Machining product quality online detection precision evaluation method based on multi-factor fusion

Abstract

The invention discloses a machining product quality online detection precision evaluation method based on multi-factor fusion. The method comprises the following steps: firstly, determining machining process input factor characteristics of online detection precision evaluation and an output machining precision grade; secondly, historical samples of the factor characteristics are analyzed, an atomization sample plate set is generated, and a random fuzzy sample plate set is constructed; an online input feature sample is converted into a random fuzzy to-be-detected set, then the random fuzzy sample set is matched with the random fuzzy to-be-detected set, and reliability distribution describing the nonlinear mapping relation between the multi-environment-factor features and the detection precision is obtained. And finally, fusing the reliability distribution, determining the precision grade of online detection of the quality of the machined product according to a fusion result, and judging whether the product is qualified or not. According to the method, a complex non-linear relationship can be processed, reliability distribution related to the detection precision can be efficiently obtained, and the online detection precision can be evaluated more accurately in practical application.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent manufacturing and quality control, and specifically relates to a method for evaluating the online detection accuracy of machined product quality based on multi-factor fusion. Background Art

[0002] In modern manufacturing, the processing accuracy of products directly affects their performance and service life. Especially in the field of high-precision manufacturing, the dimensional accuracy level of processed products is a key indicator to ensure product quality. However, due to many uncertain factors in the processing process, such as the stability of processing equipment, fluctuations in ambient temperature, changes in the characteristics of workpiece materials, etc., traditional dimensional accuracy detection methods are difficult to cope with these complex working conditions, often resulting in inaccurate and unreliable detection results. Statistics show that the product failure rate due to insufficient dimensional accuracy in production is about 10%, of which the defective rate caused by detection errors accounts for a large proportion.

[0003] The precision assessment (calibration) method of intelligent machined products is to dynamically adjust the product dimensional accuracy according to the real-time data, environmental changes and equipment operating status during the processing. In terms of dimensional detection, this method can keep the calibration parameters consistent with the actual processing status, provide more accurate accuracy level judgment, and effectively improve the processing quality of the product. Different from the traditional precision detection method, the online detection precision assessment method of machined product quality based on multi-factor fusion can automatically adjust the calibration strategy and correct the detection accuracy in real time by sensing the operating status of the processing equipment (such as spindle speed, feed speed, cutting force, etc.), environmental conditions (such as temperature, humidity, etc.), material properties (such as hardness, elasticity, etc.) and changes in processing tasks (such as finishing, roughing, etc.). This can overcome the detection errors caused by complex working conditions or environmental changes, ensure the stability of accuracy during the processing process, and guarantee product quality.

[0004] In this context, in order to improve the quality inspection accuracy of processed products and reduce the defective rate of products caused by dimensional errors, an online inspection method that can accurately calibrate the dimensional accuracy level of processed products in real time is urgently needed. The proposed online inspection accuracy assessment method for machined product quality based on multi-factor fusion can dynamically adjust and calibrate the dimensional changes during the processing process by combining multiple influencing factors to ensure high-precision quality control in complex environments. This is of great significance for improving product quality, reducing rework costs, and improving production efficiency. Summary of the invention

[0005] In view of the above problems, the present invention proposes a method for online detection accuracy assessment of machined product quality based on multi-factor fusion. The method first constructs a random fuzzy template set through historical sample data. When the actual detection data is obtained online, a random fuzzy set to be tested is generated, and it is matched with the random fuzzy template set in real time to obtain the reliability distribution of the accuracy level. Subsequently, the reliability distribution is comprehensively analyzed through the reliability fusion model, and finally the accuracy assessment result of the online data is given. The actual measurement value is calibrated according to the accuracy level obtained by the evaluation, and the calibration value is compared with the upper and lower tolerance range of the product geometric size to determine whether the product meets the quality standard.

[0006] The method for online detection accuracy assessment of machining product quality based on multi-factor fusion proposed by the present invention comprises the following steps:

[0007] (1) Determine the input factor characteristics and output accuracy level for multi-factor fusion of online detection accuracy assessment;

[0008] (2) Perform a coarse-grained analysis on historical samples to obtain the key parameters of the atomization sample set, including expected value (Ex), entropy (En), and hyperentropy (He), and generate an atomization sample set;

[0009] (3) Based on the atomized sample set generated in step (2), the KL divergence between the atomized sample set and the historical sample set is calculated to further construct a random fuzzy sample set;

[0010] (4) Obtaining the input factor feature data of product quality inspection online and constructing a random fuzzy set of input factor feature data to be inspected;

[0011] (5) Based on step (3) and step (4), the random fuzzy sample set is matched with the random fuzzy test set to obtain a reliability distribution describing the nonlinear mapping relationship between the characteristics of multiple environmental factors and the detection accuracy;

[0012] (6) Fuse the reliability distribution of multiple environmental factor characteristics and determine the accuracy level of online product quality detection based on the fusion results;

[0013] (7) After evaluating the actual measured value using the accuracy grade, compare the evaluated value with the upper and lower tolerances of the measured geometric dimensions of the product to determine whether the product is qualified.

[0014] The present invention proposes a method for online detection accuracy assessment of machined product quality based on multi-factor fusion. First, a multi-factor fusion model for online detection accuracy assessment is determined, including input factor characteristics and output accuracy level. Through coarse-grained analysis of historical samples, the key parameters of the atomization sample set are extracted to generate an atomization sample set. Based on the set, the KL divergence between it and the historical sample set is calculated, and a complete random fuzzy sample set is further constructed. After the input factor characteristic data of product quality detection is obtained online, a random fuzzy set to be tested corresponding to it is generated. By matching the random fuzzy sample set with the random fuzzy set to be tested, a reliability distribution describing the nonlinear mapping relationship between the characteristics of multiple environmental factors and the detection accuracy is obtained. Subsequently, the reliability distribution of multiple factor characteristics is fused to finally determine the accuracy level of online product quality detection. According to the accuracy level, the actual measured value is evaluated, and the evaluation value is compared with the upper and lower tolerance ranges of the product geometric dimensions to comprehensively judge whether the product meets the quality standards.

[0015] The present invention provides a method for online detection accuracy assessment of machined product quality based on multi-factor fusion, which has the following beneficial effects:

[0016] (1) Improve the accuracy of online detection accuracy assessment. The present invention can accurately evaluate the online detection accuracy of machined product quality by constructing multiple input factor features of a multi-factor fusion model. The key parameter extraction technology of the atomized sample set is used, and the random fuzzy sample set is further constructed in combination with the calculation of KL divergence, so that the model can handle complex nonlinear relationships, thereby improving the accuracy of detection accuracy assessment.

[0017] (2) Effectively handle the correlation between historical samples and real-time data. By performing a coarse-grained analysis of historical samples and constructing a random fuzzy template set based on KL divergence, the relationship between historical sample data and real-time detection data can be effectively mapped. In real-time detection, based on the matching of the random fuzzy test set and the random fuzzy template set, the reliability distribution related to the detection accuracy can be efficiently obtained, thereby more accurately evaluating the online detection accuracy in practical applications.

[0018] (3) Supports fusion analysis of multiple environmental factors. The present invention adopts a multi-factor fusion method to comprehensively consider multiple factors that affect the detection accuracy. By fusing the reliability distribution, it can effectively evaluate the detection accuracy under complex working conditions. This method is particularly suitable for production scenarios where environmental factors change frequently, ensuring that reliable detection results can still be obtained under diverse environmental conditions.

[0019] (4) Improve the reliability of quality assessment and judgment. By comparing with the upper and lower tolerance ranges of the product's geometric dimensions, the present invention can not only accurately assess the detection accuracy of product quality, but also further ensure whether the product meets the quality standards. This provides a more stable and reliable basis for quality assessment for the manufacturing industry, helping enterprises to promptly discover and correct product defects during the production process.

[0020] (5) It has good applicability and promotion value. The method of the present invention can be widely used in the fields of quality control, quality management and process optimization of machined products, and is particularly suitable for precision manufacturing and high-demand detection scenarios. Through the comprehensive evaluation of multi-factor characteristics, the present invention has strong adaptability and can adjust the model parameters according to different process requirements. It has good versatility and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flowchart of the online detection accuracy assessment process of machining product quality based on the fusion of multiple factors;

[0022] Figure 2 It is a structural diagram of online detection accuracy assessment of machining product quality based on multi-factor fusion;

[0023] Figure 3 It is the matching graph between the random fuzzy template set and the random fuzzy test set. DETAILED DESCRIPTION

[0024] The embodiment of the present application proposes a method for online detection accuracy calibration of machined product quality based on multi-factor fusion, and its flow chart is as follows: Figure 1 As shown, it includes the following steps:

[0025] (1) Determine the input factor characteristics and output accuracy level of multi-factor fusion for online detection accuracy assessment;

[0026] (2) Perform a coarse-grained analysis on historical samples to obtain the key parameters of the atomization sample set, including expected value (Ex), entropy (En), and hyperentropy (He), and generate an atomization sample set;

[0027] (3) Based on the atomized sample set generated in step (2), the KL divergence between the atomized sample set and the historical sample set is calculated to further construct a random fuzzy sample set;

[0028] (4) Obtaining the input factor feature data of product quality inspection online and constructing a random fuzzy set of input factor feature data to be inspected;

[0029] (5) Based on step (3) and step (4), the random fuzzy sample set is matched with the random fuzzy test set to obtain a reliability distribution describing the nonlinear mapping relationship between the characteristics of multiple environmental factors and the detection accuracy;

[0030] (6) Fuse the reliability distribution of multiple environmental factor characteristics and determine the accuracy level of online product quality detection based on the fusion results;

[0031] (7) After evaluating the actual measured value using the accuracy grade, compare the evaluated value with the upper and lower tolerances of the measured geometric dimensions of the product to determine whether the product is qualified.

[0032] The specific description of step (1) is as follows:

[0033] Determine the factor feature vector according to the geometric size type and processing technology of the machined product to be tested As the input of the multi-factor fusion model for online detection accuracy assessment, they represent the external environment temperature during the processing , external humidity , Detection unit mechanical amplitude , Detection unit mechanical vibration frequency and detection unit temperature ; Set the accuracy level As the output of the fusion model, they represent high accuracy , medium accuracy , low accuracy Then, by The feature sample set consisting of historical samples is recorded as , Indicates A set of factor feature vectors.

[0034] For ease of understanding, step (1) is illustrated here as an example. The process is as follows:

[0035] In the process of machining a mechanical component, determine the factor characteristic vector of the online detection accuracy assessment of the component machining quality , respectively represent the external environment temperature during the processing , external humidity , Detection unit mechanical amplitude , Detection unit mechanical vibration frequency and detection unit temperature ; Set the accuracy level As the output of the fusion model, they represent high accuracy , medium accuracy , low accuracy . In the processing of this mechanical component, Historical samples, which constitute the feature sample set, are recorded as .

[0036] The specific description of step (2) is as follows:

[0037] (2-1) According to the factor eigenvector The corresponding historical samples are filtered at the same accuracy level , we get the factor characteristics The accuracy level is of A collection of samples , Indicates the first samples.

[0038] (2-2) Based on the data obtained in step (2-1) , we can use formula (1) to formula (3) to obtain the set Atomization sample collection Key parameters :

[0039] (1)

[0040] (2)

[0041] (3)

[0042] (2-3) As expected, Generate a normal random number with standard deviation , which satisfies , and then For expectations, is the standard deviation, generate another normal random number ,satisfy ,and The corresponding degree of certainty It can be calculated by formula (4). So far, a subset of the atomization sample set can be obtained. :

[0043] (4)

[0044] (2-4) Repeat step (3-3) times until it produces Normal random numbers and their corresponding degrees of certainty constitute the atomization sample set ,in .

[0045] For ease of understanding, step (2) is illustrated here as an example. The process is as follows:

[0046] The historical sample The 180 sample data in the were screened and the accuracy levels under different factor characteristics were obtained. , , Data collection According to steps (2-4), we can get Corresponding atomization sample set .

[0047] The specific description of step (3) is as follows:

[0048] (3-1) Atomization sample set obtained based on step (2-4) , which contains The normal random numbers are arranged in ascending order according to their values, and the sorted normal random number set can be obtained. ,satisfy .

[0049] (3-2) The sorted normal random number set Evenly divided into two parts, which are low-order subsets in turn and high-order subsets ,in , .

[0050] (3-3) Similarly, In Based on steps (3-1) and (3-2), the sample sets can be successively obtained into low-order subsets and high-order subsets ,in, .like is an odd number, assign the middle element to any of the groups.

[0051] (3-4) Based on the low-order subset obtained in step (3-2) and high-order subsets , calculate their means respectively , and standard deviation , , and calculate the KL divergence value by formula (5) :

[0052] (5)

[0053] At the same time, we can get the low-order subset and high-order subsets The mean , Standard Deviation , , further calculation and The KL divergence value of :

[0054] (6)

[0055] Finally, we can use formula (7) to determine The degree of atomization :

[0056] (7)

[0057] (3-5) In The samples are randomly grouped with replacement to obtain a total of The number of samples in each group is , among which The group data can be expressed as ,in and From equation (8) and equation (9), we can obtain:

[0058] (8)

[0059] (9)

[0060] in represents the number of normal random numbers in step (3-4), and .

[0061] (3-6) For Group Data , calculate its expected value and standard deviation :

[0062] (10)

[0063] (11)

[0064] Similarly, all The mean of the group data and standard deviation .

[0065] (3-7) According to the estimation of the atomization degree in step (3-4), if It is judged as a non-atomized state, and the key parameters of the random fuzzy template set are obtained through equations (12) to (14): :

[0066] (12)

[0067] (13)

[0068] (14)

[0069] in , , , ;

[0070] like If it is judged to be in a fogging state, the key parameters of the random fuzzy template set are obtained through equations (15) to (18): :

[0071] (15)

[0072] (16)

[0073] (17)

[0074] (18)

[0075] in express The sample variance of express The sample fourth-order central moment of .

[0076] (3-8) Based on the method in step (3-7), we can obtain the factor characteristics The accuracy level is A random fuzzy template set Corresponding key parameters , then for the three accuracy levels under the five factor characteristics, the corresponding 15 key parameters of the random fuzzy sample set can be obtained, as shown in Table 1:

[0077] Table 1 Random fuzzy sample set Key parameters

[0078]

[0079] (3-9) Key parameters of the random fuzzy template set obtained based on step (3-8) , replace the ones in steps (2-3) in sequence , and According to steps (2-3) and (2-4), we can get 15 random fuzzy sample sets corresponding to 3 accuracy levels under 5 factor characteristics. :

[0080] Temperature characteristics of the external environment The random fuzzy sample sets of three levels of accuracy are ;

[0081] Humidity characteristics in the external environment Below, a set of random fuzzy templates with three levels of accuracy ;

[0082] Mechanical amplitude characteristics in the detection unit The random fuzzy sample sets of three levels of accuracy are ;

[0083] Mechanical vibration frequency characteristics in the detection unit Below, a set of random fuzzy templates with three levels of accuracy ;

[0084] Temperature characteristics in the detection unit Below, a set of random fuzzy templates with three levels of accuracy .

[0085] For ease of understanding, step (3) is illustrated here as an example. The process is as follows:

[0086] Calculate the historical sample screening The KL divergence value of The KL divergence value is compared to judge The atomization state is further obtained The key parameters of the random fuzzy template set , as shown in Table 2:

[0087] Table 2 Random fuzzy sample set Key parameters

[0088]

[0089] Based on the key parameters of the random fuzzy sample set in Table 2, 15 random fuzzy sample sets can be obtained: .

[0090] The specific description of step (4) is as follows:

[0091] (4-1) When a set of data to be evaluated is obtained online , the random fuzzy test set can be calculated by formula (19): Key parameters , and is determined based on the key parameters of the random fuzzy template set and :

[0092] (19)

[0093] (20)

[0094] (twenty one)

[0095] (4-2) Key parameters of the random fuzzy test set obtained based on step (4-1) , replace the steps (2-3) in sequence , and According to steps (2-3) and (2-4), we can get the random fuzzy test set under the characteristics of 5 factors. .

[0096] For ease of understanding, here is an example of online assessment data: When the random fuzzy test set The acquisition process is as follows:

[0097] According to formula (19)~(21), we can obtain the random fuzzy test set: The key parameters are shown in Table 3.

[0098] Table 3 Key parameters of random fuzzy test set

[0099]

[0100] based on The key parameters of the random fuzzy test set can be obtained .

[0101] The specific description of step (5) is as follows:

[0102] (5-1) Constructing a random fuzzy template set The expected curve and entropy expected curve are and , random fuzzy test set The expected curve and entropy expected curve are and .

[0103] (5-2) In the factor characteristics Under this condition, we can obtain a set of random fuzzy templates with three levels of accuracy. and random fuzzy test set The boundary value of the intersection area is calculated as follows:

[0104] calculate and The horizontal coordinate value of the intersection point , and the curve and The horizontal coordinate value of the intersection point , and obtain the curve and The ordinate value of the intersection point , and the curve and The vertical coordinate value of the intersection point , through the above four points, we can determine the random fuzzy template set and random fuzzy test set The subset extent of the intersection area.

[0105] (5-3) According to the range determined in step (5-2), from the random fuzzy template set Select the one that satisfies formula (22) Subset elements :

[0106] (twenty two)

[0107] At the same time, from the random fuzzy test set Select the one that satisfies formula (23) Subset elements :

[0108] (twenty three)

[0109] Finally, according to formula (24), we can get Next, the random fuzzy template set and random fuzzy test set The matching value of :

[0110] (twenty four)

[0111] (5-4) Obtain the matching values ​​of the three accuracy levels , , By normalizing according to formula (26), we can obtain the reliability distribution of the data to be evaluated at different accuracy levels as shown in formula (25): :

[0112] (25)

[0113] (26)

[0114] (5-5) Similarly, for different input factor characteristics, the reliability distribution shown in Table 4 can be obtained;

[0115] Table 4 Reliability distribution under different factor characteristics

[0116]

[0117] For ease of understanding, step (5) is illustrated here as an example. The process is as follows:

[0118] Construct a set of 15 random fuzzy templates The expected curve and entropy expected curve of The expected curve of is matched with the entropy expected curve, and the obtained matching value is normalized to the reliability distribution, as shown in Table 5:

[0119] Table 5 Reliability distribution under different factor characteristics

[0120]

[0121] The specific description of step (6) is as follows:

[0122] (6-1) Based on historical feature samples , construct the initial judgment matrix :

[0123] (27)

[0124] (6-2) Normalize the initial judgment matrix to obtain the normalized discriminant matrix :

[0125] (28)

[0126] In the formula .

[0127] (6-3) Calculation The probability of each element in , and obtain the probability matrix :

[0128] (29)

[0129] In the formula .

[0130] (6-4) Calculate the information entropy of each input feature through formula (30) , and calculate the fusion factor of each attribute through formula (31) :

[0131] (30)

[0132] (31)

[0133] (6-5) Based on the reliability obtained in step (5-5) ,in ,set up Credibility Factor The evidence obtained under different features is fused using the credibility fusion model to obtain the fused credibility distribution :

[0134] (32)

[0135] (33)

[0136] (34)

[0137] (35)

[0138] in for The power set of To obtain the final reliability distribution after fusing the five reliability distributions, For any one different from A collection of .

[0139] (6-6) Based on the fusion reliability obtained in step (6-5) , the accuracy level corresponding to the maximum confidence is selected as the online detection accuracy assessment result.

[0140] For ease of understanding, step (6) is illustrated here as an example. The process is as follows:

[0141] Based on the analysis of historical samples, the fusion factors under the characteristics of each factor can be obtained ,set up Credibility Factor , integrating the reliability distribution under different factor characteristics to obtain , select the accuracy level corresponding to 0.6249 As the result of online detection accuracy assessment.

[0142] The specific description of step (7) is as follows:

[0143] (7-1) According to the online detection accuracy evaluation results of step (6-6) , the accuracy level represents the measurement error range that may be produced by the measuring equipment, denoted as ,in and are the upper and lower bounds of the accuracy level respectively.

[0144] (7-2) Use the measuring equipment to measure the product to obtain the actual geometric dimensions of the product, recorded as , and based on the measurement error interval corresponding to the accuracy level, calculate the evaluation size interval of the product to be tested .

[0145] (7-3) According to product design requirements, the ideal tolerance range of qualified products is set as ,in, is the lower tolerance, Represents the upper tolerance. According to equations (33) and (34), the actual tolerance interval of the product to be tested is obtained. :

[0146] (36)

[0147] (37)

[0148] (7-4) Based on the actual tolerance interval of the product to be tested obtained in step (7-3), if both , If the product is found to be qualified, it is a qualified product; otherwise, it is an unqualified product.

[0149] For ease of understanding, step (7) is illustrated here as an example. The process is as follows:

[0150] Given the actual product processing information as shown in Table 6:

[0151] Table 6 Product processing information

[0152]

[0153] The online detection accuracy evaluation results are: , the corresponding measurement error interval is [-0.001, +0.001], the evaluation size interval of the product to be tested is calculated to be [9.9381-0.001, 9.9381+0.001], and the actual tolerance interval of the product to be tested is [-0.0156, -0.0136] according to formula (33) and formula (34), which meets the tolerance requirements and the product to be tested is a qualified product.

[0154] The following is combined with Figure 2 and attached Figure 3 , an embodiment of the method of the present invention is described in detail:

[0155] The core part of the present invention is: first, determine the multi-factor fusion model of online detection accuracy assessment, including input factor characteristics and output accuracy level. Through coarse-grained analysis of historical samples, extract the key parameters of the atomization sample set, and generate the atomization sample set. Based on the set, calculate its KL divergence with the historical sample set, and further construct a complete random fuzzy sample set. After obtaining the input factor characteristic data of product quality detection online, generate the corresponding random fuzzy set to be tested. By matching the random fuzzy sample set with the random fuzzy set to be tested, obtain the reliability distribution that describes the nonlinear mapping relationship between the characteristics of multiple environmental factors and the detection accuracy. Subsequently, the reliability distribution of multiple factor characteristics is fused to finally determine the accuracy level of online product quality detection. According to the accuracy level, the actual measured value is evaluated, and the evaluation value is compared with the upper and lower tolerance ranges of the product geometric dimensions to comprehensively judge whether the product meets the quality standards. Figure 2 The structural block diagram of online detection accuracy evaluation is given in.

[0156] The following describes in detail each step of the method of the present invention in combination with actual evaluation data.

[0157] 1. Determine the input factor characteristics and output accuracy level of the multi-factor fusion model for online detection accuracy assessment. During the processing of a certain mechanical component, determine the factor feature vector As the input of the multi-factor fusion model for online detection accuracy assessment, they represent the external environment temperature during the processing , external humidity , Detection unit mechanical amplitude , Detection unit mechanical vibration frequency and detection unit temperature ; Set the accuracy level As the output of the fusion model, they represent high accuracy , medium accuracy , low accuracy .here Represents the detection error interval [-0.001, +0.001], Represents the detection error interval [-0.002, +0.002], Represents the detection error range [-0.003, +0.003]. Historical samples, which constitute the feature sample set, are recorded as .

[0158] 2. Generate atomization sample set:

[0159] The historical sample The 180 sample data in the were screened and the accuracy levels under different factor characteristics were obtained. , , Data Collection ,For example:

[0160]

[0161]

[0162]

[0163] Other collections have similar forms.

[0164] According to steps (2-4), we can get Corresponding atomization sample set ,For example:

[0165] ={(25.9526,0.7260),(46.148,0.4373),…,(47.0014,0.4005)}

[0166] ={(35.3798,0.9703),(35.126,0.9923),…,(35.5798,0.5990)}

[0167] ={(41.1739,0.4462),(40.0881,0.7946),…,(42.9997,0.4119)}

[0168] Other collections have similar forms.

[0169] 3. Construct a random fuzzy sample set:

[0170] Calculate the historical sample screening The KL divergence value of The KL divergence value is compared to judge The atomization state is further obtained The key parameters of the random fuzzy template set , as shown in Table 7:

[0171] Table 7 Random fuzzy sample set Key parameters

[0172]

[0173] Based on the key parameters of the random fuzzy sample set in Table 7, 15 random fuzzy sample sets can be obtained: ,For example:

[0174] ={(30.9484,0.9319),(34.0306,0.8247),…,(33.5275,0.9026)}

[0175] ={(37.4174,0.2999),(29.5385,0.1882),…,(32.7606,0.8906)}

[0176] ={(32.0331,0.5947),(33.6100,0.9184),…,(37.0759,0.6972)}

[0177] Other collections have similar forms.

[0178] 4. Construct a random fuzzy set to be tested:

[0179] Online evaluation data is When the random fuzzy test set The acquisition process is as follows:

[0180] According to formula (19)~(21), we can obtain the random fuzzy test set: The key parameters are shown in Table 8:

[0181] Table 8 Key parameters of random fuzzy test set

[0182]

[0183] based on The key parameters of the random fuzzy test set can be obtained ,For example:

[0184] ={(29.0937,0.7665),(24.9778,0.7329),…,(25.6533,0.8547)}

[0185] Other collections have similar forms.

[0186] 5. Random fuzzy template set matches random fuzzy test set:

[0187] Construct a set of 15 random fuzzy templates The expected curve and entropy expected curve of The expected curve of is matched with the entropy expected curve, and the obtained matching value is normalized to the reliability distribution, as shown in Table 9. Figure 3 The matching graph between the random fuzzy template set and the random fuzzy test set is given in .

[0188] Table 9 Reliability distribution under different factor characteristics

[0189]

[0190] 6. The reliability distribution is integrated and the accuracy level of online product quality detection is determined based on the integration results:

[0191] Based on the analysis of historical samples, the fusion factors under the characteristics of each factor can be obtained ,set up Credibility Factor , integrating the reliability distribution under different factor characteristics to obtain =[0.6249,0.2849,0.0902], select the accuracy level corresponding to 0.6249 As the result of online detection accuracy assessment.

[0192] 7. The accuracy grade is evaluated on the actual measured value, and the product is judged to be qualified by comparing the evaluated value with the upper and lower tolerances of the measured geometric dimensions of the product.

[0193] Given the actual product processing information as shown in Table 10:

[0194] Table 10 Product processing information

[0195]

[0196] The online detection accuracy evaluation results are: , the corresponding measurement error interval is [-0.001, +0.001], the evaluation size interval of the product to be tested is calculated to be [9.9381-0.001, 9.9381+0.001], and the actual tolerance interval of the product to be tested is [-0.0156, -0.0136] according to formula (33) and formula (34), which meets the tolerance requirements and the product to be tested is a qualified product.

Claims

1. The online detection accuracy assessment method for machining product quality based on multi-factor fusion is characterized by: The following steps are involved: Step 1: Determine the input factor characteristics and output accuracy level for online detection accuracy assessment; Step 2: Analyze historical samples, obtain key parameters, and generate atomization sample sets; Step 3: Calculate the KL divergence between the atomized sample set and the historical samples to construct a random fuzzy sample set; Step 4: Obtain the input factor features of product quality inspection online and construct a random fuzzy set to be inspected; Step 5: Match the random fuzzy sample set with the random fuzzy test set to obtain a reliability distribution describing the nonlinear mapping relationship between the characteristics of multiple environmental factors and the detection accuracy; Step 6: Fuse the reliability distribution and determine the accuracy level of online product quality detection based on the fusion result; Step 7: Use the accuracy level to evaluate the actual measured value and determine whether the product is qualified.

2. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 1 is characterized in that: The input factor characteristics and output accuracy level described in step 1 are as follows: Determine the factor feature vector according to the geometric dimension type and processing technology of the machined product to be tested , as the input of online detection accuracy assessment, represents the external environment temperature during the processing , external humidity , Detection unit mechanical amplitude , Detection unit mechanical vibration frequency and detection unit temperature ; Set the accuracy level , as output, represents high accuracy , medium accuracy , low accuracy ;Depend on The feature sample set consisting of historical samples is recorded as , For the A set of factor feature vectors.

3. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 2 is characterized in that: The specific implementation process of step 2 is as follows: Step 2-1: According to The corresponding historical samples are filtered at the same accuracy level , get the factor characteristics The accuracy level is of A collection of samples ; Step 2-2, based on , find the value of Atomization sample collection The key parameters, including the expected value ,entropy and super entropy , Super Entropy Entropy one third of Step 2-3: For expectations, is the standard deviation, generating normal random numbers ; and then For expectations, is the standard deviation, generating normal random numbers , The corresponding degree of certainty is , get a subset of the atomization template set : Repeat steps 2-4 and 2-3 times until it produces Normal random numbers and their corresponding degrees of certainty constitute the atomization sample set .

4. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 3 is characterized in that: The specific implementation process of step 3 is as follows: Step 3-1: Include the atomization sample set Normal random numbers are arranged in ascending order of value to obtain a sorted set , ; Step 3-2: sort the normal random number set Evenly partition into subsets and subset ,in , ; Step 3-3, similarly, In The sample sets are then sub-sets and subset ,in, ;like is an odd number, assign the middle element to any of the groups; Step 3-4: Based on subset and subset , calculate the KL divergence value : Subset-based and subset , calculate the KL divergence value ,judge The degree of atomization : ； Step 3-5: In The samples are randomly grouped with replacement, and we get The number of samples in each group is , among which The group data is represented as ,in and ; Steps 3-6: Group Data , calculate its expected value and standard deviation , similarly obtain all The mean of the group data and standard deviation ; Step 3-7, obtaining key parameters of the random fuzzy sample set according to the degree of atomization; Step 3-8: Based on the factor characteristics obtained in step 3-7 The accuracy level is A random fuzzy template set Corresponding key parameters , then for the three accuracy levels under the five factor characteristics, the corresponding 15 key parameters of the random fuzzy template set are obtained; Step 3-9: Based on the key parameters of the random fuzzy template set, replace the key parameters in step 2-3 in turn. , and According to steps 2-3 and 2-4, we get 15 random fuzzy sample sets corresponding to 3 accuracy levels under 5 factor characteristics. .

5. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 4 is characterized in that: The specific implementation process of steps 3-7 is as follows: like It is judged as non-atomized state, and the key parameters of the random fuzzy sample set are obtained: ； ； ； in , , , ; like It is judged as a fogged state, and the key parameters of the random fuzzy sample set are obtained: ； ； ； ； in express The sample variance of express The sample fourth-order central moment of .

6. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 5 is characterized in that: The specific implementation process of step 4 is as follows: Step 4-1: Obtain the data to be assessed online , calculate the random fuzzy test set Key parameters , and is determined based on the key parameters of the random fuzzy template set and , and are the means of three corresponding key parameters respectively; Step 4-2: Based on key parameters , replacing , and According to steps 2-3 and 2-4, we get the random fuzzy set to be tested under the characteristics of 5 factors. .

7. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 6 is characterized in that: The specific implementation process of step 5 is as follows: Step 5-1: Construct a random fuzzy sample set The expected curve and the entropy expectation curve , random fuzzy test set The expected curve and the entropy expectation curve ; Step 5-2: Factor characteristics Under this condition, we can obtain a set of random fuzzy templates with three levels of accuracy. and random fuzzy test set Boundary value of the intersection area; Step 5-3: According to the range determined in step 5-2, select the random fuzzy template set Select Subset elements ,satisfy: ; From the random fuzzy test set Select Subset elements ,satisfy: ; In the factor characteristics Down, and The matching value of ; Step 5-4: Obtain the matching values ​​of the three accuracy levels , , Normalize and combine the three normalized matching values ​​to obtain the reliability distribution of the data to be evaluated at different accuracy levels. ; Step 5-5: For different input factor characteristics, obtain the corresponding reliability distribution.

8. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 7 is characterized in that: The specific implementation process of step 6 is as follows: Step 6-1: Take the factors of each sample in the historical sample as a row in the matrix and construct the initial judgment matrix , and normalize it to obtain the normalized discriminant matrix ; Step 6-2, Calculation The probability of each element in , and obtain the probability matrix , the information entropy of the input features is calculated by the elements in the probability matrix , get the fusion factor of each attribute : ； Step 6-3: Based on the reliability obtained in step 5-5 ,in ,set up Credibility Factor , and obtain the fused reliability distribution : ； ， ； ； in for The power set of is the confidence distribution after fusion, For any one different from The accuracy level corresponding to the maximum fusion confidence is selected as the online detection accuracy assessment result.

9. The method for online detection accuracy assessment of machining product quality based on multi-factor fusion according to claim 8 is characterized in that: The specific implementation process of step 7 is as follows: Step 7-1: Based on the online detection accuracy evaluation results , the accuracy level represents the measurement error interval produced by the measuring equipment, denoted as ,in and are the upper and lower bounds of the accuracy level respectively; Step 7-2: Measure the product to be tested and obtain the actual geometric dimensions of the product , and based on the measurement error interval corresponding to the accuracy level, calculate the evaluation size interval of the product to be tested ; Step 7-3: Set the ideal tolerance range of qualified products as , is the lower tolerance, Indicates the upper tolerance and obtains the actual tolerance range of the product to be tested : ； ； Step 7-4: Based on the actual tolerance range of the product to be tested, if both , If the product is found to be qualified, it is a qualified product; otherwise, it is an unqualified product.

Images