Seismic interpretation is a common method for predicting deepwater shallow geological hazards before drilling. Many onsite application results show that this method has the following limitations such as hard to quantitatively classify the impact of shallow geological hazards to drilling operation based on the data and very expensive. Based on the laboratory experiment study, this paper proposes a new prediction model by using a hybrid computational approach and the prediction model can effectively solve the problems. The prediction model proposed in this paper has been applied in the South China Sea Lingshui 17–2 Deepwater, and the prediction accuracy is more than 90%.


Due to the diapiric structure of the Lishui area in the western part of the South China Sea, the shallow gas of the shallow formation in this area is active. Due to the shallow burial depth of shallow gas, the drilling work at that stage does not involve blowout prevention. Once the shallow gas blows out during drilling work that the well cannot be shut down, drilling becomes very dangerous. In addition, the deepwater drilling work in the South China Sea is full of challenges, and the well structure is very complicated. If the prediction of shallow gas is not accurate, the shallow casing level increases, the work space is subsequently reduced, and the work risks and difficulty increase. Therefore, drilling risk assessment and prevention technology with a higher accuracy are key to the safety assessment and risk control technology of well control in deepwater drilling.

The characteristics of the seismic velocity response differ in normal formations and shallow gas formations. Based on this principle, this paper designs a simulation experiment. Through the simulation experiment, this paper investigates the differences between the longitudinal wave velocity trend lines of the shallow gas formation and a normal formation and analyzes the amplitude and range by which the shallow gas longitudinal wave velocity deviates from that of the normal formation.

This content is only available via PDF.
You can access this article if you purchase or spend a download.