CN112801392A - Intelligent prediction method for speed of corrugated paper machine - Google Patents
Intelligent prediction method for speed of corrugated paper machine Download PDFInfo
- Publication number
- CN112801392A CN112801392A CN202110162647.4A CN202110162647A CN112801392A CN 112801392 A CN112801392 A CN 112801392A CN 202110162647 A CN202110162647 A CN 202110162647A CN 112801392 A CN112801392 A CN 112801392A
- Authority
- CN
- China
- Prior art keywords
- data
- vehicle speed
- vibration
- speed
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 69
- 238000004519 manufacturing process Methods 0.000 claims abstract description 26
- 238000010801 machine learning Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000012216 screening Methods 0.000 claims description 11
- 238000001228 spectrum Methods 0.000 claims description 11
- 238000012795 verification Methods 0.000 claims description 8
- 238000010183 spectrum analysis Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000003595 spectral effect Effects 0.000 claims description 5
- 230000001133 acceleration Effects 0.000 claims description 4
- 238000011144 upstream manufacturing Methods 0.000 claims 1
- 238000004806 packaging method and process Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- Manufacturing & Machinery (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The embodiment of the invention provides an intelligent prediction method for the speed of a corrugated paper machine, which comprises the following steps: the method comprises the following steps: collecting vibration data by using an MCM100 device; step two: reading parameter data of a production management system; step three: processing and analyzing the vibration data and the parameter data; step four: based on the characteristic engineering, clustering the vibration data and the parameter data by using a clustering algorithm to obtain characteristic training data; step five: training the characteristic training data by using a machine learning algorithm so as to predict the speed of the corrugated paper machine; the embodiment of the invention can predict the speed of the corrugating machine to reduce the fluctuation of the speed and the vibration, thereby improving the quality of paper and further realizing the improvement of the production efficiency.
Description
Technical Field
The invention relates to the technical field of intelligent manufacturing, in particular to an intelligent prediction method for the speed of a corrugated paper machine.
Background
The running state of the unable real time monitoring flute machine of traditional paper mill and the production state of paper, the abnormal conditions can not be handled, the paper quality effect difference that can lead to producing is big, can lead to the corrugated paper to play fold and damage sometimes even, traditional corrugated paper machine speed control mode is the manual adjustment method, traditional manual adjustment method is unstable, speed fluctuation is great, can lead to corrugated paper machine's vibration fluctuation great, also can lead to the corrugated paper to play fold and damage, the method of manual adjustment speed of a motor vehicle does not reach this corrugated paper machine at the material of certain corrugated paper, the optimum speed of a motor vehicle under the breadth condition most of time, influence production efficiency, the human cost of control corrugated paper machine speed of a motor vehicle also is a significant spending of mill.
Summary of the invention
In order to overcome the defects of the prior art, the invention provides an intelligent speed prediction method of a corrugated paper machine, which is used for solving the technical problems of unstable speed of the corrugated paper machine, low paper quality, easy damage of paper and low paper production efficiency.
The technical scheme adopted by the invention for solving the technical problems is as follows: the method for intelligently predicting the speed of the corrugated paper machine comprises the following steps:
the method comprises the following steps: collecting vibration data by using an MCM100 device;
step two: reading parameter data of a production management system;
step three: processing and analyzing the vibration data and the parameter data;
step four: based on the characteristic engineering, clustering the vibration data and the parameter data by using a clustering algorithm to obtain characteristic training data;
step five: and training the characteristic training data by using a machine learning algorithm so as to predict the speed of the corrugating machine.
Specifically, the method for collecting vibration data by using the MCM100 device comprises the following steps:
using an MCM100 to access a vibration sensor to collect time domain data;
converting time domain data into FFT frequency spectrum data through Fourier transform;
calculating an OA value through FFT spectrum data;
specifically, the OA value is calculated by FFT spectrum data, and the steps include:
by calculation of
Evaluating the vibration acceleration of a conveying belt of the corrugated paper machine;
wherein OA is the OA value of the full bandwidth or a certain frequency band, n is the number of the full bandwidth or in-band spectrum analysis pieces, AiFor amplitude values on the spectral analysis line, i.e. FFT spectral data, NBFFor the weighted window factor, the Hanning window is 1.5.
Specifically, reading production management system parameter data, the steps include:
reading a text file stored by a corrugated paper machine production management system;
and acquiring target data corresponding to the corrugated medium paper from a text file stored by a production management system, wherein the target data comprises material data, width data, ridge data and vehicle speed data.
Specifically, the vibration data and the parameter data are processed and analyzed, and the steps include:
acquiring vehicle speed data, material data and width data at regular time according to a time sequence;
selecting data with the vehicle speed greater than 50, selecting data with the material not being empty and selecting data with the width greater than 0, and carrying out data numeralization on the material data and the ridge data;
fixing the material, ridge and width, and visualizing the vehicle speed data and the vibration data according to the time sequence.
Specifically, based on feature engineering, clustering vibration data and parameter data by using a clustering algorithm, the steps comprising:
fixing the material, ridge, width and speed, and calculating the mean value of the vibration data;
fixing the material, ridge and width, and counting the frequency of the occurrence of the vehicle speed;
and sequentially arranging according to the vehicle speed frequency, and screening the vehicle speed data with the frequency ranking being 12 th.
Preferably, the vehicle speed data are arranged in sequence according to the vehicle speed frequency, and after the vehicle speed data with the frequency ranking 12 are screened, the steps further include:
screening out the vehicle speed smaller than the vibration mean value from the 12 vehicle speeds according to the calculated mean value of the vibration data of each vehicle speed;
using a Kmeans clustering algorithm to cluster the vehicle speeds smaller than the vibration mean value into two types of vehicle speeds;
and selecting the maximum value and the minimum value of the two types of vehicle speeds to form a first vehicle speed range and a second vehicle speed range.
Preferably, after the maximum value and the minimum value of the two vehicle speeds are selected to form a first vehicle speed range and a second vehicle speed range, the steps further include:
judging whether the current vehicle speed is in an uplink vehicle speed range or a downlink vehicle speed range according to the first vehicle speed range and the second vehicle speed range;
when the current vehicle speed is in the uplink vehicle speed range, the label of the uplink vehicle speed is the average value of the uplink vehicle speed range;
and when the current vehicle speed is in the next vehicle speed range, taking the average value of the next vehicle speed range by the label of the next vehicle speed.
In particular, the characteristic training data is trained by using a machine learning algorithm so as to predict the speed of the corrugating machine,
the steps include:
according to the labels of the uplink vehicle speed and the labels of the downlink vehicle speed, the feature training data are divided into uplink vehicle speed training data and downlink vehicle speed training data, the uplink vehicle speed training data and the downlink vehicle speed training data are respectively divided into 10 parts, the divided training data account for 8 parts, and the verification data account for 2 parts.
Specifically, after dividing into uplink vehicle speed training data and downlink vehicle speed training data, the steps further include:
training the divided uplink vehicle speed training data and the divided downlink vehicle speed training data by using an Lgb gradient lifting tree algorithm;
and verifying the predicted uplink vehicle speed and the predicted downlink vehicle speed by using the verification data.
The embodiment of the invention has the following beneficial effects: the method comprises the following steps: collecting vibration data by using an MCM100 device; step two: reading parameter data of a production management system; step three: processing and analyzing the vibration data and the parameter data; step four: based on the characteristic engineering, clustering the vibration data and the parameter data by using a clustering algorithm to obtain characteristic training data; step five: the characteristic training data are trained by utilizing a machine learning algorithm so as to predict the speed of the corrugated paper machine and reduce the fluctuation of the speed and the vibration, thereby improving the quality of paper and further realizing the improvement of the production efficiency.
Drawings
Fig. 1 is a flow chart diagram of a corrugating machine speed intelligent prediction method.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following detailed description of specific implementations of the present invention is provided in conjunction with specific embodiments:
the first embodiment is as follows:
fig. 1 shows an implementation flow of a method for intelligently predicting the speed of a corrugating machine according to a first embodiment of the present invention, and for convenience of description, only the parts related to the embodiment of the present invention are shown, which is detailed as follows:
in step S101, collecting vibration data with the MCM100 device;
specifically, the method for collecting vibration data by using the MCM100 device comprises the following steps:
using an MCM100 to access a vibration sensor to collect time domain data;
converting time domain data into FFT frequency spectrum data through Fourier transform;
calculating an OA value through FFT spectrum data;
specifically, the OA value is calculated by FFT spectrum data, and the steps include:
by calculation of
Evaluating the vibration acceleration of a conveying belt of the corrugated paper machine;
wherein OA is the OA value of the full bandwidth or a certain frequency band, n is the number of the full bandwidth or in-band spectrum analysis pieces, AiFor amplitude values on the spectral analysis line, i.e. FFT spectral data, NBFFor the weighted window factor, the Hanning window is 1.5.
In the embodiment of the application, the vibration data is collected, the running state of the corrugated paper machine can be monitored in real time, the fluctuation of speed and vibration is reduced, the quality of paper is improved, and the condition of paper breakage is reduced.
In step S102, reading production management system parameter data;
specifically, reading production management system parameter data, the steps include:
reading a text file stored by a corrugated paper machine production management system;
and acquiring target data corresponding to the corrugated medium paper from a text file stored by a production management system, wherein the target data comprises material data, width data, ridge data and vehicle speed data.
In the embodiment of the application, the parameter data of the production management system is read, the production state of a paper mill can be monitored in real time, the prediction of the suggested speed of the corrugated paper machine is facilitated, and various abnormal production conditions can be processed in time;
in step S103, the vibration data and the parameter data are processed and analyzed;
specifically, the vibration data and the parameter data are processed and analyzed, and the steps include:
acquiring vehicle speed data, material data and width data at regular time according to a time sequence;
selecting data with the vehicle speed greater than 50, selecting data with the material not being empty and selecting data with the width greater than 0, and carrying out data numeralization on the material data and the ridge data;
fixing the material, ridge and width, and visualizing the vehicle speed data and the vibration data according to the time sequence.
In the embodiment of the application, the statistical characteristics in the historical production data can be rapidly calculated, and the training accuracy of the corrugating machine suggested speed algorithm is improved.
In step S104, clustering the vibration data and the parameter data by using a clustering algorithm based on the characteristic engineering to obtain characteristic training data;
specifically, based on feature engineering, clustering vibration data and parameter data by using a clustering algorithm, the steps comprising:
fixing the material, ridge, width and speed, and calculating the mean value of the vibration data;
fixing the material, ridge and width, and counting the frequency of the occurrence of the vehicle speed;
and sequentially arranging according to the vehicle speed frequency, and screening the vehicle speed data with the frequency ranking being 12 th.
Preferably, the vehicle speed data are arranged in sequence according to the vehicle speed frequency, and after the vehicle speed data with the frequency ranking 12 are screened, the steps further include:
screening out the vehicle speed smaller than the vibration mean value from the 12 vehicle speeds according to the calculated mean value of the vibration data of each vehicle speed;
using a Kmeans clustering algorithm to cluster the vehicle speeds smaller than the vibration mean value into two types of vehicle speeds;
and selecting the maximum value and the minimum value of the two types of vehicle speeds to form a first vehicle speed range and a second vehicle speed range.
Preferably, after the maximum value and the minimum value of the two vehicle speeds are selected to form a first vehicle speed range and a second vehicle speed range, the steps further include:
judging whether the current vehicle speed is in an uplink vehicle speed range or a downlink vehicle speed range according to the first vehicle speed range and the second vehicle speed range;
when the current vehicle speed is in the uplink vehicle speed range, the label of the uplink vehicle speed is the average value of the uplink vehicle speed range;
and when the current vehicle speed is in the next vehicle speed range, taking the average value of the next vehicle speed range by the label of the next vehicle speed.
In the embodiment of the application, the reasonable method can be used for screening the vehicle speed smaller than the vibration mean value, the vehicle speed larger than the vibration mean value in operation is avoided, the paper is prevented from wrinkling or being damaged, the clustering algorithm can be used for quickly finding out the optimal suggested vehicle speed, the circular traversal method is avoided, and the operation speed of a program is improved.
In step S105, the feature training data is trained using a machine learning algorithm to predict the corrugator machine speed.
Specifically, the method comprises training feature training data by using a machine learning algorithm so as to predict the speed of the corrugator machine, and comprises the following steps:
according to the labels of the uplink vehicle speed and the labels of the downlink vehicle speed, the feature training data are divided into uplink vehicle speed training data and downlink vehicle speed training data, the uplink vehicle speed training data and the downlink vehicle speed training data are respectively divided into 10 parts, the divided training data account for 8 parts, and the verification data account for 2 parts.
Specifically, after dividing into uplink vehicle speed training data and downlink vehicle speed training data, the steps further include:
training the divided uplink vehicle speed training data and the divided downlink vehicle speed training data by using an Lgb gradient lifting tree algorithm;
and verifying the predicted uplink vehicle speed and the predicted downlink vehicle speed by using the verification data.
In the embodiment of the application, the data of the corrugated machine suggested speed algorithm model training is convenient to predict, and then the suggested speed prediction model can be trained.
Example two:
vibration data were collected using MCM 100:
the acquisition of vibration data is important for vehicle speed control of the corrugating machine, the vibration is too large, corrugated paper can be damaged or wrinkled, and the control of the vibration data below the vibration mean value is very necessary. The vibration data acquisition mainly considers the position of a driving side and an operation side of a corrugating machine, equipment used for acquiring the vibration data is an MCM100, the MCM100 is a mechanical equipment state monitoring edge computing platform, four channels for receiving signals are arranged, an interface end of a vibration sensor is connected into one channel for receiving the signals in the MCM100, a sensing end is installed on the driving side of the corrugating machine, the vibration sensor is used for acquiring time domain data, then the time domain data are converted into FFT frequency spectrum data through Fourier transform, an OA Value (Overall Value) is calculated through the FFT frequency spectrum data, the OA Value is a total energy expression method, the unit is g, the vibration size can be evaluated according to the OA Value, and the calculation formula is as follows:
OA: full bandwidth or OA value of a certain frequency band.
n: analyzing the number of the frequency spectrums; refers to full bandwidth or in-band.
Ai: the amplitude values on the spectral analysis line, i.e. the FFT spectral data.
NBF: weighted window factor, Hanning window 1.5.
Reading production management system data using software:
the corrugated paper machine parameter data acquisition method comprises the following steps: reading a text file stored by a production management system in an industrial personal computer of a paper mill by using software written by a C # programming language, acquiring data such as the material and width of corrugated medium paper, the flute shape and the running speed (vehicle speed) of a corrugating machine and the like, and storing the data into an Sql Server database. Different materials and widths can be used for different corrugating machine edge shapes, and corresponding optimal running speeds are different.
The data were processed and analyzed using Python program:
step one, connecting the data collected every day through a time field, and sequencing according to a time sequence.
Step two, data screening: selecting data with the vehicle speed greater than 50 (the data with the vehicle speed less than 50 is in an acceleration stage and should not be used as training data of an algorithm, the vehicle speed range is 0-173), selecting data with the material not being empty (the material is empty when not producing), and selecting data with the width greater than 0 (the width is 0 when not producing).
And step three, digitizing the two types of data, namely the material type data and the ridge type data, so that the algorithm can be trained conveniently, and the algorithm can only recognize the digitized data.
And step four, fixing the material, ridge and width, visualizing the vehicle speed and vibration according to time, and analyzing the relation between the vehicle speed and the vibration so as to optimize the statistical characteristics.
Statistical features were calculated using the python program, and data were clustered using a clustering algorithm:
firstly, calculating the vibration mean value of each vehicle speed, wherein the calculation method comprises the following steps: firstly, fixing the material, edge, width and speed, and under the condition, averaging the vibration data to obtain the vibration average value under the condition.
Step two, counting the frequency of each vehicle speed, wherein the counting method comprises the following steps: the material, ridge and width are fixed, vibration data are not considered temporarily, one row is added once when a certain vehicle speed appears, and the like.
And thirdly, arranging the vehicle speed data according to the vehicle speed frequency from large to small, screening the vehicle speed data of the first 12 in the frequency ranking, and if the screening ranking is smaller than the first 12, losing the characteristic data under partial conditions.
And step four, screening out the vehicle speeds smaller than the vibration mean value from the 12 vehicle speeds according to the vibration mean value of each vehicle speed calculated in the step one, wherein the vehicle speeds larger than the vibration mean value can possibly cause damage to corrugating machines, and the vehicle speeds exclude the prediction range of the suggested vehicle speed.
And step five, clustering the vehicle speeds smaller than the vibration mean value into two types by using a Kmeans clustering algorithm, and selecting a first vehicle speed range and a second vehicle speed range of the maximum value and the minimum value of the two types of vehicle speeds.
Judging whether the current vehicle speed is an uplink vehicle speed range or a downlink vehicle speed range according to the first vehicle speed range and the second vehicle speed range;
and seventhly, when the current vehicle speed is in the upper vehicle speed range, the label of the upper vehicle speed takes the average value of the upper vehicle speed range, and when the current vehicle speed is in the lower vehicle speed range, the label of the lower vehicle speed takes the average value of the lower vehicle speed range.
Step eight, the generated data has seven characteristics of date, material, ridge, width, drive side OA value, vibration mean value and vehicle speed, two labels of the recommended uplink vehicle speed and the recommended downlink vehicle speed are provided, the seven characteristic data and the two label data are collectively referred to as characteristic training data, and the example data is as follows:
training a machine learning algorithm:
the method comprises the steps of firstly, dividing feature training data into uplink vehicle speed training data and downlink vehicle speed training data according to two labels, dividing the uplink vehicle speed training data into seven features and uplink vehicle speeds, dividing the uplink vehicle speed training data and the downlink vehicle speed training data into 10 parts, wherein the downlink vehicle speed training data comprises the seven features and the downlink vehicle speeds, the divided training data accounts for 8 parts, and the verification data accounts for 2 parts.
And step two, training the divided uplink vehicle speed training data and the divided downlink vehicle speed training data respectively by using an Lgb gradient lifting tree algorithm, and finely adjusting and modifying algorithm parameters to adapt to the two divided training data.
Step three, verifying by using verification data, verifying the predicted uplink vehicle speed and downlink vehicle speed results, taking mean-square error (MSE) as a standard for evaluating the accuracy of the algorithm, wherein the mean-square error is a measure for reflecting the difference degree between the predicted result and the real result, the smaller the mean-square error is, the higher the accuracy of the algorithm is, and when the mean-square error is not reduced any more, the algorithm training is stopped; algorithm training continues while the mean square error is still decreasing.
Storing the trained model:
and storing the trained model into an h5 file, so as to facilitate prediction, packaging and deployment.
Packaging the model into a service: and packaging the data into Web API service for the proportional valve control system to call.
And packing the model into a Web service by using a flash framework of Python for being called by other application programs.
The control of the corrugating machine speed by the proportional valve is realized: the proportional valve sends a request to a message queue through an MQTT protocol, a proportional valve control system obtains the request from the message queue through the MQTT protocol, Python-packed Web API service is called, the suggested uplink and downlink vehicle speed is obtained, data are placed into the message queue, the proportional valve is called, and the vehicle speed of the corrugating machine is controlled.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the embodiments described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation.
Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. An intelligent prediction method for the speed of a corrugated paper machine is characterized by comprising the following steps:
the method comprises the following steps: collecting vibration data by using an MCM100 device;
step two: reading parameter data of a production management system;
step three: processing and analyzing the vibration data and the parameter data;
step four: based on the characteristic engineering, clustering the vibration data and the parameter data by using a clustering algorithm to obtain characteristic training data;
step five: and training the characteristic training data by using a machine learning algorithm so as to predict the speed of the corrugating machine.
2. The corrugator machine speed intelligent prediction method as claimed in claim 1, wherein the MCM100 equipment is used for collecting vibration data, and the steps comprise:
using an MCM100 to access a vibration sensor to collect time domain data;
converting time domain data into FFT frequency spectrum data through Fourier transform;
OA values are calculated from the FFT spectral data.
3. The intelligent corrugating machine speed predicting method according to claim 2, wherein the OA value is calculated by FFT spectrum data, and the steps comprise:
by calculation of
Evaluating the vibration acceleration of a conveying belt of the corrugated paper machine;
wherein OA is the OA value of the full bandwidth or a certain frequency band, n is the number of the full bandwidth or in-band spectrum analysis pieces, AiFor amplitude values on the spectral analysis line, i.e. FFT spectral data, NBFFor the weighted window factor, the Hanning window is 1.5.
4. The intelligent corrugator machine speed predicting method according to claim 3, wherein the production management system parameter data is read, and the steps comprise:
reading a text file stored by a corrugated paper machine production management system;
and acquiring target data corresponding to the corrugated medium paper from a text file stored by a production management system, wherein the target data comprises material data, width data, ridge data and vehicle speed data.
5. The intelligent corrugating machine speed predicting method as claimed in claim 4, wherein the vibration data and the parameter data are processed and analyzed, and the steps comprise:
acquiring vehicle speed data, material data and width data at regular time according to a time sequence;
selecting data with the vehicle speed greater than 50, selecting data with the material not being empty and selecting data with the width greater than 0, and carrying out data numeralization on the material data and the ridge data;
fixing the material, ridge and width, and visualizing the vehicle speed data and the vibration data according to the time sequence.
6. The intelligent prediction method of the speed of the corrugating machine as claimed in claim 5, wherein clustering is performed on the vibration data and the parameter data by using a clustering algorithm based on feature engineering, and the steps comprise:
fixing the material, ridge, width and speed, and calculating the mean value of the vibration data;
fixing the material, ridge and width, and counting the frequency of the occurrence of the vehicle speed;
and sequentially arranging according to the vehicle speed frequency, and screening the vehicle speed data with the frequency ranking being 12 th.
7. The intelligent corrugated paper machine speed prediction method according to claim 6, wherein the vehicle speed data are arranged in sequence according to vehicle speed frequency, and after the vehicle speed data with the frequency ranking 12 are screened, the method further comprises the following steps:
screening out the vehicle speed smaller than the vibration mean value from the 12 vehicle speeds according to the calculated mean value of the vibration data of each vehicle speed;
using a Kmeans clustering algorithm to cluster the vehicle speeds smaller than the vibration mean value into two types of vehicle speeds;
and selecting the maximum value and the minimum value of the two types of vehicle speeds to form a first vehicle speed range and a second vehicle speed range.
8. The intelligent corrugating machine speed predicting method according to claim 7, wherein after the maximum value and the minimum value of the two types of vehicle speeds are selected to form a first vehicle speed range and a second vehicle speed range, the steps further comprise:
judging whether the current vehicle speed is in an uplink vehicle speed range or a downlink vehicle speed range according to the first vehicle speed range and the second vehicle speed range;
when the current vehicle speed is in the uplink vehicle speed range, the label of the uplink vehicle speed is the average value of the uplink vehicle speed range;
and when the current vehicle speed is in the next vehicle speed range, taking the average value of the next vehicle speed range by the label of the next vehicle speed.
9. The intelligent corrugator speed predicting method according to claim 8, wherein the feature training data is trained by a machine learning algorithm to predict the corrugator speed, and the method comprises the following steps:
according to the labels of the uplink vehicle speed and the labels of the downlink vehicle speed, the feature training data are divided into uplink vehicle speed training data and downlink vehicle speed training data, the uplink vehicle speed training data and the downlink vehicle speed training data are respectively divided into 10 parts, the divided training data account for 8 parts, and the verification data account for 2 parts.
10. The intelligent corrugating machine speed predicting method according to claim 9, wherein after dividing into upstream vehicle speed training data and downstream vehicle speed training data, the method further comprises:
training the divided uplink vehicle speed training data and the divided downlink vehicle speed training data by using an Lgb gradient lifting tree algorithm;
and verifying the predicted uplink vehicle speed and the predicted downlink vehicle speed by using the verification data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110162647.4A CN112801392B (en) | 2021-02-05 | 2021-02-05 | Intelligent speed prediction method for corrugated paper machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110162647.4A CN112801392B (en) | 2021-02-05 | 2021-02-05 | Intelligent speed prediction method for corrugated paper machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112801392A true CN112801392A (en) | 2021-05-14 |
| CN112801392B CN112801392B (en) | 2023-07-04 |
Family
ID=75814445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110162647.4A Active CN112801392B (en) | 2021-02-05 | 2021-02-05 | Intelligent speed prediction method for corrugated paper machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112801392B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117521423A (en) * | 2024-01-05 | 2024-02-06 | 山东鑫林纸制品有限公司 | Intelligent prediction method and device for machine speed of corrugated paper processing and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6087796A (en) * | 1998-06-16 | 2000-07-11 | Csi Technology, Inc. | Method and apparatus for determining electric motor speed using vibration and flux |
| US20200272122A1 (en) * | 2019-02-27 | 2020-08-27 | Fanuc Corporation | Chatter vibration determination device, machine learning device, and system |
-
2021
- 2021-02-05 CN CN202110162647.4A patent/CN112801392B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6087796A (en) * | 1998-06-16 | 2000-07-11 | Csi Technology, Inc. | Method and apparatus for determining electric motor speed using vibration and flux |
| US20200272122A1 (en) * | 2019-02-27 | 2020-08-27 | Fanuc Corporation | Chatter vibration determination device, machine learning device, and system |
Non-Patent Citations (1)
| Title |
|---|
| 解梦涛等: "基于频谱筛选的航空发动机振动总量提取方法", 《现代机械》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117521423A (en) * | 2024-01-05 | 2024-02-06 | 山东鑫林纸制品有限公司 | Intelligent prediction method and device for machine speed of corrugated paper processing and storage medium |
| CN117521423B (en) * | 2024-01-05 | 2024-04-05 | 山东鑫林纸制品有限公司 | Intelligent prediction method and device for machine speed of corrugated paper processing and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112801392B (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111637964B (en) | Rail corrugation identification method | |
| CN111623868B (en) | Convolutional neural network construction method for rail corrugation identification | |
| CN113657221B (en) | Power plant equipment state monitoring method based on intelligent sensing technology | |
| CN118214168B (en) | Real-time state monitoring method for power distribution circuit | |
| CN112173636B (en) | Method for detecting faults of belt conveyor carrier roller by inspection robot | |
| CN112801392A (en) | Intelligent prediction method for speed of corrugated paper machine | |
| CN107844067A (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN102607641A (en) | Cluster anomaly detection method of combustion gas turbine | |
| CN111125895A (en) | On-line assessment method for service power characteristics of complete wind driven generator | |
| CN118114186B (en) | Equipment vibration fault diagnosis method based on artificial intelligence | |
| CN104713730B (en) | Method for determining degeneration rate of aircraft engine according to vibration signal | |
| CN109598309B (en) | Detection system and monitoring method of metal packaging punching machine | |
| CN117466153B (en) | Fault detection method, device, computer equipment and readable storage medium | |
| CN117423345A (en) | Voiceprint recognition monitoring system for power equipment | |
| CN113505898A (en) | Equipment predictive maintenance method based on AI technology | |
| CN113139430A (en) | Acoustic signal analysis method for fault detection, and method and device for fault detection | |
| CN116502043A (en) | Finish rolling motor state analysis method based on isolated forest algorithm | |
| CN113627885A (en) | Power grid power supply abnormity monitoring system and monitoring method thereof | |
| CN112699609A (en) | Diesel engine reliability model construction method based on vibration data | |
| CN113658603A (en) | Intelligent fault diagnosis method for belt conveyor carrier roller based on audio frequency | |
| CN113052232A (en) | Motor fault analysis method and system | |
| CN112417966A (en) | Multi-dimensional machine fault detection method based on audio signal characteristics and AI algorithm | |
| CN110929425B (en) | Soft measurement method for longitudinal elongation of paper based on Lasso algorithm | |
| CN117577137B (en) | Cutter health assessment method, device, equipment and storage medium | |
| JP7306659B1 (en) | MONITORING SYSTEM, MONITORING METHOD, MONITORING PROGRAM |
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 |