Prediction and Optimization of ROP Assisted by Adjacent Well Data Based on Geological and Engineering Driving

The main lithology of the Shahejie Formation reservoir in the central Bohai Sea is mudstone and dark sandy mudstone, and the rate of penetration (ROP) is generally low when drilling through this reservoir, significantly influencing the drilling cycle and costs. To address this issue, a ROP prediction and optimization model integrating geology and engineering was proposed. This model consisted of two parts: ROP prediction and ROP optimization. The ROP prediction leveraged geological and engineering data to establish an ROP prediction model of the drilled well assisted by adjacent well data. After completing the ROP prediction, a feature contribution coefficient was defined to quantify the influence of different feature parameters on the final result. This feature contribution coefficient allowed for both an interpretation of the predicted results and the identification of controllable parameters that significantly affect ROP. For these controllable parameters, the ROP optimization model used a grid search optimization algorithm to explore the optimal parameter combination, thereby improving ROP. The ROP optimization results based on this model show that the ROP of the test well increases by 6.34% on average, with the three parameters contributing most to the prediction results being gamma values, weight on bit, and bit drilling time. This model effectively integrates geological and engineering parameters, achieving high-accuracy ROP predictions and substantial ROP improvements and providing valuable guidance for ROP enhancement in two development wells where it has been applied.