Accelerated testing and statistical lifetime

Application of Accelerated Testing and Statistical Lifetime Modeling to Membrane Electrode Assembly Development... 

Accelerated testing and statistical lifetime modeling are important tools in the development of durable membrane electrode assemblies (MEAs). There are several reasons for using accelerated tests and three of the main reasons are listed here ... 

Utilizing tools such as accelerated tests and statistical lifetime modeling allow for early estimates of durability. There are three basic types of accelerated tests (1) screening tests are appropriate for measuring incremental improvements in component... 

ABSTRACT There are many properties correlated with maintenance costs. Due to minimize the maintenance costs we should recognize them and assess the level of influence and choose them of which the mitigation is the most cost effective. One of the most important property is lifetime which is assessed using statistical methods during accelerated tests. And may be improved based on real data. The parametric and non-parametric statistics could be applied to assessment the lifetime of aviation items. In the paper their advantages and disadvantages were recognized. 

Figure 1.15(a). Statistical analysis of accelerated test data with MEA lifetime prediction done by 3M (Symbols—data points lines—model fits) [47]. (Reprinted with permission from Hicks M. Membrane and catalyst durability under accelerated testing. In Conference proceedings of fuel cell durability stationary, automotive, portable. 2005 Knowledge Press.)... 

Again, as for the section on time dependent yielding it is necessary to consider how to obtain data that can be useful in preventing failure over the lifetime of a structure that may be intended to last for 20 to 100 years. The only rational means for this is some type of accelerated testing such as that offered by the time-temperature or time-stress-superposition procedure. Also, it is reiterated that failure is a stochastic event and of necessity a reliable statistical analysis should be performed. 

SIR tests are accelerated aging tests. Accelerated life tests (ALT) are used to determine the expected lifetime in the use environment. To obtain this data, a statistical number of test coupons are exposed to accelerated conditions and tested to failure. Several temperature conditions are needed to determine the activation energy for the process, as indicated in the Arrhenius equation helow ... 

Fig. 9 Statistical MEA lifetime predictions from accelerated test data using SPLIDA. The symbols represent actual data points and the lines are the statistictil model fit to the data solid lines load profile A, dashed lines load profile B, dotted lines load profile C...

Lifetime is a probability distribution and therefore the data analysis needs to account for this fact. When the results of lifetime tests are combined with statistical lifetime analysis, it is possible to predict product lifetime at end-use conditions from accelerated degradation conditions. This can often lead to faster development of MEAs and components because critical lifetime information can be learned without waiting for lengthy lifetime tests to complete under end use conditions. 

---