Abstract: The safety accident of prefabricated cabin type lithium iron phosphate battery power station will cause serious economic losses and social impact. It is very necessary to develop an intelligent fire warning system for timely intervention and rapid response. In this paper, a fire intelligent warning system based on characteristic gas (H2, CO) concentration and BMS temperature signal is developed. However, there are some abnormal problems such as burr and drift in monitoring signal of gas detector, which maybe led to false alarm of the system. The effects of these abnormal problems are eliminated by means of time delay and comparison of monitoring data from multiple detectors, which verified the effectiveness and reliability of the accurate intervention of the fire warning system by the analysis of practical application cases.

Key words: lithium iron phosphate battery power station, fire warning system, gas detector, drift, burr suppression