Assembly quality evaluation method and system for product, storage medium and equipment

Abstract

The embodiment of the invention discloses an assembly quality evaluation method for a product, and the method comprises the steps: monitoring the quality fluctuation of an assembly process according to a quality fluctuation model, and obtaining the quality data of the assembly process; de-noising the quality data based on Kalman filtering to obtain characteristic parameter data; determining a joint quality control chart according to the characteristic parameter data; determining a capability index value of the assembly process according to the joint quality control chart; and evaluating the assembly quality of the product according to a matching relationship between the capability index value of the assembly process and a preset range. According to the method, the evaluation effect of the product assembly quality is comprehensively improved through monitoring of the quality fluctuation of the assembly process, efficient data processing, a visual quality control view and a quantitative quality evaluation standard, the product quality is ensured by an enterprise, and powerful support is provided for quality control and continuous improvement of the enterprise.

Description

Technical Field

[0001] The present invention relates to the technical field of production management, and in particular to a method, system, storage medium and device for evaluating the assembly quality of a product. Background Art

[0002] With the continuous development of manufacturing as the foundation of the national economy, my country's manufacturing industry faces multiple challenges such as high manufacturing costs, unstable product quality, low production efficiency, irregular workshop management and insufficient corporate innovation capabilities. Especially in high-end manufacturing fields such as aircraft, weapon systems, and complex electromechanical products, the production and manufacturing process involves a highly complex knowledge system and strict quality requirements. Improving product manufacturing capabilities and quality levels has become the key to the transformation and upgrading of the manufacturing industry.

[0003] Quality problems arising from the manufacturing process account for the majority of product quality problems, and the quality control of the assembly link, as the core of production and manufacturing, directly affects the quality of the final product. However, the traditional assembly production control model relies on manual judgment and post-inspection, and has problems such as data lag, low efficiency, and untimely problem discovery. Under the trend of integration of informatization and industrialization, although the digital level of assembly processes has been improved, most companies still lack scientific and effective quality control methods, and have failed to make full use of advanced technologies such as artificial intelligence and data mining to realize real-time analysis and online control of quality data.

[0004] At present, enterprises mostly rely on experience to manage quality and lack a systematic analysis method from qualitative to quantitative. As a result, process-level quality problems are difficult to be discovered and handled in a timely manner. These problems are often exposed only during the final assembly and testing of the product, causing significant economic losses. Therefore, there is an urgent need to implement scientific quality monitoring and process capability analysis of the assembly process to achieve real-time monitoring of assembly quality and abnormal warning, ensure that product assembly quality meets high standards, and enhance the overall competitiveness of the manufacturing industry. Summary of the invention

[0005] Based on this, it is necessary to propose a method for product assembly quality assessment to address the above issues.

[0006] A method for evaluating the assembly quality of a product, the method comprising the following steps:

[0007] Monitoring the quality fluctuation of the assembly process according to the quality fluctuation model and obtaining the quality data of the assembly process;

[0008] Performing denoising processing on the quality data based on Kalman filtering to obtain characteristic parameter data;

[0009] Determining a joint quality control chart based on the characteristic parameter data;

[0010] Determining the capability index value of the assembly process according to the joint quality control chart;

[0011] The assembly quality of the product is evaluated based on the matching relationship between the capability index value of the assembly process and a preset range.

[0012] In the above scheme, the quality fluctuation of the assembly process is monitored according to the quality fluctuation model, and the quality data of the assembly process is obtained, which specifically includes:

[0013] Modeling the mass fluctuation based on the relationship between the noise information and the mass fluctuation measurement data and the fluctuation state data;

[0014] A state space model of mass fluctuation is obtained, wherein the state space model is composed of a state equation, an observation equation and initial information, which are respectively expressed as:

[0015] μ t =μ t-1 +w t ,w t ~N[0,W t ]

[0016] y t =μ t +v t ,v t ~N[0,V t ]

[0017] X0|D0~N[μ0,C0]

[0018] The quality fluctuation of the assembly process is monitored according to the state equation, the observation equation and the initial information, and the quality data of the assembly process is obtained.

[0019] In the above solution, the denoising process is performed on the quality data based on Kalman filtering to obtain characteristic parameter data, which specifically includes:

[0020] The state equation x(k) and observation equation z(k) of the time discrete system are determined according to the linear stochastic difference equation:

[0021] x(k)=Ax(k-1)+Bu(k)+w(k)

[0022] z(k)=Hx(k)+v(k)

[0023] Among them, A is the state transfer matrix, B is the input gain matrix, H is the measurement matrix, u(k) is the process input at time k, w(k) is the system excitation noise information, and v(k) is the measured noise information;

[0024] Determining an optimal estimate of the state equation x(k) based on an observed value of the observation equation z(k);

[0025] The initial state variable X0 of the state equation x(k), the covariance P0 of X0, the process excitation noise covariance Q t , measurement noise covariance R t Pre-estimate and determine the characteristic parameter data.

[0026] In the above scheme, the initial state variables X0 and X0 of the state equation x(k) have a covariance P0, and the process excitation noise covariance Q t , measurement noise covariance R t Pre-estimating and determining the characteristic parameter data specifically includes:

[0027] The estimated value of the initial state variable X0 is determined according to the following formula: Covariance P0 of X0:

[0028]

[0029] P0=E[)x(0)-u0)(x(0)-u0) T ] = Var(x(0));

[0030] The state equation x(k) is observed n times independently to obtain the time data sequence {x(0), x(1), ... x(n)} and calculate the average value Get the corresponding standard deviation S(x);

[0031] Get Type A standard uncertainty:

[0032]

[0033] The process excitation noise covariance Q is calculated according to the discount factor δ t Estimate the estimated value Q of x(k) to obtain the prior estimation accuracy of x(k):

[0034] p(k-1|k-1)=δp(k|k-1),0<δ<1

[0035] The a priori estimation accuracy of x(k) is processed and solved to obtain the process excitation noise covariance Q t The estimated value of Q is:

[0036] Q=(1-δ) 2 Rδ -1 .

[0037] In the above solution, determining the joint quality control chart according to the characteristic parameter data specifically includes:

[0038] Determine the joint quality control chart based on the Shewhart mean control chart and the exponentially weighted moving average control chart:

[0039] When the assembly process of the product is under control, μ0 = μ; assuming X1, X2, ... are independent random variable sequences, the exponentially weighted moving average statistic Z t for:

[0040] Z t =λX t +(1-λ)Z t-1 ,0<λ≤1,t=0,1,2,3…

[0041] Among them, λ is the smoothing coefficient;

[0042] The recursive formula and the cumulative sum of data items obtained through iteration are:

[0043]

[0044] Among them, Z0 is the starting value;

[0045] Take the mean value of the data as the starting value Z0=μ0 and obtain the sample statistic Z t The expectation and variance of are:

[0046]

[0047] Where n is the number of samples.

[0048] In the above solution, determining the joint quality control chart according to the characteristic parameter data also includes:

[0049] The control limits for obtaining the joint quality control chart are:

[0050]

[0051] Where L is the control limit of the exponentially weighted moving average control chart, k is the parameter for determining the control limit, UCL is the upper limit of the joint quality control chart, and LCL is the lower limit of the joint quality control chart.

[0052] In the above scheme, the determination of the capability index value of the assembly process according to the joint quality control chart specifically includes:

[0053] Determine the capability index of the process

[0054] The capability index C of the process p Make modifications to obtain the corrected process capability index C pk :

[0055]

[0056] in, is the mean of the time series mean samples of the joint quality control chart, is the standard deviation of the mean sample, USL is the upper limit of the joint quality control chart, and LSL is the lower limit of the joint quality control chart.

[0057] The present application also proposes an assembly quality assessment system for a product, the system comprising: a data acquisition unit, a data processing unit, a capability index value acquisition unit and an assessment unit;

[0058] The data acquisition unit is used to monitor the quality fluctuation of the assembly process according to the quality fluctuation model and acquire the quality data of the assembly process;

[0059] The data processing unit is used to perform denoising on the quality data based on Kalman filtering to obtain characteristic parameter data;

[0060] The capability index value acquisition unit is used to determine a joint quality control chart according to the characteristic parameter data; and determine a capability index value of an assembly process according to the joint quality control chart;

[0061] The evaluation unit is used to evaluate the assembly quality of the product according to the matching relationship between the capability index value of the assembly process and a preset range.

[0062] The present application also proposes a readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the following steps:

[0063] Monitoring the quality fluctuation of the assembly process according to the quality fluctuation model and obtaining the quality data of the assembly process;

[0064] Performing denoising processing on the quality data based on Kalman filtering to obtain characteristic parameter data;

[0065] Determining a joint quality control chart based on the characteristic parameter data;

[0066] Determining the capability index value of the assembly process according to the joint quality control chart;

[0067] The assembly quality of the product is evaluated based on the matching relationship between the capability index value of the assembly process and a preset range.

[0068] The present application also proposes a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the following steps of the computer program:

[0069] Monitoring the quality fluctuation of the assembly process according to the quality fluctuation model and obtaining the quality data of the assembly process;

[0070] Performing denoising processing on the quality data based on Kalman filtering to obtain characteristic parameter data;

[0071] Determining a joint quality control chart based on the characteristic parameter data;

[0072] Determining the capability index value of the assembly process according to the joint quality control chart;

[0073] The assembly quality of the product is evaluated based on the matching relationship between the capability index value of the assembly process and a preset range.

[0074] The embodiment of the present invention has the following beneficial effects: first, the quality fluctuation of the assembly process is monitored according to the quality fluctuation model, and the quality data of the assembly process is obtained; the quality data is denoised based on the Kalman filter to obtain characteristic parameter data; a joint quality control chart is determined according to the characteristic parameter data; the capability index value of the assembly process is determined according to the joint quality control chart; and the assembly quality of the product is evaluated according to the matching relationship between the capability index value of the assembly process and the preset range. This method comprehensively improves the evaluation effect of product assembly quality through efficient data processing, intuitive quality control views, and quantitative quality evaluation standards, which not only helps enterprises ensure product quality, but also provides strong support for enterprise quality control and continuous improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0076] in:

[0077] Figure 1 A schematic flow chart of a method for evaluating product assembly quality in one embodiment;

[0078] Figure 2 Schematic diagram of the basic form of the Shewhart mean control chart in one embodiment. DETAILED DESCRIPTION

[0079] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0080] In the following description, a large number of specific details are given in order to provide a more thorough understanding of the present invention; however, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details; in other examples, in order to avoid confusion with the present invention, some technical features known in the art are not described, and it should be understood that the present invention can be implemented in different forms and should not be interpreted as limited to the embodiments presented here; on the contrary, providing these embodiments will make the disclosure thorough and complete and fully convey the scope of the present invention to those skilled in the art.

[0081] The purpose of the terms used herein is only to describe specific embodiments and is not intended to be limiting of the present invention. When used herein, the singular forms "a", "an" and "said / the" are also intended to include the plural forms, unless the context clearly indicates otherwise, and it should also be understood that the terms "consisting of" and / or "comprising", when used in this specification, determine the presence of the features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0082] In order to thoroughly understand the present invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by the present invention; optional embodiments of the present invention are described in detail as follows, but in addition to these detailed descriptions, the present invention may also have other implementation methods.

[0083] like Figure 1 As shown, in one embodiment, a method for evaluating the assembly quality of a product is provided, and the method for evaluating the assembly quality of a product includes steps S101 to S106, which are described in detail as follows:

[0084] S101. Monitor the quality fluctuation of the assembly process according to the quality fluctuation model, and obtain the quality data of the assembly process;

[0085] The use of the quality fluctuation model can systematically consider various factors that affect assembly quality, such as material properties, process parameters, environmental conditions, etc., thereby ensuring the comprehensiveness and accuracy of the data and helping to improve the reliability of the evaluation results.

[0086] In some embodiments, the quality fluctuation of the assembly process is monitored according to the quality fluctuation model, and the quality data of the assembly process is obtained, which specifically includes:

[0087] Modeling the mass fluctuation based on the relationship between the noise information and the mass fluctuation measurement data and the fluctuation state data;

[0088] The state space model of mass fluctuation is obtained. The state space model consists of state equation, observation equation and initial information, which are expressed as:

[0089] μ t =μ t-1 +w t ,w t ~N[0,W t ]

[0090] y t =μ t +v t ,v t ~N[0,V t ]

[0091] X0|D0~N[μ0,C0]

[0092] Monitor the quality fluctuation of the assembly process based on the state equation, observation equation and initial information, and obtain the quality data of the assembly process

[0093] Specifically, by modeling quality fluctuations, the relationship between noise information and quality fluctuation measurement data and fluctuation state data is established, so as to achieve effective monitoring of quality fluctuations and keep them within an appropriate range. The quality fluctuation model of the manufacturing process is established using the Bayesian state space method. The state space model of the assembly offset is composed of state equations, observation equations and initial information.

[0094] For all the above moments, v at any moment t t and w t , and v at other times s and w s are independent of each other, assuming that the measurement noise v t and state noise w t It obeys Gaussian distribution and Laplace distribution. In this quality fluctuation model, the observation equation reflects the observation status of the quality characteristics of the assembly process at time t; the state equation reflects the assembly quality fluctuation status of the process or manufacturing system at time t.

[0095] S102, performing denoising processing on the quality data based on Kalman filtering to obtain characteristic parameter data;

[0096] Kalman filtering is used to process quality data. Kalman filtering is an efficient recursive filter that can estimate the state of a dynamic system in the presence of measurement noise and process noise. This method can effectively remove noise from the data and extract characteristic parameter data that is valuable for assembly quality assessment, which helps to more accurately reflect the actual state of the assembly process and provide strong support for the subsequent establishment and evaluation of quality control charts.

[0097] In some embodiments, the quality data is processed based on Kalman filtering to obtain characteristic parameter data, specifically including:

[0098] The state equation x(k) and observation equation z(k) of the time discrete system are determined according to the linear stochastic difference equation:

[0099] x(k)=Ax(k-1)+Bu(k)+w(k)

[0100] z(k)=Hx(k)+v(k)

[0101] Among them, A is the state transfer matrix, B is the input gain matrix, H is the measurement matrix, u(k) is the process input at time k, w(k) is the system excitation noise information, and v(k) is the measured noise information;

[0102] Determine the optimal estimate of the state equation x(k) based on the observed value of the observation equation z(k);

[0103] For the initial state variable X0 of the state equation x(k), the covariance P0 of X0, and the covariance Q of the process excitation noise t , measurement noise covariance R t Make an estimate and determine the characteristic parameter data.

[0104] The Kalman filter algorithm is an optimized recursive data processing state estimation algorithm proposed by REKalman, which reduces the noise of heterogeneous disturbances from a series of observation data according to a certain statistical point of view, and then uses a series of gray data information feedback control recursive form to obtain the optimal estimated state of a dynamic complex system.

[0105] For the Kalman problem, the noise information of the assembly measurement data basically obeys the Gaussian distribution. Therefore, the Kalman filtering method can effectively remove most of the noise information, restore the true value well, and provide an effective data source for subsequent control chart analysis. Therefore, the complex product assembly workshop is a typical discrete manufacturing system, and the state equation x(k) and observation equation z(k) of the above time discrete system can be described by linear random difference equations.

[0106] Preferably, under the interference of noise w(k) and v(k), observation is performed, and based on the above observation values: z(1), z(2), z(3) ... z(k), after obtaining the optimal estimate of x(k) under the minimum mean square error, it is assumed that w(k) and v(k) are uncorrelated white noises with zero mean and variance matrices Q and R respectively. According to the established quality fluctuation model, the Kalman filter algorithm is introduced for data preprocessing, in order to remove the noise and correlation of the data, and obtain the optimal estimate of the assembly quality X t sequence.

[0107] In some embodiments, before implementing the Kalman filter on the time series process quality characteristic data, the four coefficients in the algorithm need to be estimated in advance, namely, the initial state variable X0, the covariance P0 of X0, the process excitation noise covariance Q t , measurement noise covariance R t ; Among them, Q t and R t The noise covariance R may change with iterative calculations. t It can generally be observed that the process excitation noise Q t The value of is difficult to determine. In the production of complex products, common process equipment is often used. In the short term, the entire process system is relatively stable. Therefore, it is assumed that the mean square error is a constant, that is, Q t =Q, R t =R, and define the signal-to-noise ratio β = Q / R.

[0108] In some embodiments, the initial state variable X0 of the state equation x(k), the covariance P0 of X0, the process excitation noise covariance Q t , measurement noise covariance R t Make an estimate and determine the characteristic parameter data, including:

[0109] (1) Determine the estimated value of the initial state variable X0 according to the following formula: Covariance P0 of X0:

[0110]

[0111] P0=E[(x(0)-u0)(x(0)-u0) T ] = Var(x(0));

[0112] Specifically, the initial state variables have the smallest impact, and only affect the filtering effect of the first few values. The initial value will converge quickly if the filtering can converge. The value of the initial variance matrix has little effect on the filtering effect. As long as the value is not zero, it can converge quickly. You can take any non-zero matrix. When the value is close to the mean, the overall filtering effect is better, and the corresponding state estimation value is Covariance P0.

[0113] In fact, the smaller the value of the measurement noise covariance R is, the faster the convergence is, but the filtering effect is not necessarily good. The noise covariance needs to be measured before filtering.

[0114] (2) This application uses the standard "Evaluation and Expression of Measurement Uncertainty" to estimate R offline. The specific evaluation steps are as follows:

[0115] Make n independent observations of the state equation x(k), obtain the time data sequence {x(0), x(1), …x(n)} and calculate the average value Get the corresponding standard deviation S(x);

[0116] Get Type A standard uncertainty:

[0117]

[0118] The parameter R is estimated by comparing the Class A standard uncertainty with the standard "Evaluation and Expression of Uncertainty in Measurement".

[0119] The process excitation noise covariance Q is calculated according to the discount factor δ t Estimate the estimated value Q of x(k) to obtain the prior estimation accuracy of x(k):

[0120] p(k-1|k-1)=δp(k|k-1),0<δ<1

[0121] Preferably, when R is determined, a discount coefficient δ is introduced to estimate Q offline. Assuming that the a priori estimation accuracy of x(k) is a compromise of the a posteriori accuracy, the recursive formula is:

[0122] p(k|k-1)=p(k-1|k-1)+Q(k)

[0123] The prior estimation accuracy of x(k) is processed and solved to obtain the process excitation noise covariance Q t The estimated value of Q is:

[0124] Q=(1-δ) 2 Rδ -1 .

[0125] It can be seen that the value of δ will affect the prediction results. The smaller the discount coefficient δ is, the less dependent the model is on prior information. Since the assembly process fluctuates slowly in the short term and has a small value, it has a greater correlation with prior information. Here, δ is set to 0.3-0.5.

[0126] S103, determining a joint quality control chart according to the characteristic parameter data;

[0127] The combined quality control chart is determined based on the characteristic parameter data. This method can comprehensively consider multiple quality characteristics to form a comprehensive quality control view. The combined quality control chart can more comprehensively reflect the overall quality status of the assembly process than a single quality characteristic control chart, making it easier to discover potential quality problems in a timely manner and improve the efficiency of quality control.

[0128] Specifically, joint quality control chart To combine the Shewhart mean control chart and the EWMA exponentially weighted moving average control chart Combined quality control charts are used to analyze assembly process capabilities and improve the accuracy of process judgment of assembly characteristic values; Shewhart The mean control chart only focuses on the statistics related to the current observation value, which is suitable for analyzing the situation where the quality data fluctuates greatly. The EWMA control chart accumulates the difference between the observation value and the standard value, which is sensitive to small deviations in the process. The control chart can make up for the detection delay problem of the EWMA control chart.

[0129] For a certain assembly process, a time series data set of assembly error of a certain quality characteristic is obtained through continuous collection, and it is imported into the Kalman filter model to remove noise. After the assembly quality time series data is processed by the Kalman filter model, the optimal estimation sequence that is closer to the true value of the actual working condition can be obtained. And provide data support for the joint quality control chart model.

[0130] like Figure 2 As shown, the following introduces the Shewhart mean control chart and the EWMA exponentially weighted moving average control chart:

[0131] a) Shewhart mean control chart

[0132] When the quality characteristic value is a continuous value, the most common typical distribution is the normal distribution, that is, X~N(μ,σ 2 ), then no matter what the mean μ and standard deviation σ take, according to the normal distribution probability theory formula:

[0133] 1-P{μ-3σ <X<μ+3σ}=0.27%

[0134] That is, the probability that the product quality characteristic value falls within (μ±3σ) is 99.73%, that is, the probability of falling outside (μ±3σ) is 100%-99.73%=0.27%. According to the principle of small probability, the probability of a small probability event is extremely small, or basically impossible to occur. If it occurs, it means that the quality characteristic value may be abnormal.

[0135] Specifically, the Shewhart mean control chart can be obtained by transforming the normal distribution curve within the actual distribution range of the product quality characteristic value. The typical Shewhart mean control chart has a center line (control line, UCL), an upper control limit (Upper Control Limit, UCL) and a lower control limit (Lower Control Limit, LCL), and samples of variables are collected in a time series; among them, the mean sample is the average of the sampling values ​​obtained by multiple sampling of a certain number of sampling subgroups in accordance with the time series, that is, the average of the mean of multiple batches of samples is used as the estimated value of μ. According to the central limit theorem, the mean sample sequence approximately obeys the normal distribution. Therefore, the quality characteristic value falls within the upper and lower UCL to LCL of the overall mean of the controlled variable with a probability of 99.73%, and the quality characteristic value within the control limit is considered to be a stable process.

[0136] b) EWMA Exponential Weighted Moving Average Control Chart

[0137] The exponential weighted moving average (EWMA) control chart was proposed by Roberts in 1959. The EWMA control chart amplifies small fluctuations in the process by giving certain weights to past measurements, and the weight ratio decreases exponentially with time. The statistics of the EWMA control chart comprehensively consider the impact of current sample data and historical sample data.

[0138] In some embodiments, let X be a random variable time series, assuming that they are independent of each other and obey the same normal distribution X~N(μ,σ 2 ), μ0 is the target value of the process mean, when the process is in a controlled state, μ0=μ, and the joint quality control chart is determined according to the characteristic parameter data, specifically including:

[0139] Determine the joint quality control chart based on the Shewhart mean control chart and the exponentially weighted moving average control chart:

[0140] When the assembly process of the product is under control, μ0 = μ; assuming X1, X2, ... are independent random variable sequences, the exponentially weighted moving average (EWMA) statistic Z t for:

[0141] Z t =λX t +(1-λ)Z t-1 ,0<λ≤1,t=0,1,2,3…

[0142] Among them, λ is the smoothing coefficient;

[0143] Specifically, the sensitivity of the EWMA control chart to small variations is mainly affected by the parameter r. When λ=1, the EWMA control chart is equivalent to the Shewhart control chart; when λ=0, the EWMA control chart is a cumulative sum control chart. The closer r is to 1, the smaller the impact of samples before time t on the EWMA control chart. The closer λ is to 0, the greater the impact of samples before time t on the EWMA control chart. λ is an important parameter that affects the performance of the EWMA control chart.

[0144] The recursive formula and the cumulative sum of data items obtained through iteration are:

[0145]

[0146] Among them, Z0 is the starting value;

[0147] Take the mean value of the data as the starting value Z0=μ0 and obtain the sample statistic Z t The expectation and variance of are:

[0148]

[0149] Where n is the number of samples. Generally, when t>70, it can be ignored.

[0150] In some embodiments, determining the joint quality control diagram according to the characteristic parameter data further comprises:

[0151] The control limits for obtaining the joint quality control chart are:

[0152]

[0153] Where L is the control limit of the exponentially weighted moving average control chart, k is the parameter for determining the control limit, UCL is the upper limit of the joint quality control chart, and LCL is the lower limit of the joint quality control chart.

[0154] It can be seen that the performance of the EWMA control chart is related to the smoothing coefficient r, the control limit parameter k and the process mean shift σ. Different parameters are selected according to the different process mean fluctuations. At present, the more common one is the parameter selection strategy within the 5σ shift proposed by Lucas and Saccucci. Before establishing the EWMA control chart, appropriate parameters should be selected according to the expected standardized shift.

[0155] S104. Determine the capability index value of the assembly process according to the joint quality control chart;

[0156] The capability index value of the assembly process is determined by combining the quality control chart. This value can quantitatively reflect the degree to which the assembly process meets the quality requirements. The capability index value provides a quantitative evaluation standard, making the evaluation of assembly quality more objective and accurate. The main purpose of the joint control chart is to monitor the assembly process. Therefore, the standard for abnormal judgment is the upper and lower limits of the control chart. When the data points collected in the assembly process fall outside the control limits, it is considered that the process is abnormal.

[0157] Preferably, after performing SPC quality control analysis on the assembly process, it is possible to determine from a statistical perspective whether the process is stable. When the process is in a stable state, it is necessary to further determine the capability of the assembly process in the stable state. When the process is stable and the process capability reaches the required capability range, it indicates that the assembly process is good. If the process capability of the process is not within the allowable range, measures must be taken to adjust the process so that the process is in a stable state in real time.

[0158] In some embodiments, determining the capability index value of the assembly process according to the joint quality control chart specifically includes:

[0159] Determine the capability index of the process

[0160] Specifically, assume that the quality index of the assembly process currently being tested obeys a normal distribution with a mean of μ and a standard deviation of σ, and the standard upper limit of the process quality characteristic value is S U , the standard lower limit is S L , then the tolerance range T = S U -S L When the distribution center coincides with the tolerance center, that is, μ = T, the capability index C of the process is determined. p .

[0161] Preferably, in actual production, the distribution center and the tolerance center will often be offset. In this case, μ≠T, and the modified process capability index C pk Defined as:

[0162]

[0163] k represents the degree of deviation of the distribution center.

[0164] Capability index C for the process p Make modifications to obtain the corrected process capability index C pk :

[0165]

[0166] in, is the mean of the time series mean samples of the joint quality control chart, is the standard deviation of the mean sample, UCL is the upper limit of the joint quality control chart, LCL is the lower limit of the joint quality control chart, and the calculated assembly process capability index can represent different process capabilities in different ranges.

[0167] S105. Evaluate the assembly quality of the product according to the matching relationship between the capability index value of the assembly process and the preset range.

[0168] By comparing the matching relationship with the preset range, it can be determined whether the assembly process meets the standard, providing a basis for subsequent improvements. The initial stage of the assembly process is generally unstable. The upper and lower control limits of the control chart in the assembly process are determined by the data of the initial process. When continuously analyzing and adjusting various influencing factors and continuously improving the assembly process, it is necessary to continuously adjust the control limits. When the assembly process in the workshop is in a stable state after multiple adjustments, the calculation Process capability index C of joint control chart p , and check the process capability corresponding to the process capability index. When the process is stable and the process capability C p It meets the requirements of assembly. At this time, the analysis control chart uses the upper and lower limits to determine the control chart and transfer it to the control chart for control. In the control chart stage, the assembly process data is no longer used for control chart analysis or control limit adjustment, but is only used as process status monitoring of output points. The assembly output points are marked and displayed to determine whether they are within the stable range.

[0169] Specifically, when C p When >1.67, it indicates that the process capability far exceeds the demand, and it is recommended to consider relaxing management to reduce costs. At this time, the assembly quality assessment is A+ grade; when 1.0 <C p When ≤1.33, the process capability meets the requirements and can maintain and allow small fluctuations. At this time, the assembly quality is evaluated as grade A. When 0.67 <C p When ≤1.00, the process capability can meet the basic requirements, but it is recommended to be strictly controlled. At this time, the assembly quality assessment is B level; when 0.67 <C p When ≤1.00, it indicates that the process capability is insufficient and measures need to be taken to improve it. In this case, the assembly quality assessment is C grade. p When <0.67, the process capability is seriously insufficient and the work should be stopped immediately to investigate the cause. At this time, the assembly quality assessment is D level.

[0170] The present application also proposes an assembly quality assessment system for a product, the system comprising: a data acquisition unit, a data processing unit, a capability index value acquisition unit and an assessment unit;

[0171] A data acquisition unit, used to monitor the quality fluctuation of the assembly process according to the quality fluctuation model and acquire the quality data of the assembly process;

[0172] A data processing unit, used for processing the quality data based on Kalman filtering to obtain characteristic parameter data;

[0173] A capability index value acquisition unit is used to determine a joint quality control chart according to the characteristic parameter data; and determine a capability index value of an assembly process according to the joint quality control chart;

[0174] The evaluation unit is used to evaluate the assembly quality of the product according to the matching relationship between the capability index value of the assembly process and the preset range.

[0175] The present application also proposes a readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the following steps:

[0176] According to the quality fluctuation model, the quality fluctuation of the assembly process is monitored and the quality data of the assembly process is obtained;

[0177] De-noising the quality data based on Kalman filtering to obtain characteristic parameter data;

[0178] Determine a joint quality control chart based on characteristic parameter data;

[0179] Determine the capability index value of the assembly process based on the joint quality control chart;

[0180] The assembly quality of the product is evaluated based on the matching relationship between the capability index value of the assembly process and the preset range.

[0181] The present application also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor to perform the following steps:

[0182] According to the quality fluctuation model, the quality fluctuation of the assembly process is monitored and the quality data of the assembly process is obtained;

[0183] De-noising the quality data based on Kalman filtering to obtain characteristic parameter data;

[0184] Determine a joint quality control chart based on characteristic parameter data;

[0185] Determine the capability index value of the assembly process based on the joint quality control chart;

[0186] The assembly quality of the product is evaluated based on the matching relationship between the capability index value of the assembly process and the preset range.

[0187] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0188] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0189] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present application. It should be pointed out that, for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. The above disclosure is only a preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope covered by the present invention.

Claims

1. A method for evaluating the assembly quality of a product, characterized in that: The method comprises: Monitoring the quality fluctuation of the assembly process according to the quality fluctuation model and obtaining the quality data of the assembly process; Performing denoising on the quality data based on Kalman filtering to obtain characteristic parameter data; Determining a joint quality control chart based on the characteristic parameter data; Determining the capability index value of the assembly process according to the joint quality control chart; The assembly quality of the product is evaluated based on the matching relationship between the capability index value of the assembly process and a preset range.

2. The method for evaluating product assembly quality according to claim 1, characterized in that: The method of monitoring the quality fluctuation of the assembly process according to the quality fluctuation model and obtaining the quality data of the assembly process specifically includes: Modeling the mass fluctuation based on the relationship between the noise information and the mass fluctuation measurement data and the fluctuation state data; A state space model of mass fluctuation is obtained, wherein the state space model is composed of a state equation, an observation equation and initial information, which are respectively expressed as: m t =μ t-1 +w t ,w t ~N[0,W t ] y t =μ t +v t ,v t ~N[0,V t ] X0|D0~N[μ0,C0] The quality fluctuation of the assembly process is monitored according to the state equation, the observation equation and the initial information, and the quality data of the assembly process is obtained.

3. The method for evaluating product assembly quality according to claim 2, characterized in that: The denoising process of the quality data based on Kalman filtering to obtain characteristic parameter data specifically includes: The state equation x(k) and observation equation z(k) of the time discrete system are determined according to the linear stochastic difference equation: x(k)=Ax(k-1)+Bu(k)+w(k) z(k)=Hx(k)+v(k) Among them, A is the state transfer matrix, B is the input gain matrix, H is the measurement matrix, u(k) is the process input at time k, w(k) is the system excitation noise information, and v(k) is the measured noise information; Determining an optimal estimate of the state equation x(k) based on an observed value of the observation equation z(k); The initial state variable X0 of the state equation x(k), the covariance P0 of X0, the process excitation noise covariance Q t , measurement noise covariance R t Pre-estimate and determine the characteristic parameter data.

4. The method for evaluating product assembly quality according to claim 3, characterized in that: The initial state variables X0 of the state equation x(k), the covariance P0 of X0, the process excitation noise covariance Q t , measurement noise covariance R t Pre-estimating and determining the characteristic parameter data specifically includes: The estimated value of the initial state variable X0 is determined according to the following formula: Covariance P0 of X0: P0=E[(x(0)-u0)(x(0)-u0) T ]=Var(x(0)); The state equation x(k) is observed n times independently to obtain the time data sequence {x(0), x(1), .... x(n)} and calculate the average value Get the corresponding standard deviation S(x); Get Type A standard uncertainty: The process excitation noise covariance Q is calculated according to the discount factor δ t Estimate the estimated value Q of x(k) to obtain the prior estimation accuracy of x(k): p(k-1|k-1)=δp(k|k-1),0<δ<1 The a priori estimation accuracy of x(k) is processed and solved to obtain the process excitation noise covariance Q t The estimated value of Q is: Q=(1-δ) 2 Rδ -1 。 5. The method for evaluating product assembly quality according to claim 4, characterized in that: Determining a joint quality control chart according to the characteristic parameter data specifically includes: Determine the joint quality control chart based on the Shewhart mean control chart and the exponentially weighted moving average control chart: When the assembly process of the product is under control, μ0 = μ; assuming X1, X2, ... are independent random variable sequences, the exponentially weighted moving average statistic Z t for: Z t =λX t +(1-λ)Z t-1 ,0<λ≤1,t=0,1,2,3… Among them, λ is the smoothing coefficient; The recursive formula and the cumulative sum of data items obtained through iteration are: Among them, Z0 is the starting value; Take the mean value of the data as the starting value Z0=μ0 and obtain the sample statistic Z t The expectation and variance of are: Where n is the number of samples.

6. The method for evaluating product assembly quality according to claim 1, characterized in that: Determining a joint quality control diagram according to the characteristic parameter data further comprises: The control limits for obtaining the joint quality control chart are: Where L is the control limit of the exponentially weighted moving average control chart, k is the parameter for determining the control limit, UCL is the upper limit of the joint quality control chart, and LCL is the lower limit of the joint quality control chart.

7. The method for evaluating product assembly quality according to claim 1, characterized in that: Determining the capability index value of the assembly process according to the joint quality control chart specifically includes: Determine the capability index C of the process p : The capability index C of the process p Make modifications to obtain the corrected process capability index C pk : in, is the mean of the time series mean samples of the joint quality control chart, is the standard deviation of the mean sample, USL is the upper limit of the joint quality control chart, and LSL is the lower limit of the joint quality control chart.

8. A system for evaluating the assembly quality of a product, characterized in that: The system comprises: a data acquisition unit, a data processing unit, a capability index value acquisition unit and an evaluation unit; The data acquisition unit is used to monitor the quality fluctuation of the assembly process according to the quality fluctuation model and acquire the quality data of the assembly process; The data processing unit is used to perform denoising on the quality data based on Kalman filtering to obtain characteristic parameter data; The capability index value acquisition unit is used to determine a joint quality control chart according to the characteristic parameter data; and determine a capability index value of an assembly process according to the joint quality control chart; The evaluation unit is used to evaluate the assembly quality of the product according to the matching relationship between the capability index value of the assembly process and a preset range.

9. A readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor is caused to execute the steps of the method according to any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 7.

Images