CN107578105A - System parameter design space optimization method and device - Google Patents
System parameter design space optimization method and device Download PDFInfo
- Publication number
- CN107578105A CN107578105A CN201710771774.8A CN201710771774A CN107578105A CN 107578105 A CN107578105 A CN 107578105A CN 201710771774 A CN201710771774 A CN 201710771774A CN 107578105 A CN107578105 A CN 107578105A
- Authority
- CN
- China
- Prior art keywords
- key
- design space
- coefficient
- parameter
- procedure parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 208
- 238000013461 design Methods 0.000 title claims abstract description 88
- 238000005457 optimization Methods 0.000 title claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 82
- 238000012216 screening Methods 0.000 claims abstract description 35
- 238000004519 manufacturing process Methods 0.000 claims abstract description 32
- 230000000153 supplemental effect Effects 0.000 claims abstract description 7
- 238000004458 analytical method Methods 0.000 claims description 21
- 238000010219 correlation analysis Methods 0.000 claims description 13
- 241000208340 Araliaceae Species 0.000 claims description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 235000008434 ginseng Nutrition 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 abstract description 2
- 239000000047 product Substances 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000012795 verification Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 238000004886 process control Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 241000205585 Aquilegia canadensis Species 0.000 description 2
- 238000003070 Statistical process control Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000013401 experimental design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 239000013067 intermediate product Substances 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- 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
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/14—Quality control systems
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Automation & Control Theory (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- Medical Informatics (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Manufacturing & Machinery (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- General Factory Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of System Parameter Design space optimization method and device, the optimization method includes：Acquisition process supplemental characteristic, the procedure parameter include the mass parameter of the intermediate of a upper workshop section；According to workshop section's condition of production, the type of selection Key Quality attribute；The screening procedure parameter related to the Key Quality attribute, as key process parameters；The relational model established between key process parameters and the Key Quality attribute；According to the relational model, design space is obtained, the design space is the specific interval range corresponding to the Key Quality attribute.The present invention establishes both relational models by filtering out the key process parameters related to Key Quality attribute, optimization System Parameter Design space, and a kind of effective, reliable solution is provided for parametric release.
Description
Technical field
The present invention relates to PROCESS KNOWLEDGE SYSTEM field, more particularly to a kind of System Parameter Design space optimization method and dress
Put.
Background technology
In PROCESS KNOWLEDGE SYSTEM (Process Knowledge System, abbreviation PKS), the setting in parameter designing space
It is that one kind of PKS system parametric release is set.
In the prior art, the setting result in parameter designing space is often single-point parameter area, and temperature is extracted with honeysuckle
Spend exemplified by parameter, collect the honeysuckle extraction workshop section temperature parameter data of multiple batches first, line number secondly is entered to each batch
According to processing, the key point of reaction temperature change is extracted, again, each key point is handled, at the process of multiple batches
The key point normal distribution fitting of reason, final parameter designing point is used as using some degree of fitting.
And the design space of single-point causes the scope of PKS parametric releases narrow, can not also be made even if process capability is relatively low
The optimization of design space, it is highly detrimental to PKS and is played a role in intelligence manufacture.There is no still one kind to be directed to PKS parameters at present
The optimization method of design space.
The content of the invention
In order to solve problem of the prior art, the invention provides a kind of System Parameter Design space optimization method and dress
Put, by filtering out the key process parameters related to Key Quality attribute, and establish both relational models, optimization system ginseng
Number design space, a kind of effective, reliable solution is provided for parametric release.The technical scheme is as follows：
On the one hand, the invention provides a kind of System Parameter Design space optimization method, including：
Acquisition process supplemental characteristic, the procedure parameter include the mass parameter of the intermediate of a upper workshop section；
According to workshop section's condition of production, the type of selection Key Quality attribute；
The screening procedure parameter related to the Key Quality attribute, as key process parameters；
The relational model established between key process parameters and the Key Quality attribute；
According to the relational model, design space is obtained, the design space is corresponding to the Key Quality attribute
Specific interval range.
Further, the screening procedure parameter related to the Key Quality attribute includes：
Relative standard deviation analysis is carried out to procedure parameter, obtains the procedure parameter for meeting deviation threshold；
Correlation analysis is carried out to the procedure parameter for meeting deviation threshold, obtains procedure parameter and Key Quality attribute
Between coefficient correlation；
According to the coefficient correlation, the notable coefficient between procedure parameter and the Key Quality attribute is obtained；
According to the notable coefficient, screening obtains key process parameters.
Further, described according to the notable coefficient, screening, which obtains key process parameters, to be included：
It is determined that notable coefficient threshold, the notable coefficient of target of the significantly coefficient threshold is yielded less than；
According to the notable coefficient of the target, procedure parameter corresponding to acquisition, as key process parameters.
Further, the relational model established between key process parameters and the Key Quality attribute includes：
Data normalization operation is carried out to key process parameters and Key Quality attribute；
Using stepwise regression method, key process parameters and Key Quality attribute to standardization are fitted operation, obtained
To relational model.
Further, also include after the acquisition design space：The design space is verified, including：
A number of point is respectively taken the design space scope is inner and outer；
Obtain qualitative data corresponding to the point, by the qualitative data of the inner and outer point of the design space scope respectively with
Default quality standard compares, and is verified result；And/or
Using the design space as parametric release scope, qualitative attribute data are collected；
Calculate the process capability of qualitative attribute；
Compared with original procedure ability, result is verified.
Further, the relative standard deviation analysis is realized by below equation：
Wherein, xiFor the value of certain procedure parameter of each batch sample, n holds for sample
Amount,For the average value of sample procedure parameter.
Further, include during successive Regression：
Obtain the coefficient of determination of relational model；
Compare the size of the coefficient of determination and decision degree threshold value；
If the coefficient of determination is less than the decision degree threshold value, expand selected confidence level, until the coefficient of determination
More than decision degree threshold value.
On the other hand, the invention provides a kind of System Parameter Design space optimization device, including with lower module：
Procedure parameter module, for acquisition process supplemental characteristic, the procedure parameter includes the intermediate of a upper workshop section
Mass parameter；
CQA selecting modules, for according to workshop section's condition of production, the type of selection Key Quality attribute；
CPP screening modules, for screening the procedure parameter related to the Key Quality attribute, join as critical process
Number；
Model module, for establishing the relational model between key process parameters and the Key Quality attribute；
Design space module, for according to the relational model, obtaining design space, the design space is corresponding to institute
State the specific interval range of Key Quality attribute.
Further, the CPP screening modules include：
RSD analytic units, for carrying out relative standard deviation analysis to procedure parameter, obtain the process for meeting deviation threshold
Parameter；
Dependency analysis unit, for carrying out correlation analysis to the procedure parameter for meeting deviation threshold, obtained
Coefficient correlation between journey parameter and Key Quality attribute；
Notable coefficient elements, for according to the coefficient correlation, obtaining between procedure parameter and the Key Quality attribute
Notable coefficient；
Screening unit, for obtaining key process parameters according to the notable coefficient, screening.
Further, described device also includes authentication module, and the authentication module includes：
Dot element is taken, for respectively taking a number of point the design space scope is inner and outer；
Quality standard comparing unit, for obtaining qualitative data corresponding to the point, by the range of the design space and
The qualitative data of outer point compared with default quality standard, is verified result respectively；And/or
Clearance unit, for using the design space as parametric release scope, collecting qualitative attribute data；
Capacity calculation unit, for calculating the process capability of qualitative attribute；
Ability comparing unit, for compared with original procedure ability, being verified result.
What technical scheme provided by the invention was brought has the beneficial effect that：
1) by RSD analyses and correlation analysis, the key process parameters related to Key Quality attribute is filtered out, are mould
The foundation of type provides reliable parameter material；
2) relational model established by stepwise regression method between key process parameters and Key Quality attribute, avoid more
The influence of weight synteny, the reliability of the adjustment model are high；
3) appropriate to expand selected confidence level, the coefficient of determination of Controlling model be higher than threshold value, guarantee final design space can
By property；
4) design space is further verified, verifies the accuracy of design space optimum results.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is System Parameter Design space optimization method flow diagram provided in an embodiment of the present invention；
Fig. 2 is the method flow diagram of screening key process parameters provided in an embodiment of the present invention；
Fig. 3 is the method flow diagram of founding mathematical models provided in an embodiment of the present invention；
Fig. 4 is the method flow diagram of expansion clearance parameter during successive Regression provided in an embodiment of the present invention；
Fig. 5 is the method flow diagram of checking design space provided in an embodiment of the present invention；
Fig. 6 is the module frame chart of System Parameter Design space optimization device provided in an embodiment of the present invention；
Fig. 7 is the software interface schematic diagram of design space provided in an embodiment of the present invention；
Fig. 8 is verification method flow chart in design space provided in an embodiment of the present invention.
Embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under the premise of creative work is not made, it should all belong to the model that the present invention protects
Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " etc. be for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that so use
Data can exchange in the appropriate case, so as to embodiments of the invention described herein can with except illustrating herein or
Order beyond those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, be not necessarily limited to for example, containing the process of series of steps or unit, method, apparatus, product or equipment
Those steps or unit clearly listed, but may include not list clearly or for these processes, method, product
Or the intrinsic other steps of equipment or unit.
Embodiment 1
In the embodiment of the present invention, by taking heavy concentration process as an example, process for producing data are collected by PKS, pass through successive Regression
The mathematical modeling that method is established between key process parameters and key quality parameters, design space finally is calculated with MODDE softwares, and
The standard that design space is let pass as workshop section's key process parameters, non-key procedure parameter can be put by nowadays production standards
OK, attempt to establish a kind of method that procedure parameter suitable for Chinese Traditional Medicine is let pass.
In one embodiment of the invention, there is provided a kind of System Parameter Design space optimization method, referring to Fig. 1, institute
Stating method includes below scheme：
S1, acquisition process supplemental characteristic.
Specifically, input process parameter mainly includes process parameter and input quality parameter is (described herein defeated
Enter procedure parameter and be referred to as procedure parameter), input process parameter is the mass parameter of upper workshop section's intermediate.
Acquisition data are mainly carried out by PKS data export function, averaged as needed after data export.Chinese medicine
Production process is complicated, and it is numerous to influence the factor of product quality, statistic process control theory (Statistical Process
Control, SPC) in the reason for thinking to cause product or intermediate quality fluctuation element essentially from 6 aspects, i.e. people, machine
Device, material, method, measurement, environment (5M1E), in order to cover the factor for causing product quality to fluctuate as far as possible, it is recommended to use cause and effect
Figure combines 5M1E analytic approach preliminary screening procedure parameters.
S2, the type for selecting Key Quality attribute.
The selection of Key Quality attribute (Critical QualityAttributes, CQA) should have according to different workshop sections
System is determined.Workshop section is such as extracted, component content is main CQA；In concentration section, admittedly contain, medicinal extract proportion is CQA；In alcohol precipitation and
Workshop section is extracted, the rate of transform of index components, the clearance of impurity are CQA, etc..
Pre-established relation mapping table, after according to actual workshop section, match corresponding Key Quality attribute.
S3, screening key process parameters.
CQA is the quality objective of refinement, and key process parameters (Critical Process Parameter, CPP) are
The process control condition for realizing CQA, therefore, it is necessary to screen the procedure parameter of (reaching to a certain degree) related to the CQA, work
For key process parameters (hereinafter referred to as CPP).
In the present embodiment, actual production is screened using RSD (relative standard deviation) analyses and correlation analysis combination
The middle key process parameter for influenceing product quality, data processing are entered in the RSD analysis modules and correlation module of PKS system
OK.It is more convenient with reference to the software use such as SPSS, Minitab.
S4, the relational model established between key process parameters and the Key Quality attribute.
In order to describe relations of the CPPs (Critical Process Parameters) between CQA, CPPs model is formulated
Enclose, it is necessary to the mathematical modeling established between CPPs and CQA.
S5, according to the relational model, obtain design space.
The design space is the scope corresponding to the CPP in the range of the specific section of the Key Quality attribute.According to
The mathematical modeling of foundation, the CPP scopes that CQA is realized under particular probability are obtained, as design space, with this control system parameter
Clearance, be advantageous to improve the accuracy of process control of the PROCESS KNOWLEDGE SYSTEM to manufacturing process, improve product manufacturing quality.
Embodiment 2
In one embodiment of the invention, there is provided a kind of method for screening key process parameters, it is described referring to Fig. 2
Method includes below scheme：
S21, relative standard deviation analysis is carried out to procedure parameter, obtain the procedure parameter for meeting deviation threshold.
In actual production, not all procedure parameter has break-up value, because in the technique of actual production,
Some procedure parameters are fixed or simply changed in small scope that these parameters are very small due to changing, can be with
It is considered as constant, does not include analysis.Herein, using relative standard deviation (relative standard deviation, RSD)
The purpose of analysis is exactly that preliminary screening obtains fluctuating bigger procedure parameter, that is, obtains the procedure parameter for meeting deviation threshold.
RSD calculation formula is：
Wherein, xiFor the value of certain procedure parameter of each batch sample, n holds for sample
Amount,For the average value of sample procedure parameter.
The parameter for being typically chosen RSD larger (>=3%) carries out correlation analysis, and RSD threshold value may be configured as 3%, but only
Refer to, actual threshold also needs to multiple analysis to determine.
S22, correlation analysis is carried out to the procedure parameter for meeting deviation threshold, obtain procedure parameter and Key Quality
Coefficient correlation between attribute.
In actual production, different procedure parameters is different, some parameters to the quality influence degree of intermediate
Decisive role is played to the quality of final products, influence of some parameters to intermediate or end product quality is then smaller,
The purpose for carrying out correlation analysis is the key process parameters and its contribution degree that can intuitively find out influence product, finds out life
The inherent law of production process, deepen the understanding to production process, autotelic control preferably is carried out to the parameters in production
System.
In embodiments of the present invention, the Pearson's coefficient in Statistics Application is measured between CQA and CPP (or CPPs)
Correlation, its value is between -1 to 1.The mathematical definition of coefficient correlation between two variables is：
Wherein, σXFor the standard deviation of X attributes, X attributes are the related features of procedure parameter, are such as pressed
Power average value, temperature averages, variance and temporal characteristics；σYFor the standard deviation of Y attributes, Y CQA, such as the rate of transform, the rate of transform
With solid content etc..Cov (X, Y) is X and Y covariance, and it is defined as, cov (X, Y)=E [(X- μX)(Y-μY)], wherein, μX,
μYRespectively attribute X and Y average value.
S23, according to the coefficient correlation, obtain the notable coefficient between procedure parameter and the Key Quality attribute.
We are influenceed using RSD (relative standard deviation) analyses and correlation analysis combination to screen in actual production herein
The key process parameter of product quality, data processing are carried out in the RSD analysis modules and correlation module of PKS system.Knot
It is more convenient to close the software use such as SPSS, Minitab.Recommend SPSS, Minitab.
Pearson's coefficient is calculated by taking Minitab as an example, the item to be analyzed is chosen in Minitab softwares and selects skin
The inferior Y-factor method Y of that, shows P values, as meets the notable coefficient between the procedure parameter of deviation threshold and Key Quality attribute.
S24, according to the notable coefficient, screening obtains key process parameters.
In correlation analysis, notable FACTOR P value is selected to be less than the ginseng of 0.05 (or carrying out analysis selection according to actual conditions)
Number is used as CPP.Pass through above-mentioned two step, you can obtain workshop section difference CQA CPP.
Method by screening CPP in the present embodiment, find out CQA and the preliminary rules of CPP in production process, preliminary intensification pair
The understanding of production process, good sample basis is provided for following founding mathematical models.
Embodiment 3
In one embodiment of the invention, there is provided a kind of method of founding mathematical models, referring to Fig. 3, methods described
Including below scheme：
S41, data normalization operation is carried out to key process parameters and Key Quality attribute.
The mode of standardization is a lot, has standard deviation standardization, extreme difference standardization etc., can the more attempt, compare in research
As a result difference.
S42, using stepwise regression method, key process parameters and Key Quality attribute to standardization are fitted behaviour
Make, obtain relational model.
The method of recurrence also has a lot, including linear processes.Used herein is that the method for successive Regression (avoids
The influence of multicollinearity), attemptable method is also including SVM, MLR, BP etc..In embodiments of the present invention, use
Minitab carries out successive Regression, and ' statistics ' → ' recurrence ' → ' model of fit ' is specially chosen in minitab, selectes response
Variable and independent variable, in ' progressively ' option, it is set into from α and deletions α and progressively type, returns after terminating, counted
Learn model, i.e. regression equation.
In general, compared with laboratory data, in lab scale research, when CQA is more, the side of overall merit can be used
Method, such as partial regression coefficient method, comprehensive multi-index rating method screens CPP, but actual production is due to no experimental design, it is difficult to take
Obtain R in lab scale research2Higher partial regression equation, actual production data are not particularly suited in this way.And production technology by
In relative maturity, Variable Factors are often fixed several factors, therefore, the relatively simple RSD of principle and correlation analysis connection
With, it is more preferable in actual production data screening CPP effect, but also have drawback, because it is more still to lack a kind of overall merit screening
The key process parameters of index, if finding design space with CQA one by one, it is more likely that design space occur does not have what is occured simultaneously
Situation, i.e., compared compared with laboratory data, the coefficient of determination R in manufacturing process2It is less than normal, therefore still need to find a kind of suitable actual life
The screening CPP of production method.(CPPs for using comprehensive each CQA corresponding CPPs as overall evaluation index is attempted at present, by
Step expands when recurrence into from α, until R2More than 0.85), i.e., as shown in figure 4, including during successive Regression：
S421, the coefficient of determination for obtaining relational model；
The coefficient of determination described in S422, comparison and the size of decision degree threshold value, if the coefficient of determination is less than the decision degree
Threshold value, then perform S423；
S423, expand selected confidence level, until the coefficient of determination is more than decision degree threshold value.
Expand confidence level to be obtained beyond CPP less than 0.05 screening according to notable FACTOR P value i.e. in upper one embodiment, expand
Enter to be selected in P values greatly more than or equal to the procedure parameter under 0.05 scope, on condition that expanding selected procedure parameter makes coefficient of determination R2
Increase, according in embodiments of the present invention, the decision degree threshold value is preferably 0.85, or can be divided according to actual condition
Analysis selection.
Embodiment 4
In one embodiment of the invention, there is provided a kind of method verified to design space, referring to Fig. 5, institute
Stating method includes below scheme：
S611, respectively take a number of point the design space scope is inner and outer；
S612, qualitative data corresponding to the point is obtained, by the qualitative data of the inner and outer point of the design space scope
Respectively compared with default quality standard, result is verified.
Such as：The point in the range of 10 design spaces is taken in actual production process, and obtains corresponding quality in production
Data, compared with default quality standard；Similarly, the point outside 10 design space scopes is taken in actual production process,
And corresponding mass data in production are obtained, compared with default quality standard.Comparative result if in the range of institute a little (or
Person's major part point, such as 9 points) corresponding mass data be more than default quality standard, and the institute outside scope is a little (or big
Partial dot, such as 9 points) corresponding mass data be less than default quality standard, then verify design space optimum results standard
True property, otherwise, the design space scope are inaccurate.
In another embodiment of the present invention, there is provided the method verified to design space of another kind, referring to Fig. 8,
Methods described includes below scheme：
S621, using the design space as parametric release scope, collect qualitative attribute data；
S622, the process capability for calculating qualitative attribute；
S623, compared with original procedure ability, be verified result.
If after carrying out parametric release according to design space, the process capability of qualitative attribute increases, then checking design is empty
Between optimum results accuracy, otherwise, the design space scope is inaccurate.The verification method and Fig. 8 in Fig. 7 can also be used
In verification method combine, to the design space carry out double verification.
The searching of design space is based primarily upon the progress of MODDE softwares.When response variable is more, ' 4D Design are selected
Space ', as a result as shown in fig. 7, the region divided according to contour in figure, can carry out selecting accordingly according to actual selection
Parametric release is implemented in design space, such as, implementation success probability 99% can meet system requirements, then select mark in Fig. 7 and be
Contour, the contour interior zone can be used as the foundation and scope of parametric release.
Wherein, it is a little acquisition process parameter that in design space, scope inside/outside, which takes, and S62 is core procedure, in corresponding mistake
Quality sample is collected under journey parameter, and the quality sample is handled, according to certain method formula calculating process ability,
Do not deploy to describe herein.
Design space is verified in the embodiment of the present invention, further ensure that the reliability of parametric release and accurate
Degree, laid a good foundation for intelligence manufacture control.
Embodiment 5
In embodiments of the present invention, there is provided a kind of System Parameter Design space optimization device, referring to Fig. 6, including it is following
Module：
Procedure parameter module 610, for acquisition process supplemental characteristic, the procedure parameter includes the intermediate of a upper workshop section
Mass parameter；
CQA selecting modules 620, for according to workshop section's condition of production, the type of selection Key Quality attribute；
CPP screening modules 630, for screening the procedure parameter related to the Key Quality attribute, as critical process
Parameter；
Model module 640, for establishing the relational model between key process parameters and the Key Quality attribute；
Design space module 650, for according to the relational model, obtaining design space, the design space is corresponding
In the specific interval range of the Key Quality attribute.
Wherein, the CPP screening modules 630 include：
RSD analytic units 631, for carrying out relative standard deviation analysis to procedure parameter, obtain meeting deviation threshold
Procedure parameter；
Dependency analysis unit 632, for carrying out correlation analysis to the procedure parameter for meeting deviation threshold, obtain
Coefficient correlation between procedure parameter and Key Quality attribute；
Notable coefficient elements 633, for according to the coefficient correlation, obtain the procedure parameter and Key Quality attribute it
Between notable coefficient；
Screening unit 634, for obtaining key process parameters according to the notable coefficient, screening.
In another preferred embodiment, the System Parameter Design space optimization device also includes authentication module 660, described
Authentication module includes：
Dot element 661 is taken, for respectively taking a number of point the design space scope is inner and outer；
Quality standard comparing unit 662, for obtaining qualitative data corresponding to the point, in the range of the design space
With the qualitative data of outer point respectively compared with default quality standard, result is verified；And/or
Clearance unit 663, for using the design space as parametric release scope, collecting qualitative attribute data；
Capacity calculation unit 664, for calculating the process capability of qualitative attribute；
Ability comparing unit 665, for compared with original procedure ability, being verified result.
It should be noted that：The System Parameter Design space optimization device that above-described embodiment provides is to be designed space excellent
, can be as needed and by above-mentioned function only with the division progress of above-mentioned each functional module for example, in practical application during change
Distribution is completed by different functional modules, i.e., the internal structure of System Parameter Design space optimization device is divided into different work(
Energy module, to complete all or part of function described above.In addition, the System Parameter Design space that the present embodiment provides is excellent
Change the design space optimization method that device embodiment provides with above-described embodiment and belong to same design, its specific implementation process refers to
Embodiment of the method, repeat no more here.
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (10)
- A kind of 1. System Parameter Design space optimization method, it is characterised in that including：Acquisition process supplemental characteristic, the procedure parameter include the mass parameter of the intermediate of a upper workshop section；According to workshop section's condition of production, the type of selection Key Quality attribute；The screening procedure parameter related to the Key Quality attribute, as key process parameters；The relational model established between key process parameters and the Key Quality attribute；According to the relational model, design space is obtained, the design space is corresponding to the specific of the Key Quality attribute Interval range.
- 2. according to the method for claim 1, it is characterised in that the screening process related to the Key Quality attribute Parameter includes：Relative standard deviation analysis is carried out to procedure parameter, obtains the procedure parameter for meeting deviation threshold；Correlation analysis is carried out to the procedure parameter for meeting deviation threshold, obtained between procedure parameter and Key Quality attribute Coefficient correlation；According to the coefficient correlation, the notable coefficient between procedure parameter and the Key Quality attribute is obtained；According to the notable coefficient, screening obtains key process parameters.
- 3. according to the method for claim 2, it is characterised in that described that crucial mistake is obtained according to the notable coefficient, screening Journey parameter includes：It is determined that notable coefficient threshold, the notable coefficient of target of the significantly coefficient threshold is yielded less than；According to the notable coefficient of the target, procedure parameter corresponding to acquisition, as key process parameters.
- 4. according to the method for claim 1, it is characterised in that described to establish the key process parameters and Key Quality category Relational model between property includes：Data normalization operation is carried out to key process parameters and Key Quality attribute；Using stepwise regression method, key process parameters and Key Quality attribute to standardization are fitted operation, closed It is model.
- 5. according to the method for claim 1, it is characterised in that also include after the acquisition design space：Set to described Space is counted to be verified, including：A number of point is respectively taken the design space scope is inner and outer；Qualitative data corresponding to the point is obtained, by the qualitative data of the inner and outer point of the design space scope respectively with presetting Quality standard compare, be verified result；And/orUsing the design space as parametric release scope, qualitative attribute data are collected；Calculate the process capability of qualitative attribute；Compared with original procedure ability, result is verified.
- 6. according to the method for claim 2, it is characterised in that the relative standard deviation analysis is real by below equation It is existing：Wherein, xiFor the value of certain procedure parameter of each batch sample, n is sample capacity, For the average value of sample procedure parameter.
- 7. according to the method for claim 4, it is characterised in that include during successive Regression：Obtain the coefficient of determination of relational model；Compare the size of the coefficient of determination and decision degree threshold value；If the coefficient of determination is less than the decision degree threshold value, expand selected confidence level, until the coefficient of determination is more than Decision degree threshold value.
- 8. a kind of System Parameter Design space optimization device, it is characterised in that including with lower module：Procedure parameter module, for acquisition process supplemental characteristic, the procedure parameter includes the quality of the intermediate of a upper workshop section Parameter；CQA selecting modules, for according to workshop section's condition of production, the type of selection Key Quality attribute；CPP screening modules, for screening the procedure parameter related to the Key Quality attribute, as key process parameters；Model module, for establishing the relational model between key process parameters and the Key Quality attribute；Design space module, for according to the relational model, obtaining design space, the design space is corresponding to the pass The specific interval range of key qualitative attribute.
- 9. device according to claim 8, it is characterised in that the CPP screening modules include：RSD analytic units, for carrying out relative standard deviation analysis to procedure parameter, obtain the process ginseng for meeting deviation threshold Number；Dependency analysis unit, for carrying out correlation analysis to the procedure parameter for meeting deviation threshold, obtain process ginseng Coefficient correlation between number and Key Quality attribute；Notable coefficient elements, it is aobvious between procedure parameter and the Key Quality attribute for according to the coefficient correlation, obtaining Write coefficient；Screening unit, for obtaining key process parameters according to the notable coefficient, screening.
- 10. device according to claim 8, it is characterised in that also include including authentication module, the authentication module：Dot element is taken, for respectively taking a number of point the design space scope is inner and outer；Quality standard comparing unit, it is for obtaining qualitative data corresponding to the point, the design space scope is inner and outer The qualitative data of point compared with default quality standard, is verified result respectively；And/orClearance unit, for using the design space as parametric release scope, collecting qualitative attribute data；Capacity calculation unit, for calculating the process capability of qualitative attribute；Ability comparing unit, for compared with original procedure ability, being verified result.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710771774.8A CN107578105B (en) | 2017-08-31 | 2017-08-31 | System Parameter Design space optimization method and device |
| PCT/CN2018/099272 WO2019042097A1 (en) | 2017-08-31 | 2018-08-08 | Optimization method and device for system parameter design space |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710771774.8A CN107578105B (en) | 2017-08-31 | 2017-08-31 | System Parameter Design space optimization method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107578105A true CN107578105A (en) | 2018-01-12 |
| CN107578105B CN107578105B (en) | 2019-03-26 |
Family
ID=61030770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710771774.8A Active CN107578105B (en) | 2017-08-31 | 2017-08-31 | System Parameter Design space optimization method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107578105B (en) |
| WO (1) | WO2019042097A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019042097A1 (en) * | 2017-08-31 | 2019-03-07 | 江苏康缘药业股份有限公司 | Optimization method and device for system parameter design space |
| CN110333330A (en) * | 2019-06-13 | 2019-10-15 | 江苏康缘药业股份有限公司 | A kind of detection method of tablet disintegration times |
| CN110353995A (en) * | 2019-06-13 | 2019-10-22 | 江苏康缘药业股份有限公司 | A kind of detection method of capsule preparations loading amount deviation |
| CN110955196A (en) * | 2018-09-27 | 2020-04-03 | 北京国双科技有限公司 | Processing method and system for production process index data |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464685A (en) * | 2007-12-21 | 2009-06-24 | 费舍-柔斯芒特系统股份有限公司 | A method to present recipe progress status information |
| CN102305792A (en) * | 2011-07-22 | 2012-01-04 | 浙江农林大学 | Nonlinear partial least square optimizing model-based forest carbon sink remote sensing evaluation method |
| CN102567812A (en) * | 2012-01-09 | 2012-07-11 | 红云红河烟草(集团)有限责任公司 | Method for predicting control index by using processing parameter in tobacco processing procedure |
| CN103279837A (en) * | 2013-06-05 | 2013-09-04 | 浙江中控软件技术有限公司 | Production data analysis method and system |
| CN103756980A (en) * | 2014-01-10 | 2014-04-30 | 中南林业科技大学 | Culture medium for fermenting panus rudis to produce laccase and laccase production method |
| CN104991984A (en) * | 2015-05-15 | 2015-10-21 | 广西大学 | Data monitoring method and system used for boiling sugar crystallization process |
| CN106650774A (en) * | 2016-10-11 | 2017-05-10 | 国云科技股份有限公司 | Method for obtaining the regression relationship between the dependant variable and the independent variables during data analysis |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8666922B2 (en) * | 2011-03-18 | 2014-03-04 | Battelle Memorial Institute | Information processing systems, reasoning modules, and reasoning system design methods |
| CN103439288B (en) * | 2013-08-24 | 2016-04-13 | 浙江大学 | A kind of real-time release detection method for ginkgo leaf medicinal material |
| CN104050612A (en) * | 2014-03-13 | 2014-09-17 | 无锡香象生物科技有限公司 | Safety monitoring method for unbalance of odds ratio of medicine on basis of cloud computing |
| CN104714537B (en) * | 2015-01-10 | 2017-08-04 | 浙江大学 | A kind of failure prediction method based on the relative mutation analysis of joint and autoregression model |
| CN107578105B (en) * | 2017-08-31 | 2019-03-26 | 江苏康缘药业股份有限公司 | System Parameter Design space optimization method and device |
-
2017
- 2017-08-31 CN CN201710771774.8A patent/CN107578105B/en active Active
-
2018
- 2018-08-08 WO PCT/CN2018/099272 patent/WO2019042097A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464685A (en) * | 2007-12-21 | 2009-06-24 | 费舍-柔斯芒特系统股份有限公司 | A method to present recipe progress status information |
| CN102305792A (en) * | 2011-07-22 | 2012-01-04 | 浙江农林大学 | Nonlinear partial least square optimizing model-based forest carbon sink remote sensing evaluation method |
| CN102567812A (en) * | 2012-01-09 | 2012-07-11 | 红云红河烟草(集团)有限责任公司 | Method for predicting control index by using processing parameter in tobacco processing procedure |
| CN103279837A (en) * | 2013-06-05 | 2013-09-04 | 浙江中控软件技术有限公司 | Production data analysis method and system |
| CN103756980A (en) * | 2014-01-10 | 2014-04-30 | 中南林业科技大学 | Culture medium for fermenting panus rudis to produce laccase and laccase production method |
| CN104991984A (en) * | 2015-05-15 | 2015-10-21 | 广西大学 | Data monitoring method and system used for boiling sugar crystallization process |
| CN106650774A (en) * | 2016-10-11 | 2017-05-10 | 国云科技股份有限公司 | Method for obtaining the regression relationship between the dependant variable and the independent variables during data analysis |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019042097A1 (en) * | 2017-08-31 | 2019-03-07 | 江苏康缘药业股份有限公司 | Optimization method and device for system parameter design space |
| CN110955196A (en) * | 2018-09-27 | 2020-04-03 | 北京国双科技有限公司 | Processing method and system for production process index data |
| CN110333330A (en) * | 2019-06-13 | 2019-10-15 | 江苏康缘药业股份有限公司 | A kind of detection method of tablet disintegration times |
| CN110353995A (en) * | 2019-06-13 | 2019-10-22 | 江苏康缘药业股份有限公司 | A kind of detection method of capsule preparations loading amount deviation |
| CN110353995B (en) * | 2019-06-13 | 2021-10-01 | 江苏康缘药业股份有限公司 | Method for detecting loading deviation of capsule preparation |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019042097A1 (en) | 2019-03-07 |
| CN107578105B (en) | 2019-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107578105A (en) | System parameter design space optimization method and device | |
| CN105425779B (en) | ICA-PCA multi-state method for diagnosing faults based on local neighborhood standardization and Bayesian inference | |
| CN111695626B (en) | High-dimensionality unbalanced data classification method based on mixed sampling and feature selection | |
| CN108762228B (en) | Distributed PCA-based multi-working-condition fault monitoring method | |
| Cai et al. | Estimating sparse precision matrix: Optimal rates of convergence and adaptive estimation | |
| WO2018068360A1 (en) | Method for obtaining regression relationships between dependent variables and independent variables during data analysis | |
| CN109299698B (en) | Wireless channel scene recognition method based on support vector machine | |
| CN108243435B (en) | Parameter optimization method and device in LTE cell scene division | |
| CN106646339A (en) | Online matching and positioning method in wireless position fingerprint indoor positioning | |
| CN103258130A (en) | Method for predicting sinter tumbler strength | |
| WO2023207453A1 (en) | Traditional chinese medicine ingredient analysis method and system based on spectral clustering | |
| CN113255222B (en) | Intelligent auxiliary debugging method and device for cavity filter | |
| CN107563515A (en) | Latent process parameter method for digging and device | |
| CN107664621B (en) | Abnormal sample identification method and system, server and electronic equipment | |
| CN111475892B (en) | Steady equilibrium design method for dynamic characteristics of key component structure of complex equipment | |
| CN112232429A (en) | Blast furnace overall state judgment method, terminal equipment and storage medium | |
| CN108169162A (en) | A kind of fast appraisement method of Tea Garden Soil Fertility level | |
| CN110659691A (en) | Method and device for monitoring quality and style of tobacco leaf raw material and computer readable storage medium | |
| CN115034443A (en) | Boiler performance prediction method based on working condition division and neural network | |
| CN107506824A (en) | Bad the observation data detection method and device of a kind of power distribution network | |
| CN104360948A (en) | IEC 61850 configuration file engineering consistency test method based on fuzzy algorithm | |
| CN111340361B (en) | Data-driven model analysis method and device for solving yellow wine raw material index range | |
| Xin et al. | ICB-MOEA/D: An interactive classification-based multi-objective optimization algorithm | |
| Wang et al. | An evaluation method for electromagnetic environment complexity | |
| Singgih et al. | Comparison of M Estimation, S Estimation, with MM Estimation to Get the Best Estimation of Robust Regression in Criminal Cases in Indonesia: Perbandingan Estimasi M, Estimasi S, dengan Estimasi MM |
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 |