A method, system and device for rapid detection of ammonia content in fire smoke

Abstract

The application relates to the technical field of gas detection, in particular to a method, system and device for rapidly detecting the ammonia content in fire smoke, which comprises the following steps: acquiring a voltage response signal in the fire smoke dominated by the ammonia concentration change; analyzing the peak value distribution characteristics at different wave peaks in the voltage response signal, extracting effective wave peaks and corresponding effective signal segments; determining first evaluation values and second evaluation values of the effective signal segments; determining an evaluation coefficient representing that the effective signal segments conform to the typical characteristics of the heat convection phenomenon, adjusting the filter window size according to the evaluation coefficient, and filtering the effective signal segments by using a filtering algorithm to obtain a filtered voltage response signal, which is input into a pre-trained neural network model to realize real-time detection of the ammonia content in the fire smoke. The application improves the detection accuracy of the ammonia content in the fire smoke.

Images