Laser methane emission monitoring based on harmonic ratio feature and estimation of absorption pulse parameters

Abstract Monitoring of methane from oil and natural gas operations is essential to mitigate environmental risks and promote sustainable practices. The method is an effective tool to detect methane presence and compute methane concentration under condition of low signal to noise ratio. The proposed adaptive time synchronous moving average, multiband decomposition and pulse signal fusion of laser signal allow to reduce measurement noise. The harmonic ratio features and time-amplitude features improve taking into consideration the properties of the absorption signal and pulse allow to improve the methane estimation accuracy. Experimental results show that the methane detection accuracy and classification accuracy are equal to 100% and 86.5%, respectively. The estimation accuracy of methane concentration in terms of R² value equal to 0.995.