Today and more than ever before, there is a necessity to reduce drilling time and optimize operations. This paper discusses the implementation of a system that monitors rig and equipment parameters and applies machine learning algorithms to reduce NPT, improve equipment efficiency, optimize procedures, forecast equipment life cycles, and conduct preventive maintenance.
The first step was to develop the platform that would gather the rig data and process it before sending it to be analysed. The rig was therefore connected to a platform that uses Internet of Things (IoT) protocols. The noisy and redundant data sets coming from the rig were standardized, filtered and the outliers were removed. Then, a feature selection method was applied to pick out of the data pool the most important parameters for optimization and forecasting. The resulting parameters were then used to train and test the machine learning model.
The processed data was then fed to an in-house developed system that works on four levels to extract additional information:
• The Rig State Classifier allows an identification of the operational phase of the rig.
• The maintenance detector helps identify whenever a maintenance has been conducted.
• The equipment efficiency monitor tracks the variations in equipment efficiencies.
The results from the pilot project shows how operators were able to use the system to predict possible failures, organize preventive maintenance and pre-schedule spare parts delivery. In the end this allowed the rig to operate at a higher level of efficiency increasing the working hours of the equipment while maintaining the load within the design conditions. This ultimately resulted in a decrease in operational time and equipment failures meaning a major reduction in rig down time.
Digital revolution, Industry 4.0, IoT, Machine Learning, Artificial Intelligence. All these terms have become buzzwords in today's energy industry. More specifically, in the drilling industry, the ideas of digitalization and the use of computers to improve operations and facilitate work and communication have been attracting a lot of attention. Today, more than ever, there is a need to reduce operational downtime and maintenance costs, improve drilling efficiency and optimize rig logistics. In short, drill faster, drill better, and all this at a lower cost. In this paper, a Data Management System was developed and installed as the focal point for a Real-Time Operations Centre (RTOC). The system uses IoT technology that applies machine learning to analyse rig sensor data. The system essentially turns drilling rigs and equipment into smart devices and finally brings these key tools of the energy industry into the digital age.
