Machine learning prediction of wellhead growth in gas well during production stage

• Main Authors: Ismail, Zool Hilmi, Elfakharany, Ahmed, Risal, Abdul Rahim
• Format: Conference or Workshop Item
• Published: 2022
• Subjects: QD Chemistry
• Online Access: http://eprints.utm.my/id/eprint/98901/; http://dx.doi.org/10.23919/ASCC56756.2022.9828232

In this paper, a machine learning approach is developed to predict wellhead growth in High-Pressure-High-Temperature (HPHT) gas well. The method relies on measurements from three laser range finding sensors to calculate the tilting angles and growth of the wellhead. The three laser sensors are pointed at the wellhead at three different positions. One of the laser sensors is used to measure the tilting around the X axis, the second one is used to measure the tilting around the y axis and the measurement of the final sensor is combined with the tilting angles to estimate the growth. The growth is estimated using a deep neural network model. To evaluate the proposed ML algorithm, we built a simulation environment that simulates the movement of the wellhead and the measurements from the sensor nodes.