Multi-detector combined fire alarm judgment method suitable for transformer substation

Abstract

The invention discloses a multi-detector combined fire alarm judgment method suitable for a transformer substation. The method comprises the following steps: 1, arranging a plurality of fire alarm detectors near an easy-to-ignite point of the transformer substation; 2, traversing easily-ignited points, igniting a fire source at the easily-ignited points, and obtaining monitoring data of the fire alarm detector; 3, performing homogeneous fusion on the monitoring data to obtain a fusion state value, and storing the fusion state value and the fire alarm category to a simulation database; 4, constructing a BP neural network model by taking the fusion state value as an input layer and the fire alarm category as an output layer, and carrying out training learning by utilizing a simulation database; 5, acquiring monitoring data of the fire alarm detector in real time, and inputting the monitoring data into the BP neural network model to evaluate the fire alarm category. Data are obtained by simulating a fire scene, homogeneous fusion is performed on the data, fusion of data layers is realized, then the BP neural network is constructed to judge the fire alarm category, the model calculation speed is high, and the pre-judgment result precision is high.

Description

technical field [0001] The invention relates to a fire alarm judging method, in particular to a multi-detector combined fire alarm judging method suitable for substations. Background technique [0002] With the rapid development of my country's economy, the scale and number of substations have increased sharply. In recent years, there have been frequent accidents caused by fires in substations that caused huge losses, and even caused casualties. The fire monitoring system of the substation is widely used. [0003] The existing substation fire monitoring system mainly relies on single-point alarms of smoke detectors, infrared flame detectors and temperature-sensing cables to determine whether a fire has occurred. The biggest disadvantage of this method is that the false alarm rate is too high, the fault tolerance is not good, and The rich detection information is limited, and multi-sensor information fusion can provide richer information for diagnosis and enhance the robustn...

AI Technical Summary

Problems solved by technology

[0003] The existing substation fire monitoring system mainly relies on single-point alarms of smoke detectors, infrared flame detectors and temperature-sensing cables to determine whether a fire has occurred. The biggest disadvantage of this method is that the false alarm rate is too high, the fault tolerance is not good, and The rich detection information is limited, and multi-sensor information fusion can provide richer information for diagnosis and enhance the robustness and reliability of the diagnosis system. Therefore, in order to make up for the lack of single-point detection and alarm, it is necessary to propose a Multi-detector combined fire alarm judgment method in substation

Images