Back to results
US 2020/0370423 A1Published application
Controller optimization via reinforcement learning on asset avatar
AssigneeSchlumberger Technology Corporation
InventorsZiyao Li, Paul Muller, Prasanna Nirgudkar
Life of this patent
Estimating the term…
Abstract
A method can include receiving sensor data from a system; encoding the sensor data to a latent space representation via a trained encoder; generating a control action using the latent space representation; and issuing an instruction that corresponds to the control action for control of the system.
Claims20
A method comprising:
- receiving sensor data from a system;
- encoding the sensor data to a latent space representation via a trained encoder;
- generating a control action using the latent space representation;
- and
- issuing an instruction that corresponds to the control action for control of the system.
2 more elements
A system comprising:
- a real time data interface;
- a compressor operatively coupled to the real time data interface;
- and
- a controller operatively coupled to the compressor, wherein the compressor receives real time data via the real time data interface and outputs reduced dimensionality data of the real time data and wherein the controller generates a control action using the reduced dimensionality data.
1 more element
One or more computer-readable storage media comprising processor-executable instructions executable to instruct a computing system to:
- receive sensor data from a system;
- encode the sensor data to a latent space representation via a trained encoder;
- generate a control action using the latent space representation;
- and
- issue an instruction that corresponds to the control action for control of the system.
2 more elements
Dependencies are inferred from the claim text.