State-of-the-art, hybrid-AI models for any physical system
The second step is all about uncovering the real dynamics, correlations and uncertainties in your production system - to truly understand its behavior and continuous development.
Squashed and Washed real-time data is leveraged by our state-of-the-art hybrid-AI solution, which orchestrates multiple machine learning models working in parallel. We combine hierarchical neural networks with first-principle physics to learn the real dynamics, correlations and uncertainties in the system. In operation, as new data becomes available, online learning ensures that the models are automatically calibrated.
Our solution provides situational awareness and support production enhancing applications in many domains, from the oil- and gas industry to fish farming.
In oil and gas production system, our hybrid-AI solutions can be used to assess the quality of the production setting, analyze the well response to control changes, study and track the result and uncertainty of well tests in real-time, flag underperforming wells, monitor and predict slugging, and calculate allocated flow rates for all wells in your production system. The simple interface of our models also facilitate connections to conventional flow simulators, forming hybrid-AI solution and expert knowledge, reinforcing the predictive capabilities of the hybrid-AI.