Use Case - Reservoir Predictive Model
The reservoir & production engineers key challenge is to maximize hydrocarbon recovery by optimally minimizing both operation time and incurred cost.
However the prior to undertaking any recovery operation, the management is more keen to know answers to the following questions;
- How large are the reserves?
- What will the primary recover be?
- What grade of crude will be produced?
- What will the market pay for it?
Now prestack seismic data and well information allow user to undertake quantitative interpretation to predict lithology and fluid content of the seismic away from the well.
IPMS helps clients to answer above questions by anticipating volumetric and phase behavior of the produced hydrocarbons as they travel from the reservoir, up to the tubing, through surface separators and finally into pipelines. IPMS mines the data collected from relevant sensors and performs a data driven simulation to conduct a predictive analysis modeling for selected duration and provide visibility in terms of financial impact as well as key operational parameters.
How does IPMS resolve the solution?
We divide collected data into two data sets, one is considered as sample (benchmark) dataset and other is marked as test dataset, by implementing intellectual proprietary mathematical and engineering complex algorithms on sample dataset along with machine learning & deep learning on sample data to perform predictive analytics, to verify model accuracy and reliability, we apply designed model on test datasets, by comparing results on test and sample dataset. This helps to evaluate model accuracy and reliability. After conducting successful predictive modeling using various scenarios and parameters, IPMS deploys finalized predictive modelling on live streaming of the data, which automates reservoir & production analysis and allows management to monitor trends with one click by superimposing the actual production data stream onto the predictive model.