中文摘要：

在现有的步进应力加速退化试验中,应力水平的改变时间通常被事先确定,这对高可靠性产品存在一定的不合理性。用Wiener过程描述产品性能退化轨道,在试验中当产品性能退化到某一阈值时,改变其应力水平,从而产品的应力水平改变时间是服从逆高斯分布的随机变量。基于该情形建立产品的步进应力加速退化模型。考虑到模型的计算复杂性,采用贝叶斯马尔科夫链蒙特卡罗方法获取模型参数的极大似然估计,最后通过仿真试验验证文中的模型和方法。

英文摘要：

During the step-stress accelerated degradation test（SSADT） experiment,the time of changing stress level is usually determined in advance.It may be unreasonable to highly reliable products.The degradation path is described as a Wiener process.During the experiment,the stress level is changed when the degradation value crosses a pre-specified value.Therefore,the time of changing stress level is regarded as a random variable which follows inverse Gaussian distribution.Based on this situation,an SSADT model is proposed.Due to the computational complexity of the model,the Bayesian Markov chain Monte Carlo（MCMC） method for the parameters is applied to obtain the maximum likelihood estimation.Finally,some simulation examples are presented to validate the proposed model and method.