Signal processing for temperature measurement in harsh environments using diode laser absorption spectroscopy

The environment of engine combustion features strong vibration, strong self luminous and large numbers of scattering particles. These factors affect the TDLAS signal quality severely, and appropriate signal processing is extremely important to improve the signal to noise ratio and the reliability of TDLAS measurement result in harsh environments. In this paper, we analyze the noise sources and their properties of TDLAS technique in engine combustion and present post signal processing method both for direct absorption and wavelength modulation techniques. The general signal processing procedures include digital filter, gross error elimination and multi-cycle averaging. For direct absorption signal, multi-peaks Lorentzian function is used to fit the line shapes. And for wavelength modulation absorption signal, the calibration-free method by extracting laser intensity and detector background signal is presented. The signal processing method is demonstrated in the harsh environments of a ground test scramjet combustor and it is suitable for developing the in-situ TDLAS sensors.