论文部分内容阅读
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.
In order to solve the problem of device status diagnosis, a hidden Markov model (SAEMHMM) based on Simulated Annealing and Expectation Maximization (SAEMHMM) algorithm is proposed. The proposed model is sensitive to initial value and maximizes expected expectation. , The simulated annealing algorithm is combined with the expectation maximization algorithm to optimize the parameter estimation process of the hidden Markov model by using the global convergence of the former with probability to overcome the local optimal problem.At last, Feasibility and effectiveness.