Implementation of Digital Signal Processing Techniques in the Design of Thermal Pulse Flowmeters

Various digital signal processing methods, which could be applicable to the design of a microcomputer based thermal pulse flowmeter, were examined. Two excitation methods were investigated: a single thermal pulse and a pseudorandom binary sequence signal (PRBS). The signal recovered downstream was processed by two alternative numerical algorithms to recover the “time of flight”, i.e., by peak detection of the signal itself and the peak of the differentiated signal. The recovered thermal pulse and the “time of flight” were then used to test the validity of two models: a diffusion-advection model and a simple time delay model. The delay model was found compatible with the data especially when the peak of the output signal derivative was used as a marker for determining the “time of flight.” The single pulse injection method was found, in general, superior to the PRBS cross-correlation technique except for the ability of the latter to provide early indication of flow-rate variations.