Sand rate model and data processing method for non-intrusive ultrasonic sand monitoring in flow pipeline

Abstract Sand production is a critical issue during oil and gas production from sandstone reservoirs. Uncontrolled sand production not only poses the risk of well failure, but also can cause extensive damage to surface and subsurface facilities such as tubing, pumps, valves and pipelines. In recent decades, research on sand production has been conducted in several fronts including sanding prediction, sand monitoring, sand control and well-bore integrity analysis to prevent or alleviate sand production and its consequences. This paper mainly focuses on sand monitoring based on non-intrusive ultrasonic sensor which produces real-time information that can be used for maximizing the safe production of hydrocarbon. We used non-intrusive ultrasonic sensor to monitor the acoustic signals generated by sand particles impacting the pipe wall, and developed a methodology for processing acoustic signals based on the kinetic energy of sand particles in the pipeline. Further, we developed a procedure for identifying and filtering acoustic noise from unrelated events. We validated the proposed methodology for signal processing against experimental data. The results indicated that the de-noising algorithm could filter out the noise from the acoustic data and the model was effective for assessing the sand rate.