CN-121998484-A - Airborne equipment product quality risk assessment method based on process parameters

CN121998484ACN 121998484 ACN121998484 ACN 121998484ACN-121998484-A

Abstract

The invention provides a method for evaluating the product quality risk of an airborne device based on a process parameter, and relates to the technical field of product quality risk evaluation. The method comprises the following steps of S1, screening out typical quality problems of airborne equipment products according to historical detection data, identifying key factors affecting occurrence of the quality problems, S2, constructing an airborne equipment product quality risk assessment index system, S3, constructing an airborne product quality risk assessment model, S4, combining a set quality risk level threshold, and dividing the airborne equipment quality into a plurality of risk levels according to the output quality risk value. According to the invention, the characteristics of the development and production process of the airborne equipment products are considered, the quality risk assessment model considering the process data characteristics is established, the comprehensive, real-time and accurate monitoring of the quality risk of the delivery and use processes of the airborne equipment products is realized, the assessment dimension of the airborne equipment products is more comprehensive, and a more accurate assessment result can be provided.

Inventors

GUO LINXIA
XU YUNTIAN
LIANG ZHAOLEI
JIANG ZIHAO

Assignees

中国航空综合技术研究所

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (10)

1. An on-board equipment product quality risk assessment method based on process parameters is characterized by comprising the following steps: s1, screening out typical quality problems of airborne equipment products according to historical detection data, and identifying key factors affecting the occurrence of the quality problems; S2, constructing an airborne equipment product quality risk assessment index system based on key factors; s3, constructing an airborne product quality risk assessment model based on a risk assessment index system, and outputting a quality risk value, wherein the method comprises the following substeps: S31, according to a risk assessment index system, determining the acquired data as an input value of an assessment model and the quality risk value as an output value; s32, constructing a quality risk assessment model based on a support vector machine, selecting a kernel function and model parameters, and comprising the following steps: S321, determining an evaluation model: ; in the formula, Is that Lagrange multiplier of (A), corresponding to Is that Is a function of the optimized Lagrange multiplier, Is a kernel function; Constant terms that are decision functions; S322, determining a kernel function, namely selecting a discriminant function form of a support vector machine constructed by a radial basis function: ; Where s is the number of centers, s support vectors may determine the center position of the radial basis function, And Comprising an input sample of the sample, Including the square of the euclidean distance between two samples, Including parameters controlling the range of influence; S323, determining model parameters, namely penalty parameters C of a loss function, radial basis and function parameters based on the radial basis function Parameters of insensitive loss function ; S33, setting a data acquisition range based on a quality risk assessment model, acquiring product development and production process data, and preprocessing the process data; S34, optimizing the super parameters of the support vector machine based on a particle swarm algorithm, and searching for an optimal parameter combination; S35, inputting the preprocessed process data as a training set to a quality risk assessment model for training, inputting test set data after training, and outputting quality risk values of each risk dimension of the product; s4, combining a set quality risk level threshold, and dividing the quality of the airborne equipment into a plurality of risk levels by the quality risk value.
2. The method for evaluating the quality risk of an on-board equipment product based on process parameters as set forth in claim 1, wherein step S3 further comprises the step of S36, after the evaluation is completed, evaluating the performance level of the quality risk evaluation model by selecting the use of a mean square error MSE and a correlation coefficient As an evaluation index: ; ; the smaller the MSE, the smaller the fitting error of the model to the training data, and the closer R is to 1, the closer the evaluation result of the model is to the actual value; and optimizing the quality risk assessment model by adjusting the evaluation index according to the assessment result.
3. The on-board equipment product quality risk assessment method based on process parameters of claim 1, wherein step S2 comprises the sub-steps of: S21, dividing the identified key influence factors into four types of machines, materials, methods and rings: the machine risk factor identification comprises the verification and maintenance of a tool fixture and a grinding tool, the state of product production equipment, the electronization of a product file management system, informatization, standardization, the capability of checking equipment, the periodic verification and inspection capability of measuring tools and the periodic maintenance of the equipment; The identification of the material risk factors comprises the steps of packaging material quality, whether packaging is verified or not, product material compliance rate, key component quality grade, warehouse position management, annual module production failure rate and material validity period management; the method risk factor identification comprises the steps of manufacturing process rejection number, standard operation instruction correctness and integrity, rework rate, complete degree of function inspection list items, factory inspection flow and fault processing flow; the ring risk factor identification comprises the condition that the storage environment of the product meets the standard and the condition that the protection environment of the product transportation meets the standard; S22, determining weights occupied by key factors and quantifying the key factors by combining with the development and production process and historical data of the airborne equipment products, summarizing the historical data risks, the appearance bad risks and the functional performance risks, and constructing an airborne equipment product quality risk assessment index system.
4. The process parameter-based on-board equipment product quality risk assessment method according to claim 1, wherein the quality risk value in step S31 includes a historic material risk, an appearance bad risk, a functional bad risk, and the quality risk value is measured by a qualification rate.
5. The process parameter-based on-board equipment product quality risk assessment method according to claim 1, wherein the method of determining the assessment model in step S321 is: the support vector regression is expressed as a given training sample, Regression objective is to find the optimal regression function Introduction of As a function of risk: ; and then Lagrange coefficients are introduced: ; ; Wherein the method comprises the steps of Is that Lagrange multiplier of (A), corresponding to Is that Is obtained by solving the optimal problem by using the optimized Lagrange multiplier And : ; The stability is considered and the average value of the support vector is adopted to determine a regression equation: ; Wherein the method comprises the steps of As a function of the kernel, Is a constant term of the decision function.
6. The method for evaluating the quality risk of an on-board equipment product based on process parameters according to claim 1, wherein the preprocessing in step S34 comprises normalizing the input data of the model by normalizing, filling in missing values and processing outliers.
7. The on-board equipment product quality risk assessment method based on process parameters of claim 1, wherein step S4 comprises the sub-steps of: S41, determining a quality risk threshold value according to the quality risk type of the airborne equipment products, comparing the evaluation accuracy of the batches of the airborne equipment products under different thresholds by evaluating the quality risk value, and selecting the threshold value with the highest accuracy as the quality risk threshold value of the airborne equipment products; S42, establishing an association relationship between the quality risk threshold and the quality risk level.
8. The process parameter-based on-board equipment product quality risk assessment method according to claim 4, wherein the quality risk value of the historic data risk is a historic data qualification rate, and the obtaining method comprises the following steps: 。
9. the process parameter-based on-board equipment product quality risk assessment method of claim 4, wherein: The quality risk value of the appearance bad risk is appearance qualification rate, and the acquisition method comprises the following steps: 。
10. The process parameter-based on-board equipment product quality risk assessment method of claim 4, wherein: The quality risk value of the functional failure risk is the functional qualification rate, and the acquisition method comprises the following steps: 。

Description

Airborne equipment product quality risk assessment method based on process parameters Technical Field The invention relates to the technical field of product quality risk assessment, in particular to a method for assessing product quality risk of airborne equipment based on process parameters. Background The airborne equipment includes all instruments, meters, systems and sub-systems thereof, including meters, navigation, self-control, communication, radar, high-altitude rescue and weapon systems, etc. that are installed on the aircraft for performing flight control, navigation, communication, monitoring of flight conditions, ensuring flight safety, improving flight efficiency, performing specific tasks, and providing passenger comfort and crew working conditions. The development and production of the airborne equipment have the characteristics of high technical complexity, multiple matched levels, strict quality requirements, complex product composition and the like, the complexity of the development and production process and the diversification of products lead to frequent occurrence of quality problems of the airborne equipment products, the occurrence frequency is higher than that of other equipment, and the aircraft equipment has the advantages of complex structure and independent functions, and if the quality problems occur, the aircraft equipment possibly has more serious consequences than the quality problems of components and standard parts. The quality risk of the on-board equipment product refers to the possibility of quality problems in the process of checking or using the equipment product in factories, and is mostly exposed in the assembly and use links. However, because the product quality risks of different aviation equipment products in the process of entrance inspection or use have great difference, the current aviation field supply chain risk prevention and control is focused on the aspects of operation capability, delivery timeliness, financial conditions and the like of the products, the control of the product quality of the airborne equipment is basically to strictly inspect standard parameters such as various functional properties, technical parameters, safety, environmental suitability, appearance quality, file data and the like of the products when the products are delivered, so that the reliability and the safety of the use of the airborne equipment are ensured, and risk assessment and monitoring on the development and production processes of the products of the equipment are lacked. The quality risk prevention and control of the airborne equipment mainly based on the outcome test is not three-dimensional enough, and the quality of the airborne equipment cannot be effectively guaranteed, and meanwhile, the parameter test of the airborne product based on the outcome is performed during delivery, so that the test efficiency is low, the delivery of the product can be influenced, the property loss is caused, or the service life of the airborne product is delayed. Disclosure of Invention In order to solve the defects in the prior art, the invention aims to provide a method for evaluating the quality risk of an airborne equipment product based on process parameters, which is characterized in that a scientific quality risk evaluation and evaluation method is constructed by considering the research and production process management and process inspection and detection results of the product, and the quality risk of the airborne equipment is evaluated in real time, systematically and effectively so as to ensure the stability of an aviation equipment supply chain and the reliability of the product quality. Specifically, the invention provides an airborne equipment product quality risk assessment method based on process parameters, which comprises the following steps: s1, screening out typical quality problems of airborne equipment products according to historical detection data, and identifying key factors affecting the occurrence of the quality problems; S2, constructing an airborne equipment product quality risk assessment index system based on key factors; s3, constructing an airborne product quality risk assessment model based on a risk assessment index system, and outputting a quality risk value, wherein the method comprises the following substeps: S31, according to a risk assessment index system, determining the acquired data as an input value of an assessment model and the quality risk value as an output value; s32, constructing a quality risk assessment model based on a support vector machine, selecting a kernel function and model parameters, and comprising the following steps: S321, determining an evaluation model: ; in the formula, Is thatLagrange multiplier of (A), corresponding toIs thatIs a function of the optimized Lagrange multiplier,Is a kernel function; Constant terms that are decision functions; S322, determining a kernel function, namely selecting a discriminant function