Glass product processing technology optimization method based on composition data analysis

Abstract

The present invention discloses a glass product processing technology optimization method based on component data analysis, and relates to the technical field of glass product processing. By obtaining the quantitative evaluation values ​​of the contents of chemical components, processing parameters and performance indicators in the processing of finished glass samples, the correlation coefficients of the chemical components in the finished glass samples and the quantitative evaluation values ​​of the performance indicators are calculated, and the chemical components whose absolute values ​​of the correlation coefficients of the contents of chemical components and the quantitative evaluation values ​​of the performance indicators are in the range of [0.6, 1) are screened as performance-influencing components, and the obtained performance-influencing components and processing parameters are used as optimization variables of the target optimization performance indicators, and a multi-objective optimization model of the target optimization performance indicators is constructed. The optimal values ​​of the optimization variables are obtained through model calculation, and the relevant process parameters and chemical components can be adjusted in a targeted manner according to different usage scenarios and performance requirements to prepare borosilicate glass products that meet the requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of glass product processing, and in particular to a glass product processing technology optimization method based on component data analysis. Background Art

[0002] The types of heat-resistant glass currently include borosilicate glass, glass-ceramic and quartz glass. At present, borosilicate glass is still the most widely used heat-resistant glass, especially in the production of daily necessities such as tableware and cooking utensils. As a special glass, heat-resistant glass has many excellent performance characteristics compared to ordinary glass, such as high temperature resistance, thermal shock resistance, good chemical stability, good mechanical resistance, etc.

[0003] The chemical composition of borosilicate glass is: SiO2 content 70% to 80%, B2O3 content 8% to 13%, Al2O3 content 2% to 4%, R2O content 4% to 10%. In the traditional production process, the horseshoe flame kiln is used to melt silicate glass. With the development of refractory materials and glass electric melting technology, electric melting and full electric melting methods are now widely used to produce borosilicate glass, and its melting temperature can reach above 1600℃. The main factors affecting the performance of borosilicate glass include its components and processing parameters. The differences between components and processing parameters, such as melting temperature and melting time, have a great impact on the performance of the finished product. At present, the proportion of borosilicate glass components and processing parameters are mostly determined by empirical methods during the processing process. For borosilicate glass products with different usage scenarios, due to the differences in their performance requirements, the traditional empirical method cannot timely and accurately adjust the relevant process parameters for different products. To this end, we propose a glass product processing technology optimization method based on component data analysis. Summary of the invention

[0004] The main purpose of the present invention is to provide a glass product processing technology optimization method based on component data analysis, which can effectively solve the problems in the background technology.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] Glass product processing technology optimization method based on composition data analysis, including:

[0007] Obtaining the quantitative evaluation values ​​of the chemical component contents, processing parameters and performance indicators of the finished glass sample during processing, wherein the processing parameters include melting temperature and melting time;

[0008] Calculate the correlation coefficient between the i-th chemical component and the j-th performance index in the finished glass sample. The calculation formula is:

[0009]

[0010] In the formula, R ij It is expressed as the correlation coefficient between the ith chemical component and the jth performance index in the finished glass sample; c if It is expressed as the fth data value of the content of the i-th chemical component in the finished glass sample; It is expressed as the mean value of the content of the i-th chemical component in the finished glass sample; pe jf The fth data value is expressed as the quantitative evaluation value of the jth performance index in the finished glass sample; It is expressed as the mean of the quantitative evaluation value of the jth performance index in the glass finished product sample; F is the total amount of data, and the chemical components whose absolute value of the correlation coefficient between the content of chemical components and the quantitative evaluation value of the performance index is in the range of [0.6,1) are selected as the performance influencing components, i=1,2,...,n; j=1,2,...,m; wherein the performance index is at least one of high temperature resistance, thermal shock resistance, thermal expansion coefficient, and mechanical properties, and the method for obtaining the quantitative evaluation value of the performance index is as follows:

[0011] S01: Set the state parameter values ​​corresponding to the jth performance indicator in the best state and the worst state to be p respectively. jmax and p jmin ;

[0012] S02: Calculate the distance d between the state parameter values ​​corresponding to the jth performance indicator in the best state and the worst state j , the calculation formula is:

[0013]

[0014] S03: Obtain the real-time status parameter value p of the jth performance indicator j , respectively calculate the real-time state parameter value p j The distance p between the state parameter values ​​at the best state and the worst state j1 and p j2 , the calculation formula is:

[0015]

[0016] S04: Calculate the quantitative evaluation value of the corresponding performance indicator according to the performance indicator type and the obtained distance value, specifically:

[0017] When the performance indicator type is thermal expansion coefficient, the calculation formula is: pe j =p j2 / d j ;

[0018] When the performance index type is any one of high temperature resistance, thermal shock resistance, and mechanical properties, the calculation formula is: pej =p j1 / d j ;

[0019] Determine the target optimization performance index, and obtain the performance influencing components and processing parameters as the optimization variables of the target optimization performance index: U k =(c k1 , c k2 , ..., c kh , T k , t k ) T , where U k It is represented as the set of optimization variable types corresponding to the k-th target optimization performance index, k∈m; c kh It is expressed as the content of the hth performance influencing component corresponding to the kth target optimization performance index, h∈n; T k It represents the melting temperature value corresponding to the kth target optimization performance index; t k It represents the smelting time corresponding to the k-th target optimization performance index;

[0020] A multi-objective optimization model of the target optimization performance index is constructed, and the optimal value of the corresponding optimization variable is obtained through the model, wherein the expression of the multi-objective optimization model is:

[0021]

[0022]

[0023] In the formula, is the objective function of the multi-objective optimization model; st{ is the constraint condition of the multi-objective optimization model; c khmin 、c khmax are the minimum and maximum values ​​of the content of the hth performance influencing component corresponding to the kth target optimization performance index; T kmin , T kmax are the minimum and maximum values ​​of the melting temperature corresponding to the kth target optimization performance index; t kmin ,t kmax are the minimum and maximum values ​​of the melting time corresponding to the kth target optimization performance index; maxpe k It is expressed as the maximum value of the quantitative evaluation value of the k-th target optimization performance index; pe r It is expressed as the quantitative evaluation value of the rth performance index other than the target optimization performance index; ω r It is expressed as the weight of the rth performance indicator other than the target optimization performance indicator; ω k11 It is represented as the weight of the melting temperature corresponding to the k-th target optimization performance index; ω k12It is represented as the weight of the smelting time corresponding to the k-th target optimization performance index; ω kh It is represented as the weight of the hth performance influencing component corresponding to the kth target optimization performance index; a k Correction coefficient for optimizing performance index for the kth objective.

[0024] The present invention has the following beneficial effects:

[0025] Compared with the prior art, the technical solution of the present invention obtains the quantitative evaluation values ​​of the contents of each chemical component, processing parameters and performance indicators of the sample during the processing of the finished glass product sample, calculates the correlation coefficient between the chemical components in the finished glass product sample and the quantitative evaluation values ​​of the performance indicators, screens the chemical components whose absolute values ​​of the correlation coefficients between the contents of the chemical components and the quantitative evaluation values ​​of the performance indicators are in the range of [0.6,1) as the performance influencing components, determines the target optimized performance indicators, and uses the obtained performance influencing components and processing parameters as the optimization variables of the target optimized performance indicators, constructs a multi-objective optimization model for the target optimized performance indicators, obtains the optimal values ​​of the optimization variables through model calculation, and can adjust the relevant process parameters and chemical components in a targeted manner according to different usage scenarios and performance requirements to prepare borosilicate glass products that meet the requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The present invention is a flow chart of a method for optimizing glass product processing technology based on composition data analysis. DETAILED DESCRIPTION

[0027] The present invention will be further described below in conjunction with specific implementation methods, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limiting the present invention. In order to better illustrate the specific implementation methods of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product.

[0028] Example 1

[0029] like Figure 1 As shown, the specific implementation process of the technical solution of the present invention includes the following steps:

[0030] Step 1: Obtain the quantitative evaluation values ​​of the chemical component content, processing parameters and performance indicators of the finished glass sample during processing. The processing parameters include melting temperature and melting time.

[0031] It should be noted that in the existing processing flow, the chemical composition and content range of each component of borosilicate glass and its products are shown in the following table;

[0032] Table 1 Chemical composition of borosilicate glass and content range of each component

[0033]

[0034] The performance index is at least one of high temperature resistance, thermal shock resistance, thermal expansion coefficient, and mechanical properties. Among them, the thermal properties of glass such as expansion coefficient and thermal stability are closely related to the network structure of glass. The network structure of borosilicate glass is more complete and compact, and has good thermal properties. Usually, its thermal shock resistance is greater than 150°C. The mechanical properties are generally characterized by the strength of the material, and the thermal expansion coefficient is not higher than 60×10 -7 / ℃ (20~300℃), for the high temperature resistance, thermal shock resistance and mechanical properties in the performance indicators, they are all positive indicators, that is, the quality of the product performance is proportional to the parameter value of the indicator, and for the thermal expansion coefficient, the quality of the product performance is inversely proportional to its parameter value, so the quantitative evaluation value of each performance indicator is determined by the following method, specifically: S01: Set the state parameter values ​​corresponding to the jth performance indicator in the best state and the worst state to be p jmax and p jmin ;

[0035] S02: Calculate the distance d between the state parameter values ​​corresponding to the jth performance indicator in the best state and the worst state j , the calculation formula is:

[0036]

[0037] S03: Obtain the real-time status parameter value p of the jth performance indicator j , respectively calculate the real-time state parameter value p j The distance p between the state parameter values ​​at the best state and the worst state j1 and p j2 , the calculation formula is:

[0038]

[0039] S04: Calculate the quantitative evaluation value of the corresponding performance indicator according to the performance indicator type and the obtained distance value, specifically:

[0040] When the performance indicator type is thermal expansion coefficient, the calculation formula is: pe j =p j2 / d j ;

[0041] When the performance index type is any one of high temperature resistance, thermal shock resistance, and mechanical properties, the calculation formula is: pe j =p j1 / d j ;

[0042] Step 2: Calculate the correlation coefficient between the ith chemical component and the jth performance index in the finished glass sample. The calculation formula is:

[0043]

[0044] In the formula, R ij It is expressed as the correlation coefficient between the ith chemical component and the jth performance index in the finished glass sample; c if It is expressed as the fth data value of the content of the i-th chemical component in the finished glass sample; It is expressed as the mean value of the content of the i-th chemical component in the finished glass sample; pe jf The fth data value is expressed as the quantitative evaluation value of the jth performance index in the finished glass sample; It is expressed as the mean of the quantitative evaluation value of the jth performance index in the glass product sample; F is the total amount of data, and the chemical components whose absolute value of the correlation coefficient between the content of chemical components and the quantitative evaluation value of the performance index is in the range of [0.6,1) are selected as the performance influencing components, i = 1, 2, ..., n; j = 1, 2, ..., m;

[0045] It should be noted that the main factors affecting the performance of borosilicate glass include its composition and processing parameters. The differences between the components and processing parameters, such as the differences in melting temperature and melting time, have a greater impact on the performance of the finished product. The differences between the components have different degrees of influence on the performance of borosilicate glass. Therefore, when the absolute value of the correlation coefficient between the content of the chemical component and the quantitative evaluation value of the performance index is in the range of [0.6,1), it means that there is a strong correlation between the content of the chemical component and the quantitative evaluation value of the corresponding performance index. Therefore, the corresponding chemical component is selected as the performance influencing component.

[0046] Step 3: Determine the target optimization performance index, and use the obtained performance-influencing components and processing parameters as optimization variables of the target optimization performance index. The specific expression is:

[0047] U k =(c k1 , c k2 , ..., c kh , T k , t k ) T

[0048] Among them, U k It is represented as the set of optimization variable types corresponding to the k-th target optimization performance index, k∈m; c kh It is expressed as the content of the hth performance influencing component corresponding to the kth target optimization performance index, h∈n; T kIt represents the melting temperature value corresponding to the kth target optimization performance index; t k It represents the smelting time corresponding to the k-th target optimization performance index;

[0049] It should be noted that since the differences in processing parameters, such as smelting temperature and smelting time, have a great influence on the performance of the finished product, the processing parameters are one of the optimization variables. The performance-influencing components determined in step 2 are strongly correlated with the target optimization performance index, so the corresponding performance-influencing components are selected as the optimization variables.

[0050] Step 4: Construct a multi-objective optimization model of target optimization performance indicators, and obtain the optimal value of the corresponding optimization variable through the model. The construction process of the multi-objective optimization model of target optimization performance indicators includes the following steps:

[0051] Step 41: Determine the objective function of the multi-objective optimization model

[0052] For the target optimization performance index, after controlling the optimization variables, it is necessary to make the target optimization performance index reach the best state. At the same time, it should be satisfied that when the target optimization performance index reaches the best state, other performance indicators should also reach a relatively ideal state, so that the performance of the glass products is the best. Therefore, there are:

[0053]

[0054] That is, when all optimization variables make the k-th target optimization performance index reach the best state maxpe k When other performance indicators reach a relatively ideal state

[0055] Step 42: Determine the constraints of the multi-objective optimization model

[0056] 1) In the existing processing flow mentioned in step 1, the chemical composition and content range of each component of borosilicate glass and its products are: SiO2: content 70% to 80%, B2O3: content 8% to 13%, Al2O3: content 2% to 4%, R2O: content 4% to 10%, so the constraints on the content of the components affecting each performance need to meet the above content ranges, which are:

[0057] c khmin ≤c kh ≤c khmax ;

[0058] 2) For processing parameters, although there are differences between the processing parameters of products with different performance, there are relevant requirements for melting temperature and melting time in the actual production process. For example, electric melting and full electric melting are commonly used to produce borosilicate glass, and its melting temperature can reach above 1600°C. However, too high a melting temperature will not only increase power consumption and production costs, but will also directly increase the brittleness of glass products, which is not conducive to improving their performance. Therefore, there are relevant constraints on melting temperature and melting time, including:

[0059]

[0060] 3) For the target optimization performance index, the linear regression equation is used to fit it and the calculation formula for its quantitative evaluation value is obtained:

[0061]

[0062] 4) For the linear regression equation in step 3), the weight coefficients in each item need to satisfy the following relationship, specifically:

[0063]

[0064] 5) Calculation formula for other performance indicators when they reach a relatively ideal state:

[0065] The weight value in the formula needs to satisfy the following relationship, specifically:

[0066]

[0067] By determining the objective function and constraints of the multi-objective optimization model, the expression of the multi-objective optimization model is obtained, which is:

[0068]

[0069]

[0070] In the formula, is the objective function of the multi-objective optimization model; st{ is the constraint condition of the multi-objective optimization model; c khmin 、c khmax are the minimum and maximum values ​​of the content of the hth performance influencing component corresponding to the kth target optimization performance index; T kmin , T kmax are the minimum and maximum values ​​of the melting temperature corresponding to the kth target optimization performance index; t kmin ,t kmax are the minimum and maximum values ​​of the melting time corresponding to the kth target optimization performance index; maxpe kIt is expressed as the maximum value of the quantitative evaluation value of the k-th target optimization performance index; pe r It is expressed as the quantitative evaluation value of the rth performance index other than the target optimization performance index; ω r It is expressed as the weight of the rth performance indicator other than the target optimization performance indicator; ω k11 It is represented as the weight of the melting temperature corresponding to the k-th target optimization performance index; ω k12 It is represented as the weight of the smelting time corresponding to the k-th target optimization performance index; ω kh It is represented as the weight of the hth performance influencing component corresponding to the kth target optimization performance index; a k is the correction coefficient of the kth target optimization performance index. Through the above model, the optimal value of each corresponding optimization variable can be obtained when the required target optimization performance index reaches the optimal state, so that the relevant process parameters and chemical components can be adjusted according to different usage scenarios and performance requirements to prepare borosilicate glass products that meet the requirements.

[0071] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A glass product processing technology optimization method based on component data analysis, characterized in that: include: Obtaining the quantitative evaluation values ​​of the chemical component contents, processing parameters and performance indicators of the finished glass sample during processing, wherein the processing parameters include melting temperature and melting time; Calculate the correlation coefficient between the i-th chemical component and the j-th performance index quantitative evaluation value in the finished glass sample, and select chemical components with strong correlation with the quantitative evaluation values ​​of each performance index as performance influencing components, i = 1, 2, ..., n; j = 1, 2, ..., m; Determine the target optimization performance index, and obtain the performance influencing components and processing parameters as the optimization variables of the target optimization performance index: U k =(c k1 , c k2 , ..., c kh , T k , t k ) T , where U k It is represented as the set of optimization variable types corresponding to the k-th target optimization performance index, k∈m; c kh It is expressed as the content of the hth performance influencing component corresponding to the kth target optimization performance index, h∈n; T k It represents the melting temperature value corresponding to the kth target optimization performance index; t k It represents the smelting time corresponding to the k-th target optimization performance index; A multi-objective optimization model of the target optimization performance index is constructed, and the optimal value of the corresponding optimization variable is obtained through the model, wherein the expression of the multi-objective optimization model is: In the formula, is the objective function of the multi-objective optimization model; st{ is the constraint condition of the multi-objective optimization model; c khmin 、c khmax are the minimum and maximum values ​​of the content of the hth performance influencing component corresponding to the kth target optimization performance index; T kmin , T kmax are the minimum and maximum values ​​of the melting temperature corresponding to the kth target optimization performance index; t kmin ,t kmax are the minimum and maximum values ​​of the melting time corresponding to the kth target optimization performance index; maxpe k It is expressed as the maximum value of the quantitative evaluation value of the k-th target optimization performance index; pe r It is expressed as the quantitative evaluation value of the rth performance index other than the target optimization performance index; ω r It is expressed as the weight of the rth performance indicator other than the target optimization performance indicator; ω k11 It is represented as the weight of the melting temperature corresponding to the k-th target optimization performance index; ω k12 It is represented as the weight of the smelting time corresponding to the k-th target optimization performance index; ω kh It is represented as the weight of the hth performance influencing component corresponding to the kth target optimization performance index; a k Correction coefficient for optimizing performance index for the kth objective.

2. The glass product processing technology optimization method based on component data analysis according to claim 1, characterized in that: The performance index is at least one of high temperature resistance, thermal shock resistance, thermal expansion coefficient, and mechanical properties.

3. The glass product processing technology optimization method based on component data analysis according to claim 1, characterized in that: The method for obtaining the quantitative evaluation value of the performance indicator is as follows: S01: Set the state parameter values ​​corresponding to the jth performance indicator in the best state and the worst state to be p respectively. jmax and p jmin ; S02: Calculate the distance d between the state parameter values ​​corresponding to the jth performance indicator in the best state and the worst state j , the calculation formula is: S03: Obtain the real-time status parameter value p of the jth performance indicator j , respectively calculate the real-time state parameter value p j The distance p between the state parameter values ​​at the best state and the worst state j1 and p j2 , the calculation formula is: S04: Calculate the quantitative evaluation value of the corresponding performance indicator according to the performance indicator type and the obtained distance value, specifically: When the performance indicator type is thermal expansion coefficient, the calculation formula is: pe j =p j2 / d j ; When the performance index type is any one of high temperature resistance, thermal shock resistance, and mechanical properties, the calculation formula is: pe j =p j1 / d j .

4. The glass product processing technology optimization method based on component data analysis according to claim 1, characterized in that: The calculation formula for the correlation coefficient between the i-th chemical component and the quantitative evaluation value of the j-th performance index in the finished glass sample is: In the formula, R ij It is expressed as the correlation coefficient between the ith chemical component and the jth performance index in the finished glass sample; c if It is expressed as the fth data value of the content of the i-th chemical component in the finished glass sample; It is expressed as the mean value of the content of the i-th chemical component in the finished glass sample; pe jf The fth data value is expressed as the quantitative evaluation value of the jth performance index in the finished glass sample; It is expressed as the mean of the quantitative evaluation values ​​of the jth performance index in the finished glass samples; F is the total amount of data.

5. The glass product processing technology optimization method based on component data analysis according to claim 1, characterized in that: The basis for judging whether there is a strong correlation is that the absolute value of the correlation coefficient between the chemical component content and the quantitative evaluation value of the performance index is in the range of [0.6,1).

Images