基于极大似然估计的汽机功率测量数据诊断

根据热工测量数据的内在联系，基于极大似然估计理论，提出了一种结合对象特性的数据诊断方法。以火电机组汽轮机功率测量为例，利用其与调节级级后压力、温度之间的耦合关系，得到了功率参数的分布规律和应属的正常数据范围，对越出该范围的数据，判断为在一定概率条件下的异常数据，解决了功率数据的诊断问题。该方法同样可以应用于其它热工参数的数据诊断。

Diagnosis of Measured Data for Power of Turbine Based on Maximum Likelihood Estimation

According to the internal relations of measured data in thermal power engineering, a method combined with object characteristics for data diagnosis, which is based on maximum likelihood estimation, is put forward. Considering the power measuring of turbine in power plant, the regularities of distribution of parameters of power and its normal data area can be obtained by using the coupling relations exiting in power of turbine, pressure and temperature after regulating stage. The data, which are out of normal range, can be judged as abnormal data in the case of a given probability, thus the diagnosis of measured data for power of turbine can be accomplished. Furthermore, this method can be applied to the diagnosis of measured data for other thermal parameters.