Reliability-oriented key quality characteristic identification method and device of product

Abstract

The invention provides a reliability-oriented key quality characteristic identification method and device of a product, and the method comprises the steps: determining K quality characteristics of a target product based on the design data and manufacturing data of the target product; based on the K quality characteristics of the target product, initializing the positions of the N individuals, and taking the initialized N individuals as the N individuals of the first iteration; determining candidate individuals in the initialized N individuals as candidate individuals of the first iteration; repeatedly executing the iteration step until the current number of iterations reaches the preset maximum number of iterations; wherein candidate individuals in the current number of iterations are determined from N individuals in each iteration step; and determining the quality characteristic of which the code is 1 in the candidate individual with the maximum number of iterations as the reliability-oriented key quality characteristic. The reliability-oriented key quality characteristics of the product can be identified, and the method has the advantages of high identification accuracy and high convergence speed.

Description

Technical Field

[0001] The present application relates to the technical field of product reliability, and in particular to a method and device for identifying key quality characteristics of a product for reliability. Background Art

[0002] Built-in reliability is a method that takes key quality characteristics as the core and implements product reliability work in all stages. It improves product quality and reliability by controlling the deviation of key quality characteristics. As an important carrier of product quality, the key quality characteristics of products run through all stages of the product life cycle. They are in line with the built-in reliability method and are the core of the built-in reliability method. Identifying and controlling the key quality characteristics of products is an important means to ensure product reliability.

[0003] However, traditional key quality characteristics are often limited to the product manufacturing process, ignoring the large amount of quality and reliability data that exists in the product design process and use process, and are unable to be driven by reliability and achieve built-in reliability. Summary of the invention

[0004] In view of this, the present application provides a method and device for identifying key quality characteristics of a product for reliability to solve the above technical problems.

[0005] In a first aspect, an embodiment of the present application provides a method for identifying key quality characteristics of a product for reliability, comprising:

[0006] Based on the design data and manufacturing data of the target product, determine K quality characteristics of the target product;

[0007] Based on the K quality characteristics of the target product, the positions of N individuals are initialized, and the initialized N individuals are used as the N individuals of the first iteration; wherein the position of each individual is a binary number consisting of K 0s or 1s, wherein 0 indicates that the quality characteristic is not a key quality characteristic for reliability, and 1 indicates that the quality characteristic is a key quality characteristic for reliability;

[0008] Determine the candidate individuals among the initialized N individuals as the candidate individuals for the first iteration;

[0009] Repeat the following steps:

[0010] Update the positions of N individuals in the previous iteration to obtain the positions of N individuals in the current iteration;

[0011] The position of each individual of the current iteration is processed using a pre-trained classifier to obtain the error rate of each individual, and the error rate of each individual and the number of key quality characteristics for reliability are brought into the fitness function to obtain the fitness value of each individual; wherein the classifier is trained using the quality characteristic data and failure data of multiple samples of the target product;

[0012] The individual with the smallest fitness value is determined as the candidate individual for the current iteration;

[0013] Based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, the candidate individuals of the current iteration are updated;

[0014] Until the current number of iterations reaches the preset maximum number of iterations;

[0015] The quality characteristics coded as 1 among the candidate individuals with the maximum number of iterations are determined as the reliability-oriented key quality characteristics.

[0016] In a possible implementation, determining a candidate individual among the initialized N individuals as a candidate individual for the first iteration includes:

[0017] Using the classifier to process the initial position of each individual to obtain the initial error rate of each individual;

[0018] Substitute the initial error rate of each individual and the number of key quality characteristics for reliability into the fitness function to obtain the initial fitness value of each individual;

[0019] The individual with the smallest initial fitness value is determined as the candidate individual among the initialized N individuals and is used as the candidate individual for the first iteration.

[0020] In a possible implementation, the positions of N individuals at the previous iteration are updated to obtain the positions of N individuals at the current iteration; including:

[0021] Determine the first individual among N individuals as the leader, and the remaining N-1 individuals as followers;

[0022] The position of the leader at the mth iteration for:

[0023]

[0024] Among them, P m is the candidate individual in the current iteration number m, c1 is the convergence factor, M is the preset maximum number of iterations; c2 and c3 are both random numbers in [0,1];

[0025] The position of the nth follower at the mth iteration for:

[0026]

[0027] Among them, 2≤n≤N; is the position of the nth individual in the m-1th iteration, is the position of the n-1th individual in the m-1th iteration.

[0028] In a possible implementation, the method further includes:

[0029] Obtain a training set of a target product, wherein the training set includes multiple groups of training samples, each group of training samples includes measurement data of multiple target products and corresponding annotated fault data, wherein the measurement data includes measurement values ​​of H reliability-oriented key quality characteristics; the annotated fault data is up to standard or not up to standard;

[0030] The measured data of multiple target products are processed by the classifier to obtain the predicted target data of multiple target products: meet the target or fail to meet the target;

[0031] Using the failure data of multiple target products, we can obtain the real compliance data of multiple target products.

[0032] Determine the error rates of multiple target products using predicted and actual target data;

[0033] The error rates of multiple target products and H values ​​are used to update the parameters of the classifier.

[0034] In a possible implementation, determining the candidate individuals for the next iteration number based on the candidate individuals for the previous iteration number and the candidate individuals for the current iteration number includes:

[0035] Determine whether the fitness value of the candidate individual of the current iteration number is less than the fitness value of the candidate individual of the previous iteration number. If yes, determine the candidate individual of the next iteration number as the candidate individual of the current iteration number.

[0036] Otherwise, calculate the probability value p:

[0037]

[0038] Where f(X) is the fitness value of candidate individual X at the current iteration, f(X0) is the fitness value of candidate individual X0 at the previous iteration; T is the annealing temperature;

[0039] A random number in the interval [0,1] is randomly generated. If p is greater than the random number, the candidate individual for the next iteration is determined as the candidate individual for the current iteration; otherwise, the candidate individual for the next iteration is determined as the candidate individual for the previous iteration.

[0040] In a possible implementation, the design data of the target product includes: design parameters and reliability indicators in the design process; the manufacturing data of the target product includes: process parameters and product characteristic data in the manufacturing process.

[0041] In a possible implementation, the method further includes:

[0042] Get the manufacturing process of the target product i The dimension value vector of the reliability-oriented key quality characteristics of the sample at time t, where i The dimension vector of the moment: X i =(x 1i ,x 2i ,...,x Hi ), where H is the number of reliability-oriented key quality characteristics of the target product, x hi is the dimension value of the hth reliability-oriented key quality characteristic, 1≤h≤H;

[0043] Calculate t i The MEWMA control chart vector Z of the sample at time i :

[0044] Z i =RX i +(IR)Z i-1

[0045] Where R is a diagonal matrix whose elements are the offset coefficients r;

[0046] Then t i The MEWMA statistic Q of the sample at time i for:

[0047]

[0048] When the statistic Q i When it is less than the upper control limit UCL, t i The size of the sample at time t is normal, otherwise, i The size value of the sample at the moment is an outlier; the upper control limit UCL is:

[0049]

[0050] Where L is the control limit parameter.

[0051] In a second aspect, an embodiment of the present application provides a device for identifying key quality characteristics of a product for reliability, including:

[0052] A first determining unit, configured to determine K quality characteristics of a target product based on design data and manufacturing data of the target product;

[0053] An initialization unit is used to initialize the positions of N individuals based on K quality characteristics of the target product, and use the initialized N individuals as the N individuals of the first iteration; wherein the position of each individual is a binary number consisting of K 0s or 1s, wherein 0 indicates that the quality characteristic is not a reliability-oriented key quality characteristic, and 1 indicates that the quality characteristic is a reliability-oriented key quality characteristic;

[0054] A processing unit, used for determining candidate individuals from the initialized N individuals as candidate individuals for the first iteration;

[0055] Iteration unit, used to repeatedly execute the following steps:

[0056] Update the positions of N individuals in the previous iteration to obtain the positions of N individuals in the current iteration;

[0057] Use the pre-trained classifier to process the position of each individual in the current iteration to obtain the error rate of each individual, and bring the error rate of each individual and the number of KQCs into the fitness function to obtain the fitness value of each individual; wherein the classifier is trained through the quality characteristic data and fault data of multiple samples of the target product;

[0058] The individual with the smallest fitness value is determined as the candidate individual for the current iteration;

[0059] Based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, the candidate individuals of the current iteration are updated;

[0060] Until the current number of iterations reaches the preset maximum number of iterations;

[0061] The second determining unit is used to determine the quality characteristics coded as 1 in the candidate individuals with the maximum number of iterations as the reliability-oriented key quality characteristics.

[0062] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of the embodiment of the present application when executing the computer program.

[0063] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions implement the method of the embodiment of the present application when executed by a processor.

[0064] The present application can identify the reliability-oriented key quality characteristics of a product and has the advantages of high recognition accuracy and fast convergence speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0066] Figure 1 A flowchart of a method for identifying key quality characteristics of a product for reliability provided in an embodiment of the present application;

[0067] Figure 2 A framework diagram for identifying reliability-oriented key quality characteristics of a shielding component provided in an embodiment of the present application;

[0068] Figure 3 A schematic diagram of an initial set of feature selection characteristics for a shielding component provided in an embodiment of the present application;

[0069] Figure 4 A control diagram of key quality characteristics of reliability for a shielding assembly provided in an embodiment of the present application;

[0070] Figure 5 A functional structure diagram of a reliability-oriented key quality characteristic identification device for a product provided in an embodiment of the present application;

[0071] Figure 6 A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0073] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0074] First, the technical terms involved in the embodiments of the present application are briefly introduced.

[0075] Built-in Reliability (BIR) refers to a reliability assurance method that is embedded in various reliability activities within the product life cycle and detects and corrects errors as early as possible in the upstream of the product life cycle. This method implements the product life cycle, systematically coordinates various roles within the product life cycle, closely integrates reliability work with the product design process, and implements product reliability work to reduce product defects and ultimately ensure product reliability. At present, BIR research either focuses only on the design stage or only on the product manufacturing process, lacks attention to the use stage, and does not grasp the connection between the various stages.

[0076] Key Quality Characteristic (KQC) refers to a collection of a small number of characteristic parameters that have the greatest impact on product quality. It is the most critical product quality characteristic. Its deviation will directly lead to a decline in product quality. If the deviation exceeds the specified range, it will directly lead to product failure and loss of its main function. As an important carrier of product quality, product KQC runs through all stages of the product life cycle, conforms to the built-in reliability method, and is the core of the built-in reliability method. Identifying and controlling product KQC is an important means to ensure product reliability. However, traditional KQC is often limited to the manufacturing process of the product, ignoring the large amount of quality and reliability data in the product design process and use process, and cannot be driven by reliability to achieve built-in reliability.

[0077] Reliability-oriented KQC (R-KQC) refers to a small number of key quality characteristics that will directly affect product reliability when deviations occur.

[0078] The SSA algorithm is a new meta-heuristic optimization algorithm, namely the salp swarm algorithm, which is a new swarm intelligence optimization algorithm that simulates the group behavior of salps to move and forage to achieve the optimization process. The SSA algorithm has the advantages of few adjustment parameters, simple and easy implementation, and strong overall search ability, but the convergence speed is slow and the local search ability is poor.

[0079] The processing flow of the SA-SSA algorithm is as follows:

[0080] 1) Initialize individuals. Initialize a salp group of size M×N.

[0081] 2) Calculate the initial fitness. Calculate the fitness values ​​of N salps.

[0082] 3) Select food. Sort the salps according to their fitness values, and set the position of the salp with the best fitness at the top as the current food position.

[0083] 4) Select leaders and followers. After the food location is selected, there are N-1 salps left in the group. According to the order of the salps group, the salps in the first half are regarded as leaders, and the rest are regarded as followers.

[0084] 5) Position update: First update the leader's position, then update the follower's position.

[0085] 6) Calculate fitness. Calculate the fitness of the updated population and perform simulated annealing to update the location of the food.

[0086] 7) Repeat 4)-6) until a certain number of iterations is reached or the fitness value reaches the termination threshold. When the termination condition is met, the current food position is output.

[0087] After introducing the technical terms involved in this application, the design concept of the embodiments of this application is briefly introduced next.

[0088] This application provides a reliability-oriented key quality characteristic identification for a product, aiming to propose a new connotation of R-KQC based on the principle of the BIR method; identify the R-KQC of the product based on the SA-SSA algorithm; finally, based on the MEWMA (exponentially weighted moving average) control chart, the monitoring and control of R-KQC is realized to achieve product reliability assurance. This application is suitable for product reliability assurance and improvement, and provides a feasible solution for manufacturing companies to ensure product reliability. The results show that this method can realize the built-in reliability of the manufacturing process and ensure the reliability of the product.

[0089] The advantages of this application are:

[0090] 1. This application can solve the limitation that activities centered on key quality characteristics in traditional product reliability assurance cannot better ensure product reliability. Aiming at product reliability, this application introduces the new concept of R-KQC to better achieve product reliability assurance.

[0091] 2. This application fully implements the system engineering concept, combines product reliability assurance work with all stages of the product life cycle, and achieves coverage of the entire product life cycle of the BIR method;

[0092] 3. The SA-SSA algorithm proposed in this application has high recognition accuracy and fast recognition convergence speed, overcomes the defect of falling into the local optimal solution, and can accurately identify R-KQC;

[0093] 4. The method of this application is scientific and reasonable, has good processability, and can be widely promoted in the industry.

[0094] After introducing the application scenarios and design concepts of the embodiments of the present application, the technical solutions provided by the embodiments of the present application are described below.

[0095] like Figure 1 As shown, the embodiment of the present application provides a reliability-oriented key quality characteristic identification of a product, including:

[0096] Step 101: Determine K quality characteristics of the target product based on the design data and manufacturing data of the target product;

[0097] For example, it is necessary to determine the identification object, determine the product or key component that needs characteristic identification, and collect relevant information of the product, including structural composition, design information, etc.

[0098] The design data of the target product include: design parameters and reliability indicators in the design process; the manufacturing data of the target product include: process parameters and product characteristic data in the manufacturing process.

[0099] Step 102: Based on the K quality characteristics of the target product, the positions of N individuals are initialized, and the initialized N individuals are used as the N individuals of the first iteration; wherein the position of each individual is a binary number consisting of K 0s or 1s, wherein 0 indicates that the quality characteristic is not a reliability-oriented key quality characteristic, and 1 indicates that the quality characteristic is a reliability-oriented key quality characteristic;

[0100] Step 103: Determine candidate individuals among the initialized N individuals as candidate individuals for the first iteration;

[0101] Step 104: Repeat the following steps: update the positions of N individuals of the previous iteration to obtain the positions of N individuals of the current iteration; use the pre-trained classifier to process the position of each individual of the current iteration to obtain the error rate of each individual, bring the error rate of each individual and the number of key quality characteristics for reliability into the fitness function to obtain the fitness value of each individual; wherein the classifier is trained by the quality characteristic data and fault data of multiple samples of the target product; determine the individual with the smallest fitness value as the candidate individual of the current iteration; based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, update the candidate individuals of the current iteration; until the current iteration reaches the preset maximum iteration number;

[0102] Step 105: Determine the quality characteristics coded as 1 in the candidate individuals with the maximum number of iterations as the reliability-oriented key quality characteristics.

[0103] The method of this embodiment is implemented based on the following assumptions: each mechanical device operates independently at the physical layer of the entire manufacturing system; the key quality characteristics corresponding to each process are considered independent of each other; and there is an absolutely reliable checkpoint after each processing is completed to check the quality status of the output product.

[0104] This application solves the limitations of traditional key quality characteristics, systematically grasps the relationship between product design, manufacturing, and use stages and product reliability, and extends product reliability assurance work to the design and use stages. It can be oriented towards user needs and product functional performance, and more accurately and comprehensively identify key characteristics for product reliability. It systematically guides manufacturers to achieve reliability built-in within the product life cycle and implement reliability work, thereby improving product quality and reliability.

[0105] In some embodiments, determining a candidate individual among the initialized N individuals as a candidate individual for a first iteration includes:

[0106] Using the classifier to process the initial position of each individual to obtain the initial error rate of each individual;

[0107] Substitute the initial error rate of each individual and the number of key quality characteristics for reliability into the fitness function to obtain the initial fitness value of each individual;

[0108] The individual with the smallest initial fitness value is determined as the candidate individual among the initialized N individuals and is used as the candidate individual for the first iteration.

[0109] In some embodiments, the positions of N individuals of the previous iteration are updated to obtain the positions of N individuals of the current iteration; including:

[0110] Determine the first individual among N individuals as the leader, and the remaining N-1 individuals as followers;

[0111] The position of the leader at the mth iteration for:

[0112]

[0113] Among them, P m is the candidate individual in the current iteration number m, c1 is the convergence factor, M is the preset maximum number of iterations; c2 and c3 are both random numbers in [0,1];

[0114] The position of the nth follower at the mth iteration for:

[0115]

[0116] Among them, 2≤n≤N; is the position of the nth individual in the m-1th iteration, is the position of the n-1th individual in the m-1th iteration.

[0117] In some embodiments, the method further comprises:

[0118] Obtain a training set of a target product, the training set comprising multiple groups of training samples, each group of training samples comprising measurement data of multiple target products and corresponding annotated fault data, wherein the measurement data comprises measurement values ​​of H reliability-oriented key quality characteristics; the annotated fault data is up to standard or not up to standard;

[0119] The measured data of multiple target products are processed by the classifier to obtain the predicted target data of multiple target products: meet the target or fail to meet the target;

[0120] Using the failure data of multiple target products, we can obtain the real compliance data of multiple target products.

[0121] Determine the error rates of multiple target products using predicted and actual target data;

[0122] The error rates of multiple target products and H values ​​are used to update the parameters of the classifier.

[0123] In some embodiments, determining the candidate individuals for the next iteration based on the candidate individuals for the previous iteration and the candidate individuals for the current iteration includes:

[0124] Determine whether the fitness value of the candidate individual of the current iteration number is less than the fitness value of the candidate individual of the previous iteration number. If yes, determine the candidate individual of the next iteration number as the candidate individual of the current iteration number.

[0125] Otherwise, calculate the probability value p:

[0126]

[0127] Where f(X) is the fitness value of candidate individual X at the current iteration, f(X0) is the fitness value of candidate individual X0 at the previous iteration; T is the annealing temperature;

[0128] A random number in the interval [0,1] is randomly generated. If p is greater than the random number, the candidate individual for the next iteration is determined as the candidate individual for the current iteration; otherwise, the candidate individual for the next iteration is determined as the candidate individual for the previous iteration.

[0129] In some embodiments, the method further comprises:

[0130] Get the manufacturing process of the target product i The dimension value vector of the reliability-oriented key quality characteristics of the sample at time t, where i The dimension vector of the moment: X i =(x 1i ,x 2i ,...,x Hi ), where H is the number of reliability-oriented key quality characteristics of the target product, x hi is the dimension value of the hth reliability-oriented key quality characteristic, 1≤h≤H;

[0131] Calculate t i The MEWMA control chart vector Z of the sample at time i :

[0132] Z i =RX i +(IR)Z i-1

[0133] Where R is a diagonal matrix whose elements are the offset coefficients r;

[0134] Then t i The MEWMA statistic Q of the sample at time i for:

[0135]

[0136] When the statistic Q i When it is less than the upper control limit UCL, t iThe size of the sample at time t is normal, otherwise, i The size value of the sample at the moment is an outlier; the upper control limit UCL is:

[0137]

[0138] Where L is the control limit parameter.

[0139] R-KQC is monitored and controlled through MEWMA control charts to promptly identify problems in the manufacturing process. When the control chart detects abnormal data, on-site engineers need to promptly check and adjust related processes and equipment to eliminate unstable factors in order to ensure product reliability.

[0140] The specific implementation process of this application is described below in conjunction with a specific application scenario.

[0141] The shielding component is a high-precision assembly product placed in the high-temperature area outside the reactor pressure vessel to absorb escaped neutrons, reduce the radiation level leaking from the core, and avoid unnecessary radioactive exposure to workers. The quality of shielding component products is directly related to the safe, stable and economical operation of the reactor. The shielding component production line has been put into operation, but quality problems such as low single rod qualification rate and low component qualification rate have gradually been exposed during the process, resulting in large-scale rework and waste of raw materials.

[0142] like Figure 2 As shown, R-KQC is identified by integrating the design parameters and reliability indicators in the shielding component design process, the process parameters and product characteristic data in the manufacturing process, and the fault data in the use process.

[0143] The shielding component design information is shown in Table 1:

[0144] Table 1 Quality characteristics of shielding component manufacturing process and its design parameters (partial)

[0145]

[0146] The shield component failure data is shown in Table 2:

[0147] Table 2 Statistics of shielding component failure and defect types and quantities

[0148]

[0149]

[0150] The design parameters are converted into actual product parameters through process parameters in the manufacturing process. The rationality of the determined process parameters is an important factor affecting product reliability. Therefore, on the basis of analyzing the design parameters, the process parameters closely related to these quality characteristics are further analyzed. By analyzing the process documents, various process parameters related to the characteristics can be analyzed. Taking the welding process that affects the characteristics of product porosity, slag inclusion, and airtightness as an example, some welding process parameters are listed below: tungsten electrode diameter 2.0mm, WL20, 30° cone angle, end Φ0.4mm platform; tungsten electrode end plug 0.1mm to 0.2mm at the weld; tungsten electrode and workpiece distance 0.6mm~08mm; welding current 87A; welding speed: 45° / s; arc starting time 0.1s; current rise time 0.1s; main current welding angle 540°; current decay time 9.9s; arc closing time 0.1s; shielding gas: argon, more than 99.999%; shielding gas (argon) flow rate (11~12)L / min, etc.

[0151] By collecting the data related to the above process parameters and product parameters, the following Figure 3 The initial set of characteristics is used to identify the R-KQC of the shielding component through the SA-SSA algorithm, and the R-KQC of the component is finally obtained, which are the length of the lower transition joint, the circular runout of the component, the diameter of the spiral groove of the pin, and the welding current.

[0152] Finally, the MEWMA control chart is used to monitor and control the four R-KQCs, and abnormal characteristics are discovered and handled in a timely manner. By collecting the above R-KQC manufacturing data, the MEWMA control chart is constructed based on Minitab software. Figure 4 It can be seen that the manufacturing process has no abnormal fluctuations, which realizes the built-in reliability of the product and well guarantees the reliability of the product.

[0153] The present application embodiment provides a device for identifying key quality characteristics of a product for reliability. Figure 5 As shown, the reliability-oriented key quality characteristic identification device 200 of a product provided in an embodiment of the present application at least includes:

[0154] A first determining unit 201 is used to determine K quality characteristics of a target product based on design data and manufacturing data of the target product;

[0155] An initialization unit 202 is used to initialize the positions of N individuals based on K quality characteristics of the target product, and use the initialized N individuals as the N individuals of the first iteration; wherein the position of each individual is a binary number consisting of K 0s or 1s, wherein 0 indicates that the quality characteristic is not a reliability-oriented key quality characteristic, and 1 indicates that the quality characteristic is a reliability-oriented key quality characteristic;

[0156] A processing unit 203 is used to determine a candidate individual from the initialized N individuals as a candidate individual for the first iteration;

[0157] The iteration unit 204 is used to repeatedly perform the following steps:

[0158] Update the positions of N individuals in the previous iteration to obtain the positions of N individuals in the current iteration;

[0159] Use the pre-trained classifier to process the position of each individual in the current iteration to obtain the error rate of each individual, and bring the error rate of each individual and the number of KQCs into the fitness function to obtain the fitness value of each individual; wherein the classifier is trained through the quality characteristic data and fault data of multiple samples of the target product;

[0160] The individual with the smallest fitness value is determined as the candidate individual for the current iteration;

[0161] Based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, the candidate individuals of the current iteration are updated;

[0162] Until the current number of iterations reaches the preset maximum number of iterations;

[0163] The second determining unit 205 is configured to determine the quality characteristics coded as 1 in the candidate individuals with the maximum number of iterations as the reliability-oriented key quality characteristics.

[0164] It should be noted that the principle of solving the technical problem by the reliability-oriented key quality characteristic identification device 200 of the product provided in the embodiment of the present application is similar to the method provided in the embodiment of the present application. Therefore, the implementation of the reliability-oriented key quality characteristic identification device 200 of the product provided in the embodiment of the present application can refer to the implementation of the method provided in the embodiment of the present application, and the repeated parts will not be repeated.

[0165] like Figure 6 As shown, the electronic device 300 provided in the embodiment of the present application includes at least: a processor 301, a memory 302, and a computer program stored in the memory 302 and executable on the processor 301. When the processor 301 executes the computer program, the reliability-oriented key quality characteristic identification method of the product provided in the embodiment of the present application is implemented.

[0166] The electronic device 300 provided in the embodiment of the present application may further include a bus 303 connecting different components (including the processor 301 and the memory 302). The bus 303 represents one or more of several types of bus structures, including a memory bus, a peripheral bus, a local bus, and the like.

[0167] The memory 302 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 3021 and / or a cache memory 3022 , and may further include a read-only memory (ROM) 3023 .

[0168] The memory 302 may also include a program tool 3024 having a set (at least one) of program modules 3025, the program modules 3025 including but not limited to: an operating subsystem, one or more applications, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0169] The electronic device 300 may also communicate with one or more external devices 304 (e.g., keyboards, remote controls, etc.), and may also communicate with one or more devices that enable a user to interact with the electronic device 300 (e.g., mobile phones, computers, etc.), and / or communicate with any device that enables the electronic device 300 to communicate with one or more other electronic devices 300 (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 305. Furthermore, the electronic device 300 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 306. Figure 6 As shown, the network adapter 306 communicates with other modules of the electronic device 300 via the bus 303. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 300, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, disk arrays (Redundant Arrays of Independent Disks, RAID) subsystems, tape drives, and data backup storage subsystems.

[0170] It should be noted that Figure 6 The electronic device 300 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0171] The embodiment of the present application also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the reliability-oriented key quality characteristic identification method of the product provided in the embodiment of the present application is implemented.

[0172] In addition, although the operations of the method of the present application are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0173] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0174] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying key quality characteristics of a product for reliability, characterized in that: include: Based on the design data and manufacturing data of the target product, determine K quality characteristics of the target product; Based on the K quality characteristics of the target product, the positions of N individuals are initialized, and the initialized N individuals are used as the N individuals of the first iteration; wherein the position of each individual is a binary number consisting of K 0s or 1s, wherein 0 indicates that the quality characteristic is not a key quality characteristic for reliability, and 1 indicates that the quality characteristic is a key quality characteristic for reliability; Determine the candidate individuals among the initialized N individuals as the candidate individuals for the first iteration; Repeat the following steps: Update the positions of N individuals in the previous iteration to obtain the positions of N individuals in the current iteration; The position of each individual of the current iteration is processed using a pre-trained classifier to obtain the error rate of each individual, and the error rate of each individual and the number of key quality characteristics for reliability are brought into the fitness function to obtain the fitness value of each individual; wherein the classifier is trained using the quality characteristic data and failure data of multiple samples of the target product; The individual with the smallest fitness value is determined as the candidate individual for the current iteration; Based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, the candidate individuals of the current iteration are updated; Until the current number of iterations reaches the preset maximum number of iterations; The quality characteristics coded as 1 among the candidate individuals with the maximum number of iterations are determined as the reliability-oriented key quality characteristics.

2. The method for identifying key quality characteristics of a product for reliability according to claim 1, characterized in that: Determine the candidate individuals among the initialized N individuals as the candidate individuals for the first iteration, including: Using the classifier to process the initial position of each individual to obtain the initial error rate of each individual; Substitute the initial error rate of each individual and the number of key quality characteristics for reliability into the fitness function to obtain the initial fitness value of each individual; The individual with the smallest initial fitness value is determined as the candidate individual among the initialized N individuals and is used as the candidate individual for the first iteration.

3. The method for identifying key quality characteristics of a product for reliability according to claim 1, characterized in that: Update the positions of N individuals in the previous iteration to obtain the positions of N individuals in the current iteration; including: Determine the first individual among N individuals as the leader, and the remaining N-1 individuals as followers; The position of the leader at the mth iteration for: Among them, P m is the candidate individual in the current iteration number m, c1 is the convergence factor, M is the preset maximum number of iterations; c2 and c3 are both random numbers in [0,1]; m≥2; The position of the nth follower at the mth iteration for: Among them, 2≤n≤N; is the position of the nth individual in the m-1th iteration, is the position of the n-1th individual in the m-1th iteration.

4. The method for identifying key quality characteristics of a product for reliability according to claim 1, characterized in that: The method further comprises: Obtain a training set of a target product, wherein the training set includes multiple groups of training samples, each group of training samples includes measurement data of multiple target products and corresponding annotated fault data, wherein the measurement data includes measurement values ​​of H reliability-oriented key quality characteristics; the annotated fault data is up to standard or not up to standard; The measured data of multiple target products are processed by the classifier to obtain the predicted target data of multiple target products: meet the target or fail to meet the target; Using the failure data of multiple target products, we can obtain the real compliance data of multiple target products. Determine the error rates of multiple target products using predicted and actual target data; The error rates of multiple target products and H values ​​are used to update the parameters of the classifier.

5. The method for identifying key quality characteristics of a product for reliability according to claim 1, characterized in that: Based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, the candidate individuals of the next iteration are determined, including: Determine whether the fitness value of the candidate individual of the current iteration number is less than the fitness value of the candidate individual of the previous iteration number. If yes, determine the candidate individual of the next iteration number as the candidate individual of the current iteration number. Otherwise, calculate the probability value p: Where f(X) is the fitness value of candidate individual X at the current iteration, f(X0) is the fitness value of candidate individual X0 at the previous iteration; T is the annealing temperature; A random number in the interval [0,1] is randomly generated. If p is greater than the random number, the candidate individual for the next iteration is determined as the candidate individual for the current iteration; otherwise, the candidate individual for the next iteration is determined as the candidate individual for the previous iteration.

6. The method for identifying key quality characteristics of a product for reliability according to claim 1, characterized in that: The design data of the target product includes: design parameters and reliability indicators in the design process; the manufacturing data of the target product includes: process parameters and product characteristic data in the manufacturing process.

7. The method for identifying key quality characteristics of a product for reliability according to claim 1, characterized in that: The method further comprises: Get the manufacturing process of the target product i The dimension value vector of the reliability-oriented key quality characteristics of the sample at time t, where i The dimension vector of the moment: X i =(x 1i ,x 2i ,...,x Hi ), where H is the number of reliability-oriented key quality characteristics of the target product, x hi is the dimension value of the hth reliability-oriented key quality characteristic, 1≤h≤H; Calculate t i The MEWMA control chart vector Z of the sample at time i : Z i =RX i +(I-R)Z i-1 Where R is a diagonal matrix whose elements are the offset coefficients r; Then t i The MEWMA statistic Q of the sample at time i for: When the statistic Q i When it is less than the upper control limit UCL, t i The size of the sample at time t is normal, otherwise, i The size value of the sample at the moment is an outlier; the upper control limit UCL is: Where L is the control limit parameter.

8. A device for identifying key quality characteristics of a product for reliability, characterized in that: include: A first determining unit, configured to determine K quality characteristics of a target product based on design data and manufacturing data of the target product; An initialization unit is used to initialize the positions of N individuals based on K quality characteristics of the target product, and use the initialized N individuals as the N individuals of the first iteration; wherein the position of each individual is a binary number consisting of K 0s or 1s, wherein 0 indicates that the quality characteristic is not a reliability-oriented key quality characteristic, and 1 indicates that the quality characteristic is a reliability-oriented key quality characteristic; A processing unit, used for determining candidate individuals from the initialized N individuals as candidate individuals for the first iteration; Iteration unit, used to repeatedly execute the following steps: Update the positions of N individuals in the previous iteration to obtain the positions of N individuals in the current iteration; Use the pre-trained classifier to process the position of each individual in the current iteration to obtain the error rate of each individual, and bring the error rate of each individual and the number of KQCs into the fitness function to obtain the fitness value of each individual; wherein the classifier is trained through the quality characteristic data and fault data of multiple samples of the target product; The individual with the smallest fitness value is determined as the candidate individual for the current iteration; Based on the candidate individuals of the previous iteration and the candidate individuals of the current iteration, the candidate individuals of the current iteration are updated; Until the current number of iterations reaches the preset maximum number of iterations; The second determining unit is used to determine the quality characteristics coded as 1 in the candidate individuals with the maximum number of iterations as the reliability-oriented key quality characteristics.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images