Prediction Model of Shallow Geological Hazards in Deepwater Drilling Based on a Hybrid Computational Approach

Deep-water drilling shallow geological hazard prediction technology is a key technology in deep-water exploration. The existing prediction methods mainly rely on deep-water seismic interpretation, which has limited accuracy and high cost. At present, there are fewer prediction models based on experimental studies. Based on experimental research in the laboratory, this paper presents a new prediction model. The ant colony optimization algorithm is used to optimize the P-wave velocity of shallow gas/shallow water flow, and then through the analysis of the control equations, the hybrid calculation achieves the optimal solution to the data discriminant degree and corresponding risk. By establishing the control equation, based on the experimental data, the hazard levels of shallow gas/shallow water flow with different pressures in the same structural area are calculated. This model takes into account shallow geologic hazard types, P-wave velocity, pressure and flow related objective functions. The model can identify data with shallow gas/shallow water flow characteristics from a large number of seismic analysis data, and perform risk level judgment and warning. In the recent 3 years, the prediction model proposed has been applied in the South China Sea. The LS field using the technology thus saving cost of detailed investigation stage.