CN114997513A - Predictive equipment maintenance method based on neural network - Google Patents

Predictive equipment maintenance method based on neural network Download PDF

Info

Publication number
CN114997513A
CN114997513A CN202210741067.5A CN202210741067A CN114997513A CN 114997513 A CN114997513 A CN 114997513A CN 202210741067 A CN202210741067 A CN 202210741067A CN 114997513 A CN114997513 A CN 114997513A
Authority
CN
China
Prior art keywords
data
representing
training
neural network
equipment
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210741067.5A
Other languages
Chinese (zh)
Inventor
胡增
江大白
汪刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Applied Technology Co Ltd
Original Assignee
China Applied Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Applied Technology Co Ltd filed Critical China Applied Technology Co Ltd
Priority to CN202210741067.5A priority Critical patent/CN114997513A/en
Publication of CN114997513A publication Critical patent/CN114997513A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/04—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
    • Y02P90/80—Management or planning

Abstract

The invention discloses a predictive maintenance method of equipment based on a neural network, which comprises the following steps of S1: acquiring equipment operation data by using an intelligent sensor through a two-point method, wherein the two points are time points of the beginning and the end of unit time; step S2: preprocessing and screening the data collected in unit time, and deleting abnormal data generated by abnormal operation of non-equipment; step S3: classifying the preprocessed data by using a classification function to obtain a training set and a test set; step S4: training the data by using a neural network; step S5: calculating a training error of the data, the error result being a description of the accuracy of the data training; step S6: the device predictive maintenance model is constructed by utilizing the radial basis function, the accuracy of the model on the device state prediction is high, manual re-inspection is not needed, the cost is reduced, and the automation degree of industrial devices is increased.

Description

Predictive equipment maintenance method based on neural network
Technical Field
The invention relates to the field of equipment maintenance, in particular to a neural network-based equipment predictive maintenance method.
Background
The existing factory equipment is mainly finished through manual inspection or planned maintenance, but the judgment by manpower is often unreliable, and once people forget to inspect or do not check in place, equipment can be shut down, so that great economic loss is brought.
Application number CN201911230761.5 discloses a cloud edge coordination-based industrial equipment predictive maintenance method, which includes the following steps: s1: the heterogeneous sensing equipment acquires equipment state data of the industrial equipment and sends the equipment state data to the edge computing platform; s2: an edge data management module of an edge computing platform obtains characteristic data required by a target device prediction task according to data uploaded by heterogeneous sensing devices; a configuration overloading module of the prediction service orchestrator acquires equipment state prediction model configuration of a cloud computing platform and equipment state prediction model configuration trained by an edge model training module, a model operation module loads a latest equipment state prediction model of target equipment, and extracted feature data is input; judging whether the target equipment has a fault risk or not according to the output data of the model operation module, if so, entering the next step, and if not, returning to the step S1; s3: and a trigger management module of the edge computing platform informs the fault early warning information of the designated responsible person according to a preset trigger. The invention can realize accurate and efficient predictive maintenance of industrial equipment.
The application number is CN201910466181.X discloses a mechanical equipment predictive maintenance method combining edge calculation and digital twins, intelligent service is provided through an edge calculation side, data are directly analyzed, the analysis efficiency is improved, the data transmission flow between an equipment terminal and a cloud center is reduced, the pressure of monitoring data analysis is relieved, and the response capability of the service is enhanced; meanwhile, the data security is improved by the distributed processing and the local storage of the data, and the data obtained by the edge calculation is locally stored; the traditional digital twin body construction based on cloud storage has time delay, particularly superwrite real simulation, real-time online adjustment cannot be completed, edge calculation is introduced to reduce time delay, edge data is input to terminal equipment, digital twin body construction and high-fidelity behavior simulation are carried out, visualization of the real-time state of mechanical equipment is achieved, quantitative and qualitative analysis is carried out by combining a constructed neural network prediction result and a virtual space high-fidelity behavior simulation result, and predictive maintenance of the mechanical equipment is comprehensively guided.
However, the model used in the predictive maintenance of the equipment in the existing method has low accuracy in predicting the state of the equipment, so that any manual check is required, and the cost is increased.
Disclosure of Invention
In order to overcome the defects and shortcomings of the prior art, the invention provides a predictive maintenance method of equipment based on a neural network.
The technical scheme adopted by the invention is that the equipment predictive maintenance method based on the neural network comprises the following steps:
step S1: acquiring equipment operation data by using an intelligent sensor through a two-point method, wherein the two points are time points of the beginning and the end of unit time;
step S2: preprocessing and screening the data collected in unit time, and deleting abnormal data generated by abnormal operation of non-equipment;
step S3: classifying the preprocessed data by using a classification function to obtain a training set and a test set;
step S4: training the data by using a neural network;
step S5: calculating a training error of the data, the error result being a description of the accuracy of the data training;
step S6: and constructing a predictive maintenance model of the equipment by using the radial basis functions.
Further, the expression of the two-point method is as follows:
Figure BDA0003715351440000031
wherein N is 1 And N 2 Respectively representing the start point and the end point of the data acquisition of the equipment in a unit time period, n 1s Is shown at the beginningCollection of point-collected data, n 2s Representing a collection of data acquired at a starting point, n 1s -n 2s Representing the running error of the same data of the equipment in unit time, v representing the type of data collected by the sensor, G H (N 1 ，N 2 ) Representing the collection of the device data set.
Further, the preprocessing and screening are expressed as follows:
Figure BDA0003715351440000032
wherein u (v) represents a data set after screening, u (v-1) represents a data set before screening, x sensor, v represents the type of data collected by the sensor, and z 1 ≠z o-1 Indicating a jump in the acquired data, z 1 ＝z o-1 Indicates that the collected data has not jumped, u max Maximum value, u, representing variation of data min Denotes the minimum value of the data change, and u denotes the total amount of data.
Further, the classification function has the expression:
q ok ＝l ok +f 1 ×rand()×(W ok -d ok )+f 2 ×rand()×(W hk -d ok )
wherein q is ok Representing a data classification function,/ ok Boundary constant, f, representing data classification 1 Representing the classification coefficients of the training set, rand () representing a random function, d ok Representing an unclassified data set, W ok Representing the amount of data already in the training set, f 2 Class coefficient, W, representing test set hk Indicating the amount of data already in the test set.
Further, the data is trained by using a neural network, and the expression is as follows:
Figure BDA0003715351440000041
wherein the content of the first and second substances,
Figure BDA0003715351440000042
representing the input of data at the next round of neural network training,
Figure BDA0003715351440000043
representing the output of data at the next round of neural network training,
Figure BDA0003715351440000044
representing the input of data during the previous neural network training round, ξ representing the input coefficient, b 1 Representing a matrix of input variables, g 1 Representing the input weight, d xy Representing the magnitude of the variable of the input, b 2 Representing the output variable matrix, g 2 Representing the output weight, d ly Representing the magnitude of the variable output.
The training error is expressed as:
Figure BDA0003715351440000045
wherein T represents a training error value, D n Means, D, representing data after training max Represents the maximum value of the trained data, D min Represents the minimum of the trained data.
Further, the radial basis function has the expression:
Figure BDA0003715351440000046
wherein, M (x) p ，x q ) Representing radial basis functions, x p Indicating the past normal operating state of the apparatus, x q Indicating a past abnormal operation state of the equipment,
Figure BDA0003715351440000047
representing the constant of the device in the radial basis function.
Further, the predictive maintenance model of the equipment has the expression:
Figure BDA0003715351440000048
wherein j (x) represents the result of predicting whether maintenance is required for the equipment, M (x) p ，x q ) Representing radial basis functions, k p Which is indicative of a varying parameter of the device,
Figure BDA0003715351440000051
the total number of parameters expressed, p ═ 1, represents the 1 st iteration, l represents the total number of iterative summations, and v represents the identity matrix affecting plant maintenance.
Has the advantages that:
the equipment predictive maintenance method based on the neural network, provided by the invention, has the advantages that the equipment data in unit time are collected by a two-point method, so that the data collection is more accurate, the collected data are preprocessed and screened, abnormal data which is not generated by abnormal operation of the equipment is deleted and then classified and trained, and an equipment predictive maintenance model is constructed, and the model has high accuracy in equipment state prediction, does not need to be manually checked again, reduces the cost, increases the automation degree of industrial equipment, and can be popularized and applied in a large range.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of the overall steps of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments can be combined with each other without conflict, and the present application will be further described in detail with reference to the drawings and specific embodiments.
As shown in fig. 1, a method for predictive maintenance of a neural network-based device, the method comprising the steps of:
step S1: acquiring equipment operation data by using an intelligent sensor through a two-point method, wherein the two points are time points of the beginning and the end of unit time;
the intelligent sensor selects sensors with different types, sizes and dimensions according to different monitoring equipment, and completes data packaging and unpacking by adopting a custom communication protocol mode according to the communication requirements of different types of sensor data. The transmission data packet in the protocol consists of a message header and a message body, wherein the message header consists of an initial separator, a message type, a message length and a CRC (cyclic redundancy check) code; the message body is data information acquired by the sensing layer. When receiving data, firstly, a message header of a data packet needs to be completely read, information of the data packet to be received is judged according to the data length, the type and other related information in the message header, then, the data packet is circularly received until the received data length is equal to the data length in the message header, and finally, the complete data packet is obtained. The application layer data receiving end adopts a STREAM SOCKET (SOCK _ STREAM) mechanism, after an embedded mainboard network module sends an access request, the embedded mainboard and a data server are respectively used as a client and a server by utilizing the SOCKET technology, corresponding ports and IP addresses are configured, data connection is established after 3 times of handshake, and the safety and the stability of data transmission are ensured.
The data collected by the two-point method can intensively display the operation condition of the equipment in unit time, and the unit time is different according to different equipment and can be different units such as seconds, minutes, hours, days, months and the like. But the normal operation of the equipment is ensured as much as possible when data acquisition in unit time is carried out, so that the acquired data is stable as much as possible, and the workload is reduced for screening abnormal data in subsequent steps.
Step S2: preprocessing and screening the data collected in unit time, and deleting abnormal data generated by abnormal operation of non-equipment;
the preprocessing screening in the invention refers to screening data generated by abnormal operation of the equipment, such as: abnormal data generated by sudden abnormal operation of parts of equipment comprises: the purpose of the preprocessing screening is to extract data generated by the equipment under the overall normal condition as much as possible.
Step S3: classifying the preprocessed data by using a classification function to obtain a training set and a test set;
the training set is used for fitting a model, training a classification model by setting parameters of the classifier, and selecting different values of the same parameter to fit a plurality of classifiers when subsequently combining the functions of the verification set; after the test set obtains the optimal model through the training set and the verification set, the test set is used for model prediction to measure the performance and classification capability of the optimal model, namely, the test set can be used as a data set which never exists, and after model parameters are determined, the test set is used for model performance evaluation.
Step S4: training the data by using a neural network;
the step is a continuous iteration process, data are continuously trained by continuously taking the output of the previous round as the input of the next round, and the more rounds are trained, the more accurate the training result is obtained.
Step S5: calculating a training error of the data, the error result being a description of the accuracy of the data training;
the training error is related to the mean of the data, the maximum of the trained data, and the minimum of the trained data.
Step S6: and constructing a predictive maintenance model of the equipment by using the radial basis functions.
The prediction of the unit time of the equipment by the predictive maintenance model is the same as the unit time proposed in step S1.
Two-point method, the expression is:
Figure BDA0003715351440000081
wherein N is 1 And N 2 Respectively representing the starting point and the ending point of the data acquisition of the equipment in a unit time period, n 1s Representing a collection of data acquired at a starting point, n 2s Representing a collection of data acquired at a starting point, n 1s -n 2s Representing the running error of the same data of the equipment in unit time, v representing the type of data collected by the sensor, G H (N 1 ，N 2 ) Indicating that a device data set was acquired.
Preprocessing and screening, wherein the expression is as follows:
Figure BDA0003715351440000082
wherein u (v) represents a data set after screening, u (v-1) represents a data set before screening, x sensor, v represents the type of data collected by the sensor, and z 1 ≠z o-1 Indicating a jump in the acquired data, z 1 ＝z o-1 Indicates that the collected data has not jump, u max Maximum value, u, representing variation of data min Represents the minimum value of the data change, and u represents the total amount of data.
The classification function, the expression is:
q ok ＝l ok +f 1 ×rand()×(W ok -d ok )+f 2 ×rand()×(W hk -d ok )
wherein q is ok Representing a data classification function,/ ok Boundary constant, f, representing data classification 1 Representing the classification coefficients of the training set, rand () representing a random function, d ok Representing an unclassified data set, W ok Representing the amount of data already in the training set, f 2 Class coefficient, W, representing test set hk Indicating the amount of data already in the test set.
Training data by using a neural network, wherein an expression is as follows:
Figure BDA0003715351440000091
wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0003715351440000092
representing the input of data at the next round of neural network training,
Figure BDA0003715351440000093
representing the output of data at the next round of neural network training,
Figure BDA0003715351440000094
representing the input of data during the previous round of neural network training, xi representing the input coefficient, b 1 Representing a matrix of input variables, g 1 Representing input weights, d xy Representing the magnitude of the variable of the input, b 2 Representing a matrix of output variables, g 2 Representing the output weight, d ly Representing the magnitude of the variable output.
Training error, the expression is:
Figure BDA0003715351440000095
wherein T represents a training error value, D n Mean, D, representing data after training max Represents the maximum value of the trained data, D min Represents the minimum of the trained data.
Radial basis function, the expression is:
Figure BDA0003715351440000096
wherein, M (x) p ，x q ) Representing radial basis functions, x p Indicating the past normal operating state of the apparatus, x q Indicating a past abnormal operation state of the equipment,
Figure BDA0003715351440000097
representing the apparatus in radial basisConstant in number.
The equipment predictive maintenance model comprises the following expression:
Figure BDA0003715351440000098
wherein j (x) represents the result of predicting whether maintenance is required for the equipment, M (x) p ，x q ) Denotes the radial basis function, k p Which is indicative of a varying parameter of the device,
Figure BDA0003715351440000099
the total number of parameters expressed, p ═ 1, represents the 1 st iteration, l represents the total number of iterative summations, and v represents the identity matrix affecting plant maintenance.
The equipment predictive maintenance method based on the neural network provided by the invention has the advantages that the equipment data in unit time are collected by a two-point method, so that the data are more accurately collected, the collected data are preprocessed and screened, abnormal data which is not generated by abnormal operation of the equipment is deleted and then classified and trained, an equipment predictive maintenance model is constructed, the model has high accuracy in equipment state prediction, manual re-inspection is not needed, the cost is reduced, the automation degree of industrial equipment is increased, and the method can be widely popularized and applied.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made herein without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims (7)

1. A method for predictive maintenance of a device based on a neural network, the method comprising the steps of:
step S1: acquiring equipment operation data by using an intelligent sensor through a two-point method, wherein the two points are time points of the beginning and the end of unit time;
step S2: preprocessing and screening the data collected in unit time, and deleting abnormal data generated by abnormal operation of non-equipment;
step S3: classifying the preprocessed data by using a classification function to obtain a training set and a test set;
step S4: training the data by using a neural network;
step S5: calculating a training error of the data, the error result being a description of the accuracy of the data training;
step S6: and constructing a predictive maintenance model of the equipment by using the radial basis functions.
2. The neural network-based predictive maintenance method for devices of claim 1, wherein said two-point method is expressed as:
Figure FDA0003715351430000011
wherein N is 1 And N 2 Respectively representing the starting point and the ending point of the data acquisition of the equipment in a unit time period, n 1s Representing a collection of data acquired at a starting point, n 2s Representing a collection of data acquired at a starting point, n 1s -n 2s The running error of the same kind of data in unit time of the equipment is represented, v represents the type of data collected by the sensor, G H (N 1 ，N 2 ) Indicating that a device data set was acquired.
3. The method of claim 1, wherein the preprocessing screen is expressed as:
Figure FDA0003715351430000021
wherein u (v) represents the data set after screening, u (v-1) represents the data set before screening, x sensor, and v represents the data collected by the sensorClass, z 1 ≠z o-1 Indicating a jump in the acquired data, z 1 ＝z o-1 Indicates that the collected data has not jump, u max Maximum value, u, representing variation of data min Represents the minimum value of the data change, and u represents the total amount of data.
4. The neural network-based predictive maintenance method for devices as claimed in claim 1, wherein said classification function is expressed as:
q ok ＝l ok +f 1 ×rand( )×(W ok -d ok )+f 2 ×rand( )×(W hk -d ok )
wherein q is ok Representing data classification functions,/ ok Boundary constant, f, representing data classification 1 Representing the classification coefficients of the training set, rand () representing a random function, d ok Representing an unclassified data set, W ok Representing the amount of data already in the training set, f 2 Class coefficient, W, representing test set hk Indicating the amount of data already in the test set.
5. The method for predictive maintenance of neural network-based devices of claim 1, wherein said training of data using a neural network is expressed as:
Figure FDA0003715351430000022
wherein the content of the first and second substances,
Figure FDA0003715351430000023
representing the input of data at the next round of neural network training,
Figure FDA0003715351430000024
representing the output of data at the next round of neural network training,
Figure FDA0003715351430000025
representing the input of data during the previous round of neural network training, xi representing the input coefficient, b 1 Representing a matrix of input variables, g 1 Representing input weights, d xy Representing the magnitude of the variable of the input, b 2 Representing a matrix of output variables, g 2 Representing the output weight, d ly A variable magnitude representing an output;
the training error is expressed as:
Figure FDA0003715351430000031
wherein T represents a training error value, D n Means, D, representing data after training max Represents the maximum value of the trained data, D min Represents the minimum of the trained data.
6. The method for predictive maintenance of neural network-based devices, as claimed in claim 1, wherein said radial basis functions are expressed as:
Figure FDA0003715351430000032
wherein, M (x) p ，x q ) Representing radial basis functions, x p Indicating the normal operating state of the plant, x q Indicating a past abnormal operation state of the equipment,
Figure FDA0003715351430000033
representing the constant of the device in the radial basis function.
7. The neural network-based device predictive maintenance method of claim 1, wherein the device predictive maintenance model is expressed as:
Figure FDA0003715351430000034
wherein j (x) represents the result of predicting whether maintenance is required for the equipment, and M (x) p ，x q ) Representing radial basis functions, k p Which is indicative of a varying parameter of the device,
Figure FDA0003715351430000035
the total number of parameters expressed, p ═ 1, represents the 1 st iteration, l represents the total number of iterative summations, and v represents the identity matrix affecting plant maintenance.
CN202210741067.5A 2022-06-27 2022-06-27 Predictive equipment maintenance method based on neural network Pending CN114997513A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210741067.5A CN114997513A (en) 2022-06-27 2022-06-27 Predictive equipment maintenance method based on neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210741067.5A CN114997513A (en) 2022-06-27 2022-06-27 Predictive equipment maintenance method based on neural network

Publications (1)

Publication Number Publication Date
CN114997513A true CN114997513A (en) 2022-09-02

Family

ID=83036192

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210741067.5A Pending CN114997513A (en) 2022-06-27 2022-06-27 Predictive equipment maintenance method based on neural network

Country Status (1)

Country Link
CN (1) CN114997513A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116463878A (en) * 2023-04-01 2023-07-21 维达护理用品(广东)有限公司 Double wrinkling device for toilet paper production
TWI814661B (en) * 2022-12-09 2023-09-01 明志科技大學 Method for discriminating pipeline leak and system thereof

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI814661B (en) * 2022-12-09 2023-09-01 明志科技大學 Method for discriminating pipeline leak and system thereof
CN116463878A (en) * 2023-04-01 2023-07-21 维达护理用品(广东)有限公司 Double wrinkling device for toilet paper production
CN116463878B (en) * 2023-04-01 2024-03-19 维达护理用品(广东)有限公司 Double wrinkling device for toilet paper production

Similar Documents

Publication Publication Date Title
CN114997513A (en) Predictive equipment maintenance method based on neural network
CN103246265B (en) Electromechanical equipment detection maintaining method
CN111928888A (en) Intelligent monitoring and analyzing method and system for water pollution
CN111461555A (en) Production line quality monitoring method, device and system
US20060190960A1 (en) System and method for incorporating video analytics in a monitoring network
CN110929934A (en) Equipment failure prediction method and device, computer equipment and storage medium
CN113676534B (en) Bridge data uploading method based on edge calculation
KR101938946B1 (en) Interface Middleware System for Collecting Mass Data and Method for Collecting Data of Interface Middleware System
TWI662424B (en) Selection method of leading auxiliary parameters and method for pre-diagnosis of equipment maintenance by combining key parameters and leading auxiliary parameters
CN115269342B (en) Monitoring cloud platform based on edge computing and monitoring method thereof
CN117371337B (en) Water conservancy model construction method and system based on digital twin
CN112016828A (en) Industrial equipment health management cloud platform architecture based on streaming big data
CN116451514A (en) Bridge structure state evaluation method and device and electronic equipment
CN115204536A (en) Building equipment fault prediction method, device, equipment and storage medium
CN111899138A (en) Teaching behavior analysis system and method based on big data
CN115378928B (en) Monitoring method and system based on cloud service
CN111444233A (en) Method for discovering environmental monitoring abnormal data based on duplicator neural network model
CN115146230A (en) Ancient building health monitoring system, method and equipment
CN117596160B (en) Method and system for manufacturing industry data link communication
CN108563664A (en) A kind of real-time data processing method at industrial equipment end
CN112417446A (en) Software defined network anomaly detection architecture
CN117808272B (en) Industrial system productivity optimization and scheduling method and system based on machine learning
JP7345006B1 (en) Learning model generation method and testing device
CN115604319B (en) Intelligent community management platform based on multi-dimensional sensor
CN116660761B (en) Lithium ion battery detection method and system

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