Kick Warning and Pressure Inversion Method Based on Wellhead Pressure Signal

ABSTRACT: Kick detection seriously restricts safe and efficient drilling, so accurate and timely warning and analysis are very important. This paper is based on drilling data and optimizes parameters of wavelet analysis such as mother wavelet function, data frequency, and data length. Using the wavelet analysis method, the wellhead pressure signal undergoes decomposition. The obtained signals are combined with logic rules to constrain the intelligent kick warning model, and the final model accuracy reaches 90.08%. To further analyze the kick parameters, spectral analysis is used to quantitatively extract the wellhead kick pressure fluctuations which is used to set the wellhead boundary. By integrating perturbation theory with the governing equation, an analytical solution for pressure propagation and attenuation in the liquid phase is proposed. This analytical solution is applied to the flow model, enabling real-time solutions and analyses of bottom-hole kick fluctuation pressure and volume. The final model demonstrates a relative error of 0.4%. This model is expected to provide new methods for the detection and control of kick. 1. INTRODUCTION Drilling is the only way to create oil and gas channels, and the process is complex and changeable. Due to factors such as inaccurate formation information prediction and improper operations, drilling risks such as the kick, lost circulation, and blowout can easily occur (Vandenbussche et al., 2012). Among them, kick is one of the common drilling risks. The fundamental reason for the kick is that the bottom hole pressure is lower than the formation pressure during the drilling process. Then the formation fluid invades the wellbore, causing the volume of fluid in the wellbore to instantly increase and kick from the wellhead. When a kick occurs, if it is not detected in time or dealt with too late, it is likely to induce a blowout accident(Aime et al., 1967; Han et al., 2017). In addition, when there is gas in the invading fluid, the bottom hole gas continues to rise with the drilling fluid, the pressure and temperature decrease, and the gas expansion accelerates. As a result, the closer to the wellhead, the greater the drilling fluid flow rate, and the more difficult it is to control the wellbore pressure (Alhuthali et al., 2010). For example, the "12.23" blowout accident in Kaixian, Chongqing, China(2003), and the blowout in the Gulf of Mexico in the United States(2010), both caused serious consequences such as equipment damage, casualties, environmental pollution, and scrapping of oil wells (Rathnayaka et al., 2012). Therefore, kick has become one of the recognized high-frequency and high-risk accidents in the drilling industry (Fu et al., 2015). Rapidly detecting kick signals and analyzing the kick process are important means to help quickly deal with kick risks.