基于SAEM-HMM的设备状态诊断模型研究

针对设备状态诊断问题,提出了基于模拟退火和期望最大化算法的隐马尔可夫模型(SAEMHMM).该模型针对改进传统隐马尔可夫模型对初值敏感及期望最大化算法容易陷入局部最优的不足,将模拟退火算法与期望最大化算法结合,利用前者具有概率的全局收敛性,克服局部最优问题,实现隐马尔可夫模型参数估计过程的优化.最后通过算例分析验证了该模型的可行性与有效性.

Machinery diagnosis model based on SAEM-HMM

Aiming at the problem of machinery diagnosis, simulated annealing (SA) algorithm and expectation maximization (EM) algorithm are considered to improve hidden Markov model (HMM). As the traditional HMM is sensitive to initial values and EM is easy to trap into a local optimization, SA is combined to improve HMM which can overcome local optimization searching problem. The proposed new HMM has strong ability of global convergence, and can optimize the process of parameters estimation. Finally, through the case study, the computation results illustrate this proposed HMM has high efficiency and accuracy, which could help machinery diagnosis in practical.