Injection molding product quality evaluation and improvement method

Abstract

The invention relates to an injection molding product quality evaluation and improvement method, and relates to the field of injection molding machine production, and the method comprises the following steps: collecting injection molding process parameters and product target quality; performing a single-factor preliminary test on the injection molding process parameters to screen out main influence factors; constructing a product regression equation based on the main influence factors and the product target quality, and calculating correlation parameters based on the main influence factors and the product target quality; obtaining correlation strength according to the correlation parameters; combining the product regression equation and the correlation strength to generate an improvement parameter; and improving parameters and a preset response surface method are applied to verify the adjustment effect, so that the product quality is stabilized within a preset threshold range. The method has the effect of efficiently and stably controlling the product quality within the preset threshold value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of injection molding machine production, in particular to an injection molding product quality evaluation and improvement method. BACKGROUND

[0002] Injection molding product quality evaluation and improvement is a whole-process technical means in the field of injection molding machine production, which is around the preset quality standard of injection molding products, through systematic data collection, key influencing factors screening, parameter quantification and quality correlation, optimization scheme generation and verification.

[0003] Currently, manufacturers will first determine the specifications of polymer injection raw materials according to injection molding needs, then screen key process parameters, and collect parameter feedback data in real time through the host computer to dynamically adjust core parameter values such as barrel temperature, storage back pressure, and storage speed. To verify the impact of parameters on quality, a four-order covariance matrix is introduced to analyze the correlation between parameters and performance indicators, and the influence coefficient of each parameter is calculated by combining the regression model. After repeated testing and verification to determine the final coefficient, a response surface model is constructed to control the quality of injection molding products within the preset threshold range.

[0004] However, the above operation does not screen the main influencing factors through standardized single-factor preliminary test, and does not generate targeted improvement parameters combined with product regression equation and correlation strength, resulting in lack of precision in parameter optimization and difficulty in efficiently and stably controlling product quality within the preset threshold, which needs to be improved. SUMMARY

[0005] In order to efficiently and stably control the product quality within the preset threshold, the present application provides an injection molding product quality evaluation and improvement method.

[0006] The present application provides an injection molding product quality evaluation and improvement method, which adopts the following technical solution: An injection molding product quality evaluation and improvement method, comprising: Collecting injection molding process parameters and product target quality; Performing single-factor preliminary test on injection molding process parameters to screen out main influencing factors; Based on the main influencing factors and the product target quality, a product regression equation is constructed, and based on the main influencing factors and the product target quality, a correlation parameter is calculated; According to the correlation parameter, the correlation strength is obtained; Combining the product regression equation and the correlation strength to generate improvement parameters; Applying the improvement parameters and the preset response surface method to verify the adjustment effect, so that the product quality is stably controlled within the preset threshold range.

[0007] By adopting the technical scheme, the core parameters having a significant influence on product quality are accurately screened out through the standardized single-factor preliminary test, so as to avoid irrelevant parameter interference in the optimization process; the product regression equation containing main effect, interaction and quadratic term is constructed based on the main influencing factors, and the correlation degree parameters are calculated and the correlation strength is divided, so as to comprehensively and quantitatively reflect the complex relationship between the parameters and the quality; finally, the targeted improvement parameters are generated in combination with the regression equation and the correlation strength, and the adjustment effect is verified by the response surface method, so as to realize the accurate optimization of the parameters, effectively solve the problem of lack of accuracy in parameter optimization in the prior art, and finally efficiently and stably control the product quality within the preset threshold.

[0008] Optionally, the specific screening method of the main influencing factors further comprises: collecting historical production data; based on the historical production data, retrieving an initial reference value and a variation interval; dividing a test gradient according to the initial reference value and the variation interval; based on the test gradient and the injection molding process parameters, performing injection molding production, and collecting product production quality of the produced products; combining the test gradient, the injection molding process parameters and the product production quality to obtain a range coefficient and a variation coefficient; when the range coefficient exceeds a preset range threshold or the variation coefficient exceeds a preset variation coefficient threshold, marking the parameter as a main influencing factor.

[0009] Optionally, the construction method of the product regression equation further comprises: based on the injection molding production, collecting factor actual values of the main influencing factors, target actual values of a product target quality, and production actual values of the product production quality; according to the factor actual values, the target actual values and the production actual values, calculating main effect coefficients, interaction coefficients, quadratic term coefficients and equation constant terms; combining the main effect coefficients, the interaction coefficients, the quadratic term coefficients, the equation constant terms and a preset error term to generate the product regression equation.

[0010] Optionally, the specific product regression equation further comprises: Y=β0+β1F1+β2F2+β3F3+β 12 F1F2+β 13 F1F3+β 23 F2F3+β 11 F1 2 +β 22 F2 2 +β 33 F3 2 +ϵ; Wherein, Y is the product target quality, β0 is a constant term, β1, β2, β3 are main effect coefficients of main influencing factor one F1, main influencing factor two F2, and main influencing factor three F3 on Y respectively, β 12 , β 13 , β 23 are interaction coefficients of F1 and F2, F1 and F3, and F2 and F3 respectively, β 11 , β 22 , β 33 are quadratic coefficients of F1, F2, and F3 respectively, and ϵ is an error term.

[0011] Optionally, the specific calculation method of the correlation degree parameter is further included: A four-order covariance matrix is constructed based on the main influencing factors and the product target quality; The sample variance and the sample covariance are calculated based on the actual values of the factors, the actual values of the target, the actual values of the production, and the four-order covariance matrix; The correlation coefficient matrix is obtained according to the sample variance and the sample covariance; The correlation degree parameter is obtained based on the correlation coefficient matrix.

[0012] Optionally, the specific construction of the four-order covariance matrix and the detailed calculation method of the sample variance and the sample covariance are further included: The expression of the four-order covariance matrix is: ; Wherein, Fi is a main influencing factor, F1 is a main influencing factor one, F2 is a main influencing factor two, and F3 is a main influencing factor three; The calculation formula of the sample variance is: ; Wherein, n is the sample number, is the mean value of the observed samples of the main influencing factor Fi, is the sample variance on the diagonal line of the four-order covariance matrix; The sample covariance includes the sample covariance between factors and the sample covariance between factors and quality, the calculation formula of the sample covariance between factors is: ; The calculation formula of the sample covariance between factors and quality is: ; Wherein, is the mean value of the observed samples of the main influencing factor Fj, yk is the kth value of the product target quality, is the mean value of the observed samples of the product target quality.

[0013] Optionally, the detailed calculation method of the correlation coefficient matrix is further included: ; ; wherein, is the correlation coefficient of factor i and factor j, is the correlation coefficient of factor i and index Y, is the sample standard deviation of Fi, is the sample standard deviation of Fj, and SY is the sample standard deviation of Y; in combination and to generate a correlation coefficient matrix expression; wherein R is the correlation coefficient matrix expression.

[0014] In summary, the present application includes at least one of the following beneficial technical effects: 1. First, the core parameters that have a significant impact on product quality are accurately screened through standardized single-factor preliminary experiments, avoiding irrelevant parameter interference in the optimization process; then, based on the main influencing factors, a product regression equation containing main effects, interactions and quadratic terms is constructed, and the correlation parameters are calculated and the correlation strength is divided, which can comprehensively and quantitatively reflect the complex relationship between parameters and quality; finally, combined with the regression equation and the correlation strength, targeted improvement parameters are generated, and the adjustment effect is verified by response surface method, which can realize the accurate optimization of parameters, effectively solve the problem of lack of accuracy in parameter optimization in the prior art, and finally efficiently and stably control the product quality within the preset threshold; 2. Using the regression model and solving the regression coefficients, the main effect coefficients are obtained, which quantize the parameter influence and reduce the loss rate of the product; 3. The initial reference value and the variation interval are determined based on historical production data, which can ensure that the test gradient division is consistent with the actual production scene; by carrying out injection molding production according to the test gradient and collecting product production quality, the influence data of different process parameter values on quality can be systematically obtained; then, the influence degree of the parameter on the quality is quantified by the range coefficient and the variation coefficient, and when either of them exceeds the preset threshold, it is marked as the main influencing factor, which not only avoids the subjective bias caused by artificial experience judgment, but also accurately locks the core parameters that have a significant impact on quality. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a method flowchart of a method for evaluating and improving injection molded product quality; Figure 2 is a method flowchart of a specific screening method for main influencing factors; Figure 3 is a method flowchart of a construction method for a product regression equation; Figure 4 is a method flowchart of a specific calculation method for correlation parameters. DETAILED DESCRIPTION

[0016] The application will be further described in details below with reference to the accompanying drawings and embodiments.

[0017] With reference to Figure 1 The embodiments of the present application disclose a method for evaluating and improving quality of injection molding products, comprising the following steps: S1: collecting injection molding process parameters and product target quality.

[0018] The injection molding process parameters refer to key technical parameters that can be regulated and controlled in the injection molding production process. The product target quality refers to preset quality standards that the injection molding products need to reach. Both the injection molding process parameters and the product target quality are pre-inputted by on-site technical personnel, which will not be described here.

[0019] S2: performing single-factor preliminary test on the injection molding process parameters to screen out main influencing factors.

[0020] The main influencing factors refer to the injection molding process parameters that have significant influence on the product target quality. The main influencing factors in the injection molding process parameters can be screened out through the single-factor preliminary test. The specific operation steps are described in detail in subsequent S20 to S25, which will not be described here.

[0021] S3: constructing a product regression equation based on the main influencing factors and the product target quality, and calculating a correlation parameter based on the main influencing factors and the product target quality.

[0022] The product regression equation refers to a mathematical model that quantifies the relationship between the main influencing factors and the product target quality. The product regression equation is formed by combining the main influencing factors and the product target quality. The specific construction method is described in detail in subsequent S30 to S301, which will not be described here.

[0023] The correlation parameter refers to a quantitative index for measuring the degree of linear correlation between the main influencing factors and between the main influencing factors and the product target quality. The correlation parameter can be calculated by understanding the main influencing factors and the product target quality. The specific calculation steps of the correlation parameter are described in detail in subsequent S31 to S312, which will not be described here.

[0024] S4: obtaining the correlation strength according to the correlation parameter.

[0025] The correlation strength refers to a hierarchical description of the correlation degree between the main influencing factors or between the main influencing factors and the quality based on the correlation parameter.

[0026] First, extract the specific value from the correlation parameter, and divide the correlation strength level according to the absolute value of the value: the absolute value close to 1 (such as greater than 0.8) is "strong correlation", 0.5 to 0.8 is "moderate correlation", 0.3 to 0.5 is "weak correlation", and close to 0 (less than 0.3) is "no correlation"; determine the correlation direction according to the value sign: positive indicates "positive correlation" (one factor becomes larger, the other factor or quality also tends to become larger), and negative indicates "negative correlation" (one factor becomes larger, the other factor or quality tends to become smaller); integrate the level and direction to obtain the correlation strength.

[0027] S5: Combine the product regression equation and the correlation strength to generate the improvement parameter.

[0028] The improvement parameter refers to the specific value of the adjusted main influencing factor.

[0029] First, extract the main effect coefficient and interaction coefficient of each factor from the product regression equation, sort them according to the absolute value of the coefficient, and determine which factors have more significant impact on quality (the larger the absolute value, the higher the priority), and then combine the results of the correlation strength to adjust the factors with strong correlation and high priority first. For example, a certain factor is positively correlated with quality (for example, the larger the factor, the better the quality), and the current quality is not up to standard, so the value of the factor is increased according to the priority; if it is negatively correlated (for example, the larger the factor, the worse the quality), the value of the factor is reduced.

[0030] Finally, automatically correct it by machine (such as PID control or intelligent algorithm), and fine-tune the parameter according to the real-time production quality data to avoid excessive adjustment (the initial value is not more than ±20% in this embodiment), and the final specific value is the improvement parameter.

[0031] S6: Apply the improvement parameter and the pre-set response surface method to verify the adjustment effect, so that the product quality is stable within the pre-set threshold range.

[0032] The response surface method refers to a test verification method that visually displays the influence of factor interaction on quality by constructing a three-dimensional response surface and contour map of "main influencing factors-product quality", and verifies whether the improvement parameter can make the quality stable in the target interval.

[0033] The threshold range refers to the allowable fluctuation interval of the target quality of the product.

[0034] The response surface method and the threshold range are set by those skilled in the art in advance, and will not be described here.

[0035] The obtained improvement parameter is used in the actual production of the injection molding machine, and the response surface method is used for verification to know whether the product quality produced under the improvement parameter falls within the threshold range.

[0036] If it falls into, it shows that the product quality produced under the improved parameters is qualified.

[0037] Referring to Figure 2 Also includes the specific screening method of the main influencing factors: S20: Collect historical production data.

[0038] The historical production data refers to the associated data of the injection molding process parameter value and the corresponding product production quality data recorded in the past injection molding production process. The historical production data is obtained by calling the preset production record library. The injection molding production data of each time is recorded in the production record library. The production record library is set by the person skilled in the art in advance, which is not described here.

[0039] S21: Based on the historical production data, the initial reference value and the variation interval are called.

[0040] The initial reference value refers to the default value of the injection molding process parameter that can make the product quality basically qualified. The variation interval refers to the range that the injection molding process parameter can be safely adjusted. The initial reference value and the variation interval can be obtained by understanding the historical production data. The initial reference value and the variation interval are contained in the historical production data.

[0041] S22: Divide the test gradient according to the initial reference value and the variation interval.

[0042] The test gradient refers to the gradient used in the single factor test to test the effect of different parameters on the quality one by one. By dividing the parameter test points in the variation interval with the initial reference value as the center and equally spaced, the test gradient is obtained. The specific interval is set by the person skilled in the art in advance, which is not described here.

[0043] S23: Based on the test gradient and the injection molding process parameter, the injection molding production is carried out, and the product production quality of the produced product is collected.

[0044] The product production quality refers to the actual quality detection value of the product produced under a certain process parameter gradient in the single factor test. The product production quality is obtained by the on-site technical personnel detecting the quality of the produced product. The method of quality detection is selected by the on-site technical personnel, which is not described here.

[0045] The test gradient and the injection molding process parameter are obtained, and the product production quality of the produced product is collected, so as to facilitate the subsequent steps.

[0046] S24: Combine the test gradient, the injection molding process parameter, and the product production quality to obtain the range coefficient and the variation coefficient.

[0047] The range coefficient refers to the difference between the maximum value and the minimum value of the product production quality under different gradients of a parameter in the single factor test.

[0048] The coefficient of variation refers to the ratio of the standard deviation of the product quality to the mean value of the parameter at different gradients in a single factor experiment.

[0049] First, record the product quality data corresponding to each injection molding process parameter at all test gradients, then sort these data by numerical value, mark the maximum value, minimum value and all original data, and finally calculate the range coefficient and the coefficient of variation. Since the calculation formulas of the range coefficient and the coefficient of variation are common knowledge in the art, they will not be described here.

[0050] S25: When the range coefficient exceeds the preset range threshold or the coefficient of variation exceeds the preset coefficient of variation threshold, mark the parameter as a main influencing factor.

[0051] The range threshold refers to the range critical value for determining whether the parameter is a main influencing factor. When the range coefficient of a certain parameter exceeds this value, it indicates that its impact on product quality is significant, and it needs to be marked as a main influencing factor.

[0052] The coefficient of variation threshold refers to the coefficient of variation critical value for determining whether the parameter is a main influencing factor. When the coefficient of variation of a certain parameter exceeds this value, it indicates that its impact on product quality stability is significant, and it needs to be marked as a main influencing factor.

[0053] The range threshold and the coefficient of variation threshold are both set by the skilled person in advance, and will not be described here.

[0054] Reference Figure 3 It also includes a method for constructing a product regression equation: S30: Based on collecting the actual value of the main influencing factor, the actual value of the target quality of the product, and the actual value of the production quality of the product during injection molding production.

[0055] The actual value of the factor refers to the actual test value of the main influencing factor. The actual value of the factor is obtained by real-time collection through the parameter monitoring system (such as pressure sensor, temperature sensor) of the injection molding machine during injection molding production.

[0056] The actual value of the target refers to the preset standard value of the target quality of the product. The actual value of the target is obtained by pre-inputting by the skilled person in the art, and will not be described here.

[0057] The actual value of the production refers to the actual detection value of the product quality under the actual value of the factor. The actual value of the production is obtained by detecting 3 to 5 random samples according to the preset detection standard (such as weighing with an electronic scale with an accuracy of 0.01g for weight, and measuring with a micrometer for size) in a constant temperature and humidity environment for each batch of products, and taking the average value as the actual value of the production of the batch.

[0058] In the injection molding production, the actual values of the main influencing factors, the target actual values of the product target quality and the production actual values of the product production quality are collected first, so as to facilitate the subsequent steps.

[0059] S300: According to the actual values of the factors, the target actual values and the production actual values, the main effect coefficients, the interaction coefficients, the quadratic term coefficients and the equation constant term are calculated.

[0060] The main effect coefficient refers to the quantitative value of the independent influence degree of a single main influencing factor on the product target quality.

[0061] The interaction coefficient refers to the quantitative value of the additional influence of two main influencing factors on the product target quality when they act together.

[0062] The quadratic term coefficient refers to the quantitative value of the nonlinear influence of the square term of a single main influencing factor on the product target quality.

[0063] The equation constant term refers to the theoretical basic value of the product target quality when the values of all main influencing factors are 0.

[0064] The collected actual values of the factors, the target actual values and the production actual values are arranged into structured data sets, and the abnormal values (such as data deviating from the mean by 3 times the standard deviation) are removed to ensure that the sample size is not less than 30 groups (meeting the statistical requirements of regression analysis).

[0065] First, a simplified regression model (used to solve the main effect coefficient) is constructed as follows: ; Wherein, is the product target quality, β0 is the equation constant term, β1, β2, β3 are the main effect coefficients, is the error term of the simplified regression model.

[0066] Secondly, the covariance matrix of the main influencing factors part is extracted from the four-order covariance matrix as follows: ; Wherein, is the covariance matrix of the main influencing factors part, , , is the sample variance, , , is the sample covariance between factors.

[0067] At the same time, the covariance vector between the main influencing factors and the product quality is constructed as follows: ; wherein, is the covariance vector between the main factors and the product quality, 、 、 is the sample covariance between the factors and the quality.

[0068] Finally, the main effect coefficients are solved by matrix inversion operation, and the formula is: ; wherein, is the inverse matrix of the covariance matrix, β1, β2, β3 are the main effect coefficients.

[0069] The interaction coefficients and the quadratic term coefficients can be solved by extending the independent variable matrix and using the least squares method simultaneously according to the above logic, and the constant term of the equation is derived by substituting the mean value into the regression equation on the basis of the known other coefficients. The mean value substitution regression equation is well known in the art and will not be described here.

[0070] S301: Combine the main effect coefficients, interaction coefficients, quadratic term coefficients, equation constant terms, and preset error terms to generate a product regression equation.

[0071] The error term is a random variable in the regression equation that represents the quality fluctuation that is not explained by the main factors. The error term is set by those skilled in the art in advance and will not be described here.

[0072] The product regression equation can be obtained by combining the main effect coefficients, interaction coefficients, quadratic term coefficients, equation constant terms, and error terms, as follows: Y=β0+β1F1+β2F2+β3F3+β 12 F1F2+β 13 F1F3+β 23 F2F3+β 11 F1 2 +β 22 F2 2 +β 33 F3 2 +ϵ。

[0073] wherein, Y is the target quality of the product, β0 is the constant term, β1, β2, β3 are the main effect coefficients of the main factor one F1, the main factor two F2, and the main factor three F3 on Y, respectively, β 12 , β 13 , β 23 are the interaction coefficients of F1 and F2, F1 and F3, and F2 and F3, respectively, β 11 , β 22 , β 33The quadratic coefficients of F1, F2, and F3, respectively, and ε is an error term.

[0074] In this embodiment, F1 is the back pressure of the storage, F2 is the injection pressure, and F3 is the temperature of the barrel.

[0075] Referring to Figure 4 The specific calculation method of the correlation degree parameter is also included. S31: Based on the main influencing factors and the product target quality, a four-order covariance matrix is constructed.

[0076] The four-order covariance matrix refers to a 4x4 matrix constructed with three main influencing factors F1, F2, and F3 and one product target quality as dimensions.

[0077] The four-order covariance matrix can be obtained by combining the main influencing factors and the product target quality. The expression is as follows: ; Wherein, F i is the main influencing factor, F1 is the first main influencing factor, F2 is the second main influencing factor, F3 is the third main influencing factor, and Y is the product target quality.

[0078] S310: Combine the actual value of the factor, the actual value of the target, the actual value of the production, and the four-order covariance matrix to calculate the sample variance and the sample covariance.

[0079] The sample variance refers to the elements on the diagonal line of the four-order covariance matrix. It is used to quantify the dispersion degree of the value of a single variable.

[0080] The sample variance is calculated by the following formula: .

[0081] Wherein, n is the sample number, is the mean of the observed sample of the main influencing factor F i , and is the sample variance on the diagonal line of the four-order covariance matrix.

[0082] The sample covariance refers to the elements on the non-diagonal line of the four-order covariance matrix. It is used to quantify the degree of cooperative change of the values of two variables.

[0083] The sample covariance includes the sample covariance between factors and the sample covariance between factors and quality. It can be calculated by the following formula: Wherein, the sample covariance between factors is calculated by the following formula: .

[0084] The sample covariance between factors and quality is calculated by the following formula: .

[0085] Wherein, F j is the mean of the observation sample of the factor F k is the kth value of the target quality of the product, is the mean of the observation sample of the target quality of the product. is the sample covariance between factors, is the sample covariance between factors and quality.

[0086] S311: obtaining a correlation coefficient matrix according to the sample variance and the sample covariance.

[0087] The correlation coefficient matrix refers to a 4x4 matrix formed by converting the "covariance / variance" in the fourth-order covariance matrix into "correlation coefficient".

[0088] Wherein, the correlation coefficient is calculated by the following formula: ; ; Wherein, is the correlation coefficient of factor i and factor j, is the correlation coefficient of factor i and index Y, is the sample standard deviation of F i , is the sample standard deviation of F j , S Y is the sample standard deviation of Y; The correlation coefficient matrix expression is generated by combining and ; Wherein, R is the correlation coefficient matrix expression.

[0089] S312: obtaining a correlation parameter based on the correlation coefficient matrix.

[0090] The correlation parameter is all the non-diagonal elements in the correlation coefficient matrix. The correlation parameter can be obtained by querying the obtained correlation coefficient matrix.

[0091] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of functional modules is taken as an example, and in actual application, the above-mentioned functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0092] The above merely describes the preferred embodiments of the present application, and the protection scope of the present application is not limited to the above-described embodiments. Any technical solution falling within the concept of the present application shall fall within the protection scope of the present application. It should be noted that, for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application shall also be considered as falling within the protection scope of the present application.

Claims

1. A method for evaluating and improving the quality of injection-molded articles, characterized in that, The method comprises the following steps: Collecting injection molding process parameters and product target quality; Performing single-factor preliminary test on injection molding process parameters to screen out main influencing factors; Based on main influencing factors and product target quality, a product regression equation is constructed, and a correlation degree parameter is calculated based on main influencing factors and product target quality; According to the correlation degree parameter, the correlation strength is obtained; Combining the product regression equation and the correlation strength, the improvement parameter is generated; The improvement parameter and the preset response surface method are applied to verify the adjustment effect, so that the product quality is stabilized within the preset threshold range.

2. The method for evaluating and improving the quality of an injection-molded article according to claim 1, characterized in that, The method further comprises a specific screening method for main influencing factors: Collecting historical production data; Based on historical production data, initial reference values and variation intervals are retrieved; According to the initial reference values and variation intervals, test gradients are divided; Based on the test gradients and injection molding process parameters, injection molding production is performed, and the product production quality of the produced products is collected; Combining the test gradients, injection molding process parameters and product production quality, the range coefficient and variation coefficient are obtained; When the range coefficient exceeds the preset range threshold or the variation coefficient exceeds the preset variation coefficient threshold, the parameter is marked as a main influencing factor.

3. The method for evaluating and improving the quality of an injection-molded article according to claim 1, characterized in that, The method further comprises a construction method for the product regression equation: Based on the actual values of the main influencing factors, the actual values of the product target quality and the actual values of the product production quality during injection molding production; According to the actual values, the actual values and the actual values, the main effect coefficient, the interaction coefficient, the quadratic term coefficient and the equation constant term are calculated; Combining the main effect coefficient, the interaction coefficient, the quadratic term coefficient, the equation constant term and the preset error term, the product regression equation is generated.

4. The method for evaluating and improving the quality of an injection-molded article according to claim 3, characterized in that, The method further comprises a specific product regression equation: Y = β0+ β1F1+ β2F2+ β3F3+ β 12 F1F2+ β 13 F1F3+ β 23 F2F3+ β 11 F1 2 + β 22 F2 2 + β 33 F3 2 + ε; Wherein, Y is the target quality of the product, β0 is a constant term, β1, β2, β3 are main effect coefficients of main influencing factor one F1, main influencing factor two F2, and main influencing factor three F3 on Y, respectively, β 12 , β 13 , β 23 are interaction coefficients of F1 and F2, F1 and F3, and F2 and F3, respectively, β 11 , β 22 , β 33 are quadratic coefficients of F1, F2, and F3, respectively, and ϵ is an error term.

5. The method of claim 3, wherein The method further comprises a specific calculation method for the correlation degree parameter: Based on the main influencing factors and the product target quality, a four-order covariance matrix is constructed; Combining the actual values, the actual values, the actual values and the four-order covariance matrix, the sample variance and the sample covariance are calculated; According to the sample variance and the sample covariance, the correlation coefficient matrix is obtained; Based on the correlation coefficient matrix, the correlation degree parameter is obtained.

Images