To enhance the accuracy of oxygen concentration measurement and solve the error drift issue in tunable diode laser absorption spectroscopy(TDLAS)oxygen sensor

an information fusion model is designed with temperature

pressure and output voltage as input variables to accurately compensate for the error of TDLAS oxygen sensor.In this paper

the basic principle of the Beer-Lambert law (BLL) is described

and the theoretical model of TDLAS oxygen sensor is constructed.The oxygen concentration error compensation model is designed

and the method for solving model parameters based on the least square method(LSM)is gi

ven

which can effectively reduce the measurement error of the sensor.The calibration experiment and test experiment of the system show that the maximum error of oxygen concentration after compensation is 0.7167%

the average error is 0.2066%

the root mean square error (RMSE) is 0.2719%

and the adjusted

R

2

is 0.9997

all of which are lower than the errors before compensation.Meanwhile

the error compensation model has good goodness of fit and prediction ability.The validity of the error compensation model based on the least square method is verified.