Process reliability parameter optimization method for aviation equipment product

Abstract

The present application relates to a kind of process reliability parameter optimization methods of aviation equipment product, it includes the following steps: S1, according to the process of aviation equipment product process and key characteristics, determine the aviation equipment product process reliability influencing factor;S2, based on the aviation equipment product process reliability key influencing factor determined in S1, establish process reliability influence relationship model;S3, process parameter is optimized and verified based on the process reliability influence relationship model determined in S2.It is according to the process of equipment product and key characteristics, analyzes equipment product process reliability influencing factor, and based on process reliability key influencing factor, carries out process data processing analysis, establishes process reliability influence relationship model;After based on process reliability influence relationship model, through the process control requirement of equipment product, carries out process parameter optimization and verifies optimization effect, proposes a kind of energy quantization and actual execution method for process reliability parameter optimization.

Description

Technical Field

[0001] This invention relates to the field of parameter optimization, specifically to a method for optimizing process reliability parameters of aerospace equipment products. Background Technology

[0002] Preliminary results have been achieved in the research on ensuring the reliability of equipment products during the manufacturing process. For example, robust design is achieved through Design for Six Sigma (DFSS), multi-source information is integrated, and various statistical methods are applied to reliability analysis according to different engineering backgrounds. However, the research proposed by these studies are all macro-level "post-event" measures. How to implement them "in advance" in the development process of specific products still lacks in-depth research.

[0003] The research on process reliability has gone through three stages of development: from manufacturing equipment and manufacturing systems to process systems ensuring product quality. To date, industry and academia generally recognize the impact of manufacturing processes on product reliability, but they focus on controlling the fluctuation characteristics of manufacturing process parameters, making it difficult to quantify the impact results.

[0004] Ensuring product reliability during the manufacturing process involves multiple factors (system parameters, key performance indicators, process characteristics, etc.), and these factors are interconnected, with many relationships difficult to quantify accurately. Furthermore, due to the trend towards small-batch, short-cycle manufacturing, equipment manufacturing processes often suffer from limited data or small sample sizes (few trial runs). Analysis of process influencing factors primarily follows two approaches: based on the interaction between process steps and work-in-process hole location characteristics (deviation flow, QR chains, etc.), and based on the physical failure mechanisms of equipment products (process defect research, etc.). Therefore, research needs to combine equipment product processing and failure mechanisms with various data to enrich the correlation studies among influencing factors.

[0005] Current research on process reliability models focuses on data-driven methods and the quantitative analysis of the root causes of process defects, emphasizing the application of process quality data such as quality inspection data, experimental data, and field failure data. However, there is a lack of research and methods on using the constructed models to improve the quality stability and consistency of equipment product development processes. Most studies focus on evaluating the product manufacturing process. Therefore, how to control and optimize existing processes through equipment product process reliability-related models is a key focus of future research.

[0006] In summary, the existing technology lacks a systematic understanding of process reliability for equipment product quality, lacks research on the relationship model between product reliability and manufacturing process parameters, and has not formed a process and method for optimizing the product development process. There is an urgent need to study a process reliability parameter optimization method that can be quantitatively implemented. Summary of the Invention

[0007] To address the deficiencies mentioned in the background section, this invention provides a method for optimizing process reliability parameters for aerospace equipment products. Based on the connotation of process reliability, and according to the processes and key characteristics of the equipment products, it analyzes the influencing factors on process reliability. Based on these key influencing factors, it performs process data processing and analysis to establish a process reliability influence relationship model. Then, based on this model, it optimizes process parameters according to the process control requirements of the equipment products and verifies the optimization effect. This provides a complete process methodology for in-depth research on the connotation, influence relationship model, and process optimization control of equipment product process reliability, guiding the assurance of product reliability during the equipment development phase. Specifically, it addresses the problems of significant quality degradation and poor stability in the manufacturing process of key characteristics of typical equipment products during the transition from development to mass production. It identifies key process reliability influencing factors, optimizes the existing process parameter control range through the process reliability influence relationship model, ensures the realization of key quality characteristics of typical products in actual manufacturing, and improves the production efficiency and quality stability and consistency of typical equipment products.

[0008] Specifically, the present invention provides a method for optimizing process reliability parameters of aerospace equipment products, which includes the following steps:

[0009] S1. Based on the processes and key characteristics of aviation equipment products, determine the factors affecting the reliability of aviation equipment product processes. This includes the following sub-steps:

[0010] S11. Collect information on quality anomalies (A) in the production processes of aviation equipment products and analyze the process influencing factors (B) that cause these anomalies. Establish a process reliability influencing factor matrix (K) for a specific process.

[0011]

[0012] Where A represents quality anomalies in the process, B represents the process influencing factors corresponding to A, K represents the process reliability influencing factor matrix of the process, and i represents the number of quality anomalies.

[0013] S12. Assign importance values ​​to the process influencing factors and reorder them according to importance to obtain the process reliability influencing factor matrix K' of the sorted process.

[0014]

[0015] Where K' is the process reliability influencing factor matrix, and m is the importance ranking of the process influencing factors;

[0016] S2. Based on the key influencing factors of the process reliability of aerospace equipment products determined in step S1, establish a process reliability influence relationship model, which specifically includes the following sub-steps:

[0017] S21. Perform data preprocessing on the raw data of quality anomaly problem A and process influencing factor B of quality anomaly to form parameter indicators for modeling input and output.

[0018] S22. Based on the importance ranking of key influencing factors of process reliability in the process reliability influencing factor matrix K', determine the number and type of input parameters and the type of output parameters. Use the determined input parameters as the input values ​​of the neural network algorithm, and add activation function, penalty factor and random seed to output the following process reliability influence relationship model:

[0019] R = f(v1, v2, ..., v n )=f(V);

[0020] Where f is the process reliability impact model, v is the input parameter, and R is the output parameter;

[0021] S23. Using the fitting coefficient r 2 The effectiveness of the established process reliability impact model is evaluated using the following formula:

[0022]

[0023] If 0 < r 2 If the value is less than 1, it indicates that the model fits better than the mean model, meaning the process reliability impact model is effective. Otherwise, return to step S22 to redetermine the algorithm configuration and repeat step S23 until 0 < r. 2 <1, output the final process reliability impact relationship model;

[0024] S3. Based on the process reliability impact relationship model determined in step S2, optimize and verify the model input parameters, i.e., the process parameters. This includes the following sub-steps:

[0025] S31. Determine the new model output parameter range, and use a greedy algorithm to obtain the model input parameter range [min, max] that satisfies the optimized output parameter range, thus obtaining the optimized process reliability parameter range.

[0026] S32. Verify the optimized process reliability parameter range. If the verification is successful, the optimized process reliability parameter range shall be determined as the final process reliability parameter range.

[0027] Preferably, the specific steps of data preprocessing in step S21 are as follows: using data cleaning, dimensionality reduction and normalization methods to process missing values, remove noise and identify outliers in the original process data.

[0028] Preferably, the input parameter is the top-ranked process influencing factor in the process reliability influencing factor matrix K' of the process.

[0029] Preferably, step S31 specifically includes the following sub-steps:

[0030] S311. Determine the new model output parameter range R' based on process control requirements;

[0031] S312. Extract each input parameter v i The range [min, max] in the process;

[0032] S313. Set the number of searches n, and sort the input parameters v in descending order of importance. i The value of is {v i1 ,v i2 ,v i3 ,…,v in}, at this point, the remaining input parameters are set to the sample median;

[0033] Among them, V i1 V is the minimum value within the range of input parameters. in The maximum value within the range of input parameters;

[0034] S314, Transfer multiple input parameters v i Substitute each input parameter v into the process reliability impact model and calculate the result. i The corresponding model output parameter prediction values;

[0035] S315. Determine and filter the input parameters v that make the output parameter R satisfy the new model output parameter range R'. i The range of values ​​for each input parameter is used to obtain the optimized range of values ​​for each input parameter, i.e., the process parameter.

[0036] Preferably, in step S32, the process capability index C is used. p and C pk The stability and consistency of the processing quality of the process system equipment before and after process parameter optimization were quantitatively evaluated, and the process capability index C was determined. p and C pk The calculation formula is:

[0037]

[0038]

[0039] In the formula, σ and μ are the standard deviation and mean of the parameter, respectively, and U and L are the upper tolerance limit and lower tolerance limit of the parameter, respectively; if the process capability index C p C pk If the value improves after optimization, it indicates that the optimization of process parameters is effective; otherwise, return to step S31 to re-optimize.

[0040] Preferably, in step S22, it is determined whether to use the standardized basis function and the conjugate gradient descent function based on the type of the input parameters.

[0041] Preferably, the method for determining the new model output parameter range V' in step S311 is to narrow down the model output parameter range.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] (1) The present invention proposes a method for optimizing the process reliability parameters of aviation equipment products. Specifically, it proposes the connotation of process reliability of equipment products, clarifies the key process reliability influencing factors in the research and development stage, and optimizes the existing process parameter control by constructing a process reliability influence relationship model to ensure the realization of key quality characteristics of typical products in actual manufacturing. This is of great significance for improving the production efficiency and quality stability and consistency of typical equipment products.

[0044] (2) Based on the process reliability influence relationship model, this invention optimizes process parameters and verifies the optimization effect through the process control requirements of equipment products. It proposes a method for quantifying and actually implementing process reliability parameter optimization, which can ensure the accuracy of the execution and results of process parameter optimization, thereby providing a guarantee for process reliability and filling the gap in the existing technology that cannot optimize the reliability of the process flow. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the method for optimizing the process reliability parameters of aviation equipment products according to the present invention.

[0046] Figure 2 This is a flowchart illustrating the method for optimizing the process reliability parameters of aviation equipment products according to the present invention. Detailed Implementation

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Specifically, this invention provides a method for optimizing process reliability parameters of aerospace equipment products, such as... Figure 1 and Figure 2 As shown, it includes the following steps:

[0049] S1. Based on the processes and key characteristics of aviation equipment products, determine the factors affecting the reliability of aviation equipment product processes. This includes the following sub-steps:

[0050] S11. Collect information on quality anomalies (A) in the production processes of aviation equipment products and analyze the process influencing factors (B) that cause these anomalies. Establish a process reliability influencing factor matrix (K) for each process.

[0051]

[0052] Where A represents quality anomalies in the process, B represents the process influencing factors corresponding to A, K represents the process reliability influencing factor matrix of the process, and i represents the number of quality anomalies.

[0053] S12. Assign importance values ​​to multiple process influencing factors and reorder them according to importance to obtain the process reliability influencing factor matrix K' of the sorted process.

[0054]

[0055] Where K' is the matrix of factors affecting the reliability of the process, and m is the ranking of the importance of the factors affecting the process.

[0056] In practical implementation, considering the role of the process system in ensuring product reliability, the definition of process reliability was proposed: the ability of the process system to meet design requirements for the key characteristics of the processed product under specified conditions and within a specified time. Among these,

[0057] 1) A process system is an organic combination of people, machines, materials, methods, environment, and measurement;

[0058] 2) The specified conditions refer to the operating environment of the process system, that is, the requirements for personnel, machines, materials, methods, environment and measurement, including the specified personnel capabilities, processing and testing equipment capabilities, raw materials, process flow methods, working environment conditions and quality management requirements;

[0059] 3) The specified time is the task time allocated to the process system;

[0060] 4) The decline in process reliability is manifested in the process system's inability to complete the product processing task and the failure of the process system to meet the design requirements for the key characteristics of the products processed.

[0061] S2. Based on the key influencing factors of the process reliability of aerospace equipment products determined in step S1, establish a process reliability influence relationship model, which specifically includes the following sub-steps:

[0062] S21. Perform data preprocessing on the data in the process reliability influencing factor matrix K' to form parameter indicators for modeling input and output.

[0063] S22. Based on the key influencing factors of process reliability, determine the number and type of input parameters and the type of output parameters. Use the determined input parameters as the input values ​​of the neural network algorithm, and add activation functions, penalty factors, and random seeds. Determine whether to use standardized basis functions and conjugate gradient descent functions based on the type of input parameters. Output the following process reliability influence relationship model:

[0064]

[0065] Where R is the process reliability impact model, v is the input parameter, n is the number of input parameters, and V is the output parameter.

[0066] S23. Using the fitting coefficient r 2 The effectiveness of the established process reliability impact model is evaluated using the following formula:

[0067]

[0068] If 0 < r 2 If the value is less than 1, it indicates that the model fits better than the mean model, meaning the process reliability impact model is effective. Otherwise, return to step S22 to redetermine the number and type of input parameters, and repeat step S23 until 0 < r. 2 <1, output the final process reliability impact relationship model.

[0069] S3. Based on the process reliability impact relationship model determined in step S2, optimize and verify the model input parameters, i.e., the process parameters. This includes the following sub-steps:

[0070] S31. Determine the new model output parameter range and use a greedy algorithm to obtain the model input parameter range [min, max] that satisfies the optimized output parameter range, thus obtaining the optimized process reliability parameter range.

[0071] S32. Verify the optimized process reliability parameter range. If the verification is successful, the optimized process reliability parameter range shall be determined as the final process reliability parameter range.

[0072] Preferably, step S31 specifically includes the following sub-steps:

[0073] S311. Determine the new model output parameter range R' based on process control requirements.

[0074] S312. Extract each input parameter v i The range [min, max] in the process.

[0075] S313. Set the number of searches n, and sort the input parameters v in descending order of importance. i The value of is {v i1 ,v i2 ,v i3 ,…,v in}, at this point, the remaining input parameters are set to the sample median;

[0076] Among them, V i1 V is the minimum value within the range of input parameters. in This represents the maximum value within the range of the input parameters.

[0077] S314, Transfer multiple input parameters v i Substitute each input parameter v into the process reliability impact model and calculate the result. i The corresponding model output parameter prediction values.

[0078] S315. Determine and filter the input parameters v that make the output parameter R satisfy the new model output parameter range R'. i The range of values ​​for each input parameter is used to obtain the optimized range of values ​​for each input parameter, i.e., the process parameter.

[0079] Preferably, in step S32, the process capability index C is used. p and C pk The stability and consistency of the processing quality of the process system equipment before and after process parameter optimization were quantitatively evaluated, and the process capability index C was determined. p and C pk The calculation formula is:

[0080]

[0081]

[0082] In the formula, σ and μ are the standard deviation and mean of the parameter, respectively, and U and L are the upper tolerance limit and lower tolerance limit of the parameter, respectively; if the process capability index C p C pk If the value improves after optimization, it indicates that the optimization of process parameters is effective; otherwise, return to step S31 to re-optimize.

[0083] Preferably, in step S22, it is determined whether to use the standardized basis function and the conjugate gradient descent function based on the type of the input parameters.

[0084] Preferably, the method for determining the new model output parameter range V' in step S311 is to narrow down the model output parameter range. Specific Implementation

[0086] S1. Based on the processes and key characteristics of aviation equipment products, determine the factors affecting the reliability of aviation equipment product processes and form a matrix. In this embodiment, for easier description, the matrix is ​​presented in tabular form.

[0087] In this embodiment, after analyzing the equipment product process and key characteristics, typical quality anomalies in the equipment product process are summarized, the process causes leading to typical product quality problems are analyzed, and a process FMECA analysis table as shown in Table 1 is formed. The process reliability influencing factors of each process are extracted, and the importance of each influencing factor is calculated and ranked using the risk priority number, and a process reliability influencing factor analysis result table as shown in Table 2 is established.

[0088] Table 1. Process FMECA Analysis

[0089]

[0090]

[0091] Table 2 Analysis Results of Factors Affecting Process Reliability

[0092]

[0093] S2. Based on the key influencing factors of process reliability, process data processing and analysis are conducted to establish a process reliability influence relationship model;

[0094] Based on the key factors affecting process reliability ranked by importance in Table 2, relevant data indicators of the equipment product processing process were collected to form the process data collection list shown in Table 3.

[0095] Table 3. List of Process Data Collection

[0096]

[0097] Subsequently, conventional data cleaning, dimensionality reduction, and normalization methods were used to preprocess the collected raw process data, including handling missing values, removing noise, and identifying outliers. This process transformed the data in the process data collection list into input and output parameters that can be used for modeling and calculation.

[0098] The number and type of input parameters and the type of output parameters are determined based on the key factors affecting process reliability. In this embodiment, the process is a welding process. Therefore, the input parameters can be key factors such as flux flow rate, welding speed, track tilt angle, flux specific gravity, solder pot temperature, or defect rate. The output parameters can be product characteristic processing indicators, such as the defect percentage of batch products or the rework ratio.

[0099] In this embodiment, a neural network algorithm is used to establish a model of the influence relationship on process reliability. The algorithm parameters are selected as follows: scale optimization type (2), quadratic penalty factor value (0.01), and tolerance parameter of incremental value (10). -6 The model is constructed as follows: (1) using a random seed, (2) using standardized basis functions, and (3) using the conjugate gradient descent algorithm.

[0100] R=0.0015312v1+7.0600520v2+2.1433360v3-4.8978062v4-7.5019256v5

[0101] f(v1,v2,…,v n )=f(V)

[0102] Correlation coefficient r 2 The value of 0.3765 indicates the effectiveness of the model construction.

[0103] S3. Based on the process reliability impact model, optimize process parameters and verify the optimization effect by considering the process control requirements of equipment products.

[0104] Based on the existing model output parameters, process control requirements for equipment products are proposed to improve the control of equipment product quality, i.e., a new range of output parameters. In this embodiment, the model output parameter is the defect percentage of batch products. Currently, the defect percentage range is [0, 0.5], and it is necessary to control the defect percentage range to [0, 0.35]. Therefore, [0, 0.35] is the process control requirement for equipment products.

[0105] Based on the collected process data, the process reliability impact relationship model established in step S2 and passed the effectiveness evaluation is used to obtain the model input parameter range that meets the output parameter range using a greedy algorithm. This provides an optimization reference for the control range of the process parameters belonging to the model input. In this embodiment, there are three model input parameters, a, b, and c, with original ranges of [1.25, 3.34], [102.23, 114.09], and [160.98, 248.90], respectively. After calculation, the new parameter range that meets the control requirements is [1.35, 3.30], [102.82, 114.08], and [169.78, 248.90], which are the parameter ranges that need to be adjusted in the process.

[0106] The analysis results of the factors affecting the product process in this embodiment are shown in Table 4. The top 5 factors in terms of importance (v1, v2, v3, v4, v5) are selected as the model input parameters. Meanwhile, the product batch defect problem and proportion statistics are shown in Table 5. The defect problem percentage R of the product batch is used as the model output parameter.

[0107] Table 4. Analysis Results of Factors Affecting the Process Reliability of a Certain Product

[0108]

[0109] Table 5. Statistics on Defects and Proportions in Product Batches

[0110]

[0111]

[0112] After updating the process control requirements, a greedy algorithm is executed. In this embodiment, the percentage of defect problems and the output parameter R need to be controlled within [0, 0.35], that is, the percentage of defect problems should not exceed 0.35%. The number of optimization columns (5) and the number of searches (20) in the greedy algorithm are specified. The greedy algorithm is executed by Python. The results before and after optimization of the input parameters of each model are shown in Table 6.

[0113] Table 6 Comparison of Model Input Parameters Before and After Optimization

[0114] Parameter categories <![CDATA[v1]]> <![CDATA[v2]]> <![CDATA[v3]]> <![CDATA[v4]]> <![CDATA[v5]]> Before optimization, the upper limit 1.25 109.86 -0.041 160.98 33.2 Lower limit before optimization 3.302 165.36 0.065 248.90 141.2 Optimized upper limit 1.35 113.63 -0.03 169.78 35.0 Optimized lower limit 3.30 165.36 0.06 248.9 90.0

[0115] Calculate the process capability index (Cp, Cpk) before and after optimization: Incorporate the new process parameter control range into the product's manufacturing process and collect the percentage of defects (R) after process parameter optimization. For R before and after optimization, use Python to statistically analyze the standard deviation, mean, and upper and lower tolerance limits (see Table 7). This yields the process reliability evaluation index before optimization. The optimized corresponding evaluation indicators are: It can be seen that the new process parameter control range has greatly improved the stability and consistency of the processing quality of the process system, proving that the optimization is effective.

[0116] Table 7 Comparison of statistical values ​​of model output parameters before and after optimization.

[0117] Statistical value categories Upper limit U Lower limit L mean μ Standard deviation σ Before optimization 1.98 0 0.621 0.3762 After optimization 0.35 0 0.235 0.0311

[0118] This invention proposes a method for optimizing process reliability parameters of aerospace equipment products. Specifically, it defines the connotation of process reliability of equipment products, clarifies the key process reliability influencing factors in the development stage, and optimizes the existing process parameter control by constructing a process reliability influence relationship model. This ensures the realization of key quality characteristics of typical products in actual manufacturing, which is of great significance for improving the production efficiency, quality stability and consistency of typical equipment products. Furthermore, it provides a new method for quantitatively analyzing process parameter optimization.

[0119] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for optimizing process reliability parameters of aerospace equipment products, characterized in that: It includes the following steps: S1. Based on the processes and key characteristics of aviation equipment products, determine the factors affecting the reliability of aviation equipment product processes. This includes the following sub-steps: S11. Collect information on quality anomalies (A) in the production processes of aviation equipment products and analyze the process influencing factors (B) that cause these anomalies. Establish a process reliability influencing factor matrix (K) for a specific process. ； Where A represents quality anomalies in the process, B represents the process influencing factors corresponding to A, K represents the process reliability influencing factor matrix of the process, and i represents the number of quality anomalies. S12. Assign importance values ​​to the process influencing factors and reorder them according to importance to obtain the process reliability influencing factor matrix K´ of the sorted process. ； Where K´ is the process reliability influencing factor matrix, and m is the importance ranking of the process influencing factors; S2. Based on the factors affecting the process reliability of aerospace equipment products determined in step S1, establish a process reliability impact relationship model, which specifically includes the following sub-steps: S21. Perform data preprocessing on the raw data of quality anomaly problem A and process influencing factor B of quality anomaly to form parameter indicators for modeling input and output; S22. Based on the importance ranking of process reliability influencing factors in the process reliability influencing factor matrix K´, determine the number and type of input parameters and the type of output parameters. Use the determined input parameters as the input values ​​of the neural network algorithm, and add activation function, penalty factor and random seed to output the following process reliability influencing relationship model: ； Where f is the process reliability impact model, v is the input parameter, and R is the output parameter; S23. Using the fitting coefficients The effectiveness of the established process reliability impact model is evaluated using the following formula: ； like This indicates that the model's fit is better than the mean model, meaning the process reliability impact model is effective. Otherwise, return to step S22 to redetermine the algorithm configuration and repeat step S23 until... Output the final process reliability impact model; S3. Based on the process reliability impact relationship model determined in step S2, optimize and verify the model input parameters, i.e., the process parameters. This includes the following sub-steps: S31. Determine the new model output parameter range, and use a greedy algorithm to obtain the model input parameter range [min, max] that satisfies the optimized output parameter range, thus obtaining the optimized process reliability parameter range. S32. Verify the optimized process reliability parameter range. If the verification is successful, the optimized process reliability parameter range shall be determined as the final process reliability parameter range. In step S32, the process capability index C is used. p and C pk The stability and consistency of the processing quality of the process system equipment before and after process parameter optimization were quantitatively evaluated, and the process capability index C was determined. p and C pk The calculation formula is: ； ； In the formula , Let C be the standard deviation and mean of the parameter, and U and L be the upper and lower tolerance limits of the parameter, respectively; if the process capability index C p C pk If the value improves after optimization, it indicates that the optimization of process parameters is effective; otherwise, return to step S31 to re-optimize.

2. The method for optimizing process reliability parameters of aerospace equipment products according to claim 1, characterized in that: The specific steps of data preprocessing in step S21 are as follows: using data cleaning, dimensionality reduction and normalization methods, missing value processing, noise removal and outlier identification are performed on the original process data.

3. The method for optimizing process reliability parameters of aerospace equipment products according to claim 1, characterized in that: Step S31 specifically includes the following sub-steps: S311. Determine the new model output parameter range R´ based on process control requirements; S312. Extract all input parameters The range [min, max] in the process; S313. Set the number of searches n, and sort them according to their importance from largest to smallest. Let a certain input parameter... The value is At this point, the remaining input parameters are set to the sample median; in, The minimum value within the range of input parameters. The maximum value within the range of input parameters; S314, Transfer multiple input parameters Substitute each input parameter into the process reliability impact model in turn and calculate it. The corresponding model output parameter prediction values; S315. Determine and filter output parameters. Input parameters that satisfy the new model output parameter range R´ The range of values ​​for each input parameter is used to obtain the optimized range of values ​​for each input parameter, i.e., the process parameter.

4. The method for optimizing process reliability parameters of aerospace equipment products according to claim 1, characterized in that: In step S22, it is determined whether to use the standardized basis function and the conjugate gradient descent function based on the type of the input parameters.

5. The method for optimizing process reliability parameters of aerospace equipment products according to claim 4, characterized in that: In step S311, the method for determining the new model output parameter range R´ is to narrow it down based on the model output parameter range.

Images