CN115058807B - Intelligent control method and system for spinning machine - Google Patents
Intelligent control method and system for spinning machine Download PDFInfo
- Publication number
- CN115058807B CN115058807B CN202210986970.8A CN202210986970A CN115058807B CN 115058807 B CN115058807 B CN 115058807B CN 202210986970 A CN202210986970 A CN 202210986970A CN 115058807 B CN115058807 B CN 115058807B
- Authority
- CN
- China
- Prior art keywords
- cleaning
- spinning machine
- image information
- analysis result
- analysis
- 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
Images
Classifications
-
- D—TEXTILES; PAPER
- D01—NATURAL OR MAN-MADE THREADS OR FIBRES; SPINNING
- D01H—SPINNING OR TWISTING
- D01H11/00—Arrangements for confining or removing dust, fly or the like
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30124—Fabrics; Textile; Paper
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Textile Engineering (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Spinning Or Twisting Of Yarns (AREA)
Abstract
The invention provides an intelligent control method and system for a spinning machine, which relate to the technical field of spinning, and the method comprises the following steps: spinning by adopting a target spinning machine according to a preset spinning flow; after at least one preset spinning flow is completed, a first image information set is obtained; inputting the data into a spinning machine cleaning analysis model to obtain an output result; obtaining a pollution analysis result of the target spinning machine; inputting a cleaning strategy analysis model to obtain cleaning strategy information; the cleaning strategy information is adopted to control the target spinning machine to clean, and then image information is acquired to obtain a second image information set; analyzing, evaluating and cleaning strategy information according to the first image information set and the second image information set to obtain an analysis result; and updating the cleaning strategy analysis model. The technical problem that the spinning machine cannot be cleaned timely and effectively is solved, the spinning machine cleaning strategy can be generated timely and accurately, and the technical effect of intelligent control over the cleaning of the spinning machine is achieved.
Description
Technical Field
The invention relates to the technical field of spinning, in particular to an intelligent control method and system for a spinning machine.
Background
Automatic change control technology's rapid development, automatic spinning machine's degree of automation is higher and higher, it is common if full automatization rotor type open-end spinning machine, its operation process need not manual operation almost, greatly reduced the product processingquality that manual production processing brought unstable, but because of there is a large amount of yarn wastes in the spinning machine processing environment, the dust, miscellaneous bits, untimely yarn wastes in the clearance machine, the dust, miscellaneous bits can lead to spinning machine's operational failure, need urgently the clearance integration intelligence control system of spinning machine, carry out intelligent control to spinning machine's clearance, in order to guarantee spinning machine's steady operation.
The technical problem that a spinning machine cannot be cleaned timely and effectively exists in the prior art.
Disclosure of Invention
The application solves the technical problem that the spinning machine cannot be cleaned timely and effectively by providing the intelligent control method and the intelligent control system for the spinning machine, and achieves the technical effects of accurately and timely generating a spinning machine cleaning strategy and intelligently controlling the cleaning of the spinning machine.
In view of the above problems, the present application provides an intelligent control method and system for a spinning machine.
In a first aspect, the present application provides an intelligent control method for a spinning machine, wherein the method comprises: spinning by adopting a target spinning machine according to a preset spinning process, wherein the preset spinning process comprises a plurality of spinning steps; after the target spinning machine finishes at least one spinning work of the preset spinning flow, acquiring and acquiring image information of a plurality of positions in the target spinning machine to obtain a first image information set; inputting the first image information set into a pre-constructed spinning machine cleaning analysis model to obtain an output result; obtaining a pollution analysis result of the target spinning machine according to the output result; inputting the pollution analysis result into a cleaning strategy analysis model to obtain cleaning strategy information for cleaning the target spinning machine; controlling the target spinning machine to clean by adopting the cleaning strategy information, and acquiring image information of a plurality of positions in the target spinning machine after cleaning to acquire a second image information set; analyzing and evaluating the cleaning strategy information according to the first image information set and the second image information set to obtain an analysis result; and updating the cleaning strategy analysis model according to the analysis result.
In a second aspect, the present application provides an intelligent control system for a spinning machine, wherein the system comprises: the spinning process executing unit is used for spinning by adopting a target spinning machine according to a preset spinning process, wherein the preset spinning process comprises a plurality of spinning steps; the image information acquisition unit is used for acquiring and acquiring image information of a plurality of positions in the target spinning machine after the target spinning machine finishes at least one spinning work of the preset spinning flow, so as to acquire a first image information set; the output result acquisition unit is used for inputting the first image information set into a pre-constructed spinning machine cleaning analysis model to obtain an output result; a pollution analysis result acquisition unit for acquiring a pollution analysis result of the target spinning machine according to the output result; a cleaning strategy information obtaining unit, configured to input the pollution analysis result into a cleaning strategy analysis model, and obtain cleaning strategy information for cleaning the target spinning machine; the image information acquisition unit is used for adopting the cleaning strategy information to control the target spinning machine to clean, acquiring and acquiring image information of a plurality of positions in the target spinning machine after cleaning, and acquiring a second image information set; an analysis result obtaining unit, configured to analyze and evaluate the cleaning policy information according to the first image information set and the second image information set, to obtain an analysis result; and the data updating unit is used for updating the cleaning strategy analysis model according to the analysis result.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
the method comprises the steps of adopting a target spinning machine to spin according to a preset spinning flow, obtaining a first image information set after finishing at least one spinning work of the preset spinning flow, inputting the first image information set into a pre-constructed spinning machine cleaning analysis model, obtaining an output result, obtaining a pollution analysis result of the target spinning machine, inputting a cleaning strategy analysis model, obtaining cleaning strategy information, controlling the target spinning machine to clean, acquiring image information of a plurality of positions in the target spinning machine, obtaining a second image information set, analyzing and evaluating the cleaning strategy information according to the first image information set and the second image information set, obtaining an analysis result, and updating the cleaning strategy analysis model. The embodiment of the application achieves the technical effects of accurately and timely generating the cleaning strategy of the spinning machine and intelligently controlling the cleaning of the spinning machine.
Drawings
FIG. 1 is a schematic flow chart of an intelligent control method for a spinning machine according to the present application;
FIG. 2 is a schematic flow chart of a cleaning strategy analysis model constructed and obtained in the intelligent control method for the spinning machine according to the application;
FIG. 3 is a schematic flow chart illustrating the determination of whether to update the cleaning strategy analysis model in the intelligent control method for the spinning machine according to the present application;
fig. 4 is a schematic structural diagram of an intelligent control system for a spinning machine according to the present application.
Description of reference numerals: the spinning process executing unit 11, the image information acquiring unit 12, the output result acquiring unit 13, the pollution analysis result acquiring unit 14, the cleaning strategy information acquiring unit 15, the image information acquiring unit 16, the analysis result acquiring unit 17 and the data updating unit 18.
Detailed Description
The application solves the technical problem that the spinning machine cannot be cleaned timely and effectively by providing the intelligent control method and the intelligent control system for the spinning machine, and achieves the technical effects of accurately and timely generating a spinning machine cleaning strategy and intelligently controlling the cleaning of the spinning machine.
Example one
As shown in fig. 1, the present application provides an intelligent control method for a spinning machine, wherein the method comprises:
s100: spinning by adopting a target spinning machine according to a preset spinning process, wherein the preset spinning process comprises a plurality of spinning steps;
s200: after the target spinning machine finishes at least one spinning work of the preset spinning flow, acquiring and acquiring image information of a plurality of positions in the target spinning machine to obtain a first image information set;
specifically, the target spinning machine is any automatic spinning machine, the target spinning machine can be a spinning frame, a ring spinning frame, a full-automatic rotor spinning machine or any other spinning machine, the type of the target spinning machine is not limited, the preset spinning process comprises a plurality of spinning steps, the preset spinning process is a spinning processing process of the target spinning machine, the target spinning machine is adopted to carry out spinning according to the preset spinning process, technical support is provided for determining a cleaning node of the spinning machine, after the target spinning machine finishes at least one spinning operation of the preset spinning process, the cleaning node of the spinning machine is determined, the cleaning node comprises a cleaning time point, a cleaning program is started, image information of a plurality of positions in the target spinning machine is acquired and obtained through an image acquisition device, the image acquisition device is an image acquisition device such as a camera, the plurality of positions are multi-angle positions of the target spinning machine, a first image information set is obtained, an image arrangement rule of the first image information set corresponds to a structural diagram of the target spinning machine, and data support is provided for subsequent cleaning.
S300: inputting the first image information set into a pre-constructed spinning machine cleaning analysis model to obtain an output result;
further, in the process of constructing the cleaning and analyzing model of the spinning machine, the step S300 further includes:
s310: obtaining a preset time span;
s320: acquiring image information of a plurality of positions in the target spinning machine within the preset time span in historical time to obtain a plurality of historical first image information sets;
s330: collecting pollution analysis results of a plurality of positions in the target spinning machine within the preset time span in historical time to obtain a plurality of historical pollution analysis result sets;
s340: and constructing the spinning machine cleaning analysis model by taking the plurality of historical first image information sets and the plurality of historical pollution analysis result sets as construction data sets.
Specifically, an information storage module, such as a register, is integrated in the intelligent control system, information related to the target spinning machine image is recorded based on the register, information retrieval and extraction are carried out through identifiers, a plurality of historical first image information sets are obtained, the spinning machine cleaning analysis model is constructed based on the plurality of historical first image information sets, the first image information sets are input into the pre-constructed spinning machine cleaning analysis model, an output result is obtained, and a model basis is provided for subsequent data analysis.
More specifically, the preset time span includes a preset starting time point and a preset ending time point, the preset starting time point can be set as a time point corresponding to the cleaning completed for the first ten times, the preset ending time point is set as a time point corresponding to the cleaning program started, the preset time span is not uniquely set, and related technicians can obtain the preset time span by self-defining the setting in combination with actual conditions; acquiring image information of a plurality of positions in the target spinning machine within the preset time span in the historical time based on an image acquisition device at the plurality of positions to obtain a plurality of historical first image information sets, if the preset starting time point is set as a time point corresponding to the completion of cleaning for the first ten times and a preset ending time point is set as a time point corresponding to the starting of a cleaning program, namely the historical first image information sets are ten, the image arrangement rules of the historical first image information sets and the first image information sets correspond to a whole machine structure diagram of the target spinning machine, namely the image distribution arrangement of the historical first image information sets and the image distribution arrangement of the first image information sets are consistent, acquiring and extracting pollution analysis results of the plurality of positions in the target spinning machine within the preset time span in the historical time, wherein the pollution analysis results can include but are not limited to waste yarn pollution information, dust pollution information and miscellaneous pollution information, obtaining a plurality of historical pollution analysis result sets, the historical pollution analysis result sets correspond to the historical first image information sets, specifically, any element of the historical first image information sets is a first image, and only one element of the pollution information sets is a pollution of the historical pollution information, and only one element of the historical pollution information sets is a pollution of the first image; and carrying out data binding on the plurality of historical first image information sets and the plurality of historical pollution analysis result sets, constructing the cleaning analysis model of the spinning machine by taking the bound data set as a construction data set, and providing technical support for guaranteeing the stability of the model by combining with related historical data information.
Further, the constructing the spinning machine cleaning analysis model, the step S340 further includes:
s341: constructing a cleaning analysis model of the spinning machine based on a convolutional neural network;
s342: dividing and identifying the construction data set to obtain a training data set, a verification data set and a test data set, wherein identification information in the training data set, the verification data set and the test data set has an identification mapping relation with data in the construction data set;
s343: performing supervision training on the spinning machine cleaning and analyzing model by adopting the training data set until the output result of the spinning machine cleaning and analyzing model is converged or meets the requirement of preset accuracy rate;
s344: and verifying and testing the spinning machine cleaning analysis model by adopting the verification data set and the test data set, and if the spinning machine cleaning analysis model meets the preset accuracy requirement, obtaining the constructed spinning machine cleaning analysis model.
Specifically, the convolutional neural network is a machine learning network model, the spinning machine cleaning analysis model is constructed on the basis of the convolutional neural network, the constructed data set is divided and identified, the division and the identification are executed after the data binding is carried out on the plurality of historical first image information sets and the plurality of historical pollution analysis result sets, a training data set, a verification data set and a test data set are obtained, identification mapping relations exist between identification information of the training data set and the verification data set and between identification information of the test data set and the constructed data set, the identification mapping relations are one-to-one corresponding mapping relations, and the mapping relations are irreversible; inputting the training data set into an input training end of a convolutional neural network, performing supervised training on the cleaning analysis model of the spinning machine, wherein the supervision data of the supervised training is the verification data set, the output result of the cleaning analysis model of the spinning machine converges or meets a preset accuracy requirement, the preset accuracy requirement can be that a test data set is sequentially input into the convolutional neural network after the training is finished, whether the output of the model is consistent with the historical results in the plurality of historical first image information sets or not is determined, if the output of the model is consistent, the verification is passed, if the output of the model is inconsistent, the verification is not passed, the ratio of the number of times of passing the verification to the total number of times of verification is defined as the verification accuracy, the preset accuracy requirement can be set to 90%, if the verification accuracy is not less than 90%, the output result of the cleaning analysis model of the spinning machine reaches the preset accuracy requirement, and the output result convergence means that the output result of the model tends to be stable in convergence; and verifying and testing the spinning machine cleaning analysis model by respectively adopting the verification data set and the test data set, and if the spinning machine cleaning analysis model meets the preset accuracy requirement, obtaining the constructed spinning machine cleaning analysis model, and providing technical theoretical support for ensuring the reliability of the model.
S400: obtaining a pollution analysis result of the target spinning machine according to the output result;
s500: inputting the pollution analysis result into a cleaning strategy analysis model to obtain cleaning strategy information for cleaning the target spinning machine;
s600: adopting the cleaning strategy information to control the target spinning machine to clean, acquiring and acquiring image information of a plurality of positions in the target spinning machine after cleaning, and acquiring a second image information set;
specifically, the plurality of positions include a drawing zone and a spindle zone, the contamination analysis result includes a contamination analysis result of the drawing zone and the spindle zone, and the contamination analysis result of the target spinning machine is determined by comparison according to the output result.
Further, as shown in fig. 2, in the process of constructing the cleaning policy analysis model, step S500 in the method provided in the embodiment of the present application further includes:
s510: setting a plurality of corresponding cleaning strategy information according to the plurality of historical pollution analysis result sets, wherein the plurality of positions comprise a drafting zone and a spindle zone, and each historical pollution analysis result set comprises a historical drafting zone pollution analysis result and a historical spindle zone pollution analysis result;
s520: constructing a two-dimensional coordinate system based on the drafting zone pollution analysis result and the spindle zone pollution analysis result;
s530: inputting the historical pollution analysis result sets into the two-dimensional coordinate system to obtain a plurality of coordinate points;
s540: clustering and analyzing the plurality of coordinate points by adopting the plurality of corresponding cleaning strategy information to obtain a plurality of clustering results;
s550: and taking the plurality of clustering results and the two-dimensional coordinate system as the cleaning strategy analysis model.
Specifically, the cleaning strategy information includes, but is not limited to, a cleaning scheme and a cleaning instruction start execution time, the drafting zone and the spindle zone both belong to a structural area of the target spinning machine, the plurality of positions include a drafting zone and a spindle zone, each historical contamination analysis result set includes a historical drafting zone contamination analysis result and a historical spindle zone contamination analysis result, a plurality of historical drafting zone contamination analysis results and a plurality of historical spindle zone contamination analysis results are determined according to the plurality of historical contamination analysis result sets, a plurality of corresponding cleaning strategy information are set by combining structural area distribution information of the drafting zone and the spindle zone, a two-dimensional coordinate system is constructed by using the drafting zone contamination analysis results and the spindle zone contamination analysis results as coordinate characteristic data, the plurality of historical contamination analysis result sets are sequentially input into the two-dimensional coordinate system, a plurality of coordinate points are obtained, and abscissa and ordinate of the coordinate points respectively correspond to the drafting zone contamination analysis results and the spindle zone contamination analysis results; the method comprises the steps of adopting the corresponding cleaning strategy information to carry out clustering analysis on the coordinate points to obtain a plurality of clustering results, clustering the coordinate points (namely, pollution analysis results) which can be cleaned by using the same cleaning strategy information into one type by the clustering analysis, limiting the clustering results to the cleaned coordinate points corresponding to a certain cleaning strategy information, corresponding the clustering results to the cleaning strategy information one by one, correspondingly combining the clustering results and the two-dimensional coordinate system to obtain a cleaning strategy analysis model, and providing technical support for ensuring the cleaning strategy information of the cleaning strategy analysis model and the accuracy corresponding to the regional pollution analysis of the target spinning machine.
Explaining further, the abscissa and the ordinate of the coordinate point respectively correspond to a drafting zone pollution analysis result and a spindle zone pollution analysis result, and after the plurality of coordinate points are subjected to cluster analysis, if the pollution degree is determined to be low, a cleaning scheme corresponding to the cleaning strategy information can be cleaning by blowing air; if the pollution degree is determined to be high, the cleaning scheme corresponding to the cleaning strategy information can be cleaning through water washing. Optionally, other cleaning methods in the prior art can also be combined for cleaning.
Inputting the pollution analysis result into a cleaning strategy analysis model to obtain cleaning strategy information for cleaning the target spinning machine, wherein the cleaning strategy information for cleaning the target spinning machine is the cleaning scheme corresponding to the position area required to be cleaned in the current state of the target spinning machine and the position area; and controlling the target spinning machine to clean by adopting the cleaning strategy information, acquiring and acquiring image information of a plurality of positions in the target spinning machine through the image acquisition devices at the plurality of positions after cleaning is finished, acquiring a second image information set, acquiring the cleaning strategy information to clean and acquiring the cleaned image information, wherein the second image information set is consistent with the image distribution arrangement of the first image information set, and providing technical support for ensuring the stable execution of the cleaning strategy of the spinning machine.
S700: analyzing and evaluating the cleaning strategy information according to the first image information set and the second image information set to obtain an analysis result;
further, according to the first image information set and the second image information set, analyzing and evaluating the cleaning policy information, the step S700 further includes:
s710: acquiring a plurality of historical first image information sets and a plurality of historical second image information sets of the plurality of positions before and after the plurality of times of cleaning;
s720: setting different cleaning degree analysis results for the plurality of historical first image information sets and the plurality of historical second image information sets respectively;
s730: based on a convolutional neural network, adopting the plurality of historical first image information sets, the plurality of historical second image information sets and the different cleaning degree analysis results to construct and obtain a cleaning degree analysis model;
and S740: and inputting the first image information set and the second image information set into a cleaning degree analysis model to obtain the analysis result.
Specifically, a plurality of historical first image information sets and a plurality of historical second image information sets of the plurality of positions before and after a plurality of cleaning are acquired and obtained through the image acquisition devices of the plurality of positions, the historical first image information sets and the historical second image information sets correspond to a group of image information before and after cleaning, and the corresponding relation can be determined by combining image acquisition time; comparing the plurality of historical first image information sets with the plurality of historical second image information sets, and respectively setting different cleaning degree analysis results, wherein the cleaning degree analysis results are analysis results corresponding to the pollution degree improvement conditions before and after cleaning, and the analysis results can analyze and evaluate the adaptability of the cleaning strategy information and the pollution analysis results of the target spinning machine and the cleaning effect of the cleaning strategy information; constructing a cleaning degree analysis model based on a convolutional neural network, comparing the plurality of historical first image information sets, the plurality of historical second image information sets and the different cleaning degree analysis results, sequentially inputting the comparison results into the cleaning degree analysis model, performing supervision training and verification on the comparison results, and obtaining the cleaning degree analysis model after the output results tend to be in a stable state; and inputting the current first image information set and the second image information set into a cleaning degree analysis model, and outputting the analysis result by the cleaning degree analysis model to provide technical support for optimizing and updating model parameters.
S800: and updating the cleaning strategy analysis model according to the analysis result.
Further, as shown in fig. 3, the updating the cleaning policy analysis model according to the analysis result, where the step S800 includes:
s810: obtaining a cleaning degree analysis threshold value;
s820: judging whether the analysis result meets the cleaning degree analysis threshold value;
s830: if so, not updating the cleaning strategy analysis model;
s840: and if not, updating the cleaning strategy analysis model, wherein the updating comprises parameter updating and reconstruction of the cleaning strategy analysis model.
Specifically, the cleaning degree analysis threshold may be threshold data obtained by comprehensively determining the cleaning time and the cleaning degree, which simply indicates that too long cleaning time may not ensure the processing efficiency of the spinning machine, and too high cleaning degree may cause frequent switching of the spinning machine to start the cleaning program and also affect the processing efficiency of the spinning machine, and may be obtained by comprehensively determining the cleaning degree analysis threshold by the cleaning time and the cleaning degree; judging whether the analysis result meets the cleaning degree analysis threshold, if so, not updating the cleaning strategy analysis model; and if the analysis result does not meet the cleaning degree analysis threshold, updating the cleaning strategy analysis model, wherein the updating comprises parameter updating and reconstruction of the cleaning strategy analysis model, preferably reconstruction of the cleaning strategy analysis model, so as to obtain cleaning strategy information more suitable for the current target spinning and pollution conditions, and provide technical support for balancing the cleaning degree and the processing efficiency of the spinning machine and ensuring the reasonability of an intelligent control scheme.
In summary, the intelligent control method and system for the spinning machine provided by the present application have the following technical effects:
the method comprises the steps of adopting a target spinning machine to spin according to a preset spinning flow, obtaining a first image information set after finishing at least one spinning work of the preset spinning flow, inputting the first image information set into a pre-constructed spinning machine cleaning analysis model, obtaining an output result, obtaining a pollution analysis result of the target spinning machine, inputting a cleaning strategy analysis model, obtaining cleaning strategy information, controlling the target spinning machine to clean, acquiring image information of a plurality of positions in the target spinning machine, obtaining a second image information set, analyzing and evaluating the cleaning strategy information according to the first image information set and the second image information set, obtaining an analysis result, and updating the cleaning strategy analysis model. The application provides the intelligent control method and system for the spinning machine, and achieves the technical effects of accurately and timely generating a spinning machine cleaning strategy and intelligently controlling the cleaning of the spinning machine.
The method comprises the steps of acquiring a preset time span, acquiring image information of a plurality of positions in a target spinning machine within the preset time span in historical time, acquiring a plurality of historical first image information sets, acquiring pollution analysis results of a plurality of positions in the preset time span in the historical time, acquiring a plurality of historical pollution analysis result sets, taking the plurality of historical first image information sets and the plurality of historical pollution analysis result sets as construction data sets, constructing a spinning machine cleaning analysis model, and providing technical support for guaranteeing the stability of the model by combining related historical data information.
The method comprises the steps of setting a plurality of corresponding cleaning strategy information according to a plurality of historical pollution analysis result sets, constructing a two-dimensional coordinate system based on a drafting zone pollution analysis result and a spindle zone pollution analysis result, inputting the plurality of historical pollution analysis result sets to obtain a plurality of coordinate points, carrying out cluster analysis on the plurality of coordinate points by adopting the plurality of corresponding cleaning strategy information to obtain a plurality of cluster results, and determining a cleaning strategy analysis model by combining the two-dimensional coordinate system. And providing technical support for ensuring the accuracy of the cleaning strategy information of the cleaning strategy analysis model corresponding to the regional pollution analysis of the target spinning machine.
Example two
Based on the same inventive concept as one of the intelligent control methods for the spinning machine in the foregoing embodiment, as shown in fig. 4, the present application provides an intelligent control system for a spinning machine, wherein the system includes:
a spinning flow executing unit 11, where the spinning flow executing unit 11 is configured to perform spinning according to a preset spinning flow by using a target spinning machine, where the preset spinning flow includes a plurality of spinning steps;
the image information acquisition unit 12 is configured to acquire and obtain image information of multiple positions in the target spinning machine after the target spinning machine completes at least one spinning work of the preset spinning flow, and obtain a first image information set;
an output result obtaining unit 13, wherein the output result obtaining unit 13 is configured to input the first image information set into a pre-constructed spinning machine cleaning analysis model to obtain an output result;
a contamination analysis result acquisition unit 14, the contamination analysis result acquisition unit 14 being configured to obtain a contamination analysis result of the target spinning machine according to the output result;
a cleaning strategy information obtaining unit 15, wherein the cleaning strategy information obtaining unit 15 is configured to input the pollution analysis result into a cleaning strategy analysis model, and obtain cleaning strategy information for cleaning the target spinning machine;
the image information acquisition unit 16 is used for controlling the target spinning machine to clean by adopting the cleaning strategy information, acquiring and acquiring image information of a plurality of positions in the target spinning machine after cleaning, and acquiring a second image information set;
an analysis result obtaining unit 17, where the analysis result obtaining unit 17 is configured to analyze and evaluate the cleaning policy information according to the first image information set and the second image information set to obtain an analysis result;
a data updating unit 18, where the data updating unit 18 is configured to update the cleaning strategy analysis model according to the analysis result.
Further, the system comprises:
the device comprises a preset time span acquisition unit, a time delay unit and a time delay unit, wherein the preset time span acquisition unit is used for acquiring a preset time span;
the historical image information acquisition unit is used for acquiring image information of a plurality of positions in the target spinning machine within the preset time span in historical time to obtain a plurality of historical first image information sets;
a historical pollution analysis result obtaining unit, wherein the historical pollution analysis result obtaining unit is used for collecting pollution analysis results of a plurality of positions in the target spinning machine within the preset time span in historical time to obtain a plurality of historical pollution analysis result sets;
a model construction unit for constructing the spinning machine cleaning analysis model using the plurality of historical first image information sets and the plurality of historical contamination analysis result sets as construction data sets.
Further, the system comprises:
the cleaning analysis model building unit is used for building the spinning machine cleaning analysis model based on a convolutional neural network;
the dividing and identifying unit is used for dividing and identifying the constructed data set to obtain a training data set, a verification data set and a test data set, wherein identification information in the training data set, the verification data set and the test data set has an identification mapping relation with data in the constructed data set;
the supervision training unit is used for adopting the training data set to supervise and train the cleaning and analyzing model of the spinning machine until the output result of the cleaning and analyzing model of the spinning machine is converged or meets the requirement of preset accuracy rate;
and the verification and test unit is used for verifying and testing the spinning machine cleaning analysis model by adopting the verification data set and the test data set, and if the spinning machine cleaning analysis model meets the preset accuracy requirement, the constructed spinning machine cleaning analysis model is obtained.
Further, the system comprises:
a pollution analysis result acquisition unit, configured to set a plurality of corresponding cleaning strategy information according to the plurality of historical pollution analysis result sets, where the plurality of positions include a drafting zone and an ingot zone, and each historical pollution analysis result set includes a historical drafting zone pollution analysis result and a historical ingot zone pollution analysis result;
a two-dimensional coordinate system construction unit for constructing a two-dimensional coordinate system based on the draft zone pollution analysis result and the spindle zone pollution analysis result;
a coordinate point obtaining unit, configured to input the plurality of historical pollution analysis result sets into the two-dimensional coordinate system, and obtain a plurality of coordinate points;
the cluster analysis unit is used for carrying out cluster analysis on the plurality of coordinate points by adopting the plurality of corresponding cleaning strategy information to obtain a plurality of cluster results;
a model determination unit for taking the plurality of clustering results and the two-dimensional coordinate system as the cleaning strategy analysis model.
Further, the system comprises:
a pollution analysis result obtaining unit for obtaining a pollution analysis result of the drafting zone and a pollution analysis result of the spindle zone according to the pollution analysis result;
a coordinate point corresponding unit, which is used for inputting the pollution analysis result of the drafting zone and the pollution analysis result of the spindle zone into the cleaning strategy analysis model to obtain a corresponding current coordinate point;
a clustering result obtaining unit, configured to obtain a clustering result corresponding to the current coordinate point, and obtain a current clustering result;
and the cleaning strategy information obtaining unit is used for obtaining the cleaning strategy information corresponding to the current clustering result.
Further, the system comprises:
the image information acquisition unit is used for acquiring and acquiring a plurality of historical first image information sets and a plurality of historical second image information sets of the plurality of positions before and after a plurality of times of cleaning;
a cleaning degree analysis unit for setting different cleaning degree analysis results for the plurality of historical first image information sets and the plurality of historical second image information sets respectively;
a cleaning degree analysis model obtaining unit, configured to construct and obtain a cleaning degree analysis model by using the plurality of historical first image information sets, the plurality of historical second image information sets, and the different cleaning degree analysis results based on a convolutional neural network;
an analysis result obtaining unit, configured to input the first image information set and the second image information set into a cleaning degree analysis model, and obtain the analysis result.
Further, the system comprises:
an analysis threshold value obtaining unit for obtaining a cleaning degree analysis threshold value;
a threshold judgment unit configured to judge whether the analysis result satisfies the cleaning degree analysis threshold;
the model updating judging unit is used for not updating the cleaning strategy analysis model if the model updating judging unit is used for judging whether the cleaning strategy analysis model is updated or not;
and the model updating unit is used for updating the cleaning strategy analysis model if the cleaning strategy analysis model is not updated, and the updating comprises parameter updating and reconstruction of the cleaning strategy analysis model.
The specification and drawings are merely illustrative of the present application, and various modifications and combinations can be made thereto without departing from the spirit and scope of the application. Such modifications and variations of the present application are within the scope of the claims of the present application and their equivalents, and it is intended that the present application include such modifications and variations.
Claims (4)
1. An intelligent control method for a spinning machine, characterized in that the method comprises:
spinning by adopting a target spinning machine according to a preset spinning process, wherein the preset spinning process comprises a plurality of spinning steps;
after the target spinning machine finishes at least one spinning work of the preset spinning flow, acquiring and acquiring image information of a plurality of positions in the target spinning machine to obtain a first image information set;
inputting the first image information set into a pre-constructed spinning machine cleaning analysis model to obtain an output result;
obtaining a pollution analysis result of the target spinning machine according to the output result;
inputting the pollution analysis result into a cleaning strategy analysis model to obtain cleaning strategy information for cleaning the target spinning machine;
controlling the target spinning machine to clean by adopting the cleaning strategy information, and acquiring image information of a plurality of positions in the target spinning machine after cleaning to acquire a second image information set;
analyzing and evaluating the cleaning strategy information according to the first image information set and the second image information set to obtain an analysis result;
updating the cleaning strategy analysis model according to the analysis result;
the construction process of the spinning machine cleaning analysis model comprises the following steps:
obtaining a preset time span;
acquiring image information of a plurality of positions in the target spinning machine within the preset time span in historical time to obtain a plurality of historical first image information sets;
collecting pollution analysis results of a plurality of positions in the target spinning machine within the preset time span in historical time to obtain a plurality of historical pollution analysis result sets;
taking the plurality of historical first image information sets and the plurality of historical pollution analysis result sets as construction data sets to construct the spinning machine cleaning analysis model;
constructing a cleaning analysis model of the spinning machine based on a convolutional neural network;
dividing and identifying the constructed data set to obtain a training data set, a verification data set and a test data set, wherein identification information in the training data set, the verification data set and the test data set and data in the constructed data set have an identification mapping relation;
performing supervision training on the spinning machine cleaning and analyzing model by adopting the training data set until the output result of the spinning machine cleaning and analyzing model is converged or meets the requirement of preset accuracy rate;
verifying and testing the spinning machine cleaning analysis model by adopting the verification data set and the test data set, and if the spinning machine cleaning analysis model meets the preset accuracy requirement, obtaining the constructed spinning machine cleaning analysis model;
the construction process of the cleaning strategy analysis model comprises the following steps:
setting a plurality of corresponding cleaning strategy information according to the plurality of historical pollution analysis result sets, wherein the plurality of positions comprise a drafting zone and a stator zone, and each historical pollution analysis result set comprises a historical drafting zone pollution analysis result and a historical stator zone pollution analysis result;
constructing a two-dimensional coordinate system based on the drafting zone pollution analysis result and the stator zone pollution analysis result;
inputting the historical pollution analysis result sets into the two-dimensional coordinate system to obtain a plurality of coordinate points;
clustering analysis is carried out on the plurality of coordinate points by adopting the plurality of corresponding cleaning strategy information to obtain a plurality of clustering results;
taking the plurality of clustering results and the two-dimensional coordinate system as the cleaning strategy analysis model;
inputting the pollution analysis result into a cleaning strategy analysis model to obtain cleaning strategy information for cleaning the target spinning machine, wherein the cleaning strategy information comprises the following steps:
obtaining a drafting zone pollution analysis result and a stator zone pollution analysis result according to the pollution analysis result;
inputting the analysis result of the pollution of the drafting zone and the analysis result of the pollution of the stator zone into the cleaning strategy analysis model to obtain a corresponding current coordinate point;
acquiring a clustering result corresponding to the current coordinate point, and acquiring a current clustering result;
and acquiring cleaning strategy information corresponding to the current clustering result.
2. The method of claim 1, wherein analyzing and evaluating the cleaning strategy information based on the first set of image information and the second set of image information comprises:
acquiring a plurality of historical first image information sets and a plurality of historical second image information sets of the plurality of positions before and after the plurality of times of cleaning;
setting different cleaning degree analysis results for the plurality of historical first image information sets and the plurality of historical second image information sets respectively;
based on a convolutional neural network, adopting the plurality of historical first image information sets, the plurality of historical second image information sets and the different cleaning degree analysis results to construct and obtain a cleaning degree analysis model;
and inputting the first image information set and the second image information set into a cleaning degree analysis model to obtain the analysis result.
3. The method of claim 1, wherein the updating the cleaning strategy analysis model according to the analysis result comprises:
obtaining a cleaning degree analysis threshold value;
judging whether the analysis result meets the cleaning degree analysis threshold value;
if so, not updating the cleaning strategy analysis model;
and if not, updating the cleaning strategy analysis model, wherein the updating comprises parameter updating and reconstruction of the cleaning strategy analysis model.
4. An intelligent control system for spinning machines, characterized in that it is adapted to perform the method according to claim 1, and in that it comprises:
the spinning process executing unit is used for spinning by adopting a target spinning machine according to a preset spinning process, wherein the preset spinning process comprises a plurality of spinning steps;
the image information acquisition unit is used for acquiring and acquiring image information of a plurality of positions in the target spinning machine after the target spinning machine finishes at least one spinning work of the preset spinning flow, so as to acquire a first image information set;
the output result acquisition unit is used for inputting the first image information set into a pre-constructed spinning machine cleaning analysis model to obtain an output result;
a pollution analysis result acquisition unit for acquiring a pollution analysis result of the target spinning machine according to the output result;
a cleaning strategy information obtaining unit, configured to input the pollution analysis result into a cleaning strategy analysis model, and obtain cleaning strategy information for cleaning the target spinning machine;
the image information acquisition unit is used for adopting the cleaning strategy information to control the target spinning machine to clean, acquiring and acquiring image information of a plurality of positions in the target spinning machine after cleaning, and acquiring a second image information set;
an analysis result obtaining unit, configured to analyze and evaluate the cleaning policy information according to the first image information set and the second image information set, and obtain an analysis result;
and the data updating unit is used for updating the cleaning strategy analysis model according to the analysis result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210986970.8A CN115058807B (en) | 2022-08-17 | 2022-08-17 | Intelligent control method and system for spinning machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210986970.8A CN115058807B (en) | 2022-08-17 | 2022-08-17 | Intelligent control method and system for spinning machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115058807A CN115058807A (en) | 2022-09-16 |
| CN115058807B true CN115058807B (en) | 2022-12-20 |
Family
ID=83208357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210986970.8A Active CN115058807B (en) | 2022-08-17 | 2022-08-17 | Intelligent control method and system for spinning machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115058807B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115933574B (en) * | 2023-01-09 | 2024-04-26 | 浙江聚优非织造材料科技有限公司 | Intelligent control method and system for producing non-woven product |
| CN115882758B (en) * | 2023-02-20 | 2023-05-30 | 江苏卓鹏智能机电有限公司 | Motor variable speed control method and system for colored spun yarns |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006291415A (en) * | 2005-04-14 | 2006-10-26 | Toyota Industries Corp | Method for controlling machine unit in spinning machine |
| CN101365833B (en) * | 2006-01-17 | 2012-08-29 | 欧瑞康纺织有限及两合公司 | Ring spinning frame comprising a travelling cleaner and a lint removing mechanism |
| CN103911702A (en) * | 2014-01-28 | 2014-07-09 | 韵升控股集团有限公司 | Yarn break monitoring device |
| EP3904572B1 (en) * | 2020-04-30 | 2022-04-06 | Maschinenfabrik Rieter AG | Device and method for detecting a fault in a spinning mill and for estimating one or more sources of the fault |
| CN112651320A (en) * | 2020-12-21 | 2021-04-13 | 南京掘物网络信息技术有限公司 | Intelligent power control method for dust collector of intelligent textile workshop |
| CN114687018B (en) * | 2020-12-30 | 2023-05-12 | 苏州多道自动化科技有限公司 | Self-cleaning composite yarn component adding device and intelligent cleaning method |
| CN114994289B (en) * | 2022-08-01 | 2022-10-28 | 江苏卓鹏智能机电有限公司 | Spinning product quality detection method and system |
-
2022
- 2022-08-17 CN CN202210986970.8A patent/CN115058807B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115058807A (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115058807B (en) | Intelligent control method and system for spinning machine | |
| CN115660509B (en) | Factory building control method and system based on digital twin technology | |
| CN109086999A (en) | Filling production lines remote data acquisition analysis system and its exception analysis method | |
| CN110716512A (en) | Environmental protection equipment performance prediction method based on coal-fired power plant operation data | |
| CN111475921A (en) | Tool residual life prediction method based on edge calculation and L STM network | |
| WO2018221948A1 (en) | Method for process management through big data analysis | |
| Li et al. | Bottleneck detection of manufacturing systems using data driven method | |
| CN113128764A (en) | Generation of tobacco dryer outlet water content prediction model and regulation and control method based on same | |
| CN115576284A (en) | Clothing workshop intelligent management method and system | |
| CN116216819B (en) | Intelligent water affair data analysis management system and method | |
| CN112904148A (en) | Intelligent cable operation monitoring system, method and device | |
| CN117118781A (en) | Intelligent industrial gateway design method and device | |
| CN116992346A (en) | Enterprise production data processing system based on artificial intelligence big data analysis | |
| CN109299080B (en) | Cleaning method for power production operation data and computing equipment | |
| WO2021111936A1 (en) | Prediction system, information processing device, and information processing program | |
| CN112926756A (en) | Application method of AI (AI) maintenance knowledge base of central air-conditioning equipment | |
| CN105072056A (en) | Switch automatic inspection system and inspection method thereof | |
| CN111190944A (en) | Data mining method and system | |
| CN115357637A (en) | Method for extracting sample data of specific working condition of industrial equipment | |
| CN113139332A (en) | Automatic model construction method, device and equipment | |
| CN113220771B (en) | Method for automatically acquiring operation condition based on historical data of power plant | |
| Maier et al. | Automated generation of timing models in distributed production plants | |
| CN116542502B (en) | Building construction period optimization system and method based on convolutional neural network | |
| CN114706743B (en) | Comprehensive evaluation method supporting real-time evaluation | |
| WO2022176375A1 (en) | Prediction system, information processing device, and information processing 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 |