Aiming at the detection error of catalytic combustion type methane gas sensor in different temperature and humidity environments
a method based on long short-term memory(LSTM)networks is proposed to achieve temperature and humidity compensation for methane gas concentration.To resolve the issue of indeterminate LSTM network parameters
a genetic algorithm(GA)is introduced to optimize LSTM parameters
and a GA-LSTM compensation model is constructed.Experimental simulation validation shows that the correlation coefficient(
R
2
)reaches 0.998 with excellent fitting effect
which proves superiority of this method.This method possesses potential engineering application value.