Abstract: A method and controller system are described for automatically tuning a controller of a drilling rig. The controller system includes a controller that determines a moving average of measured values, determines a moving standard deviation of the measured values, defines an outer zone of measured values with the determined moving average and the determined moving standard deviation, defines an inner zone of measured values with the determined moving average and the determined moving standard deviation, monitors the measured values for the occurrence of statistical events with respect to the outer zone and inner zone. Based on the detected statistical event, the controller adjusts the gain of the controller the based on the statistical event being associated with the outer zone or inner zone.

Abstract: Systems and methods are provided for using a Deep Reinforcement Learning (DRL) agent to provide adaptive tuning of process controllers, such as Proportional-Integral-Derivative (PID) controllers. The agent can monitor process controller performance, and if unsatisfactory, can attempt to improve it by making incremental changes to the tuning parameters for the process controller. The effect of a tuning change can then be observed by the agent and used to update the agent's process controller tuning policy. It has been unexpectedly discovered that providing adaptive tuning based on incremental changes in tuning parameters, as opposed to making changes independent of current values of the tuning parameters, can provide enhanced or improved control over a controlled variable of a process.

Abstract: Systems and methods are provided for arranging processing units in a common volume to allow for processing of a fluid flow as part of a mass and/or heat transfer process. Fuel cells are examples of processing units that include separate flow paths for processing two input fluid flows with mass and/or heat transfer between the separate flow paths. The arrangements described herein can allow a gas phase fluid flow to be delivered to a first process flow path of processing units in a common volume. The gas phase fluid flow can be delivered in a relatively uniform manner without the use of an intervening manifold to distribute gas from the common volume into the processing units.

Abstract: In various aspects, systems and methods are provided for integrating molten carbonate fuel cells with a fired heater for production of electrical power while also reducing or minimizing the amount of CO2 present in the flue gas generated by the fired heater. The molten carbonate fuel cells can be integrated for use with fired heater so that at least a portion of the flue gas from fired heater flows through cathodes of the fuel cells and at least a portion of the cathode exhaust is returned to a convection section of the fired heater.

Abstract: Systems and methods are provided for using a Deep Reinforcement Learning (DRL) agent to provide adaptive tuning of process controllers, such as Proportional-Integral-Derivative (PID) controllers. The agent can monitor process controller performance, and if unsatisfactory, can attempt to improve it by making incremental changes to the tuning parameters for the process controller. The effect of a tuning change can then be observed by the agent and used to update the agent's process controller tuning policy. It has been unexpectedly discovered that providing adaptive tuning based on incremental changes in tuning parameters, as opposed to making changes independent of current values of the tuning parameters, can provide enhanced or improved control over a controlled variable of a process.

Abstract: A controller system, comprising: a controller configured to control toward a desired process value, an arrangement to repetitively measure a value of the desired process value; wherein the controller is configured to execute a routine that: determines a moving average of the measured values, determines a moving standard deviation of the measured values, defines an outer zone of measured values with the determined moving average and a first plurality of the determined moving standard deviation, defines an inner zone of measured values with the determined moving average and a second plurality of the determined moving standard deviation, monitors the measured values for the occurrence of a first statistical event with respect to the outer zone and adjusting a gain of the controller by a first factor upon detection of the first statistical event, monitors the measured values for the occurrence of a second statistical event with respect to the inner zone and adjusting a gain of the controller by a second factor up

Abstract: In various aspects, systems and methods are provided for integrating molten carbonate fuel cells with a fired heater for production of electrical power while also reducing or minimizing the amount of CO2 present in the flue gas generated by the fired heater. The molten carbonate fuel cells can be integrated for use with fired heater so that at least a portion of the flue gas from fired heater flows through cathodes of the fuel cells and at least a portion of the cathode exhaust is returned to a convection section of the fired heater.

Abstract: Systems and methods are provided for arranging processing units in a common volume to allow for processing of a fluid flow as part of a mass and/or heat transfer process. Fuel cells are examples of processing units that include separate flow paths for processing two input fluid flows with mass and/or heat transfer between the separate flow paths. The arrangements described herein can allow a gas phase fluid flow to be delivered to a first process flow path of processing units in a common volume. The gas phase fluid flow can be delivered in a relatively uniform manner without the use of an intervening manifold to distribute gas from the common volume into the processing units.

Abstract: A method and apparatus for steady-state target calculation that explicitly accounts for model uncertainty is disclosed. In accordance with one aspect of the invention, when model uncertainty is incorporated, the linear program associated with the steady-state target calculation can be recast as a highly structured nonlinear program. In accordance with another aspect of the invention, primal-dual interior point methods can be applied to take advantage of the resulting special structure. For a system having characteristic gain parameters G having a known uncertainty description, the present invention provides a method and apparatus for selecting steady-state targets for said system-manipulated variables such that all system-controlled variables will remain feasible at steady-state for all possible values of the parameters G within the known uncertainty description.