Acoustic proximity ranging and its applications to cavity thickness monitoring.

摘要

Time-of-flight (ToF) based acoustic proximity ranging is widely used in many applications. In this dissertation, its applications to cavity thickness monitoring of a supercavitating vehicle are studied. Most of the currently available ToF based acoustic ranging systems are not directly applicable to this case due to their low measurement accuracy and low parameter update rate. New measurement schemes and the corresponding signal processing approaches need to be devised and their performance evaluated for this challenging practical problem. Based on this motivation, four proximity ranging methods, namely the phase-shift approach, the multifrequency technique, the PEARS (Parameter Estimation for Acoustic Ranging Systems) scheme, and the Multi-PEARS (Multi-echo Parameter Estimation for Acoustic Ranging Systems) algorithm, are proposed and investigated.; For the phase-shift approach, two frequencies are used and the measurements are assumed to be corrupted by colored Gaussian noise. By taking the a priori knowledge of the acoustically hard reflection into account, a new time delay estimation algorithm based on the maximum-likelihood (ML) theory is derived. It is shown that our new method outperforms the traditional method in terms of both the estimation accuracy and the robustness against data model mismatch. For the multi-frequency technique, an arbitrary number of frequencies is used. A novel time delay estimator based on the nonlinear least squares (NLS) fitting criterion is derived. To minimize the highly oscillatory cost function, an efficient two-stage estimation algorithm is proposed. Numerical examples show that for a fixed frequency interval and a fixed signal-to-noise ratio (SNR), the more frequencies used, the lower the SNR threshold and the higher the estimation accuracy that can be obtained. Inspired by the multi-frequency technique, the PEARS scheme is devised. PEARS is novel in that it is applicable to arbitrary transmitted waveforms as long as the waveform is periodic. Numerical examples and the experiments performed by using commercially available ultrasonic transducers are used to demonstrate the excellent performance of PEARS. To deal with the interference of secondary echoes observed in experiments, the Multi-PEARS algorithm is presented for the joint proximity ranging and secondary echo mitigation. Both numerical and experimental results verify that Multi-PEARS can provide very accurate distance measurements even in the presence of strong secondary echoes. Finally, ranging experiments are conducted in the air-water tunnel at different flow conditions. Experimental results show that the distance between the sensors and the air-water interface can be accurately measured by using the ranging techniques developed in this work.