CN115409267A - Data processing method, device and storage medium - Google Patents
Data processing method, device and storage medium Download PDFInfo
- Publication number
- CN115409267A CN115409267A CN202211067180.6A CN202211067180A CN115409267A CN 115409267 A CN115409267 A CN 115409267A CN 202211067180 A CN202211067180 A CN 202211067180A CN 115409267 A CN115409267 A CN 115409267A
- Authority
- CN
- China
- Prior art keywords
- product performance
- anchor point
- production process
- model
- point position
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Abstract
The application discloses a data processing method and device. The method uses a product performance prediction model to predict the product performance index of an anchor point position according to production process indexes (including production materials used at the anchor point position in the process of producing the electronic equipment, process parameters of processing the production materials and the like) collected aiming at the anchor point position of the electronic equipment. Thus, the quality problem of the anchor point position can be found without carrying out destructive test on the anchor point position. In addition, production process indexes which possibly cause anchor point position quality problems can be input and backtracked according to the model corresponding to the predicted value, so that guidance is provided for the production process, and the product quality is higher.
Description
Technical Field
The present application relates to the field of computing data processing technologies, and in particular, to a data processing method, an apparatus, and a storage medium.
Background
In the production process of dispensing and bonding of the notebook computer shell, the indexes of glue weight, glue temperature, glue exposure time, pressing time, pressure, whether the glue is stirred and the like need to be controlled so as to prevent the appearance from being influenced by glue overflow or the shell from cracking due to insufficient bonding strength.
But at present, whether the tensile value (adhesive force) of the notebook shell after dispensing meets the requirement can only be verified by performing destructive tests, so that the material and labor cost is higher, and time and labor are wasted.
Disclosure of Invention
The applicant creatively provides a data processing method, a data processing device and a storage medium.
According to a first aspect of embodiments of the present application, there is provided a data processing method, including: preprocessing the sampling data to obtain model input data, wherein the sampling data comprise production process indexes acquired aiming at the anchor point position of the electronic equipment, and the production process indexes comprise production materials used at the anchor point position in the process of producing the electronic equipment and process parameters for processing the production materials; and determining a predicted value of the product performance index of the anchor point position according to the model input data and the product performance prediction model.
According to an embodiment of the application, the method further comprises: determining at least two indexes from the production process indexes as independent variables, and taking the product performance index at the anchor point position as a dependent variable; establishing a model function of a dependent variable and an independent variable, wherein the model function comprises undetermined parameters; and training the product performance prediction model by using model training data, and determining the value of the parameter to be determined, wherein the model training data comprises sampling data and the actual value of the product performance index of the anchor point position corresponding to the sampling data.
According to an embodiment of the present application, establishing a model function of a dependent variable and an independent variable includes: establishing a multivariate linear regression equation of dependent variables and independent variables, correspondingly training a product performance prediction model by using model training data, and determining the value of a parameter to be determined, wherein the method comprises the following steps: determining sample data from the model training data; using sample data to solve undetermined coefficients in a multiple linear regression equation to minimize the sum of squares of prediction errors; and determining whether the product performance prediction model meets a preset condition, if not, adjusting independent variables or sample data of the multiple linear regression equation, and re-solving undetermined coefficients in the multiple linear regression equation.
According to an embodiment of the present application, a multiple linear regression equation for dependent variables and independent variables is established, including: establishing a multivariate linear regression equation of a dependent variable and an independent variable by using a least square method; accordingly, solving the undetermined coefficients in the multiple linear regression equation to minimize the sum of squares of the prediction errors includes: the undetermined coefficients in the multiple linear regression equation are solved using a gradient descent method to minimize the sum of squares of the prediction errors.
According to one embodiment of the application, determining at least two indexes as independent variables from the production process indexes comprises: determining the grey correlation degree between each production process index and the product performance index of the anchor point position according to the model training data; and determining at least two indexes from the production process indexes as independent variables according to the grey correlation degree.
According to an embodiment of the application, determining a gray correlation between each production process index and a product performance index of an anchor point position according to model training data includes: acquiring an initial transformation sequence of each production process index and a product performance index of an anchor point position according to model training data; obtaining a difference sequence corresponding to the initialized transformation sequence according to the initialized transformation sequence; and determining the grey correlation degree between each production process index and the product performance index of the anchor point position according to the difference sequence.
According to one embodiment of the application, determining at least two indices from production process indices as independent variables includes: drawing thermodynamic diagrams of various production process indexes and product performance indexes of anchor point positions according to model training data; determining the correlation degree of each production process index and the product performance index of the anchor point position according to the thermodynamic diagram; at least two indices are determined as independent variables from the production process indices based on the correlation.
According to an embodiment of the application, the anchor point position comprises a glue dispensing connection position, and the production process index comprises glue weight, glue temperature, glue exposure time, pressing time, pressure and/or whether the glue is stirred.
According to a second aspect of embodiments of the present application, there is provided a data processing apparatus including: the model input data acquisition module is used for preprocessing the sampling data to obtain model input data, wherein the sampling data comprises production process indexes acquired aiming at the anchor point position of the electronic equipment, and the production process indexes comprise production materials used at the anchor point position in the process of producing the electronic equipment and process parameters for processing the production materials; and the product performance prediction module is used for determining the predicted value of the product performance index of the anchor point position according to the model input data and the product performance prediction model.
According to a third aspect of embodiments herein, there is provided a computer storage medium comprising a set of computer executable instructions for performing any of the data processing methods described above when executed.
The embodiment of the application provides a data processing method, a data processing device and a storage medium. The method uses a product performance prediction model to predict the product performance index of an anchor point position according to production process indexes (including production materials used at the anchor point position in the process of producing the electronic equipment, process parameters of processing the production materials and the like) collected aiming at the anchor point position of the electronic equipment. Thus, the quality problem of the anchor point position can be found without carrying out destructive testing on the anchor point position. In addition, production process indexes which possibly cause anchor point position quality problems can be input and backtracked according to the model corresponding to the predicted value, so that guidance is provided for the production process, and the product quality is higher.
It is to be understood that the present application need not achieve all of the above-described benefits, but that certain aspects may achieve certain technical benefits, and that other embodiments of the present application may achieve benefits not mentioned above.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a schematic flow chart illustrating a data processing method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of modeling and training models according to another embodiment of the present application;
FIG. 3 is a schematic flow chart illustrating modeling according to another embodiment of the present application;
FIG. 4 is a schematic flow chart illustrating training establishment according to another embodiment of the present application;
FIG. 5 is a schematic flow chart of modeling and training models according to another embodiment of the present application;
fig. 6 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present application.
Detailed Description
In order to make the objects, features and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Moreover, various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without being mutually inconsistent.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or to implicitly indicate the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
Fig. 1 shows a flow of implementing a data processing method according to an embodiment of the present application. Referring to fig. 1, the method includes:
operation S110, preprocessing the sampled data to obtain model input data, where the sampled data includes production process indicators collected for anchor points of the electronic device, and the production process indicators include production materials used at the anchor points in the process of producing the electronic device and process parameters for processing the production materials;
the anchor point refers to a connection point of the respective parts of the product, for example, a bonding point between the parts of the product, or a connecting part, etc.
The sample data is typically sampled from the production data. The production data refers to data related to production process indexes collected in the production process of the product, particularly in anchor point position preparation.
The sampling data is preprocessed by a series of operations including cleaning, screening, duplicate removal, deficiency, formatting and the like, so that the production data becomes regular and stable input data which accords with the format of the model input data.
The process parameters typically include the nature of the production materials, the temperature of the production environment, the specific flow and/or preparation time of the production process, etc. The attributes of the production materials include the types, quality grades, weights, components, brands, manufacturers and the like of the production materials. The specific flow of the production process includes preparation steps required for preparing the anchor point position, a manufacturing process (for example, stirring, bonding, etching, grinding and the like) adopted by each step, an execution sequence of each step and the like.
And operation S120, determining a predicted value of the product performance index of the anchor point position according to the model input data and the product performance prediction model.
The performance indicators of the product at the anchor point are mainly performance indicators for evaluating the quality of the anchor point, such as adhesion, flatness, maximum heat resistance temperature, water resistance, and the like.
The product performance prediction model is a model for predicting the value of a product performance index according to model input data, namely data which is obtained by preprocessing sampling data of a production process index acquired aiming at an anchor point position of electronic equipment and accords with a model input data format.
The product performance prediction model is a machine learning model obtained by training through training data, and the model can learn the relationship and the internal relation between each data index based on historical data, and describe and quantify the quantitative relationship and the internal relation between each input data index and the product performance index through a model function and model function parameters, so as to predict the product performance index values which can be generated by different input data.
Production process indexes, such as process parameters of production materials and processing production materials, often affect the product performance indexes of anchor points and have a certain quantitative relation with the product performance indexes. For example, the better the quality of the production material used at the anchor point position, the better the product performance at the anchor point position; the use of production materials at different temperatures may affect the performance of the product at the anchor point during the mixing time of the production materials.
The product performance prediction model can be used for learning and finding out the potential relation and the internal relation between the production process index and the product performance index at the anchor point position from the existing data, quantifying the relation and predicting according to the quantified relation.
With the increasing maturity of machine learning technology, the continuous perfection of various production data and the increasing prediction accuracy of the prediction model of the accumulated product performance, the referential degree is also increased.
Thus, the quality problem of the anchor point position can be found without carrying out destructive testing on the anchor point position. In addition, production process indexes which possibly cause anchor point position quality problems can be input and backtracked according to the model corresponding to the predicted values, so that guidance is provided for a production process, and the product quality is higher.
In the embodiment of the application, the product performance model is trained in advance. The product performance model can be built by an implementer and trained by utilizing self data, can also be obtained by cooperation and joint training with a third party, and can also be a trained commercial model which is directly purchased.
Fig. 2 shows a process of establishing a product performance prediction model and training the product performance prediction model according to another embodiment of the present application, as shown in fig. 2, including:
operation S210, determining at least two indexes from the production process indexes as independent variables, and taking the product performance indexes of the anchor point position as dependent variables;
the product performance index can be a single index, if the product performance index is a single index, the relation between each production process index and a certain product performance index is easier to determine, when a certain product performance index has a problem, the source tracing is easier to perform, and the corresponding production process index is adjusted.
The product performance index can also be a plurality of indexes, and if the product performance index is a plurality of indexes, the relationship between each production process index and the comprehensive product performance index can be determined.
Operation S220, establishing a model function of the dependent variable and the independent variable, where the model function includes the undetermined parameter;
in the embodiment of the present application, the specific type of the model function is not limited, and the model function may be a model function of interpolation fitting, linear regression, gray prediction, or the like, and an implementer may flexibly determine the model function according to a specific application scenario and a service requirement.
Operation S230 is performed to train the product performance prediction model by using model training data, and determine a value of the parameter to be determined, where the model training data includes the sampling data and an actual value of the product performance index at the anchor point position corresponding to the sampling data.
The training process typically takes multiple rounds until the model converges or a desired accuracy is achieved.
Fig. 3 and 4 show a process of establishing a product performance prediction model and training the product performance prediction model according to another embodiment of the present application.
Fig. 3 shows the main process of establishing a product performance model in this embodiment, including:
operation S310, drawing thermodynamic diagrams of various production process indexes and product performance indexes of anchor points according to model training data;
in this embodiment, the product performance index at the anchor point position includes a plurality of indexes.
For example, taking the pivot between the notebook screen and the keyboard as an example, two indexes of flexibility and firmness of the pivot are included.
Therefore, the correlation coefficient of each production process index and each index in the product performance indexes of a plurality of anchor points can be calculated through the Pearson correlation coefficient, and then a thermodynamic diagram is drawn according to the correlation coefficient of each production process index and each index in the product performance indexes of the anchor points.
Thermodynamic diagrams are often used to represent distribution densities on a map, and may also be simply understood as a mapping of values in two-dimensional coordinates to colors.
In this embodiment, a thermodynamic diagram may be drawn by taking production process indexes (e.g., a distance between the rotating shafts, a brand of the rotating shafts, an angle during installation, etc.) for producing and installing the rotating shafts as an abscissa X, and taking a product performance index axis (e.g., flexibility and firmness of the rotating shafts) at an anchor point position as an ordinate Y according to a method in which the color is lighter when the correlation coefficient is smaller, and the color is darker when the correlation coefficient is larger.
The above processes of calculating the pearson correlation coefficient and drawing the correlation coefficient thermodynamic diagram can be completed in one step by a third party tool, for example, heatplot.
Operation S320, determining a correlation between each production process index and a product performance index at an anchor point position according to the thermodynamic diagram;
in the previous example, the correlation between each production process index and the product performance index at the anchor point position can be determined according to the color depth of the thermodynamic diagram.
Operation S330, determining at least two indexes from the production process indexes as independent variables according to the correlation, and taking the product performance index of the anchor point position as a dependent variable;
in the above example, at least two production process indexes with the darkest colors can be found from the correlation coefficient thermodynamic diagram as dependent variables.
In operation S340, a model function of the dependent variable and the independent variable is established using the multiple linear regression equation as the model function.
For example, the following multiple linear regression equation is established:
Y1＝β+β1X1+β2X2+……+βmXm；
Y2＝ɑ+ɑ1X1+ɑ2X2+……+ɑmXm
wherein Y1 and Y2 represent dependent variables, X1, X2, \8230;, xm represent independent variables, and β, β 1, β 2, \8230;, β m or α, a 1, a 2, \8230;, and a m represent undetermined coefficients.
In this embodiment of the present application, a least square method is used to establish a multiple linear regression equation of the dependent variable and the independent variable, and a value of a parameter to be determined in the multiple linear regression equation needs to be determined in a process of training a product performance prediction model using model training data, where a model training process is as shown in fig. 4, and includes:
operation S410, determining sample data from the model training data;
when sample data is determined from the model training data, the data may be randomly extracted or filtered according to a threshold. The implementer can flexibly confirm according to the requirement.
Operation S420, using the sample data to solve undetermined coefficients in the multiple linear regression equation, so as to minimize the sum of squares of the prediction errors;
specifically, the embodiment of the present application uses a gradient descent method, and solves the undetermined coefficients in the multiple linear regression equation to minimize the square sum of the prediction errors.
And operation S430, determining whether the product performance prediction model meets a preset condition, if not, adjusting independent variables or sample data of the multiple linear regression equation, and re-solving undetermined coefficients in the multiple linear regression equation.
The preset conditions generally include whether the pattern converges or the model accuracy has reached a desired accuracy, etc. When the independent variables of the multiple linear regression equation are adjusted, some of the independent variables can be replaced or the number of the independent variables can be increased/decreased, and the specific measures mainly depend on the expert experience and the model effect.
For the sample data, the random data extraction algorithm can be changed, or the threshold value can be tightened or loosened.
Fig. 5 shows the main processes of modeling and training a model in an early stage in order to apply the data processing method to the application scenario of adhesive force prediction at the dispensing and gluing position of the notebook computer housing, as shown in fig. 5, including:
operation S510, collecting data in the shell dispensing process, and establishing an original data sequence;
the original data sequence is a data sequence formed by each production process index in the shell dispensing process, and the production process indexes in the embodiment include: glue weight, glue temperature, glue exposure time, press time, pressure, and/or whether the glue is agitated, etc.
Operation S520, performing dimensionless processing on the original data sequence to obtain an initialized transformation sequence;
due to different physical meanings of various factors in the system, the dimensions of the data are not necessarily the same, so that the comparison is inconvenient, or a correct conclusion cannot be obtained easily during the comparison. Therefore, in the gray correlation analysis, data processing without dimensioning is generally performed.
Common dimensionless processing includes extreme, normalized, averaged, and standard deviation methods.
Operation S530, obtaining a difference sequence corresponding to the initial transformation sequence according to the initial transformation sequence;
the difference sequence is the difference, typically an absolute difference, between each point on the curve comparing the series Xi and each point on the curve referencing the series X0. Where the reference series is typically a curve fitted from an initially transformed sequence, or a curve predicted from expert experience.
Operation S540, determining the grey correlation degree between each production process index and the product performance index of the anchor point position according to the difference sequence;
for example, assume that the minimum difference of the difference sequence is Δ min and the maximum difference is Δ max.
The absolute difference between each point on the curve of each comparison series Xi and each point on the curve of the reference series X0, denoted as Δ oi (k), is then the correlation coefficient ξ (Xi) can also be simplified by the following equation:
since the correlation coefficient is the value of the degree of correlation between the comparison array and the reference array at each time (i.e., each point in the curve), there is more than one correlation coefficient and the information is too scattered to facilitate an overall comparison. It is therefore necessary to concentrate the correlation coefficients at each time (i.e. each point in the curve) to one value, i.e. to average them, as a quantitative representation of the degree of correlation between the comparison series and the reference series, the degree of correlation can be calculated using the following formula:
wherein r is i The gray degree of correlation between the comparison array Xi and the reference array X0, or referred to as the sequence degree of correlation, the average degree of correlation, and the line degree of correlation. When r is i Closer to 1, the better the correlation.
Operation S550, selecting an independent variable according to the grey correlation degree, selecting a certain amount of data as a training sample and a test sample by taking the adhesion value as a dependent variable;
in the above example, r can be selected i The production process index with a value closer to 1 is taken as an independent variable.
Operation S560, establishing a multiple regression equation, and solving the undetermined coefficient of the product performance prediction model;
operation S570, determining whether the product performance prediction model meets a preset condition, if yes, continuing to operate S580, and if not, continuing to operate S590;
operation S580, outputting a product performance prediction model;
in operation S590, the number of arguments or training sample data is adjusted.
The embodiment of the application selects data of the same pressing machine, the same glue, the same part and the latest 60 days as training data, and specific coefficients are obtained after the processing of the steps. The test data is then substituted into a multiple regression equation to yield the predicted values and errors as shown in table 1 below (this table is merely an example of data).
As can be seen from the data shown in table 1, even though the same pressing machine, the same glue and the same component are used, the values of the production process index sampled in a period of time are slightly different, and the adhesive force is changed accordingly. This shows that there is a certain correlation between each production process index and the adhesive force of the dispensing.
In the embodiment, the product performance prediction model is established by using the multiple regression equation, sampling data sampled and accumulated in a period of time is preprocessed, the sampling data is converted into model training data, and the product performance model is trained by using the training data. Therefore, the product performance model can learn and find the incidence relation between each production process index and the adhesive force for adhesive dispensing from the training data, and determine undetermined parameters in the multiple regression equation, so that the product performance prediction model can accurately predict the product performance of the adhesive force for adhesive dispensing according to each production process index value in the sampling data.
In view of the error (| difference between actual value and predicted value |/actual value × 100%) shown in table 1, the product performance prediction model established and trained by the method in the embodiment of the present application has high prediction accuracy and is reliable, and the actual result of the destructive experiment can be partially replaced in the production stage, so that the product performance prediction model can be used as a basis for a preventive measure for finding problems and a decision for adjusting the production process.
Table 1:
further, the embodiment of the application also provides a data processing device. As shown in fig. 6, the apparatus 60 includes: the model input data acquisition module 601 is configured to preprocess the sampling data to obtain model input data, where the sampling data includes production process indicators acquired for anchor points of the electronic device, and the production process indicators include production materials used at the anchor points in a process of producing the electronic device and process parameters for processing the production materials; and the product performance prediction module 602 is configured to determine a predicted value of the product performance index at the anchor point position according to the model input data and the product performance prediction model.
According to an embodiment of the present application, the apparatus 60 further comprises: the independent variable determining module is used for determining at least two indexes from the production process indexes as independent variables and taking the product performance indexes at the anchor point position as dependent variables; the model establishing module is used for establishing a model function of the dependent variable and the independent variable, and the model function comprises undetermined parameters; and the model training module is used for training the product performance prediction model by using model training data to determine the value of the parameter to be determined, wherein the model training data comprises sampling data and the actual value of the product performance index of the anchor point position corresponding to the sampling data.
According to an embodiment of the present application, the model building module is specifically configured to build a multiple linear regression equation of the dependent variable and the independent variable, and accordingly, the model training module includes: the sample data determining submodule is used for determining sample data from the model training data; the undetermined coefficient solving submodule is used for solving undetermined coefficients in the multivariate linear regression equation by using the sample data so as to minimize the square sum of the prediction error; and the model evaluation submodule is used for determining whether the product performance prediction model meets a preset condition, if not, adjusting independent variables or sample data of the multiple linear regression equation, and re-solving undetermined coefficients in the multiple linear regression equation.
According to an embodiment of the application, the model establishing module is specifically configured to establish a multivariate linear regression equation of a dependent variable and an independent variable by using a least square method; correspondingly, the undetermined coefficient solving submodule is specifically used for solving the undetermined coefficients in the multiple linear regression equation by using a gradient descent method so as to minimize the square sum of the prediction errors.
According to an embodiment of the present application, the independent variable determination module includes: the gray correlation degree determining submodule is used for determining the gray correlation degree between each production process index and the product performance index of the anchor point position according to the model training data; and the independent variable determining submodule is used for determining at least two indexes from the production process indexes as independent variables according to the grey correlation degree.
According to an embodiment of the present application, the gray relevance determining sub-module includes: the initialization transformation sequence acquisition unit is used for acquiring the initialization transformation sequences of the production process indexes and the product performance indexes of the anchor point positions according to the model training data; a difference sequence obtaining unit, configured to obtain a difference sequence corresponding to the initialized transformation sequence according to the initialized transformation sequence; and the gray correlation degree determining unit is used for determining the gray correlation degree between each production process index and the product performance index of the anchor point position according to the difference sequence.
According to an embodiment of the present application, the independent variable determination module includes: the thermodynamic diagram drawing unit draws thermodynamic diagrams of various production process indexes and product performance indexes of anchor point positions according to model training data; the correlation determination submodule is used for determining the correlation between each production process index and the product performance index of the anchor point position according to the thermodynamic diagram; and the independent variable determining submodule is used for determining at least two indexes from the production process indexes as independent variables according to the correlation.
In addition, the present application also provides a computer storage medium, where the storage medium includes a set of computer executable instructions, and when the instructions are executed, the storage medium is used for executing the data processing method of any one of the above.
It is to be noted here that: the above description of the embodiment of the data processing apparatus and the above description of the embodiment of the computer storage medium are similar to the description of the foregoing method embodiments, and have similar beneficial effects to the foregoing method embodiments, and therefore are not repeated herein. For technical details that have not been disclosed in the present application in the description of the embodiments of the data processing apparatus and the embodiments of the computer storage medium, please refer to the description of the foregoing method embodiments of the present application for understanding, and therefore will not be described again for brevity.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a component of' 8230; \8230;" does not exclude the presence of another like element in a process, method, article, or apparatus that comprises the element.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of a unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another device, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or in other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps of implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer-readable storage medium, and when executed, executes the steps including the method embodiments; and the aforementioned storage medium includes: various media capable of storing program codes, such as a removable storage medium, a Read Only Memory (ROM), a magnetic disk, and an optical disk.
Alternatively, the integrated units described above in the present application may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially implemented or portions thereof that contribute to the prior art may be embodied in the form of a software product stored in a storage medium, and including several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a removable storage medium, a ROM, a magnetic disk, an optical disk, or the like.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method of data processing, comprising:
preprocessing sampling data to obtain model input data, wherein the sampling data comprise production process indexes acquired aiming at an anchor point position of electronic equipment, and the production process indexes comprise production materials used at the anchor point position in the process of producing the electronic equipment and process parameters for processing the production materials;
and determining the predicted value of the product performance index of the anchor point position according to the model input data and the product performance prediction model.
2. The method of claim 1, further comprising:
determining at least two indexes from the production process indexes as independent variables, and taking the product performance indexes of the anchor point position as dependent variables;
establishing a model function of the dependent variable and the independent variable, wherein the model function comprises undetermined parameters;
and training the product performance prediction model by using model training data to determine the value of the undetermined parameter, wherein the model training data comprises sampling data and the actual value of the product performance index of the anchor point position corresponding to the sampling data.
3. The method of claim 2, the establishing a model function of the dependent variable and the independent variable, comprising:
establishing a multiple linear regression equation of the dependent variable and the independent variable,
correspondingly, the training the product performance prediction model by using the model training data to determine the value of the parameter to be determined includes:
determining sample data from the model training data;
using the sample data to solve undetermined coefficients in the multiple linear regression equation, and enabling the square sum of prediction errors to be minimum;
and determining whether the product performance prediction model meets a preset condition, if not, adjusting independent variables or sample data of the multiple linear regression equation, and re-solving undetermined coefficients in the multiple linear regression equation.
4. The method of claim 3, the establishing a multiple linear regression equation for the dependent variable and the independent variable, comprising:
establishing a multivariate linear regression equation of the dependent variable and the independent variable by using a least square method;
accordingly, solving the undetermined coefficients in the multiple linear regression equation to minimize a sum of squares of prediction errors includes:
solving the undetermined coefficients in the multiple linear regression equation using a gradient descent method to minimize the sum of squares of the prediction errors.
5. The method of claim 2, determining at least two indices from the production process indices as arguments, comprising:
determining gray correlation degrees between each production process index and the product performance index of the anchor point position according to the model training data;
and determining at least two indexes from the production process indexes as independent variables according to the grey correlation degree.
6. The method of claim 5, wherein determining a gray correlation between each production process metric and a product performance metric at the anchor location based on the model training data comprises:
acquiring an initial transformation sequence of each production process index and the product performance index of the anchor point position according to the model training data;
obtaining a difference value sequence corresponding to the initialization transformation sequence according to the initialization transformation sequence;
and determining the grey correlation degree between each production process index and the product performance index of the anchor point position according to the difference sequence.
7. The method of claim 2, determining at least two indices from the production process indices as arguments, comprising:
drawing thermodynamic diagrams of various production process indexes and product performance indexes of the anchor point position according to the model training data;
determining the correlation degree of each production process index and the product performance index of the anchor point position according to the thermodynamic diagram;
and determining at least two indexes from the production process indexes as independent variables according to the correlation degree.
8. The method of claim 1, wherein the anchor point location comprises a location of a glue joint, and the production process indicator comprises glue weight, glue temperature, glue exposure time, press fit time, pressure, and/or whether the glue is agitated.
9. A data processing apparatus comprising:
the system comprises a model input data acquisition module, a model output data acquisition module and a model output data acquisition module, wherein the model input data acquisition module is used for preprocessing sampling data to obtain model input data, the sampling data comprise production process indexes acquired aiming at an anchor point position of electronic equipment, and the production process indexes comprise production materials used at the anchor point position in the process of producing the electronic equipment and process parameters for processing the production materials;
and the product performance prediction module is used for determining the predicted value of the product performance index of the anchor point position according to the model input data and the product performance prediction model.
10. A computer-readable storage medium comprising a set of computer-executable instructions that, when executed, perform the method of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211067180.6A CN115409267A (en) | 2022-09-01 | 2022-09-01 | Data processing method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211067180.6A CN115409267A (en) | 2022-09-01 | 2022-09-01 | Data processing method, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115409267A true CN115409267A (en) | 2022-11-29 |
Family
ID=84162709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211067180.6A Pending CN115409267A (en) | 2022-09-01 | 2022-09-01 | Data processing method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115409267A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116307405A (en) * | 2023-05-25 | 2023-06-23 | 日照鲁光电子科技有限公司 | Diode performance prediction method and system based on production data |
-
2022
- 2022-09-01 CN CN202211067180.6A patent/CN115409267A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116307405A (en) * | 2023-05-25 | 2023-06-23 | 日照鲁光电子科技有限公司 | Diode performance prediction method and system based on production data |
| CN116307405B (en) * | 2023-05-25 | 2023-08-04 | 日照鲁光电子科技有限公司 | Diode performance prediction method and system based on production data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840671B (en) | Operation improvement effect calculation device, operation improvement effect calculation method, and recording medium | |
| JP2020522800A (en) | Computer system and method for online construction and deployment of predictive inference models | |
| JP6707920B2 (en) | Image processing apparatus, image processing method, and program | |
| EP3499375A1 (en) | Predicting quality of an information system using system dynamics modelling and machine learning | |
| CN100474878C (en) | Image quality prediction method and apparatus and fault diagnosis system | |
| CN109523188B (en) | Man-machine interface display-oriented method and system for evaluating man-machine cognitive characteristic work efficiency | |
| WO2007001527B1 (en) | Automated ultrasonic elasticity image formation with quality measure | |
| Khoshgoftaar et al. | A multiobjective module-order model for software quality enhancement | |
| CN115409267A (en) | Data processing method, device and storage medium | |
| WO2019235614A1 (en) | Relationship analysis device, relationship analysis method, and recording medium | |
| JP6989860B2 (en) | Image inspection equipment, image inspection methods and programs | |
| US8280708B2 (en) | Configuration of tools and processes for metal forming | |
| JP6093791B2 (en) | POSITIONING DEVICE, POSITIONING METHOD, POSITIONING PROGRAM, AND POSITIONING SYSTEM | |
| CN108563895A (en) | A kind of interval model modification method considering correlation | |
| US20230140696A1 (en) | Method and system for optimizing parameter intervals of manufacturing processes based on prediction intervals | |
| US20210224664A1 (en) | Relationship analysis device, relationship analysis method, and recording medium | |
| CN110751170A (en) | Panel quality detection method, system, terminal device and computer readable medium | |
| US20220163387A1 (en) | Method for optimizing output result of spectrometer and electronic device using the same | |
| JPWO2019235608A1 (en) | Analyzer, analysis method and program | |
| Defernez et al. | Modelling beer fermentation variability | |
| EP3608848A1 (en) | Method and device for use in estimating lifecycle of component | |
| JP2022173108A (en) | Information processing apparatus and program | |
| JP4308113B2 (en) | Data analysis apparatus and method, and program | |
| TW201743273A (en) | Production control device and production control program | |
| Tinungki | The election of the best autoregressive integrated moving average model to forecasting rice production in Indonesia |
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 |