Artificial Engineering Intelligence (AeI) - Connected Microservices for Drilling Optimization and Real Time Well Engineering

Abstract

Abstract As we develop autonomous drilling system the setpoint operational and control, parameters for drilling are to be determined. The decision-making control loop performance should be faster. As the system with high aspect ratio i.e., length over diameter is highly stochastic, calculating the parameters for the optimal condition is very difficult. The paper presents methodology to obtain the envelop of sufficed, satisfied, and optimized engineering conditions resulting in "optosatisficed" condition aka "artificial engineering intelligence" which is good enough or an acceptable solution or a reasonable solution under the given condition. The paper presents an adaptive model which connects different engineering models in the form of microservices for constraints rather than a complete theoretical optimal closed form solution-based model. The approach to be taken under this condition is to include the stability of the bit, drillstring and wellbore, which will result in the near optimum envelop (desired) for the drilling parameters. Important analyses are coupled - vibration-stability analysis and bit-wear performance analysis. The optimum rate of penetration limiting values are calculated as the well is drilled with specified interval by considering the bit wear and the vibration. At each drilled interval, it is assumed that the bit wears out continuously and the amount of wear depends on the uncontrolled and uncertain parameters such as formation, anisotropy, dip etc. Also, all the calculations such as hole cleaning mechanical specific energy are calculated continuously. The analysis results have shown that the convergence was very quick in obtaining optimal solution and the predictability in the test wells have shown best solution results under uncertainty. Also, it has been found that the results provide reasonable threshold values when more and more data are used as the well is drilled. The stability envelop is shown in the figure with multiple connected engineering microservices. As long as the driller stays within the operational region, the results have shown that the operating parameters are satisfying and good enough solution for the desirable outcome. In other words, near normal engineering solution is achieved. Updating of the stability plot at every discrete drilled interval provides the pattern of the envelop for further decision-making set points for the controller in the case of an autonomous system. This paper provides a new framework for faster decision strategy. Injecting life to the well design software itself will provide amazing power. When a well is designed or monitored or drilled, well plan can be made interactive so that user can optimize and take proper decision to drill ahead or automate the operation.