Safety and cost are major concerns in drilling operations, particularly when drilling for unconventional gas and oil occurs in the greater depths and harsher conditions of deepwater environments. Current drilling practices focus on controllable drilling parameters including weight-on-bit (WOB), rotational speeds (RPM) of the bit, and the hydraulic (H) power driving the drilling fluid. However, these parameters have not been thoroughly optimized for improved drilling efficiency. Drillers operate within a range of values for each parameter based on recommendations from service companies, bit manufacturers, or previous field experience. Improving the economics of deep exploration is critical in reducing well development costs and increasing domestic energy supplies. Implementation of computational models and simulation tools to optimize drilling operations will play a key role in achieving cost effective drilling exploration and well development.

The current invention describes an apparatus and method for determining optimized drilling parameters by collecting real time measurements while drilling, taking into account mechanical specific energy (MSE). The computational and simulation tools provide in-time recommendations of drilling parameters including WOB, RPM, and H, to optimize the rate of penetrations while reducing MSE expenditure. The new method addresses shortcomings of other processes used for drilling optimization through predicting MSE for key controllable parameters using combined artificial neural network simulation coupled with physics-empirical modeling to evaluate and control drilling dynamics.