Performance improvement of a near-infrared CH4 detection device using wavelet-denoising-assisted wavelength modulation technique

Abstract

By introducing wavelet-denoising (WD) in the traditional tunable diode laser absorption spectroscopy (TDLAS), a novel CH4 detection device based on WD-assisted wavelength modulation technique was experimentally demonstrated. Formulation and detection procedure were presented, and experiments were carried out to prove the normal function on the extraction of 2f signal from severely polluted differential signal with WD algorithm. With the fabricated device, sensing characteristics were measured on CH4 concentration. The response time is less than 8s within the whole detection range of 0–50kppm. By virtue of the suppression on noises with WD, the minimum detection limit (MDL) is dropped from 4 to 1ppm, the maximum detection error is minimized from 6.2% to 3.8% over the concentration range of 4–50kppm, and the decreased Allan deviation (per an average observing time of 9min for 24h) from 0.13 to 0.08ppm suggests enhanced stability. The proposed device also exhibits improved performances compared with our previously reported sensors based on simple direct spectroscopy absorption (DSA) technique and wide-band incandescence wire-source.