CN114139813A - Flow type production equipment product quality prediction method based on self-updating model - Google Patents

Flow type production equipment product quality prediction method based on self-updating model Download PDF

Info

Publication number
CN114139813A
CN114139813A CN202111470880.5A CN202111470880A CN114139813A CN 114139813 A CN114139813 A CN 114139813A CN 202111470880 A CN202111470880 A CN 202111470880A CN 114139813 A CN114139813 A CN 114139813A
Authority
CN
China
Prior art keywords
self
updating
production equipment
flow
type production
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111470880.5A
Other languages
Chinese (zh)
Inventor
左颖
柯超凡
张萌
陶飞
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.)
Beihang University
Original Assignee
Beihang University
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 Beihang University filed Critical Beihang University
Priority to CN202111470880.5A priority Critical patent/CN114139813A/en
Publication of CN114139813A publication Critical patent/CN114139813A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
    • 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
    • G06N3/044—Recurrent networks, e.g. Hopfield 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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

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

Abstract

The invention discloses a flow type production equipment product quality prediction method based on a self-updating model, which comprises the following steps: step (1), calculating the original data delay of the flow type production equipment, advancing the production parameters of each area of an original data set, and aligning the production parameters of each area in the time dimension in the original data set after the advance; step (2), extracting characteristics of the flow type production equipment; and (3) performing self-updating of the product quality prediction model of the flow production equipment, including self-updating between batches and self-updating within a batch. The method and the device can solve the problems of large time lag of the flow production equipment and large influence of different batches of materials to a certain extent, and improve the accuracy of product quality prediction of the flow production equipment.

Description

Flow type production equipment product quality prediction method based on self-updating model
Technical Field
The invention belongs to the field of computer science, and particularly relates to a flow type production equipment product quality prediction method based on a self-updating model.
Background
The process type production process refers to that materials pass through various containers and pipelines and undergo various physical, chemical or biological processes to achieve the separation effect or generate new products. The production of fermentation, synthesis, extraction, tobacco shred making and the like all belong to flow production. The quality of the product output by the flow type production equipment is effectively predicted, the production process can be effectively detected, and the relation between the environmental variable and the quality of the output product is learned, so that the production process is guided, and the value of the control parameter is adjusted in time according to the quality of the product.
The flow-type production process generally relates to a very complex production process, has the characteristics of large time lag, high coupling and nonlinearity, and the performance and the quality of the product are greatly influenced by materials of different batches, so that great difficulty is brought to the quality prediction of key products. With the development of machine learning in recent years, prediction models such as Logistic regression model, decision tree model, and ELM (extreme learning machine) model are widely applied to quality prediction in industrial and manufacturing scenes, but the characteristics of large time lag of flow type production equipment and large influence of material batches are not considered, so that the prediction effect is not ideal.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method covers the design of a delay corresponding module for original data of the flow type production equipment, the design of a characteristic extraction and data preprocessing module for the flow type production equipment and the design of a self-updating module for the product quality prediction model of the flow type production equipment, can solve the problems of large time lag and large influence of different batches of materials of the flow type production equipment to a certain extent, and improves the accuracy of the product quality prediction of the flow type equipment.
The technical problem to be solved by the invention is realized by adopting the following technical scheme: a flow type production equipment product quality prediction method based on a self-updating model comprises the following steps:
step (1), calculating the original data delay of the flow type production equipment, advancing the production parameters of each area of an original data set, and aligning the production parameters of each area in the time dimension in the original data set after the advance;
step (2), extracting characteristics of the flow type production equipment;
and (3) performing self-updating of the product quality prediction model of the flow production equipment, including self-updating between batches and self-updating within a batch.
Further, the step (1) is specifically as follows:
(1.1) dividing the process type production equipment into n areas according to the process flow;
(1.2) recording the passing time of the material in the ith area as ti；
(1.3) advancing the production parameter of the ith area of the original data set by t ═ t1+t2+…ti-1After the advance, in the original data set, the production parameters of each region are aligned in the time dimension; 1,2, … n;
(1.4) performing moving average on the parameters of the ith area, wherein the window width is ti。
Further, the step (2) is specifically as follows:
(2.1) determining input/output characteristics, determining the characteristics needing to be predicted of the flow type production equipment as output characteristics, and determining the other characteristics as input characteristics;
(2.2) carrying out correlation analysis on the input and output characteristics, and eliminating the input characteristics of which the correlation with the output characteristics is lower than a threshold value;
(2.3) carrying out correlation analysis on the input features, and removing mutually equivalent input features;
and (2.4) performing feature selection on the input features, and selecting the input features with the importance higher than a threshold value based on the XGboost algorithm, wherein the input features comprise the overall importance features and the features in each area of the flow-type production equipment.
Further, the step 3 performs self-updating of the product quality prediction model of the flow-type production equipment, including self-updating between batches and self-updating within a batch, and is specifically realized as follows:
(3.1) training a BP neural network based on historical data;
(3.2) resetting the timer t to 0, and setting the self-updating frequency parameter t between batches0；
(3.3) collecting production data of equipment at the current moment every 1 second, wherein t is t + 1;
(3.4) predicting the quality of equipment output at the current moment based on the BP neural network and the current production data;
(3.5) determining whether t is equal to t0And if so, performing in-batch self-updating: the nearest t0Adding the second data into the training set, and returning to the step (3.1); otherwise, entering the step (3.6);
and (3.6) judging whether the production of the current batch is finished. If the production is not finished, returning to the step (3.3); if the current batch production is finished, carrying out self-updating between batches: and (4) adding the current batch into the training set, deleting the farthest batch from the training set, and returning to the step (3.1).
According to another aspect of the present invention, there is also provided a flow type production equipment product quality prediction system based on a self-updating model, including:
the flow type production equipment original data delay module is used for calculating the original data delay of the flow type production equipment, advancing the production parameters of all areas of the original data set, and aligning the production parameters of all areas in the original data set in a time dimension after the advance;
the flow type production equipment feature extraction module is used for extracting the flow type production equipment features;
and the flow production equipment product quality prediction model self-updating module is used for carrying out self-updating on the flow production equipment product quality prediction model, and comprises batch-to-batch self-updating and batch-to-batch self-updating.
Compared with the prior art, the invention has the advantages that:
(1) based on the process flow of the flow type production equipment, the original data is subjected to delayed corresponding processing, so that all the characteristics are aligned on a time scale, and the problem of difficult product quality prediction caused by the large time-lag characteristic of the flow type equipment is solved to a certain extent.
(2) A model self-updating mechanism is added, the construction of a BP neural network model, the batch-to-batch self-updating and the batch-to-batch self-updating of a prediction model are realized, the real-time batch-to-batch updating and the real-time batch-to-batch updating are carried out in the production process, the problem that flow type production equipment is greatly influenced by material batches is solved to a certain extent, and the product quality prediction accuracy is improved.
(3) The correlation analysis and the XGboost algorithm are used for extracting the input features, so that the number of the features is reduced.
Drawings
FIG. 1 is a block flow diagram of the method of the present invention;
FIG. 2(a) raw data set;
fig. 2(b) delays the corresponding data set.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by a person skilled in the art based on the embodiments of the present invention belong to the protection scope of the present invention without creative efforts.
The process type production equipment is widely applied to various industrial and manufacturing scenes such as fermentation, synthesis, extraction, tobacco shred making and the like, and with the development of intelligent manufacturing and information industry, higher requirements are provided for efficient and accurate prediction of key characteristics of the process type production equipment. The flow-type production process generally relates to a very complex production process, has the characteristics of large time lag, high coupling and nonlinearity, and the performance and the quality of the product are greatly influenced by materials of different batches, so that great difficulty is brought to the prediction of the product quality.
According to an embodiment of the present invention, a flow-type production equipment product quality prediction system based on a self-updating model is provided, which includes:
the flow type production equipment original data delay module is used for calculating the original data delay of the flow type production equipment, advancing the production parameters of all areas of the original data set, and aligning the production parameters of all areas in the original data set in a time dimension after the advance;
the flow type production equipment feature extraction module is used for extracting the flow type production equipment features;
and the flow production equipment product quality prediction model self-updating module is used for carrying out self-updating on the flow production equipment product quality prediction model, and comprises batch-to-batch self-updating and batch-to-batch self-updating.
According to another embodiment of the present invention, a method for predicting product quality of a flow-type production facility based on a self-updating model is further provided, which includes the following steps:
step (1), calculating the original data delay of the flow type production equipment, advancing the production parameters of each area of an original data set, and aligning the production parameters of each area in the time dimension in the original data set after the advance;
step (2), extracting characteristics of the flow type production equipment;
and (3) performing self-updating of the product quality prediction model of the flow production equipment, including self-updating between batches and self-updating within a batch.
The method and the device can solve the problems of large time lag of the process type production equipment and large influence of different batches of materials to a certain extent, and improve the accuracy of the output quality prediction of the process type equipment.
The structural block diagram of the invention is shown in fig. 1, and the specific implementation mode is as follows:
(1) fig. 1 shows a flow-type production equipment raw data delay corresponding module, which is used to calculate a flow-type production equipment raw data delay, advance the production parameters of each region of the raw data set, and after the advance, align the production parameters of each region in the raw data set in the time dimension;
the method specifically comprises the following steps:
dividing flow type equipment into n areas according to a process flow, wherein a material sequentially passes through an area 1 and an area 2 … area n in the flow type equipment, each area corresponds to a plurality of production parameters, for example, taking oil refining equipment as an example, the material sequentially passes through an alkali refining pot, a water washing pot, a dehydration pot and other areas in production, wherein the alkali refining pot corresponds to parameters such as crude oil temperature, crude oil acidity and brine temperature, the water washing pot corresponds to parameters such as temperature and water adding amount, and the dehydration pot corresponds to parameters such as air pressure and oil temperature;
(ii) the time of the material in the ith area is ti；(i＝2,…n)；
Thirdly, advancing the production parameter of the ith area of the original data set by t ═ t1+t2+…ti-1After the advance, in the original data set, the production parameters of each region are aligned in the time dimension;
fourthly, the input parameters of the ith (i is 1,2, … n) area are subjected to sliding average, and the window width is ti。
Assuming that there is a raw data set as shown in fig. 2(a), the values of the respective production parameters at different times are recorded in time order in the longitudinal direction.
As shown in FIG. 2(a), according to the process flow, the process type production equipment is divided into a plurality of areas, the production parameters in the area 1 comprise a parameter 1 and a parameter 2, and the time for the material to pass through the area 1 is t1(ii) a The production parameters in the area 2 comprise parameters 3, 4 and 5, and the time for passing the material through the area 2 is t2(ii) a The production parameter in the area 3 has a parameter 6, and the time for the material to pass through the area 3 is t3. If the original data set is not subjected to delayed corresponding processing, the parameters of all the areas which are in the same row and collected at the same time cannot act on the same section of material, so that the result is not true. Therefore, the delay correspondence processing needs to be performed on the original data set.
After the production parameters of the regions are aligned on the time scale, the time length of the action of the regions 1,2 and 3 on the material is respectively as shown in FIG. 2(b)t1、t2、t3The data of each region is subjected to a moving average operation with the region time as a window width, and the finally obtained data can be input and output correspondingly.
(2) Fig. 1,2, a flow-type production equipment feature extraction module, which is used for performing flow-type production equipment feature extraction;
the method specifically comprises the following steps:
firstly, determining input/output characteristics, and determining the characteristics needing to be predicted of the process type production equipment as output characteristics, generally the quality of an equipment product, and determining the other characteristics as input characteristics which mainly comprise environmental parameters, control parameters, process parameters and the like;
secondly, performing Pearson correlation analysis on the input and output features, and eliminating the input features with low correlation with the output features, wherein the Pearson correlation coefficient has strong correlation when the absolute value of the Pearson correlation coefficient is more than 0.8, and has weak correlation between 0.3 and 0.8. Therefore, the features with the Pearson correlation coefficient absolute value below 0.3 in the input features and the output features are removed;
thirdly, carrying out correlation analysis on the input features, eliminating the input features which are equivalent to each other, considering the input feature pairs with the absolute value of the Pearson correlation coefficient larger than 0.8 as being equivalent to each other, and only keeping one feature;
and fourthly, selecting the characteristics of the input characteristics, calculating the importance of the input characteristics by using an XGboost algorithm, and selecting the input characteristics with higher importance ranking, wherein the input characteristics comprise the characteristics with higher overall importance and the characteristics with higher importance in each area of the process type production equipment.
(3) Fig. 1 shows a flow-type production equipment product quality prediction model self-updating module 3, configured to perform self-updating of a flow-type production equipment product quality prediction model, including inter-batch self-updating and intra-batch self-updating, and specifically execute the following steps:
and (4) training a BP neural network based on historical data.
The learning rate, the number of hidden layers and the number of nodes of the hidden layers of the neural network can be adjusted according to the actual condition of the equipment.
② resetting the timer t to be 0, setting the self-updating frequency parameter t between batches0；
T can be set appropriately based on the actual production situation of the flow type apparatus0Is reasonable.
Collecting production data of equipment at the current moment every 1 second, wherein t is t + 1;
predicting the quality of equipment output at the current moment based on the BP neural network and the current production data;
judging whether t is equal to t0And if so, performing in-batch self-updating: the nearest t0Adding the second data into a training set, and returning to the step I; otherwise, entering the step (sixthly);
sixthly, judging whether the production of the current batch is finished. If the production is not finished, returning to the step III; if the current batch production is finished, carrying out self-updating between batches: and adding the current batch into the training set, deleting the farthest batch from the training set, and returning to the step (i).
The model self-updating module for predicting the product quality of the flow-type production equipment mainly comprises two functions of self-updating in batch and self-updating between batches. As the flow type production equipment is greatly influenced by different batches of environmental parameters and material quality, the model needs to be updated in batches during the operation process of the equipment, namely real data generated by the current batch are continuously added into a training set, so that the network training can be guided, and the network prediction precision is improved. Meanwhile, as time goes on, the process type production equipment is also affected by environmental invariants (such as weather, environmental temperature and humidity, equipment health degree and the like), and production data before a long time cannot provide guidance for prediction of a current batch, so that self-updating of a model between batches is required after production of each batch is finished, and a training set is guaranteed to be always composed of a plurality of recent batches.
In summary, the invention discloses a product quality prediction method for flow type production equipment based on a self-updating model, which can solve the problems of large time lag and large influence of different batches of materials of the flow type production equipment to a certain extent and improve the accuracy of the output quality prediction of the flow type equipment.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims (5)

1. A flow type production equipment product quality prediction method based on a self-updating model is characterized by comprising the following steps:
step (1), calculating the original data delay of the flow type production equipment, advancing the production parameters of each area of an original data set, and aligning the production parameters of each area in the time dimension in the original data set after the advance;
step (2), extracting characteristics of the flow type production equipment;
and (3) performing self-updating of the product quality prediction model of the flow production equipment, including self-updating between batches and self-updating within a batch.
2. The method for predicting the product quality of the flow-type production equipment based on the self-updating model according to claim 1, wherein the step (1) is specifically as follows:
(1.1) dividing the process type production equipment into n areas according to the process flow;
(1.2) recording the passing time of the material in the ith area as ti；
(1.3) advancing the production parameter of the ith area of the original data set by t ═ t1+t2+…ti-1After the advance, in the original data set, the production parameters of each region are aligned in the time dimension; 1,2, … n;
(1.4) performing moving average on the parameters of the ith area, wherein the window width is ti。
3. The product quality prediction method for the flow-type production equipment based on the self-updating model as claimed in claim 1, wherein the step (2) is specifically as follows:
(2.1) determining input/output characteristics, determining the characteristics needing to be predicted of the flow type production equipment as output characteristics, and determining the other characteristics as input characteristics;
(2.2) carrying out correlation analysis on the input and output characteristics, and eliminating the input characteristics of which the correlation with the output characteristics is lower than a threshold value;
(2.3) carrying out correlation analysis on the input features, and removing mutually equivalent input features;
and (2.4) performing feature selection on the input features, and selecting the input features with the importance higher than a threshold value based on the XGboost algorithm, wherein the input features comprise the overall importance features and the features in each area of the flow-type production equipment.
4. The method for predicting the product quality of the process-type production equipment based on the self-updating model as claimed in claim 1, wherein the step 3 performs the self-updating of the product quality prediction model of the process-type production equipment, including self-updating between batches and self-updating within a batch, and is implemented as follows:
(3.1) training a BP neural network based on historical data;
(3.2) resetting the timer t to 0, and setting the self-updating frequency parameter t between batches0；
(3.3) collecting production data of equipment at the current moment every 1 second, wherein t is t + 1;
(3.4) predicting the quality of equipment output at the current moment based on the BP neural network and the current production data;
(3.5) determining whether t is equal to t0And if so, performing in-batch self-updating: the nearest t0Adding the second data into the training set, and returning to the step (3.1); otherwise, entering the step (3.6);
and (3.6) judging whether the production of the current batch is finished. If the production is not finished, returning to the step (3.3); if the current batch production is finished, carrying out self-updating between batches: and (4) adding the current batch into the training set, deleting the farthest batch from the training set, and returning to the step (3.1).
5. A flow type production equipment product quality prediction system based on a self-updating model is characterized by comprising:
the flow type production equipment original data delay module is used for calculating the original data delay of the flow type production equipment, advancing the production parameters of all areas of the original data set, and aligning the production parameters of all areas in the original data set in a time dimension after the advance;
the flow type production equipment feature extraction module is used for extracting the flow type production equipment features;
and the flow production equipment product quality prediction model self-updating module is used for carrying out self-updating on the flow production equipment product quality prediction model, and comprises batch-to-batch self-updating and batch-to-batch self-updating.
CN202111470880.5A 2021-12-03 2021-12-03 Flow type production equipment product quality prediction method based on self-updating model Pending CN114139813A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111470880.5A CN114139813A (en) 2021-12-03 2021-12-03 Flow type production equipment product quality prediction method based on self-updating model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111470880.5A CN114139813A (en) 2021-12-03 2021-12-03 Flow type production equipment product quality prediction method based on self-updating model

Publications (1)

Publication Number Publication Date
CN114139813A true CN114139813A (en) 2022-03-04

Family

ID=80387801

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111470880.5A Pending CN114139813A (en) 2021-12-03 2021-12-03 Flow type production equipment product quality prediction method based on self-updating model

Country Status (1)

Country Link
CN (1) CN114139813A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114819575A (en) * 2022-04-18 2022-07-29 中国兵器装备集团自动化研究所有限公司 Quality analysis and optimization method and system for multi-time scale data fusion

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109884892A (en) * 2019-02-28 2019-06-14 杭州电子科技大学 Process industry system prediction model based on crosscorrelation time lag grey correlation analysis
CN110097929A (en) * 2019-04-16 2019-08-06 北京科技大学 A kind of blast furnace molten iron silicon content on-line prediction method
CN110263997A (en) * 2019-06-19 2019-09-20 燕山大学 A kind of cement producing line flue gas NO based on deep neural networkxConcentration prediction method
CN111103420A (en) * 2019-11-20 2020-05-05 华东理工大学 Phenolic resin product quality prediction method under uncertain raw materials
CN113705888A (en) * 2021-08-27 2021-11-26 浙江工业大学 Industrial steam generation amount prediction method and system based on Pearson correlation and neural network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109884892A (en) * 2019-02-28 2019-06-14 杭州电子科技大学 Process industry system prediction model based on crosscorrelation time lag grey correlation analysis
CN110097929A (en) * 2019-04-16 2019-08-06 北京科技大学 A kind of blast furnace molten iron silicon content on-line prediction method
CN110263997A (en) * 2019-06-19 2019-09-20 燕山大学 A kind of cement producing line flue gas NO based on deep neural networkxConcentration prediction method
CN111103420A (en) * 2019-11-20 2020-05-05 华东理工大学 Phenolic resin product quality prediction method under uncertain raw materials
CN113705888A (en) * 2021-08-27 2021-11-26 浙江工业大学 Industrial steam generation amount prediction method and system based on Pearson correlation and neural network

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114819575A (en) * 2022-04-18 2022-07-29 中国兵器装备集团自动化研究所有限公司 Quality analysis and optimization method and system for multi-time scale data fusion

Similar Documents

Publication Publication Date Title
CN107451101B (en) Method for predicting concentration of butane at bottom of debutanizer by hierarchical integrated Gaussian process regression soft measurement modeling
CN111950191B (en) Rotary kiln sintering temperature prediction method based on hybrid deep neural network
CN109396576B (en) Deep learning-based electric spark machining stability and energy consumption state optimization decision system and decision method
CN114048600A (en) Digital twin-driven multi-model fusion industrial system anomaly detection method
US11966847B2 (en) Method for constructing efficient product surface defect detection model based on network collaborative pruning
CN108595803B (en) Shale gas well production pressure prediction method based on recurrent neural network
US20010051858A1 (en) Method of setting parameters for injection molding machines
CN115618196A (en) Transformer anomaly detection method based on space-time characteristics
CN114015825A (en) Method for monitoring abnormal state of blast furnace heat load based on attention mechanism
CN113591374B (en) Cement finished product specific surface area prediction method based on convolution simple circulation unit network
CN113449919B (en) Power consumption prediction method and system based on feature and trend perception
CN114282443A (en) Residual service life prediction method based on MLP-LSTM supervised joint model
CN114139813A (en) Flow type production equipment product quality prediction method based on self-updating model
CN116226739A (en) Map convolution network industrial process fault diagnosis method based on space-time fusion
CN118037112A (en) Tread quality prediction model construction method based on data driving
CN116341371A (en) Bi-LSTM-CORAL network migration learning-based method for predicting residual life of aeroengine
CN114997464A (en) Popularity prediction method based on graph time sequence information learning
CN115712777A (en) Ranking method of literature recommendation system based on logistic regression
CN112596391B (en) Deep neural network large time lag system dynamic modeling method based on data driving
CN113723015A (en) Catalytic cracking unit optimization method based on mechanism model and big data technology
Kalos et al. Hybrid model development methodology for industrial soft sensors
Li et al. Long short-term memory based on random forest-recursive feature eliminated for hot metal silcion content prediction of blast furnace
Jianxu et al. Soft sensing modeling using neurofuzzy system based on rough set theory
CN112183642A (en) Method and system for detecting coal consumption of cement firing based on random forest model
CN114330089A (en) Method and system for predicting content change of rare earth element

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