This research describes an innovative data-driven approach for estimating mechanical-specific energy (MSE) in the context of geothermal heat pump well drilling in Saxony, Germany. Due to their remarkable hardness, the geological formations in this region present particular challenges. A web-based automated simulation was constructed using datasets generously donated by Technische Universität Freiberg and real-time measurements obtained from the dedicated Wissenschaftliches Bohren in Sachsen (WiBOS) drilling rig. To improve MSE predictions, the simulation uses machine learning techniques such as uncertainty analysis. To accomplish this, this web-based software is divided into four independent phases: data pre-processing, interval automation, visualization, and energy prediction. Notably, the Data Pre-Processing phase detects and manages outliers using specialized algorithms and statistical models, whilst the Intervals Automation phase provides mechanical-specific energy values for depth intervals. Furthermore, this work investigates the complexities of energy prediction by taking geological formations, drilling depths, and MSE values into account. The technology enables drilling specialists to find correlations, interpret data, and make informed judgments in this difficult geological setting. This research contributes to more efficient and cost-effective drilling operations in Saxony by providing valuable insights for the geothermal heat pump well drilling business in this region by delivering a data-driven approach to MSE prediction and energy management.

Press the Back button in your browser, or search again.

Copyright 2024, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.