False alarm of Built-in Test(BIT) has important influence on system readiness and usage cost. Aimed at solving the problem of false alarm caused by environmental factors, a Time Stress Measurement Device(TSMD) was designed to sample the temperature and vibration of the system. The relationship between false alarm and environmental factors was built based on Support Vector Machine(SVM) which has the advantage of small sample learning. The occurring rule of false alarm was described by using Hidden Markov Model (HMM) which can process dynamic sequential signal well. Then, an intelligent method of reducing false alarm was proposed in terms of TSMD-SVM-HMM. Finally, the method was applied and validated on a helicopter heading attitude system. The experimental results show that the method can effectively recognize and reduce false alarms caused by environmental factors.