Build and validate predictive models in a matter of days that match all your historical production and injection data
Run probabilistic forecasts to estimate P10-P50-P90 production and recovery in minutes for any injection scenario
Quantitatively estimate the impact of producer to injector conversions and prioritize wells to convert
Prioritize which currently down wells to bring online that have the biggest impact on current production
Quantitatively estimate the impact of infill drilling one or multiple wells and create drilling schedules
Quantitatively estimate the impact of changing producer BHPs and thereby identify best candidates to workover
Data Physics models solve for black oil reservoir physics similar to traditional reservoir simulators but is orders of magnitude faster
Understand your field through estimates of connectivity between wells and spatial variation of reservoir quality
Quickly react to unforeseen circumstances such as unplanned downtime or well failures
Get an ensemble of models that all match your historical data to account for uncertainty and measurement errors
Get posterior distributions of reservoir parameters like porosity and permeability that match your historical data
Can be run locally or on Tachyus Cloud, with no infrastructure footprint or upfront compute costs
In traditional reservoir management, various types of predictive models have been applied over the years for quantitative decision making. Such models range from the very simple analytical models (type-curves, etc.) to the very complex reservoir simulation models. Simulation models attempt to model detailed behavior of reservoir physics and integrate all kinds of measured data. However, many issues such as the significant time and effort required to build and calibrate these models, computational complexity, etc. generally make their practical application difficult.
Tachyus’ Data Physics is the amalgamation of the state-of-the-art in machine learning and the same underlying physics present in reservoir simulators. These models can be created as efficiently as machine learning models, integrate all kinds of data, and can be evaluated orders of magnitude faster than full scale simulation models, and since they include similar underlying physics as simulators, they have long term predictive capacity. The Data Physics based waterflood management workflow is very fast and users can go from data preparation and modeling to decisions in a matter of days. Additionally, the models can be quickly updated with new data so that they always remain operationally useful.