Reliability Analysis of HEE Parameters via Progressive Type-II Censoring with Applications

A new extended exponential lifetime model called Harris extended-exponential(HEE)distribution for data modelling with increasing and decreasing hazard rate shapes has been considered.In the reliability context,researchers prefer to use censoring plans to collect data in order to achieve a compromise between total test time and/or test sample size.So,this study considers both maximum likelihood and Bayesian estimates of the Harris extended-exponential distribution parameters and some of its reliability indices using a progressive Type-II censoring strategy.Under the premise of independent gamma priors,the Bayesian estimation is created using the squared-error and general entropy loss functions.Due to the challenging form of the joint posterior distribution,to evaluate the Bayes estimates,samples from the full conditional distributions are generated using Markov Chain Monte Carlo techniques.For each unknown parameter,the highest posterior density credible intervals and asymptotic confidence intervals are also determined.Through a simulated study,the usefulness of the various suggested strategies is assessed.The optimal progressive censoring plans are also shown,and various optimality criteria are investigated.Two actual data sets,taken from engineering and veterinary medicine areas,are analyzed to show how the offered point and interval estimators can be used in practice and to verify that the proposed model furnishes a good fit than other lifetimemodels:alpha power exponential,generalized-exponential,Nadarajah-Haghighi,Weibull,Lomax,gamma and exponential distributions.Numerical evaluations revealed that in the presence of progressively Type-II censored data,the Bayes estimation method against the squared-error(symmetric)loss is advised for getting the point and interval estimates of the HEE distribution.

Analysis of detection capabilities of LEO reconnaissance satellite constellation based on coverage performance

In the design problem of low earth orbit(LEO) reconnaissance satellite constellation, optimization of coverage performance is the design goal in most current methods. However,in the using process, the user only concerns with the detection capabilities rather than coverage performance. To establish the relationship between these two aspects, the reconnaissance processes of normal stochastic targets are considered and the mathematic models of detection processes are built. The indicators of coverage performance are used to evaluate the detection probability and expectation of detection time delay, which are important factors in reconnaissance constellation estimation viewed from military intelligence discipline. The conclusions confirmed by the final simulation will be useful in LEO reconnaissance constellation design, optimization and evaluation.

Research on vacuum plume and its effects

In vacuum environment, the exhaust flow of attitude control thrusters would expand freely and produce the plume, which possibly causes undesirable contamination, aerodynamic force and heating effects to the spacecraft. Plume work station （PWS） is developed by Beihang University （BUAA） for numerically simulating the vacuum plume and its effects. An approach which combines the direct simulation Monte Carlo （DSMC） method and difference solution of Navier-Stokes （N-S） equations is applied. The internal flows in nozzles are simulated by solving the NS equations. The flow parameters at nozzle exit are used as the inlet boundary condition for the DSMC calculation. Experimental studies are carried out in a supersonic low density wind tunnel which could simulate the 60-80 km altitude environment to investigate the plume and its effects. To demonstrate the capability of PWS, numerical simulations are performed for the vacuum plume of several typical attitude control thrusters. The research results are of great help for the engineering design.