CN117252284B - Intelligent screening method and system for industrial silk oil agent raw materials - Google Patents

Intelligent screening method and system for industrial silk oil agent raw materials Download PDF

Info

Publication number
CN117252284B
CN117252284B CN202310038794.XA CN202310038794A CN117252284B CN 117252284 B CN117252284 B CN 117252284B CN 202310038794 A CN202310038794 A CN 202310038794A CN 117252284 B CN117252284 B CN 117252284B
Authority
CN
China
Prior art keywords
friction coefficient
spinning
data
information
industrial yarn
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
Application number
CN202310038794.XA
Other languages
Chinese (zh)
Other versions
CN117252284A (en
Inventor
孙冈剑
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiaxing Hongdian Application Technology Co ltd
Original Assignee
Jiaxing Hongdian Application Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiaxing Hongdian Application Technology Co ltd filed Critical Jiaxing Hongdian Application Technology Co ltd
Priority to CN202310038794.XA priority Critical patent/CN117252284B/en
Publication of CN117252284A publication Critical patent/CN117252284A/en
Application granted granted Critical
Publication of CN117252284B publication Critical patent/CN117252284B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0499—Feedforward networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • Development Economics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Primary Health Care (AREA)
  • Manufacturing & Machinery (AREA)
  • Educational Administration (AREA)
  • Spinning Methods And Devices For Manufacturing Artificial Fibers (AREA)

Abstract

The invention relates to the technical field of intelligent manufacturing, and provides an intelligent screening method and system for industrial silk oil raw materials, wherein the intelligent screening method comprises the following steps: the basic information of the spinning processing technology is obtained and comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information; obtaining a screening evaluation index and an evaluation index expected value, wherein the screening evaluation index comprises a first type friction coefficient, a second type friction coefficient and a spinning static electricity eliminating amount; generating first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static elimination amount prediction data according to basic information of a spinning processing technology; judging whether the expected value of the evaluation index is met or not; if the formula information is not satisfied, optimizing the formula information of the industrial silk oil, generating an optimization result of raw material components of the industrial silk oil, and screening raw materials. Solves the technical problem of lack of objectivity and weakness caused by more subjective screening of industrial silk oil raw materials in the prior art.

Description

Intelligent screening method and system for industrial silk oil agent raw materials
Technical Field
The invention relates to the technical field of intelligent manufacturing, in particular to an intelligent screening method and system for industrial silk oil raw materials.
Background
In order to reduce friction force and electrostatic force between spinning, the spinning process mostly adopts industrial silk oiling agent to realize the processing purpose, and the application of the industrial silk oiling agent is common nowadays. Because different industrial yarn oil agents in the spinning process have great influence on the quality of a spinning finished product, the screening of industrial yarn oil agent raw materials is a key focus of the spinning process.
At present, the screening of industrial silk oil raw materials is selected according to experience, and the screening is highly dependent on the professional level of raw material selection personnel, so that subjectivity is strong and objectivity is weak.
In the prior art, the screening of industrial silk oil raw materials is subjective, so that the technical problem of lack of objectivity and weakness exists.
Disclosure of Invention
The application provides an intelligent screening method and system for industrial silk oil raw materials, and aims to solve the technical problem that the prior art lacks objectively and weakly because of subjective screening of the industrial silk oil raw materials.
In view of the above problems, embodiments of the present application provide an intelligent screening method and system for industrial silk oil raw materials.
In a first aspect of the disclosure, an intelligent screening method for industrial silk oil raw materials is provided, where the method includes: obtaining basic information of a spinning process, wherein the basic information of the spinning process comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information; obtaining a screening evaluation index and an evaluation index expected value, wherein the screening evaluation index comprises a first type friction coefficient, a second type friction coefficient and a spinning static electricity elimination amount, wherein the first type friction coefficient represents a friction coefficient between spinning, and the second type friction coefficient represents a friction coefficient between spinning and equipment; traversing the first type friction coefficient, the second type friction coefficient and the spinning static electricity eliminating amount to predict according to the industrial yarn oil agent formula information, the spinning processing equipment information, the spinning processing condition information and the spinning type information, and generating first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static electricity eliminating amount prediction data; judging whether the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data meet the expected value of the evaluation index or not; if the information of the industrial yarn oil formulation does not meet the requirement, a first optimization instruction is acquired, and the information of the industrial yarn oil formulation is optimized to generate an industrial yarn oil formulation optimization result; obtaining an industrial yarn oil raw material component optimization result according to the industrial yarn oil formula optimization result; and (5) screening the raw materials according to the optimized result of the raw material components of the industrial yarn oiling agent.
In another aspect of the disclosure, an intelligent screening system for industrial silk oil raw materials is provided, wherein the intelligent screening system comprises: the first data acquisition module is used for acquiring basic information of a spinning process, wherein the basic information of the spinning process comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information; the second data acquisition module is used for acquiring screening evaluation indexes and evaluation index expected values, wherein the screening evaluation indexes comprise a first type friction coefficient, a second type friction coefficient and a spinning static electricity elimination amount, the first type friction coefficient represents a friction coefficient between spinning, and the second type friction coefficient represents a friction coefficient between spinning and equipment; the silk oil performance prediction module is used for traversing the first type friction coefficient, the second type friction coefficient and the spinning static electricity elimination amount to predict according to the industrial silk oil formula information, the spinning processing equipment information, the spinning processing condition information and the spinning type information, so as to generate first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static electricity elimination amount prediction data; the data judging module is used for judging whether the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data meet the expected value of the evaluation index or not; the formula optimization module is used for acquiring a first optimization instruction if the formula information is not satisfied, optimizing the industrial yarn oiling agent formula information and generating an industrial yarn oiling agent formula optimization result; the third data acquisition module is used for acquiring an industrial yarn oil raw material component optimization result according to the industrial yarn oil formula optimization result; and the task execution module is used for screening the raw materials according to the optimized result of the raw material components of the industrial yarn oiling agent.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
the basic information according to the spinning processing technology is adopted, so that the information such as the formula, processing equipment, processing conditions, spinning type and the like of the industrial yarn oiling agent to be used are determined; setting screening evaluation indexes of the industrial yarn oiling agent formula and evaluation index expected values of all indexes; sequentially evaluating screening evaluation indexes according to the information of an industrial yarn oiling agent formula, processing equipment, processing conditions, spinning types and the like to be used to obtain first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static elimination amount prediction data; further judging whether the data obtained by evaluation accords with the expected value of the evaluation index of each index; if the information is not met, optimizing the industrial silk oil formula information to obtain an industrial silk oil formula optimization result; and finally, according to the technical scheme of raw material screening according to the optimization result of the industrial yarn oil formulation, the performance data of the current industrial yarn oil formulation is determined through the automatic analysis of screening evaluation indexes based on the basic information of the spinning processing technology, if the performance data does not accord with the expected value, the raw material screening is carried out according to the optimized formulation after the optimization algorithm is optimized, and the technical effects of high automation degree and stronger objectivity are achieved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic diagram of a possible flow chart of an intelligent screening method for industrial silk oil raw materials according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a possible flow chart for predicting the performance of an industrial yarn oil agent in an intelligent screening method of industrial yarn oil agent raw materials according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a possible flow chart of training a performance prediction model in an intelligent screening method of industrial silk oil raw materials according to an embodiment of the present application;
fig. 4 is a schematic diagram of a possible structure of an intelligent screening system for industrial yarn oil materials according to an embodiment of the present application.
Reference numerals illustrate: the device comprises a first data acquisition module 11, a second data acquisition module 12, a silk oil performance prediction module 13, a data judgment module 14, a formula optimization module 15, a third data acquisition module 16 and a task execution module 17.
Detailed Description
The technical scheme provided by the application has the following overall thought:
the embodiment of the application provides an intelligent screening method and system for industrial silk oil raw materials, and aims to solve the technical problem that the prior art lacks objectively weaker due to the fact that the screening of the industrial silk oil raw materials is subjective. The performance data of the current industrial silk finish formula is determined through automatic analysis of screening evaluation indexes based on basic information of a spinning processing technology, if the performance data does not accord with an expected value, raw material screening is carried out according to the optimized formula after optimization according to an optimization algorithm, and the technical effects of high automation degree and stronger objectivity are achieved.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides an intelligent screening method for industrial silk oil raw materials, including the steps of:
s100: obtaining basic information of a spinning process, wherein the basic information of the spinning process comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information;
Specifically, the spinning process base information is any type of spinning process parameter base information including, but not limited to: the type of spinning raw materials, the content of spinning raw materials, the type of additives, the content of additives and the preparation process of the additives which are prepared before processing; the processing conditions involved in any one processing link in the processing process include, for example: processing equipment, control parameters of the processing equipment, processing environment temperature, processing environment humidity, processing environment pH value and the like.
The basic information of the spinning process can be used for extracting the formula information, the spinning equipment information, the spinning condition information and the spinning type information of the industrial yarn oil. The industrial yarn oil agent can be regarded as a certain additive, so that the industrial yarn oil agent formula information at least comprises data such as the type of the industrial yarn oil agent component, the content of the industrial yarn oil agent component, the preparation flow of the industrial yarn oil agent and the like; the spinning processing equipment information at least comprises processing equipment model information, control parameters of processing equipment and other data of each link; the spinning processing condition information at least comprises processing environment temperature, processing environment humidity, processing environment pH value and other condition parameters of each processing link; the spinning type information is the conventional classification result of spinning.
Other basic information of spinning process is not adopted because the performance evaluation effect of the industrial yarn oiling agent formula information is smaller, and the basic information of the spinning process is set to be a fixed value by staff and is regarded as a constant state. The industrial yarn oiling agent formula information, the spinning processing equipment information, the spinning processing condition information and the spinning type information are set to be in a state to be responded, and can be used as reference data for predicting the performance of the industrial yarn oiling agent in the later step, so that the method is convenient to call efficiently.
S200: obtaining a screening evaluation index and an evaluation index expected value, wherein the screening evaluation index comprises a first type friction coefficient, a second type friction coefficient and a spinning static electricity elimination amount, wherein the first type friction coefficient represents a friction coefficient between spinning, and the second type friction coefficient represents a friction coefficient between spinning and equipment;
specifically, screening and evaluating indexes are index dimensions which are customized in advance and used for evaluating the performance of the industrial wire oil agent; the expected value of the evaluation index is an ideal interval of each index which is in one-to-one correspondence with the screening evaluation index and is custom-set based on the production scene and the spinning type.
Preferably, the industrial yarn oil is mainly used for reducing friction force between spinning yarns, friction force between spinning yarns and processing equipment and static electricity of spinning yarns, so that the set screening evaluation indexes at least comprise a first type friction coefficient representing friction coefficient between spinning yarns, a second type friction coefficient representing friction coefficient between spinning yarns and equipment and static electricity eliminating quantity of spinning yarns. Further, the method for calculating the static electricity eliminating amount of spinning is preferably as follows: and counting the static electricity quantity of spinning without using the industrial yarn oiling agent for a plurality of times, calculating an average value, setting the average value as an initial static electricity quantity, counting the static electricity quantity of spinning with using the industrial yarn oiling agent, setting the static electricity quantity after treatment, and subtracting the initial static electricity quantity from the initial static electricity quantity to obtain a static electricity elimination quantity.
Determining target parameters of industrial yarn oiling agent evaluation by setting screening evaluation indexes; and determining the qualification condition of the industrial yarn oil formulation by evaluating the expected value of the index, facilitating the performance evaluation of the industrial yarn oil in a later step, judging the qualification, setting the industrial yarn oil in a state to be responded, and waiting for the later step to be used.
S300: traversing the first type friction coefficient, the second type friction coefficient and the spinning static electricity eliminating amount to predict according to the industrial yarn oil agent formula information, the spinning processing equipment information, the spinning processing condition information and the spinning type information, and generating first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static electricity eliminating amount prediction data;
further, as shown in fig. 2, based on the formula information according to the industrial yarn oil agent, the spinning processing equipment information, the spinning processing condition information and the spinning type information, the first type friction coefficient, the second type friction coefficient and the spin static electricity eliminating amount are traversed to predict, and first type friction coefficient prediction data, second type friction coefficient prediction data and spin static electricity eliminating amount prediction data are generated, and step S300 includes the steps of:
S310: training a performance prediction model according to the spinning processing equipment information, the spinning processing condition information and the spinning type information, wherein the performance prediction model comprises a first type friction coefficient prediction layer, a second type friction coefficient prediction layer and a spinning static elimination amount prediction layer;
s320: disassembling the industrial yarn oil formula information to obtain industrial yarn oil formula component information and industrial yarn oil formula preparation information;
s330: inputting the component information of the industrial yarn oil formulation and the preparation information of the industrial yarn oil formulation into the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static electricity elimination amount prediction layer to obtain the first type friction coefficient prediction data, and the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data.
Further, as shown in fig. 3, based on the information of the spinning processing equipment, the information of the spinning processing conditions and the information of the spinning type, a performance prediction model is trained, and step S310 includes the steps of:
s311: frequent excavation is carried out by taking the spinning processing equipment information, the spinning processing condition information and the spinning type information as screening scene parameters, so as to generate industrial yarn oil formula component record data, industrial yarn oil formula preparation process record data, first type friction coefficient record data, second type friction coefficient record data and spinning static elimination quantity record data;
S312: training the first type friction coefficient prediction layer according to the industrial yarn oil formula component recording data, the industrial yarn oil formula preparation process recording data and the first type friction coefficient recording data;
s313: training the second-type friction coefficient prediction layer according to the industrial yarn oil formula component recording data, the industrial yarn oil formula preparation process recording data and the second-type friction coefficient recording data;
s314: training the spinning static elimination amount prediction layer according to the recording data of the components of the industrial yarn oil formula, the recording data of the preparation process of the industrial yarn oil formula and the recording data of the spinning static elimination amount;
s315: and combining the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static elimination amount prediction layer to generate the performance prediction model.
Further, the step S311 includes the steps of:
S3111: taking the spinning processing equipment information, the spinning processing condition information and the spinning type information as screening scene parameters to acquire spinning processing big data, and acquiring the industrial yarn oiling agent formula component recording data and the industrial yarn oiling agent formula preparation process recording data;
s3112: traversing the industrial yarn oiling agent formula component record data and the industrial yarn oiling agent formula preparation process record data according to the spinning processing equipment information, the spinning processing condition information and the spinning type information to acquire industrial yarn oiling agent performance parameters, and acquiring first type friction coefficient log data, second type friction coefficient log data and spinning static elimination amount log data;
s3113: and traversing the first type friction coefficient log data, and performing frequent analysis on the second type friction coefficient log data and the spinning static elimination amount log data to obtain the first type friction coefficient record data, the second type friction coefficient record data and the spinning static elimination amount record data.
Further, the step S3113 includes steps of:
S31131: traversing the first type friction coefficient log data to obtain a first support degree, wherein the first support degree represents the triggering frequency duty ratio of any one of the first type friction coefficient log data;
s31132: when the first support degree meets a first support degree threshold value, adding the first support degree threshold value into the first type friction coefficient record data;
s31133: traversing the second-type friction coefficient log data to obtain a second support degree, wherein the second support degree represents the triggering frequency duty ratio of any one of the second-type friction coefficient log data;
s31134: when the second support degree meets a second support degree threshold value, adding the second support degree threshold value into the second type friction coefficient record data;
s31135: traversing the spinning static elimination amount log data to obtain a third support degree, wherein the third support degree represents the triggering frequency duty ratio of any spinning static elimination amount log data;
s31136: and when the third support degree meets a third support degree threshold value, adding the third support degree threshold value into the spinning static electricity eliminating quantity record data.
Specifically, after basic information of spinning process such as industrial yarn oil formulation information, spinning processing equipment information, spinning processing condition information and spinning type information and performance evaluation target parameters such as a first type friction coefficient, a second type friction coefficient and a spinning static electricity eliminating amount are determined, performance prediction is sequentially performed on the first type friction coefficient, the second type friction coefficient and the spinning static electricity eliminating amount according to the industrial yarn oil formulation information, the spinning processing equipment information, the spinning processing condition information and the spinning type information, so that first type friction coefficient prediction data representing a first type friction coefficient evaluation result, second type friction coefficient prediction data representing a second type friction coefficient evaluation result and spinning static electricity eliminating amount prediction data representing a spinning static electricity eliminating amount evaluation result are obtained.
Preferably, since the actual process is nonlinear and complex, the evaluation of the first type of friction coefficient, the second type of friction coefficient and the spin static elimination amount is achieved using an intelligent model process constructed based on machine learning, specifically as follows:
the performance prediction model refers to an intelligent model for evaluating the first type friction coefficient, the second type friction coefficient, and the spin static electricity eliminating amount. Further, since the first type friction coefficient, the second type friction coefficient and the spin static electricity eliminating amount are three types of data with larger difference, the performance prediction model comprises a first type friction coefficient prediction layer, a second type friction coefficient prediction layer and a spin static electricity eliminating amount prediction layer, wherein the first type friction coefficient prediction layer is used for evaluating the first type friction coefficient, the second type friction coefficient prediction layer is used for evaluating the second type friction coefficient, the spin static electricity eliminating amount prediction layer is used for evaluating the spin static electricity eliminating amount, and preferably, the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spin static electricity eliminating amount prediction layer are distributed in the performance prediction model as three parallel independent processing nodes, and can directly receive input data of an input layer of the performance prediction model for independent processing and output the input data to an output layer of the performance prediction model.
The industrial yarn oil formula component information and the industrial yarn oil formula preparation information refer to formula data extracted from the industrial yarn oil formula information, wherein the industrial yarn oil formula component information comprises industrial yarn oil formula component type information and industrial yarn oil formula component content information, and the industrial yarn oil formula preparation information refers to the preparation process of the industrial yarn oil and is the data to be called.
After the component information of the industrial yarn oil formulation and the preparation information of the industrial yarn oil formulation are input into a performance prediction model, the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static electricity eliminating amount prediction layer process the component information of the industrial yarn oil formulation and the preparation information of the industrial yarn oil formulation at the same time, and generate first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static electricity eliminating amount prediction data, so that reference data is provided for the later screening of the industrial yarn oil.
Preferably, the performance prediction model is trained based on a BP neural network, the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static elimination amount prediction layer are respectively BP neural network topology results, and serve as three parallel nodes of the performance prediction model, and the preferred training process is as follows:
A first step of: collecting model training data:
frequent item mining is carried out by taking spinning processing equipment information, spinning processing condition information and the spinning type information as screening scene parameters, so as to generate industrial yarn oil formula component record data, industrial yarn oil formula preparation process record data, first type friction coefficient record data, second type friction coefficient record data and spinning static electricity elimination amount record data, wherein the screening scene parameters refer to fixed and unchanged data mining conditions, and therefore a plurality of processing records are obtained, and any one processing record data comprises one-to-one correspondence: recording data of components of the industrial yarn oil formula, recording data of the industrial yarn oil formula preparation process, recording data of a first type friction coefficient, recording data of a second type friction coefficient and recording data of a spinning static elimination amount.
The above data acquisition process can be further refined into:
the spinning processing equipment information, the spinning processing condition information and the spinning type information are taken as screening scene parameters to acquire spinning processing big data, so that industrial yarn oil formula component recording data and industrial yarn oil formula preparation process recording data are obtained, and in any spinning processing, the industrial yarn oil formula component recording data and the industrial yarn oil formula preparation process recording data are fixed, so that the acquisition is easy, but even the same industrial yarn oil formula component recording data and the industrial yarn oil formula preparation process recording data, the first type friction coefficient, the second type friction coefficient and the spinning static electricity eliminating amount are possibly different, so that the industrial yarn oil formula component recording data and the industrial yarn oil formula preparation process recording data which are easy to determine are firstly determined.
And then, traversing industrial yarn oil formula component record data and industrial yarn oil formula preparation process record data to acquire industrial yarn oil performance parameters by using spinning processing equipment information, spinning processing condition information and spinning type information to obtain first type friction coefficient log data, second type friction coefficient log data and spinning static electricity eliminating amount log data, wherein the first type friction coefficient log data, the second type friction coefficient log data and the spinning static electricity eliminating amount log data represent multiple groups of data, any one group of data is ternary, and the ternary is respectively the first type friction coefficient, the second type friction coefficient and the spinning static electricity eliminating amount, and any one industrial yarn oil formula component record data and industrial yarn oil formula preparation process record data correspond to multiple groups of first type friction coefficient log data, the second type friction coefficient log data and the spinning static electricity eliminating amount log data. The final output of the model is the only first type friction coefficient, the second type friction coefficient and the evaluation data of the spinning static electricity eliminating amount, so that the industrial yarn oil agent formula component record data and the industrial yarn oil agent formula preparation process record data need to be traversed, and a plurality of groups of first type friction coefficient log data, second type friction coefficient log data and spinning static electricity eliminating amount log data corresponding to the industrial yarn oil agent formula component record data are subjected to correlation analysis, so that the most relevant first type friction coefficient log data, second type friction coefficient log data and spinning static electricity eliminating amount log data are determined, and are set to be the first type friction coefficient record data, the second type friction coefficient record data and the spinning static electricity eliminating amount record data, and the model training is facilitated.
Further, the relevance analysis process is preferably determined by adopting a custom frequent item analysis algorithm, as follows:
traversing the first type friction coefficient log data to obtain a first support degree, wherein the first support degree characterizes the triggering frequency duty ratio of any one first type friction coefficient log data, and the preferred determining mode of the triggering frequency duty ratio is as follows: and determining a plurality of groups of first-type friction coefficient log data, second-type friction coefficient log data and spinning static elimination log data corresponding to the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data according to any pair of the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data. Then determining the total quantity of first-type friction coefficient log data, wherein the first-type friction coefficient log data comprises a plurality of characteristic values, and the characteristic values are possibly repeated, and the repeated number is recorded as the triggering frequency; and calculating the ratio of the triggering frequency of any first type friction coefficient characteristic value to the total quantity of the first type friction coefficient log data, and recording the ratio as the first support degree.
The first support threshold refers to screening the lowest trigger frequency duty cycle of the first type coefficient of friction characteristic value. And storing the first type friction coefficient log data meeting the first support threshold and the triggering frequency duty ratio thereof, and further taking the triggering frequency duty ratio as a weight to calculate a weighted average of the first type friction coefficient log data so as to obtain the first type friction coefficient record data. And traversing a plurality of groups of industrial yarn oil formula component record data and industrial yarn oil formula preparation process record data to obtain a plurality of corresponding first type friction coefficient record data.
Traversing the second-type friction coefficient log data to obtain a second support degree, wherein the second support degree characterizes the triggering frequency duty ratio of any one second-type friction coefficient log data, and the triggering frequency duty ratio is preferably determined in the following manner: and determining a plurality of groups of first-type friction coefficient log data, second-type friction coefficient log data and spinning static elimination log data corresponding to the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data according to any pair of the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data. Then determining the total quantity of the first type friction coefficient log data, wherein the second type friction coefficient log data comprises a plurality of characteristic values, and the characteristic values are possibly repeated, and the repeated number is recorded as the triggering frequency; calculating the ratio of the triggering frequency of any one of the characteristic values of the second type friction coefficient to the total quantity of the log data of the second type friction coefficient, and recording the ratio as the second support degree;
the second support threshold refers to screening the lowest trigger frequency duty cycle of the first type coefficient of friction characteristic value. And storing the second-type friction coefficient log data meeting the second support threshold and the triggering frequency duty ratio thereof, and further taking the triggering frequency duty ratio as a weight to calculate a weighted average of the second-type friction coefficient log data so as to obtain second-type friction coefficient record data. Traversing a plurality of groups of industrial yarn oil formula component record data and industrial yarn oil formula preparation process record data to obtain a plurality of corresponding second type friction coefficient record data.
Traversing the log data of the spinning static elimination amount to obtain a third support, wherein the third support represents the triggering frequency duty ratio of any one of the log data of the spinning static elimination amount, and the preferred determining mode of the triggering frequency duty ratio is as follows: and determining a plurality of groups of first-type friction coefficient log data, second-type friction coefficient log data and spinning static elimination log data corresponding to the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data according to any pair of the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data. Then determining the total quantity of the spinning static electricity eliminating quantity log data, wherein the spinning static electricity eliminating quantity log data comprises a plurality of characteristic values, and the characteristic values are possibly repeated, and the repeated number is recorded as the trigger frequency; and calculating the ratio of the triggering frequency of any spinning static electricity eliminating amount log data characteristic value to the total amount of the spinning static electricity eliminating amount log data, and marking the ratio as a third support degree.
The third support threshold value refers to the lowest trigger frequency duty ratio of the characteristic value of the log data of the screening spinning static elimination amount. And storing the log data of the spinning static elimination amount meeting the third support threshold and the triggering frequency duty ratio thereof, and further taking the triggering frequency duty ratio as a weight to calculate a weighted average of the log data of the spinning static elimination amount, thereby obtaining the record data of the spinning static elimination amount. Traversing a plurality of groups of industrial yarn oil formula component record data and industrial yarn oil formula preparation process record data to obtain a plurality of spinning static elimination record data corresponding to the record data.
And a second step of: training a performance prediction model:
when training is performed by using any group of industrial yarn oil formula component recording data, industrial yarn oil formula preparation process recording data, first type friction coefficient recording data, second type friction coefficient recording data and spinning static electricity eliminating amount recording data:
the method comprises the steps of taking industrial silk oil formula component recording data and industrial silk oil formula preparation process recording data as input data, taking first type friction coefficient recording data as input identification data, and training a first type friction coefficient prediction layer based on a BP neural network; the method comprises the steps of taking recording data of components of an industrial yarn oil formula and recording data of a preparation process of the industrial yarn oil formula as input data, taking recording data of a second type friction coefficient as input identification data, and training a second type friction coefficient prediction layer based on a BP neural network; the method is characterized in that the industrial yarn oil formula component record data and the industrial yarn oil formula preparation process record data are used as input data, the spinning static elimination amount record data are used as input identification data, the spinning static elimination amount prediction layer is trained based on a BP neural network, and any conventional model training mode can be adopted in the training process.
And finally, based on the BP neural network, taking the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static electricity elimination amount prediction layer as three parallel nodes of a performance prediction model to form a net-in-net topological structure and generate the performance prediction model. The performance prediction of the industrial yarn oiling agent formula can be performed efficiently.
S400: judging whether the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data meet the expected value of the evaluation index or not;
specifically, when the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity eliminating amount prediction data are predicted, according to the evaluation index expected value, the first type friction coefficient expected value, the second type friction coefficient expected value and the spinning static electricity eliminating amount expected value are obtained; comparing the first type friction coefficient prediction data with the first type friction coefficient expected value, comparing the second type friction coefficient prediction data with the second type friction coefficient expected value, and comparing the spinning static electricity elimination amount prediction data with the spinning static electricity elimination amount expected value, if any one of the predicted data does not accord with the expected value, the predicted data is regarded as the first type friction coefficient prediction data, and the predicted data of the second type friction coefficient and the predicted data of the spinning static electricity elimination amount do not meet the estimated index expected value.
If the predicted data of the first type friction coefficient and the predicted data of the second type friction coefficient and the predicted data of the static electricity eliminating amount of the spinning meet respective expected values, outputting the predicted data to meet, namely, screening raw materials based on the current industrial silk oil formulation information. If the formula information is not satisfied, the formula information of the industrial silk oil agent needs to be optimized to screen the raw materials.
S500: if the information of the industrial yarn oil formulation does not meet the requirement, a first optimization instruction is acquired, and the information of the industrial yarn oil formulation is optimized to generate an industrial yarn oil formulation optimization result;
further, based on the failure, a first optimization instruction is obtained, the industrial yarn oil formulation information is optimized, and an industrial yarn oil formulation optimization result is generated, and step S500 includes the steps of:
s510: setting a variable constraint interval for the industrial yarn oil formula component information and the industrial yarn oil formula preparation information, and obtaining a formula component type constraint interval, a formula component proportion constraint interval and a formula preparation condition constraint interval;
s511: traversing the formula component type constraint area or/and the formula component proportion constraint area or/and the formula preparation condition constraint area, and randomly adjusting the formula component information of the industrial yarn oil agent and the preparation information of the industrial yarn oil agent to obtain a kth adjustment result of the industrial yarn oil agent formula;
S512: inputting a kth adjustment result of the industrial yarn oiling agent formula into the performance prediction model to obtain a kth adjustment result performance prediction result;
s513: judging whether the k-th adjustment result performance prediction result meets the evaluation index expected value or not;
s514: and if the k adjustment result of the industrial yarn oil formula is met, setting the k adjustment result as the industrial yarn oil formula optimization result.
Further, based on the determination as to whether the kth adjustment result performance prediction result satisfies the evaluation index expected value, step S513 further includes the steps of:
s5131: if the performance deviation degree is not met, the performance prediction result of the k-1 adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the k-1 adjustment result is obtained;
s5132: the performance prediction result of the kth adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the kth adjustment result is obtained;
s5133: judging whether the performance deviation degree of the k-1 adjustment result is smaller than the performance deviation degree of the k adjustment result or not;
s5134: if the value is smaller than the preset value, adding the k-1 adjustment result of the industrial yarn oil formula into the elimination data set, and setting the k adjustment result of the industrial yarn oil formula as comparison winning data;
S5135: if the comparison result is greater than or equal to the comparison result, adding the k-1 adjustment result of the industrial wire finish formula into the elimination data set, and setting the k-1 adjustment result of the industrial wire finish formula as the comparison winning data;
s5136: repeating optimization until k meets preset iteration times, and judging whether an industrial yarn oil formula optimization result exists;
s5137: if not, setting the comparison winning data as the optimized result of the industrial silk oil formula.
Specifically, when the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data do not meet the expected value of the evaluation index, a first optimization instruction is generated to optimize the industrial yarn oil formula information, so that an industrial yarn oil formula optimization result meeting the expected value of the evaluation index is obtained, and raw material screening and reference are facilitated.
Preferred embodiments of the optimization algorithm are as follows:
the method comprises the steps of defining and setting a representation variable constraint interval for the formula component information of the industrial silk oil agent and the preparation information of the industrial silk oil agent, and obtaining a formula component type constraint interval representing any component replaceable component set, a formula component proportion constraint interval representing any component proportion range and various control parameters representing a formula preparation process, such as: a formula preparation condition constraint interval of temperature, pH, humidity, reaction duration and the like; randomly adjusting any one or more of a formula component type constraint interval, a formula component proportion constraint interval and a formula preparation condition constraint interval to realize the random adjustment of the formula component information of the industrial yarn oil agent and the preparation information of the industrial yarn oil agent and obtain a kth adjustment result of the industrial yarn oil agent formula representing the adjustment result; inputting the component information of the industrial yarn oil formulation corresponding to the kth adjustment result of the industrial yarn oil formulation and the preparation information of the industrial yarn oil formulation into a performance prediction model, and inputting the kth adjustment result performance prediction result representing the performance prediction result; and judging whether the k-th adjustment result performance prediction result meets the expected value of the evaluation index.
If the k adjustment result of the industrial yarn oil formula is met, setting the k adjustment result as an industrial yarn oil formula optimization result;
if the performance deviation degree is not met, the performance prediction result of the k-1 adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the k-1 adjustment result is obtained; invoking a k adjustment result performance prediction result and the evaluation index expected value to perform deviation calculation to obtain a k adjustment result performance deviation degree; judging whether the performance deviation degree of the k-1 adjustment result is smaller than the performance deviation degree of the k adjustment result; if the value is smaller than the preset value, adding the k-1 adjustment result of the industrial yarn oil formula into the elimination data set, and setting the k adjustment result of the industrial yarn oil formula as comparison winning data; if the comparison result is greater than or equal to the comparison result, adding the k-1 adjustment result of the industrial wire finish formula into the elimination data set, and setting the k-1 adjustment result of the industrial wire finish formula as the comparison winning data; repeating optimization until k meets preset iteration times, and judging whether an industrial yarn oil formula optimization result exists; if the optimal value does not exist, the comparison winning data is set as an optimal result of the industrial silk oil formula, and the relatively optimal value can be selected on the premise that no ideal value exists. Thereby realizing the automatic optimization of the formula information of the industrial yarn oiling agent and providing a reference basis for the subsequent screening of the raw materials of the industrial yarn oiling agent.
S600: obtaining an industrial yarn oil raw material component optimization result according to the industrial yarn oil formula optimization result;
s700: and (5) screening the raw materials according to the optimized result of the raw material components of the industrial yarn oiling agent.
Specifically, the industrial yarn oil raw material component optimization result refers to data representing the type and the component content of the characterization component stored in the industrial yarn oil formula optimization result, and raw material screening is performed based on the industrial yarn oil raw material component optimization result, so that the degree of automation is high, and the objectivity is high.
In summary, the intelligent screening method and system for the industrial silk oil raw material provided by the embodiment of the application have the following technical effects:
the performance data of the current industrial silk finish formula is determined through automatic analysis of screening evaluation indexes based on basic information of a spinning processing technology, if the performance data does not accord with an expected value, raw material screening is carried out according to the optimized formula after optimization according to an optimization algorithm, and the technical effects of high automation degree and stronger objectivity are achieved.
Example two
Based on the same inventive concept as the intelligent screening method of an industrial yarn oil raw material in the foregoing embodiment, as shown in fig. 4, an embodiment of the present application provides an intelligent screening system of an industrial yarn oil raw material, which includes:
The first data acquisition module 11 is used for acquiring basic information of a spinning process, wherein the basic information of the spinning process comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information;
a second data acquisition module 12, configured to acquire a screening evaluation index and an evaluation index expected value, where the screening evaluation index includes a first type friction coefficient, a second type friction coefficient, and a spin static electricity eliminating amount, where the first type friction coefficient represents a friction coefficient between spinning and the second type friction coefficient represents a friction coefficient between spinning and equipment;
the filament oil performance prediction module 13 is configured to predict the first type friction coefficient, the second type friction coefficient and the spin static electricity eliminating amount according to the industrial filament oil formula information, the spin processing equipment information, the spin processing condition information and the spin type information, so as to generate first type friction coefficient prediction data, second type friction coefficient prediction data and spin static electricity eliminating amount prediction data;
a data determination module 14 for determining whether the first type of friction coefficient prediction data, the second type of friction coefficient prediction data and the spin-on-static elimination amount prediction data satisfy the evaluation index expected value;
The formula optimizing module 15 is configured to obtain a first optimizing instruction if the first optimizing instruction is not satisfied, optimize the industrial yarn oil formula information, and generate an industrial yarn oil formula optimizing result;
a third data acquisition module 16, configured to obtain an industrial yarn oil raw material component optimization result according to the industrial yarn oil formula optimization result;
and the task execution module 17 is used for screening the raw materials according to the optimized result of the raw material components of the industrial yarn oiling agent.
Further, the performance prediction module 13 performs the steps of:
training a performance prediction model according to the spinning processing equipment information, the spinning processing condition information and the spinning type information, wherein the performance prediction model comprises a first type friction coefficient prediction layer, a second type friction coefficient prediction layer and a spinning static elimination amount prediction layer;
disassembling the industrial yarn oil formula information to obtain industrial yarn oil formula component information and industrial yarn oil formula preparation information;
inputting the component information of the industrial yarn oil formulation and the preparation information of the industrial yarn oil formulation into the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static electricity elimination amount prediction layer to obtain the first type friction coefficient prediction data, and the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data.
Further, the performance prediction module 13 performs the steps of:
frequent excavation is carried out by taking the spinning processing equipment information, the spinning processing condition information and the spinning type information as screening scene parameters, so as to generate industrial yarn oil formula component record data, industrial yarn oil formula preparation process record data, first type friction coefficient record data, second type friction coefficient record data and spinning static elimination quantity record data;
training the first type friction coefficient prediction layer according to the industrial yarn oil formula component recording data, the industrial yarn oil formula preparation process recording data and the first type friction coefficient recording data;
training the second-type friction coefficient prediction layer according to the industrial yarn oil formula component recording data, the industrial yarn oil formula preparation process recording data and the second-type friction coefficient recording data;
training the spinning static elimination amount prediction layer according to the recording data of the components of the industrial yarn oil formula, the recording data of the preparation process of the industrial yarn oil formula and the recording data of the spinning static elimination amount;
and combining the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static elimination amount prediction layer to generate the performance prediction model.
Further, the performance prediction module 13 performs the steps of:
taking the spinning processing equipment information, the spinning processing condition information and the spinning type information as screening scene parameters to acquire spinning processing big data, and acquiring the industrial yarn oiling agent formula component recording data and the industrial yarn oiling agent formula preparation process recording data;
traversing the industrial yarn oiling agent formula component record data and the industrial yarn oiling agent formula preparation process record data according to the spinning processing equipment information, the spinning processing condition information and the spinning type information to acquire industrial yarn oiling agent performance parameters, and acquiring first type friction coefficient log data, second type friction coefficient log data and spinning static elimination amount log data;
and traversing the first type friction coefficient log data, and performing frequent analysis on the second type friction coefficient log data and the spinning static elimination amount log data to obtain the first type friction coefficient record data, the second type friction coefficient record data and the spinning static elimination amount record data.
Further, the performance prediction module 13 performs the steps of:
Traversing the first type friction coefficient log data to obtain a first support degree, wherein the first support degree represents the triggering frequency duty ratio of any one of the first type friction coefficient log data;
when the first support degree meets a first support degree threshold value, adding the first support degree threshold value into the first type friction coefficient record data;
traversing the second-type friction coefficient log data to obtain a second support degree, wherein the second support degree represents the triggering frequency duty ratio of any one of the second-type friction coefficient log data;
when the second support degree meets a second support degree threshold value, adding the second support degree threshold value into the second type friction coefficient record data;
traversing the spinning static elimination amount log data to obtain a third support degree, wherein the third support degree represents the triggering frequency duty ratio of any spinning static elimination amount log data;
and when the third support degree meets a third support degree threshold value, adding the third support degree threshold value into the spinning static electricity eliminating quantity record data.
Further, the recipe optimizing module 15 performs the steps of:
setting a variable constraint interval for the industrial yarn oil formula component information and the industrial yarn oil formula preparation information, and obtaining a formula component type constraint interval, a formula component proportion constraint interval and a formula preparation condition constraint interval;
Traversing the formula component type constraint area or/and the formula component proportion constraint area or/and the formula preparation condition constraint area, and randomly adjusting the formula component information of the industrial yarn oil agent and the preparation information of the industrial yarn oil agent to obtain a kth adjustment result of the industrial yarn oil agent formula;
inputting a kth adjustment result of the industrial yarn oiling agent formula into the performance prediction model to obtain a kth adjustment result performance prediction result;
judging whether the k-th adjustment result performance prediction result meets the evaluation index expected value or not;
and if the k adjustment result of the industrial yarn oil formula is met, setting the k adjustment result as the industrial yarn oil formula optimization result.
Further, the recipe optimizing module 15 performs the steps of:
if the performance deviation degree is not met, the performance prediction result of the k-1 adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the k-1 adjustment result is obtained;
the performance prediction result of the kth adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the kth adjustment result is obtained;
judging whether the performance deviation degree of the k-1 adjustment result is smaller than the performance deviation degree of the k adjustment result or not;
If the value is smaller than the preset value, adding the k-1 adjustment result of the industrial yarn oil formula into the elimination data set, and setting the k adjustment result of the industrial yarn oil formula as comparison winning data;
if the comparison result is greater than or equal to the comparison result, adding the k-1 adjustment result of the industrial wire finish formula into the elimination data set, and setting the k-1 adjustment result of the industrial wire finish formula as the comparison winning data;
repeating optimization until k meets preset iteration times, and judging whether an industrial yarn oil formula optimization result exists;
if not, setting the comparison winning data as the optimized result of the industrial silk oil formula.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any of the methods to implement embodiments of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.

Claims (8)

1. An intelligent screening method for industrial silk oil raw materials is characterized by comprising the following steps:
obtaining basic information of a spinning process, wherein the basic information of the spinning process comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information;
obtaining a screening evaluation index and an evaluation index expected value, wherein the screening evaluation index comprises a first type friction coefficient, a second type friction coefficient and a spinning static electricity elimination amount, wherein the first type friction coefficient represents a friction coefficient between spinning, and the second type friction coefficient represents a friction coefficient between spinning and equipment;
traversing the first type friction coefficient, the second type friction coefficient and the spinning static electricity eliminating amount to predict according to the industrial yarn oil agent formula information, the spinning processing equipment information, the spinning processing condition information and the spinning type information, and generating first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static electricity eliminating amount prediction data;
judging whether the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data meet the expected value of the evaluation index or not;
If the information of the industrial yarn oil formulation does not meet the requirement, a first optimization instruction is acquired, and the information of the industrial yarn oil formulation is optimized to generate an industrial yarn oil formulation optimization result;
obtaining an industrial yarn oil raw material component optimization result according to the industrial yarn oil formula optimization result;
and (5) screening the raw materials according to the optimized result of the raw material components of the industrial yarn oiling agent.
2. The method of claim 1, wherein said traversing said first type of friction coefficient, said second type of friction coefficient, and said spin static elimination amount based on said industrial yarn finish formulation information, said spin processing equipment information, said spin processing condition information, and said spin type information predicts a first type of friction coefficient prediction data, a second type of friction coefficient prediction data, and a spin static elimination amount prediction data, comprising:
training a performance prediction model according to the spinning processing equipment information, the spinning processing condition information and the spinning type information, wherein the performance prediction model comprises a first type friction coefficient prediction layer, a second type friction coefficient prediction layer and a spinning static elimination amount prediction layer;
disassembling the industrial yarn oil formula information to obtain industrial yarn oil formula component information and industrial yarn oil formula preparation information;
Inputting the component information of the industrial yarn oil formulation and the preparation information of the industrial yarn oil formulation into the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static electricity elimination amount prediction layer to obtain the first type friction coefficient prediction data, and the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data.
3. The method of claim 2, wherein said training a performance prediction model based on said spin processing equipment information, said spin processing condition information, and said spin type information comprises:
frequent excavation is carried out by taking the spinning processing equipment information, the spinning processing condition information and the spinning type information as screening scene parameters, so as to generate industrial yarn oil formula component record data, industrial yarn oil formula preparation process record data, first type friction coefficient record data, second type friction coefficient record data and spinning static elimination quantity record data;
training the first type friction coefficient prediction layer according to the industrial yarn oil formula component recording data, the industrial yarn oil formula preparation process recording data and the first type friction coefficient recording data;
Training the second-type friction coefficient prediction layer according to the industrial yarn oil formula component recording data, the industrial yarn oil formula preparation process recording data and the second-type friction coefficient recording data;
training the spinning static elimination amount prediction layer according to the recording data of the components of the industrial yarn oil formula, the recording data of the preparation process of the industrial yarn oil formula and the recording data of the spinning static elimination amount;
and combining the first type friction coefficient prediction layer, the second type friction coefficient prediction layer and the spinning static elimination amount prediction layer to generate the performance prediction model.
4. The method of claim 3, wherein said frequent mining with said spin finish equipment information, said spin finish condition information, and said spin type information as screening scene parameters generates industrial yarn finish formulation component record data, industrial yarn finish formulation preparation process record data, first type friction coefficient record data, second type friction coefficient record data, and spin static elimination amount record data, comprising:
taking the spinning processing equipment information, the spinning processing condition information and the spinning type information as screening scene parameters to acquire spinning processing big data, and acquiring the industrial yarn oiling agent formula component recording data and the industrial yarn oiling agent formula preparation process recording data;
Traversing the industrial yarn oiling agent formula component record data and the industrial yarn oiling agent formula preparation process record data according to the spinning processing equipment information, the spinning processing condition information and the spinning type information to acquire industrial yarn oiling agent performance parameters, and acquiring first type friction coefficient log data, second type friction coefficient log data and spinning static elimination amount log data;
and traversing the first type friction coefficient log data, and performing frequent analysis on the second type friction coefficient log data and the spinning static elimination amount log data to obtain the first type friction coefficient record data, the second type friction coefficient record data and the spinning static elimination amount record data.
5. The method of claim 4, wherein said traversing said first type of friction coefficient log data, said second type of friction coefficient log data and said spin-on-static-elimination log data for frequent item analysis, obtaining said first type of friction coefficient log data, said second type of friction coefficient log data and said spin-on-static-elimination log data, comprises:
Traversing the first type friction coefficient log data to obtain a first support degree, wherein the first support degree represents the triggering frequency duty ratio of any one of the first type friction coefficient log data;
when the first support degree meets a first support degree threshold value, adding the first support degree threshold value into the first type friction coefficient record data;
traversing the second-type friction coefficient log data to obtain a second support degree, wherein the second support degree represents the triggering frequency duty ratio of any one of the second-type friction coefficient log data;
when the second support degree meets a second support degree threshold value, adding the second support degree threshold value into the second type friction coefficient record data;
traversing the spinning static elimination amount log data to obtain a third support degree, wherein the third support degree represents the triggering frequency duty ratio of any spinning static elimination amount log data;
and when the third support degree meets a third support degree threshold value, adding the third support degree threshold value into the spinning static electricity eliminating quantity record data.
6. The method of claim 2, wherein if not, obtaining a first optimization instruction, optimizing the industrial wire finish formulation information, and generating an industrial wire finish formulation optimization result, comprises:
Setting a variable constraint interval for the industrial yarn oil formula component information and the industrial yarn oil formula preparation information, and obtaining a formula component type constraint interval, a formula component proportion constraint interval and a formula preparation condition constraint interval;
traversing the formula component type constraint area or/and the formula component proportion constraint area or/and the formula preparation condition constraint area, and randomly adjusting the formula component information of the industrial yarn oil agent and the preparation information of the industrial yarn oil agent to obtain a kth adjustment result of the industrial yarn oil agent formula;
inputting a kth adjustment result of the industrial yarn oiling agent formula into the performance prediction model to obtain a kth adjustment result performance prediction result;
judging whether the k-th adjustment result performance prediction result meets the evaluation index expected value or not;
and if the k adjustment result of the industrial yarn oil formula is met, setting the k adjustment result as the industrial yarn oil formula optimization result.
7. The method of claim 6, wherein the determining whether the kth adjustment result performance prediction result meets the evaluation index expected value further comprises:
if the performance deviation degree is not met, the performance prediction result of the k-1 adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the k-1 adjustment result is obtained;
The performance prediction result of the kth adjustment result is called to carry out deviation calculation with the expected value of the evaluation index, and the performance deviation degree of the kth adjustment result is obtained;
judging whether the performance deviation degree of the k-1 adjustment result is smaller than the performance deviation degree of the k adjustment result or not;
if the value is smaller than the preset value, adding the k-1 adjustment result of the industrial yarn oil formula into the elimination data set, and setting the k adjustment result of the industrial yarn oil formula as comparison winning data;
if the comparison result is greater than or equal to the comparison result, adding the k-1 adjustment result of the industrial wire finish formula into the elimination data set, and setting the k-1 adjustment result of the industrial wire finish formula as the comparison winning data;
repeating optimization until k meets preset iteration times, and judging whether an industrial yarn oil formula optimization result exists;
if not, setting the comparison winning data as the optimized result of the industrial silk oil formula.
8. An intelligent screening system for industrial yarn oiling agent raw materials, which is characterized by comprising:
the first data acquisition module is used for acquiring basic information of a spinning process, wherein the basic information of the spinning process comprises industrial yarn oiling agent formula information, spinning processing equipment information, spinning processing condition information and spinning type information;
The second data acquisition module is used for acquiring screening evaluation indexes and evaluation index expected values, wherein the screening evaluation indexes comprise a first type friction coefficient, a second type friction coefficient and a spinning static electricity elimination amount, the first type friction coefficient represents a friction coefficient between spinning, and the second type friction coefficient represents a friction coefficient between spinning and equipment;
the silk oil performance prediction module is used for traversing the first type friction coefficient, the second type friction coefficient and the spinning static electricity elimination amount to predict according to the industrial silk oil formula information, the spinning processing equipment information, the spinning processing condition information and the spinning type information, so as to generate first type friction coefficient prediction data, second type friction coefficient prediction data and spinning static electricity elimination amount prediction data;
the data judging module is used for judging whether the first type friction coefficient prediction data, the second type friction coefficient prediction data and the spinning static electricity elimination amount prediction data meet the expected value of the evaluation index or not;
the formula optimization module is used for acquiring a first optimization instruction if the formula information is not satisfied, optimizing the industrial yarn oiling agent formula information and generating an industrial yarn oiling agent formula optimization result;
The third data acquisition module is used for acquiring an industrial yarn oil raw material component optimization result according to the industrial yarn oil formula optimization result;
and the task execution module is used for screening the raw materials according to the optimized result of the raw material components of the industrial yarn oiling agent.
CN202310038794.XA 2023-01-13 2023-01-13 Intelligent screening method and system for industrial silk oil agent raw materials Active CN117252284B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310038794.XA CN117252284B (en) 2023-01-13 2023-01-13 Intelligent screening method and system for industrial silk oil agent raw materials

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310038794.XA CN117252284B (en) 2023-01-13 2023-01-13 Intelligent screening method and system for industrial silk oil agent raw materials

Publications (2)

Publication Number Publication Date
CN117252284A CN117252284A (en) 2023-12-19
CN117252284B true CN117252284B (en) 2024-03-05

Family

ID=89130090

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310038794.XA Active CN117252284B (en) 2023-01-13 2023-01-13 Intelligent screening method and system for industrial silk oil agent raw materials

Country Status (1)

Country Link
CN (1) CN117252284B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113818104A (en) * 2021-09-27 2021-12-21 江苏悦孚油品有限公司 Superfine denier filament spinning oil and preparation method thereof
CN115511398A (en) * 2022-11-23 2022-12-23 江苏未来网络集团有限公司 Welding quality intelligent detection method and system based on time sensitive network

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3105862A1 (en) * 2019-12-31 2021-07-02 Bull Sas METHOD AND SYSTEM FOR SELECTING A LEARNING MODEL WITHIN A PLURALITY OF LEARNING MODELS

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113818104A (en) * 2021-09-27 2021-12-21 江苏悦孚油品有限公司 Superfine denier filament spinning oil and preparation method thereof
CN115511398A (en) * 2022-11-23 2022-12-23 江苏未来网络集团有限公司 Welding quality intelligent detection method and system based on time sensitive network

Also Published As

Publication number Publication date
CN117252284A (en) 2023-12-19

Similar Documents

Publication Publication Date Title
CN110232203B (en) Knowledge distillation optimization RNN short-term power failure prediction method, storage medium and equipment
CN115276006A (en) Load prediction method and system for power integration system
Khoshgoftaar et al. A multiobjective module-order model for software quality enhancement
CN107644194A (en) The system and method that monitoring data is provided
CN112116198A (en) Data-driven process industrial state perception network key node screening method
CN115511398B (en) Welding quality intelligent detection method and system based on time sensitive network
CN115184193B (en) Automatic testing method, device, equipment and storage medium for cable
CN108830417A (en) A kind of residential energy consumption prediction technique and system based on ARMA and regression analysis
CN115576284A (en) Clothing workshop intelligent management method and system
JP2005519394A (en) Automatic experiment planning method and system
CN109615115B (en) Task reliability-oriented integrated production task scheduling method
CN117252284B (en) Intelligent screening method and system for industrial silk oil agent raw materials
CN114548493A (en) Method and system for predicting current overload of electric energy meter
Cousens Design and interpretation of interference studies: are some methods totally unacceptable
CN112214911A (en) Power supply health state prediction method
Baykasoğlu et al. Genetic programming based data mining approach to dispatching rule selection in a simulated job shop
CN113642209B (en) Structure implantation fault response data acquisition and evaluation method based on digital twinning
CN114116840A (en) Evaluation method for artificial precipitation effect
CN115441456A (en) Fault diagnosis method and device for power grid dispatching support system
CN115471122A (en) Energy consumption evaluation method and system based on metadata model
CN112434808B (en) Full-stack type forward neural network deep learning system security analysis and detection method
CN111459996B (en) Method and device for detecting working state of oil gun in specified time period
CN111625525B (en) Environment data repairing/filling method and system
CN113807743A (en) Power grid dispatching automation software reliability assessment method and system
WO2023241126A1 (en) Aquaculture breeding prediction system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant