CN115017700A - Cigarette shred making process variable selection and weighting method based on SCAD algorithm - Google Patents

Cigarette shred making process variable selection and weighting method based on SCAD algorithm Download PDF

Info

Publication number
CN115017700A
CN115017700A CN202210626198.9A CN202210626198A CN115017700A CN 115017700 A CN115017700 A CN 115017700A CN 202210626198 A CN202210626198 A CN 202210626198A CN 115017700 A CN115017700 A CN 115017700A
Authority
CN
China
Prior art keywords
scad
variable
cigarette
regression model
steady
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
CN202210626198.9A
Other languages
Chinese (zh)
Inventor
刘继辉
马晓龙
杨晶津
华一崑
苏丽
杨佳东
高占勇
汪显国
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.)
Hongyun Honghe Tobacco Group Co Ltd
Original Assignee
Hongyun Honghe Tobacco Group 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 Hongyun Honghe Tobacco Group Co Ltd filed Critical Hongyun Honghe Tobacco Group Co Ltd
Priority to CN202210626198.9A priority Critical patent/CN115017700A/en
Publication of CN115017700A publication Critical patent/CN115017700A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • 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

Abstract

The invention discloses a variable selection and empowerment method for a cigarette shred making process based on an SCAD algorithm, which comprises the following steps: acquiring a data sample of a cigarette shredding process; constructing a linear regression model of least squares according to the data samples; constructing an SCAD penalty function, estimating a regression coefficient of the linear regression model by using the SCAD penalty function, and compressing the regression coefficient in the estimation process to perform variable selection on an explanatory variable; and determining the influence weight of the explanatory variable according to the estimation result of the regression coefficient of the linear regression model. According to the variable selection and weighting method for the cigarette filament making process based on the SCAD algorithm, the SCAD punishment method is adopted, the L1 punishment item and the weight thereof are added on the basis of least square, the problem of multiple collinearity among variables is continuously processed in sequence, the problem of excessive compression of coefficients in the Lasso process is solved, the Oracle property is met, and variable screening and objective weighting in the cigarette filament making process are synchronously realized.

Description

Cigarette shred making process variable selection and weighting method based on SCAD algorithm
Technical Field
The invention relates to the technical field of cigarette shred manufacturing, in particular to a variable selection and weighting method for a cigarette shred manufacturing process based on an SCAD algorithm.
Background
The shred making process is an important link for highlighting the sensory style of the cigarette, stabilizing the product quality and reducing the consumption of raw materials. Under the multipoint processing layout of cigarette products, the differences of regional climate, process layout and equipment level are large, so that a set of scientific method for evaluating the process quality in the silk making process is established, and the stability and consistency of the product quality are ensured. The silk making processing equipment has various parameters, a large amount of interaction effects exist in the silk making processing equipment, in addition, the silk making processing flow is long, the quality indexes of the upstream process directly or indirectly influence the quality control of the downstream process and even the final product, so the screening of key parameters and the measurement of the weight thereof are important links for establishing a scientific evaluation method.
At present, the variable selection method adopted in the silk making process mainly has the following problems: (1) the subjective weighting method sets the weight according to the empirical judgment and the final target, can better reflect the subjective intention and preference of a decision maker to an evaluation target, but cannot overcome the problems of larger subjective randomness and the like, including an expert evaluation method, an analytic hierarchy process and the like; (2) the objective weighting method fully excavates data information contained in the sample by applying a statistical analysis method, and can effectively reduce the influence of subjective factors, including a principal component analysis method, a maximum entropy technology method and the like. The entropy weighting method is to determine the weight according to the size of the information quantity provided by the explanatory variable, and does not bring the target variable into modeling analysis; the principal component weighting method is based on multiple regression analysis, considers the influence of the explained variables on the target variables, but does not solve multiple collinearity among the explained variables, and compared with a machine learning combination algorithm, the robustness of the multiple regression model prediction is lower.
Therefore, a method for selecting and assigning weights to variables in the cigarette shred manufacturing process based on the SCAD algorithm is needed.
Disclosure of Invention
The invention aims to provide a variable selection and empowerment method for a cigarette shred making process based on an SCAD algorithm, so as to solve the problems in the prior art.
The invention provides a variable selection and empowerment method for a cigarette shred making process based on an SCAD algorithm, wherein the variable selection and empowerment method comprises the following steps:
acquiring a data sample of a cigarette shredding process;
constructing a linear regression model of least squares according to the data samples;
constructing an SCAD penalty function, estimating a regression coefficient of the linear regression model by using the SCAD penalty function, and compressing the regression coefficient in the estimation process to perform variable selection on an explanatory variable;
and determining the influence weight of the explanatory variable according to the estimation result of the regression coefficient of the linear regression model.
The method for selecting and assigning weights to variables in a cigarette making process based on the SCAD algorithm preferably includes:
obtaining an original data sample in the cigarette shredding process;
based on the raw data samples, steady-state data samples are obtained.
The method for selecting and assigning weights to variables in a cigarette making process based on the SCAD algorithm preferably includes:
and acquiring an original data sample of the cigarette making process according to the cigarette making process manufacturing execution system.
The method for selecting and assigning weights to cigarette primary process variables based on the SCAD algorithm preferably obtains steady-state data samples based on the original data samples, and specifically includes:
rejecting shutdown material-breaking batch data according to a judgment rule that the flow of a process inlet is reduced to 0 and the duration time exceeds 90s in the production process;
intercepting the effective data according to the interception rule of the effective data;
eliminating set parameters with the production process numerical values as constants;
and grouping the effective data according to the residence time of the materials, and calculating an average value to form a steady-state data sample.
The method for selecting and weighting the variables in the cigarette making process based on the SCAD algorithm preferably includes the following steps:
carrying out standardization processing on the data sample;
and constructing a least square linear regression model according to the data samples after the normalization processing.
The method for selecting and assigning weights to the variables in the cigarette making process based on the SCAD algorithm preferably includes the following steps:
each of the steady state data samples is normalized using a normal normalization method,
for steady state data sample x by the following formula 1 ,x 2 ,…,x i ,…,x n And (3) carrying out conversion:
Figure BDA0003677656830000031
where n represents the number of samples of steady-state data samples, x i Representing the ith data in steady-state data samples, z i Representing the normalized steady-state data samples,
Figure BDA0003677656830000038
represents the mean of the steady-state data samples and s represents the standard deviation of the steady-state data samples.
The method for selecting and weighting the variables in the cigarette making process based on the SCAD algorithm preferably includes the following steps:
constructing a least squares linear regression model from the normalized steady state data
Figure BDA0003677656830000032
Figure BDA0003677656830000037
Obtaining an estimated value of a variable coefficient of a linear regression model by a least square method
Figure BDA0003677656830000033
Satisfy the requirement of
Figure BDA0003677656830000034
Wherein, y i Representing the explained variable of the regression model, and taking the discharged water content or the discharged temperature of each procedure as the explained variable; x is the number of ij The explanation variables representing the regression model comprise equipment parameters and/or process parameters of each process, and j is the number of the explanation variables; beta is a j Representing a metric-interpretation variable x ij Normalized regression coefficient of relative importance, α i And beta j Is the regression coefficient to be estimated of the model; epsilon i An error term representing the model is then calculated,
Figure BDA0003677656830000035
the minimum calculation factor of the formula is represented.
The method for selecting and weighting the variables in the cigarette making process based on the SCAD algorithm preferably includes the steps of constructing an SCAD penalty function, estimating a regression coefficient of a linear regression model by using the SCAD penalty function, and compressing the regression coefficient in the estimation process to perform variable selection on an explanatory variable, and specifically includes:
constructing a SCAD penalty function P λ (|β j |) on the basis of the formula (3) and the SCAD penalty function, an estimated value of a variable coefficient of the linear regression model is calculated by the following formula
Figure BDA0003677656830000036
Figure BDA0003677656830000041
SCAD penalty function P λ (|β j |) is as follows:
Figure BDA0003677656830000042
wherein a and lambda are only formula parameters, no practical meaning exists, a is more than 2, and a parameter lambda with the minimum mean square error of the model is determined as the optimal value of lambda by using cross-over validation;
fitting the regression model path of the formula (4) through the SCAD punishment algorithm of the formula (5), and respectively estimating a regression coefficient beta j In the estimation process, the regression coefficient with small influence is shrunk to 0, so that the explanation variable x is realized ij Selecting the variable (c).
The method for selecting and weighting the variables in the cigarette making process based on the SCAD algorithm preferably determines the influence weight of the explanatory variable according to the estimation result of the regression coefficient of the linear regression model, and specifically includes:
normalizing regression coefficient beta in formula (2) of multiple regression model j As an explanatory variable x ij A measure of the degree of influence;
the explanatory variable x is calculated by the following formula ij For the explained variable y i Influence weight of (2):
Figure BDA0003677656830000043
wherein, beta j Representing a metric-interpretation variable x i Normalized regression coefficient of relative importance, | β j I represents the normalized regression coefficient beta j Taking an absolute value;
Figure BDA0003677656830000044
represents the absolute value of | β for all normalized regression coefficients j Sum, | W i Representing an explanatory variable x ij For the explained variable y i The influence weight of (c).
The invention provides a variable selection and weighting method for a cigarette filament making process based on an SCAD algorithm, which adopts an SCAD punishment method, increases an L1 punishment item and the weight thereof on the basis of least square, continuously processes the multiple collinearity problem among variables in sequence, solves the problem of excessive compression of coefficients in the Lasso process, meets Oracle properties, namely the sparsity, continuity and unbiasedness of variable selection, and synchronously realizes variable screening and objective weighting in the cigarette filament making process; the method for screening and empowering the key parameters of the silk making process based on the full sample data can provide reference for the accurate control of the key quality characteristics of silk making and the evaluation of the process quality.
Drawings
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the accompanying drawings, in which:
fig. 1 is a flowchart of an embodiment of a variable selection and weighting method for a cigarette making process based on an SCAD algorithm provided by the invention.
Detailed Description
Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is merely illustrative and is in no way intended to limit the disclosure, its application, or uses. The present disclosure may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that: the relative arrangement of parts and steps, the composition of materials, numerical expressions and numerical values set forth in these embodiments are to be construed as merely illustrative, and not as limitative, unless specifically stated otherwise.
As used in this disclosure, "first", "second": and the like, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element preceding the word covers the element listed after the word, and does not exclude the possibility that other elements are also covered. "upper", "lower", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
In the present disclosure, when a specific component is described as being located between a first component and a second component, there may or may not be intervening components between the specific component and the first component or the second component. When it is described that a specific component is connected to other components, the specific component may be directly connected to the other components without having an intervening component, or may be directly connected to the other components without having an intervening component.
All terms (including technical or scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs unless specifically defined otherwise. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail, but are intended to be part of the specification where appropriate.
As shown in fig. 1, the method for selecting and assigning weights to variables in a cigarette making process based on an SCAD algorithm provided in this embodiment specifically includes the following steps in an actual execution process:
and step S1, obtaining a data sample of the cigarette shred manufacturing process.
In an embodiment of the method for selecting and assigning weights to variables in a cigarette making process based on the SCAD algorithm of the present invention, the step S1 may specifically include:
and step S11, obtaining an original data sample of the cigarette shred manufacturing process.
Specifically, the raw data samples of the cigarette making process are collected according to a Manufacturing Execution System (MES).
And step S12, obtaining a steady-state data sample based on the original data sample.
In an embodiment of the method for selecting and assigning weights to cigarette primary processing variables based on the SCAD algorithm, the step S12 may specifically include:
and step S121, rejecting the shutdown and material-breaking batch data according to a judgment rule that the process inlet flow is reduced to 0 and the duration exceeds 90S in the production process.
And step S122, intercepting the effective data according to the interception rule of the effective data.
And S123, eliminating set parameters with the production process numerical values as constants.
And step S124, grouping the effective data according to the residence time of the material and calculating an average value to form a steady-state data sample.
And step S2, constructing a least square linear regression model according to the data samples.
In an embodiment of the method for selecting and assigning weights to cigarette primary processing variables based on the SCAD algorithm, the step S2 may specifically include:
and step S21, carrying out standardization processing on the data sample.
The data samples are normalized, so that the influence of units and the difference of self variables can be eliminated. Wherein each of the steady state data samples is normalized using a normal normalization method,
specifically, the steady state data sample x is calculated by the following formula 1 ,x 2 ,...,x i ,...,x n And (3) carrying out transformation:
Figure BDA0003677656830000061
where n represents the number of samples of steady-state data samples, x i Representing the ith data, z, in steady-state data samples i Representing the normalized steady-state data samples,
Figure BDA0003677656830000077
represents the mean of the steady-state data samples and s represents the standard deviation of the steady-state data samples.
Take the water content of the discharged material after loosening and moisture regaining as an example, z i Representing a steady state data sample of the moisture content of the loose moisture regaining discharged material after standardization,
Figure BDA0003677656830000078
and s represents the standard deviation of the steady-state data sample of the moisture content of the loose and remoistened discharging material.
And step S22, constructing a least square linear regression model according to the normalized data samples.
Specifically, a least-squares linear regression model is constructed from normalized steady-state data
Figure BDA0003677656830000071
Obtaining an estimated value of a variable coefficient of a linear regression model by a least square method
Figure BDA0003677656830000072
Satisfy the requirement of
Figure BDA0003677656830000073
Wherein, y i Representing the explained variable of the regression model, and taking the discharged water content or the discharged temperature of each procedure as the explained variable; x is the number of ij The explanation variables representing the regression model comprise equipment parameters and/or process parameters of each process, and j is the number of the explanation variables; beta is a j Representing a metric-interpretation variable x ij Normalized regression coefficient of relative importance, alpha i And beta j Is the regression coefficient to be estimated of the model; epsilon i An error term representing the model is then calculated,
Figure BDA0003677656830000074
the minimum calculation factor of the formula is represented.
And step S3, constructing an SCAD penalty function, estimating a regression coefficient of the linear regression model by using the SCAD penalty function, and compressing the regression coefficient in the estimation process to perform variable selection on the explanatory variable.
In an embodiment of the method for selecting and assigning weights to cigarette primary processing variables based on the SCAD algorithm, the step S3 may specifically include:
step S31, constructing a SCAD (synthetic penalty) penalty function P λ (|β j |) on the basis of the formula (3) and the SCAD penalty function, an estimated value of a variable coefficient of the linear regression model is calculated by the following formula
Figure BDA0003677656830000075
Figure BDA0003677656830000076
Wherein, SCAD penalty function P λ (|β j |) is as follows:
Figure BDA0003677656830000081
wherein a and lambda are only formula parameters, and have no practical meaning, a is more than 2, and usually takes a value of 3.7, and the parameter lambda with the minimum mean square error of the model is determined as the optimal value of lambda by applying cross-folding verification.
Step S32, fitting the regression model path of the formula (4) through the SCAD penalty algorithm of the formula (5), and respectively estimating the regression coefficient beta j In the estimation process, the regression coefficient with small influence is shrunk to 0, so that the explanation variable x is realized ij The variables of (2) are selected.
By introducing the SCAD penalty algorithm of the formula (5), the fitting regression model path of the formula (4) is realized, the regression coefficients are respectively estimated, and the regression coefficients with small influence are shrunk to 0 in the estimation process, so that the variable selection of the explanatory variable is realized.
And step S4, determining the influence weight of the explanatory variable according to the estimation result of the regression coefficient of the linear regression model.
In an embodiment of the method for selecting and assigning weights to cigarette primary processing variables based on the SCAD algorithm, the step S4 may specifically include:
step S41, normalizing regression coefficient beta in multiple regression model formula (2) j As an explanatory variable x ij A measure of the degree of influence.
Taking the loosening and moisture regaining process as an example, the normalized regression coefficient of the water addition amount of the loosening and moisture regaining is 0.3, so the influence degree on the discharged water content is 0.3, and the normalized regression coefficient of the flow rate of the loosening and moisture regaining process is 0.05, so the influence degree on the discharged water content is 0.05.
Step S42, calculating an explanatory variable x by the following formula ij For the explained variable y i Influence weight of (2):
Figure BDA0003677656830000082
wherein, beta j Representing a metric-interpretation variable x i Normalized regression coefficient of relative importance, | β j I represents the normalized regression coefficient beta j Taking an absolute value;
Figure BDA0003677656830000083
represents the absolute value of | β for all normalized regression coefficients j Sum, | W i Representing an explanatory variable x ij For the explained variable y i The influence weight of (c).
According to the variable selection and weighting method for the cigarette filament making process based on the SCAD algorithm, the SCAD punishment method is adopted, the L1 punishment item and the weight thereof are added on the basis of least square, the problem of multiple collinearity among variables is continuously processed in sequence, the problem of excessive compression of coefficients in the Lasso process is solved, the Oracle property is met, namely the sparsity, the continuity and the unbiased property of variable selection are met, and the variable screening and the objective weighting of the cigarette filament making process are synchronously realized; the method for screening and weighting the key parameters of the silk making process established based on the full sample data can provide reference for accurate control of the key quality characteristics of silk making and evaluation of process quality.
Thus, various embodiments of the present disclosure have been described in detail. Some details that are well known in the art have not been described in order to avoid obscuring the concepts of the present disclosure. It will be fully apparent to those skilled in the art from the foregoing description how to practice the presently disclosed embodiments.
Although some specific embodiments of the present disclosure have been described in detail by way of example, it should be understood by those skilled in the art that the foregoing examples are for purposes of illustration only and are not intended to limit the scope of the present disclosure. It will be understood by those skilled in the art that various changes may be made in the above embodiments or equivalents may be substituted for elements thereof without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims (9)

1. A cigarette cut tobacco making process variable selection and weighting method based on an SCAD algorithm is characterized by comprising the following steps:
acquiring a data sample of a cigarette shredding process;
constructing a linear regression model of least squares according to the data samples;
constructing an SCAD penalty function, estimating a regression coefficient of the linear regression model by using the SCAD penalty function, and compressing the regression coefficient in the estimation process to perform variable selection on an explanatory variable;
and determining the influence weight of the explanatory variable according to the estimation result of the regression coefficient of the linear regression model.
2. The method for selecting and weighting cigarette throwing process variables based on the SCAD algorithm according to claim 1, wherein the obtaining of the data samples of the cigarette throwing process specifically comprises:
obtaining an original data sample in the cigarette shredding process;
based on the raw data samples, steady-state data samples are obtained.
3. The method for selecting and weighting cigarette throwing process variables based on the SCAD algorithm according to claim 2, wherein the obtaining of the original data sample of the cigarette throwing process specifically comprises:
and acquiring an original data sample of the cigarette making process according to the cigarette making process manufacturing execution system.
4. The cigarette throwing process variable selecting and weighting method based on the SCAD algorithm according to claim 2, wherein the obtaining of the steady-state data sample based on the original data sample specifically comprises:
rejecting shutdown material-breaking batch data according to a judgment rule that the flow of a process inlet is reduced to 0 and the duration time exceeds 90s in the production process;
intercepting the effective data according to the interception rule of the effective data;
eliminating set parameters with the production process numerical values as constants;
and grouping the effective data according to the residence time of the materials, and calculating an average value to form a steady-state data sample.
5. The cigarette throwing process variable selecting and weighting method based on the SCAD algorithm according to claim 2, wherein the constructing a least squares linear regression model according to the data samples specifically comprises:
carrying out standardization processing on the data sample;
and constructing a least square linear regression model according to the data samples after the normalization processing.
6. The cigarette throwing process variable selecting and weighting method based on SCAD algorithm according to claim 5, wherein the data sample is standardized, specifically comprising:
each of the steady state data samples is normalized using a normal normalization method,
for steady state data sample x by the following formula 1 ,x 2 ,...,x i ,...,x n And (3) carrying out transformation:
Figure FDA0003677656820000021
where n represents the number of samples of steady-state data samples, x i Representing the ith data, z, in steady-state data samples i Representing the normalized steady-state data samples,
Figure FDA0003677656820000022
represents the mean of the steady-state data samples and s represents the standard deviation of the steady-state data samples.
7. The cigarette throwing process variable selecting and weighting method based on the SCAD algorithm according to claim 5, wherein the construction of a least-squares linear regression model according to the data samples after the standardization process specifically comprises:
constructing a least squares linear regression model from normalized steady state data
Figure FDA0003677656820000023
Figure FDA0003677656820000024
Obtaining an estimated value of a variable coefficient of a linear regression model by a least square method
Figure FDA0003677656820000025
Satisfy the requirement of
Figure FDA0003677656820000026
Wherein, y i Representing the explained variable of the regression model, and taking the discharged water content or the discharged temperature of each procedure as the explained variable; x is the number of ij The explanation variables of the regression model comprise equipment parameters and/or process parameters of each process, and j is the number of the explanation variables; beta is a j Representing a metric-interpreting variable x ij Normalized regression coefficient of relative importance, alpha i And beta j Is the regression coefficient to be estimated of the model; epsilon i An error term representing the model is then calculated,
Figure FDA0003677656820000027
the minimum calculation factor of the formula is represented.
8. The method for selecting and weighting the variables in the cigarette making process based on the SCAD algorithm according to claim 7, wherein the method comprises the steps of constructing an SCAD penalty function, estimating a regression coefficient of a linear regression model by using the SCAD penalty function, and compressing the regression coefficient in the estimation process to perform variable selection on an explanation variable, and specifically comprises the following steps:
construction of SCAD penalty function P λ (|β j |) on the basis of the formula (3) and the SCAD penalty function, an estimated value of a variable coefficient of the linear regression model is calculated by the following formula
Figure FDA0003677656820000031
Figure FDA0003677656820000032
SCAD penalty function P λ (|β j |) is as follows:
Figure FDA0003677656820000033
wherein a and lambda are only formula parameters, no practical meaning exists, a is more than 2, and a parameter lambda with the minimum mean square error of the model is determined as the optimal value of lambda by using cross-over validation;
fitting the regression model path of formula (4) through the SCAD penalty algorithm of formula (5), and respectively estimating the regression coefficient beta j In the estimation process, the regression coefficient with small influence is shrunk to 0, so that the explanation variable x is realized ij Selecting the variable (c).
9. The cigarette throwing process variable selecting and weighting method based on SCAD algorithm according to claim 8, wherein the determining of the influence weight of the explanatory variable according to the estimation result of the regression coefficient of the linear regression model specifically comprises:
normalizing regression coefficient beta in formula (2) of multiple regression model j As an explanatory variable x ij A measure of the degree of influence;
the explanatory variable x is calculated by the following formula ij For the explained variable y i Influence weight of (2):
Figure FDA0003677656820000034
wherein, beta j Representing a metric-interpretation variable x i Normalized regression coefficient of relative importance, | β j I represents the normalized regression coefficient beta j Taking an absolute value;
Figure FDA0003677656820000035
represents the absolute value of | β for all normalized regression coefficients j Sum, | W i Representing an explanatory variable x ij For the explained variable y i The influence weight of (c).
CN202210626198.9A 2022-06-02 2022-06-02 Cigarette shred making process variable selection and weighting method based on SCAD algorithm Pending CN115017700A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210626198.9A CN115017700A (en) 2022-06-02 2022-06-02 Cigarette shred making process variable selection and weighting method based on SCAD algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210626198.9A CN115017700A (en) 2022-06-02 2022-06-02 Cigarette shred making process variable selection and weighting method based on SCAD algorithm

Publications (1)

Publication Number Publication Date
CN115017700A true CN115017700A (en) 2022-09-06

Family

ID=83073200

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210626198.9A Pending CN115017700A (en) 2022-06-02 2022-06-02 Cigarette shred making process variable selection and weighting method based on SCAD algorithm

Country Status (1)

Country Link
CN (1) CN115017700A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117332923A (en) * 2023-10-09 2024-01-02 北京京航计算通讯研究所 Weighting method and system for netlike index system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117332923A (en) * 2023-10-09 2024-01-02 北京京航计算通讯研究所 Weighting method and system for netlike index system
CN117332923B (en) * 2023-10-09 2024-03-26 北京京航计算通讯研究所 Weighting method and system for netlike index system

Similar Documents

Publication Publication Date Title
CN107563672A (en) A kind of method based on analytic hierarchy process (AHP) structure cigarette primary processing Environmental Evaluation Model
CN112036701A (en) Technological parameter optimization method for multi-process industrial production process
CN111144667A (en) Tobacco conditioner discharged material water content prediction method based on gradient lifting tree
CN111160425A (en) Neural network-based flue-cured tobacco comfort classification evaluation method
CN103844344A (en) Method for regulating and controlling tobacco shred quality uniformity of different batches of cigarettes and application of method
CN112273696B (en) Method, device and equipment for controlling moisture after shredding
CN115017700A (en) Cigarette shred making process variable selection and weighting method based on SCAD algorithm
CN112273695A (en) Method, device and equipment for predicting water content of loose moisture regain outlet
CN113011491A (en) Hot continuous rolling strip steel width prediction method based on principal component analysis and random forest cooperation
CN114403487B (en) Water adding control method for loosening and dampening
CN111339478A (en) Weather data quality evaluation method based on improved fuzzy analytic hierarchy process
CN110705859A (en) PCA-self-organizing neural network-based method for evaluating running state of medium and low voltage distribution network
CN111678841A (en) Evaluation method suitable for cigarette density uniformity
CN113076692B (en) Method for inverting nitrogen content of leaf
CN110916239A (en) Primary flue-cured tobacco leaf segmentation method based on different zone bit color difference changes
CN110973687B (en) Accurate control method for moisture in silk making process
CN108614536B (en) Complex network construction method for key factors of cigarette shred making process
CN110866221B (en) Detection method for threshing and redrying data accuracy
CN109962915B (en) BQP network-based anomaly detection method
CN112101817A (en) Method for evaluating homogenization degree of batch-processed flue-cured tobacco finished products
CN114166794B (en) Method, medium and equipment for predicting tobacco flake quality
CN117493911A (en) Cut-tobacco dryer outlet moisture prediction method based on multiple linear regression model
CN117281281A (en) Tobacco threshing process parameter acquisition method, device and equipment
Anoop et al. Rice Crop Yield Prediction Using Multi-Level Machine Learning Techniques
CN116777292B (en) Defect rate index correction method based on multi-batch small sample space product

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