BUILD RATE PREDICTION USING ARTIFICIAL NEURAL NETWORK FOR POSITIVE DISPLACEMENT MOTOR APPLICATION IN FIELD X

One of the critical issue in drilling activity is well steering. Well trajectory as the result of well steering can affect well placement in reservoir, completion issue and anti-collision issue, etc. Steering wellbore to the wrong trajectory can cause damage and increase drilling cost. Directional driller performs well steering by giving steering command and controlling drilling parameters. This command is adjusted based on steering behavior of drilling BHA. Steering behavior is the ability of the drilling BHA in deviating wellbore based on given steering command and drilling parameter. By understanding the steering behavior of drilling BHA, directional driller can predict build rate and turn rate produced so accurate well trajectory can be accomplished. Several factors that affect steering behavior are steering command, formation characteristic, drilling assembly mechanism and drilling parameters. Obstacle in understanding steering behavior is the absence of correlation that connects each factor. Artificial Neural Network (ANN) is a tool that can find the relation between input parameters and output parameter without generating correlation, and use new input data to predict the value of the output. This research shows that Artificial Neural Network can be used as a tool to analyze steering behavior and predict build rate based on steering behavior. Using formation characteristic, steering mode, weight on bit, rotary speeds, jet impact force, motor bent angle and stabilizer size from 10 wells in field X as input parameters, ANN generates a model which later validated in predicting build rate from new dataset. The good agreement between prediction data and the actual data is showed in the results.