A Method of Fault Alarm Recognition based on Hidden Markov Model

BIT was first used in the field of aircraft and weapon equipment system in 1970s. It plays an import role in simplifying the maintenance process and reducing costs. But it has some problems with high fault alarm rate, which prevent its wider application, and the traditional method to recognize fault alarm cannot meet the requirement for modern equipment. The modern intelligent algorithm is widely used in many areas, and it is useful and innovative in patter recognition and fault diagnosis. It uses the data to construct the intelligent model and does not consider the complicated fault mechanism of equipment. Therefore, a method of fault alarm recognition based on HMM is proposed. This method takes full use of HMM, which takes time series into account and the BIT results is taken as the data source. A case study shows that the method is effective and gives an accurate state of the equipment.