Prediction of oil well production using multi-neural networks

This study presents an application of a multiple artificial neural network (MNN) model to estimate the future production performance of oil wells based on monthly-production time series data. Several engineering techniques that require expertise and data that are difficult to obtain are used currently to predict well performance. By comparison, a MNN model relies only on a large amount of available historical data on well production. Single artificial neural network (ANN) models have been used to predict short-term production performance. However, the accuracy of such a model may be seriously compromised when it is used for making long-term multistep predictions. Hence, instead of using single-ANN models, we propose using multiple artificial neural networks to improve accuracy of the model while also Minimizing complexity. A MNN is a group of ANNs that work together to solve a problem. Each ANN makes a Prediction for a different time period The MNN model for prediction of future well performance is applied to the time series data obtained from four pools of wells in the southwestern region of Saskatchewan, Canada The results showed that a MNN model performed well and slightly better than a single-ANN model for long term prediction.