Accelerometer-based Stroke Length Estimation of Sucker-Rod Pumping System in Smart Field

In smart field, innovative technologies and advanced equipment are used to reduce operating costs, monitor wells conditions, optimize the production and improve economic performance. Stroke length is an important parameter in oilfield. It reflects the working condition of production wells. Accurate and effective measurement of stroke length imposes great impacts on oil production. In this paper, a novel method to efficiently perform drift-free stroke length estimation of sucker-rod pumping system with acceleration sensors is proposed. This method includes two stages of integration and two stages of filtering. First, an band-pass infinite impulse response filter is applied to eliminate the drift and sensing errors of acceleration data collected by accelerometer. Then, an integration is calculated to derive velocity signal with the filtered data. After that, a high-pass filter is performed to remove the DC offset of the velocity. Finally, a periodic velocity signal is selected to perform integration operation to compute the stroke length. The proposed method is validated against a real dataset from an oilfield. The experimental results illustrate that the proposed method can reduce sensing errors, suppress interference, and improve estimation accuracy.