Signal Processing for Error Reduction in a Hot-Wire Air Flow Sensor

The hot-wire(H/W) element of an air flow sensor has a tendency to indicate lower than actual flow rate when subjected to large-amplitude oscillatory flow. First, the causes for this measurement error is investigated and a new dynamic model for H/W is proposed. The dynamic model is a combination of the static non-linearity output characteristic of the H/W element and the dynamic response of the H/W element. Second, a signal processing method is developed to correct the measurement error of the H/W by inversing the proposed H/W dynamic model. The result shows that by using this signal processing method, the measurement error of the H/W is reduced.