CN-122020963-A - Missile life prediction method under zero fault data condition

Abstract

The invention discloses a missile life prediction method under the condition of zero fault data, relates to the technical field of missile life prediction, and aims to solve the problem that life distribution fitting cannot be performed by a traditional distribution method under the condition of zero failure in test data. The method comprises the steps of collecting missile field detection data, sorting the data into periodic tail cutting data, calculating Bayesian estimation of fault probability of each point in zero fault tail cutting data by using a multi-layer Bayesian method, selecting a Weber distribution function as an empirical life distribution function of the missile, calculating the Bayesian estimation of fault probability of each point by using a weighted least square method to obtain shape parameter estimation and scale parameter estimation of the Weber distribution function, and obtaining reliability estimation of the missile according to the shape parameter estimation and the scale parameter estimation of the Weber distribution function. The invention can effectively process the zero failure problem existing in the periodic detection and obtain the service life distribution of the product.

Inventors

• JIA HAONAN
• SONG GUIFEI
• YIN HUIJIN
• WANG WEINA
• SONG XIAOYU
• WANG SHAOGUANG
• ZHANG YANGYANG
• WANG BIN
• JIANG ZHIBAO
• NIU ZHENGYI
• ZHAO XIAODONG
• LI SHAOJIE
• HAN WENBIN

Assignees

• 中国人民解放军32181部队

Dates

Publication Date

20260512

Application Date

20251224

Claims (2)

1. 1. The missile life prediction method under the condition of zero fault data is characterized by comprising the following steps: Collecting missile field detection data, sorting the missile field detection data into periodic tail cutting data, and calculating Bayesian estimation of fault probability of each point in zero fault tail cutting data by using a multi-layer Bayesian method; Selecting a Weber distribution function as an empirical life distribution function of the missile, and calculating Bayesian estimation of the fault probability of each point by using a weighted least square method to obtain shape parameter estimation and scale parameter estimation of the Weber distribution function; And obtaining the life distribution and reliability functions of the missile according to the shape parameter estimation and the scale parameter estimation of the Weber distribution function.
2. 2. The missile life prediction method in the case of zero fault data according to claim 1, wherein the calculating the bayesian estimation of the fault probability of each time point in the zero fault truncated data by using a multi-layer bayesian method specifically includes: taking the Beta distribution function as a priori distribution function of the fault probability; And combining the prior distribution function with a likelihood function of the fault probability to obtain Bayesian estimation of the fault probability.

Description

Missile life prediction method under zero fault data condition Technical Field The invention relates to the technical field of missile life prediction, in particular to a missile life prediction method under the condition of zero fault data. Background The fault data are important data sources for evaluating the service life of the missile product, and comprise shooting failure data of armies in each storage year, failure data of a part assembly, a subsystem and a whole missile obtained through a special service life test and a complete missile, failure data of an acceleration service life test of the part assembly and failure data of a special flight test, wherein the data reflect the change rule of the reliability and the service life of products in each level along with the storage time. Particularly, when the collected fault data span more storage time points, such as failure data under the actual storage environment conditions of long-term service (1-13 years) or even over-term service (more than 13 years) products, a reliable and accurate life assessment prediction result can be obtained based on the fault data. Therefore, the life assessment and prediction method research based on fault data has important significance for missile product life assessment and life-prolonging management. The field fault data is collected under the actual product storage environment condition, is extremely precious, reflects the change rule of the product reliability and service life under the actual storage environment, and represents the reliability performance of the product more than the laboratory simulation or acceleration condition. The field fault data of the missile product comprise failure data of army use, failure data of part assembly, subsystem and whole missile disassembling and inspection, failure data of part assembly accelerated life test, belongs to data obtained in an actual storage environment, is an important data source for evaluating the reliability and the service life of the missile product, and is particularly collected. Because field fault data may have the condition that the target test number of armies is small (especially the storage year is long) or the number of the overhauled samples is limited, the field fault data often has the condition of zero failure in certain years, and the life distribution fitting cannot be directly performed by using the traditional distribution method. Disclosure of Invention The invention aims to provide a missile life prediction method under the condition of zero fault data, which is used for giving out parameter estimation of products under the condition that the fault data has zero failure. In order to achieve the above object, the present invention provides the following technical solutions: a missile life prediction method under the condition of zero fault data comprises the following steps: Collecting missile field detection data, sorting the missile field detection data into periodic tail cutting data, and calculating Bayesian estimation of fault probability of each point in zero fault tail cutting data by using a multi-layer Bayesian method; Selecting a Weber distribution function as an empirical life distribution function of the missile, and calculating Bayesian estimation of the fault probability of each point by using a weighted least square method to obtain shape parameter estimation and scale parameter estimation of the Weber distribution function; And obtaining the life distribution and reliability functions of the missile according to the shape parameter estimation and the scale parameter estimation of the Weber distribution function. More specifically, the calculating the bayesian estimation of the fault probability of each time point in the zero fault tail-biting data by using the multi-layer bayesian method specifically includes: taking the Beta distribution function as a priori distribution function of the fault probability; And combining the prior distribution function with a likelihood function of the fault probability to obtain Bayesian estimation of the fault probability. Compared with the prior art, the missile life prediction method under the condition of zero fault data has the following beneficial effects: according to the invention, missile periodic detection data under the condition of zero fault data are arranged into periodic tail cutting data, fault probability estimated values of detection points are obtained by using a layered Bayesian method, and life distribution with optimal fitting effect is obtained by using a weighted least square method based on the fault probability estimated values of the points, so that missile product life prediction and analysis are carried out, the problem of zero failure existing in periodic detection can be effectively processed, and product life distribution is obtained. Drawings The accompanying drawings, which are included to provide a further understanding of the invention and are incorporat