Liquid Propulsion System Reliability Estimation using Computational Bayesian Approach with Multi-level Data

2019 ◽  
Vol 11 (4) ◽  
Author(s):  
S. Muthukumar ◽  
P. Subhash Chandra Bose ◽  
R.A. Srivardhan
Keyword(s):  
Test Data ◽  
Bayesian Approach ◽  
System Reliability ◽  
Propulsion System ◽  
Reliability Estimation ◽  
System Level ◽  
Space Applications ◽  
Unique Challenge ◽  
Cost Constraints ◽  
Level Data

Liquid propulsion system (LPS) is a highly reliable and complex system that is used for the military and space applications. It consists of many flight critical components arranged in series configuration. Reliability is the most critical parameter for this system, even one subsystem failure leads to total failure of flight vehicle. Determining the achieved reliability of a LPS is a unique challenge for designer of these systems. The system reliability needs to be estimated with limited number of tests due to the destructive nature of tests, time and cost constraints. In this paper, reliability of LPS was estimated with subsystem test data using computational Bayesian approach. Component level, subsystem level and system level data are considered and a framework is created by combing all information. The reliability of the LPS was calculated using Markov Chain Monte Carlo (MCMC) simulation which has avoided numerical integration. Results are compared with Lindstorm Madden method and Bayesian hybrid method. Computational Bayesian approach can give reasonably better reliability estimate with limited test data.

EXACT BAYESIAN ESTIMATION OF SYSTEM RELIABILITY WITH POTENTIAL MISCLASSIFICATIONS IN SAMPLING

2004 ◽  
Vol 11 (03) ◽  
pp. 223-241
Author(s):  
N. TURKKAN ◽  
T. PHAM-GIA
Keyword(s):  
Bayesian Estimation ◽  
Bayesian Approach ◽  
Posterior Distribution ◽  
System Reliability ◽  
Exact Expression ◽  
System Level ◽  
Posterior Density ◽  
Approximate Methods ◽  
Credible Intervals ◽  
Highest Posterior Density

We provide the exact expression of the reliability of a system under a Bayesian approach, using beta distributions as both native and induced priors at the system level, and allowing uncertainties in sampling, expressed under the form of misclassifications, or noises, that can affect the final posterior distribution. Exact 100(1-α)% highest posterior density credible intervals, for system reliability, are computed, and comparisons are made with results from approximate methods proposed in the literature.

TEST PLAN ALLOCATION TO MINIMIZE SYSTEM RELIABILITY ESTIMATION VARIABILITY

2004 ◽  
Vol 11 (03) ◽  
pp. 257-272 ◽  
Author(s):  
JOSE E. RAMIREZ-MARQUEZ ◽  
DAVID W. COIT ◽  
TONGDAN JIN
Keyword(s):  
System Reliability ◽  
Reliability Estimation ◽  
System Structure ◽  
System Level ◽  
Reliability Estimate ◽  
Component Level ◽  
Component Reliability ◽  
Allocation Process ◽  
Reliability Estimates ◽  
Estimate Variance

A new methodology is presented to allocate testing units to the different components within a system when the system configuration is fixed and there are budgetary constraints limiting the amount of testing. The objective is to allocate additional testing units so that the variance of the system reliability estimate, at the conclusion of testing, will be minimized. Testing at the component-level decreases the variance of the component reliability estimate, which then decreases the system reliability estimate variance. The difficulty is to decide which components to test given the system-level implications of component reliability estimation. The results are enlightening because the components that most directly affect the system reliability estimation variance are often not those components with the highest initial uncertainty. The approach presented here can be applied to any system structure that can be decomposed into a series-parallel or parallel-series system with independent component reliability estimates. It is demonstrated using a series-parallel system as an example. The planned testing is to be allocated and conducted iteratively in distinct sequential testing runs so that the component and system reliability estimates improve as the overall testing progresses. For each run, a nonlinear programming problem must be solved based on the results of all previous runs. The testing allocation process is demonstrated on two examples.

A Solution to Weibull Prior Distribution of CNC Life-Span Using Information Entropy Theory

2012 ◽  
Vol 548 ◽  
pp. 489-494
Author(s):  
Zhao Jun Yang ◽  
Wei Wang ◽  
Fei Chen ◽  
Kai Wang ◽  
Xiao Bing Li ◽  
...  
Keyword(s):  
Information Entropy ◽  
System Reliability ◽  
Prior Distribution ◽  
Maximum Entropy Method ◽  
Entropy Method ◽  
Small Sample ◽  
Reliability Estimation ◽  
System Level ◽  
Entropy Theory ◽  
Subsystem Level

By using the information entropy theory, a solution to Weibull-small sample prior distribution of system reliability is proposed, which aims at solving the reliability estimation of high-end CNC. Firstly, the prior information is converted from subsystem level into system level based on entropy theory. Then, the prior distribution is solved with the constrained maximum entropy method. Finally, multi-information is fused based on the entropy weighs. It is proved by a case example that this method can obtained the prior distribution under Webull-small sample effectively.

Testability demonstration with component level data from virtual and physical tests

2014 ◽  
Vol 229 (2) ◽  
pp. 265-272 ◽  
Author(s):  
Chao Wang ◽  
Jing Qiu ◽  
Guan-jun Liu ◽  
Yong Zhang
Keyword(s):  
Sample Size ◽  
Test Data ◽  
Flight Control ◽  
Planning Method ◽  
System Level ◽  
Data Conversion ◽  
Equivalent System ◽  
Component Level ◽  
Physical Test ◽  
Level Data

Testability demonstration plays an important role in assuring the testability capability, which can decrease the fault diagnosis time and accelerate the maintenance actions. However, testability demonstration test with classical planning method has the problems of large fault sample size, high test cost, and long test period. A new testability demonstration planning method is proposed, which takes the component level data from both virtual and physical demonstration tests as prior information. Owing to the limitations of the testability modeling technology and relevant programming tools, the virtual testability prototype of the system level cannot be established and the virtual testability test data is not totally credible. So a data conversion method based on the information entropy theory is proposed to convert the component level virtual and physical test data into equivalent system test data, in which the data credibility is taken into consideration. The equivalent system test data is then used to get the prior probability density function of the testability indexes with an empirical Bayesian method. Then, a testability demonstration planning method of Bayesian posterior risk criteria is presented. Finally, the fault detection rate demonstration tests of a flight control system and a heating controller are taken as examples to verify the proposed method. The results show that the introduction of prior test data can effectively decrease the sample size and the credibility of the virtual testability test data can affect the test plan.

scholarly journals Reliability Estimation of a Network Structure Using Generalized Trapezoidal Fuzzy Numbers

2021 ◽  
Vol 51 (1) ◽  
pp. 225-241
Author(s):  
Amit Kumar ◽  
Pooja Dhiman
Keyword(s):  
System Reliability ◽  
Fuzzy Numbers ◽  
Test System ◽  
Reliability Estimation ◽  
Trapezoidal Fuzzy Numbers ◽  
Level Of Satisfaction ◽  
Degree Of Satisfaction ◽  
Fuzzy Logic Method ◽  
Context Specific ◽  
Generalized Trapezoidal Fuzzy Numbers

Abstract Classical sets are used commonly to consider reliability. Because of the uncertainty in the data (which considered in the present paper) classical sets fail to describe the reliability accurately. Uncertainty leads to fluctuation in the actual situation of the structure. Fuzzy logic method attempts to test system reliability with the benefit of membership function. Within this context, specific problems of reasoning-based approaches are studied, explored and correlated with standard reliability approaches. In this paper Generalized Trapezoidal Fuzzy numbers (GTrFN) are used to assess the structure's fuzzy reliability. The reliability of each event is assigned with different level of satisfaction and some improved operations on the generalized trapezoidal fuzzy numbers (GTrFN) are used to calculate the fuzzy boundaries for the resultant reliability of the final event along with the degree of satisfaction. Also the results are compared to demonstrate the application of the improved operations on Generalized Trapezoidal Fuzzy Numbers (GTrFN). The obtained results converge to more precise interval values as compare to the vague fuzzy number.

Sign in / Sign up

Export Citation Format

Share Document