustering | |
| CN105588684B (en) | A kind of electronic scanner pressure measuring system automatic fault diagnosis inspection method | |
| CN117689256B (en) | Aluminum alloy casting product quality tracing method | |
| CN106595788B (en) | Based on the modified large pumping station flow monitoring method of Multi-parameter coupling | |
| CN115993531A (en) | Permanent magnet synchronous motor double closed loop fault prediction and health management method and device | |
| CN110197199B (en) | Embedded DCNN and edge calculation cracking furnace heavy tube temperature identification method | |
| CN117829821B (en) | Cloud platform-based composite material equipment maintenance and management method | |
| CN115329663A (en) | Key feature selection method and device for processing power load monitoring sparse data | |
| CN112884348B (en) | Method for diagnosing production deviation source of space detonator based on dynamic Bayesian network | |
| CN112766687B (en) | Aircraft final assembly material distribution priority distribution method based on depth CNN | |
| CN110705132B (en) | Guidance control system performance fusion evaluation method based on multi-source heterogeneous data | |
| CN110222814B (en) | Ethylene cracking furnace tube heavy tube identification method based on embedded DCNN | |
| JP6950647B2 (en) | Data determination device, method, and program | |
| CN116362036B (en) | High-speed rail equipment fault early warning method and system | |
| CN115762558B (en) | Performance detection system and method for escalator production | |
| CN116957361A (en) | Ship task system health state detection method based on virtual-real combination | |
| CN116258904A (en) | Depth order classification anomaly detection method and device based on contrast learning optimization | |
| CN114818945A (en) | Small sample image classification method and device integrating category adaptive metric learning | |
| CN115169735A (en) | Electroceramic quality prediction method and system based on sample expansion | |
| CN109753018B (en) | Error compensation system and dynamic compensation method based on cloud intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |