Acoustic-Based Edge Computing Method for Oil Well Liquid Level Depth Measurement

A novel method for measuring the dynamic liquid level in oil wells based on embedded edge computing is proposed to address the limitations of traditional acoustic methods, including high transmission bandwidth requirements, excessive storage consumption, inaccurate depth calculations, and delayed result updates. This method estimates the power spectral density (PSD) of the raw coupling echo signal using an indirect approach implemented on a microcontroller, followed by the extraction of the acoustic wave velocity in the oil well through an automatic periodic peak recognition algorithm. The liquid level echo time is then determined using the Butterworth low-pass filter (BLPF) and an energy-based peak and valley detection algorithm, enabling accurate liquid level depth calculation. The proposed method achieves real-time acquisition and processing of acoustic wave signals at the edge. Laboratory and field tests were conducted using multiple sets of acoustic wave signals to validate the method’s performance. Experimental results demonstrated that the proposed method achieves stable depth measurements under varying sound velocities, depths, and noise interference conditions. Compared to existing signal processing techniques, the proposed method achieves an average relative error of approximately 0.19% and enhances the system’s real-time performance by a factor of 4.2, effectively replacing traditional workflows involving full-wave collection, transmission, storage, and processing. This method reduces transmission load, improves measurement accuracy, and enhances real-time performance.