A computer-implemented method using a reinforcement learning trained reduced order estimator (RL-trained ROE) and a closure model is provided for controlling a heating, ventilation, and air conditioning (HVAC) system including actuators. The method uses a processor coupled with a memory storing instructions implementing the method, wherein the instructions, when executed by the processor, carry out at steps of the method, includes acquiring setpoints of the HVAC system from a user input and measurement data from sensors arranged in the HVAC system,

computing a high-dimensional state estimate using the measurement data and an estimate of reduced-order state from the RL-trained ROE, determining a controller with respect to the setpoints by using the RL-trained ROE, generating control commands based on the controller, and transmitting the control commands to the actuators of HVAC system via an output interface.

A nonlinear dynamic control system is defined by a set of equations that include a state vector and one or more control inputs. Via a machine learning method, a sub-optimal controller is derived that stabilizes the nonlinear dynamic control system at an equilibrium point. The sub-optimal controller is retrained to be used as a stabilizing controller for the nonlinear dynamic control system under general operating conditions.

The present disclosure provides a method for managing a pipe network of natural gas. The method may comprise: obtaining pipe network information of natural gas in at least one area, the pipe network information including a running time of a system of the pipe network of the natural gas and gas leakage information of the pipe network; extracting feature information based on the running time and the gas leakage information; predicting a maintenance time of the pipe network by inputting the feature information into a maintenance time prediction model.

A disturbance estimation apparatus that includes a position data receiver, a thrust data receiver, and processing circuitry is provided. The position data receiver receives position data indicating a position of a ship. The thrust data receiver receives thrust data indicating a thrust force driving the ship during navigation. The processing circuitry determines a magnitude of the thrust force based on the thrust data, and determines, based on the position data, disturbance data including a drift direction in which the ship drifts due to an external force and a drift speed of the ship while the thrust force is less than a threshold value. The processing circuitry outputs the disturbance data that indicates disturbance acting on the ship and assists to control movement of the ship for automatically maintaining a selected position or heading direction of the ship.

A computing device and/or computer-implemented method determines whether acoustic data associated with operation of a cleaning machine represents normal or anomalous operation of the cleaning machine. In some examples, upon determining that the acoustic data represents anomalous operation of the cleaning machine, the computing device further identifies a root cause(s) of the anomalous operation and/or one or more remedial actions that may be taken in response the root cause of the anomalous operation.

Implementations described and claimed herein provide systems and methods for a scripting technique to clone equation-oriented models of a modeled system for parallel simulation of the modeled system. The multiple equation-oriented models may be solved in parallel to quickly create an optimized solution for different operating conditions by providing different input variable sets to the cloned equation-oriented models. The multiple equation-oriented models may provide real-time optimization of the modeled system to provide continuous optimization of all controls or handles of the system to help achieve a target performance of the system. The equation-oriented models may also provide a nomination tool to predict the output of the system over a nomination period with different input variables and performance monitoring capabilities of the system. Offline “what-if” simulations may also be executed on the equation-oriented modeling system to aid operators in predicting performance of the modeled system and troubleshoot potential problems.

A computer-implemented method for digital product networks includes storing a usage data structure that includes usage data related to use of each digital product, a sensor data structure that includes sensor data collected from sensors related to each digital product, and a derived data structure that includes data derived from the usage data and the sensor data. The method includes generating product level data based on at least one of the usage data structure, the sensor data structure, or the derived data structure. The method includes encoding the product level data as a product level data structure configured to convey parameters indicated by the product level data across a set of digital products. The method includes writing the product level data structure to at least one of a product memory of the set of digital products or a control tower memory of a product network control tower.

A computer-implemented method for executing a quantum computing task includes providing a quantum computing system. The computer-implemented method includes receiving a request, from a quantum computing client, to execute a quantum computing task via the quantum computing system. The computer-implemented method includes executing the requested quantum computing task via the quantum computing system. The executing the requested quantum computing task includes trapping a set of ions. The computer-implemented method includes returning a response related to the executed quantum computing task to the quantum computing client.

Possible input value combinations of a prediction of an engineered system are iterated over, comprising, for a possible input value combination: selecting an action to perform on the engineered system for the possible input value combination, comprising: performing a plurality of predictions of the engineered system scored by evaluating an objective function associated with the engineered system and using the possible input value combination and a corresponding plurality of actions. The action is selected from the corresponding plurality of actions, the selection being based at least in part on scores of the plurality of predictions. A rule specifying a corresponding set of one or more rule conditions that is met when the possible input value combination is matched and a corresponding action associated with the rule as a selected action is generated. The generated set of rules to be stored or further processed is output.

A priori geo-referenced data is obtained for a worksite, along with field data that is collected by a sensor on a work machine that is performing an operation at the worksite. A predictive model is generated, while the machine is performing the operation, based on the geo-referenced data and the field data. A model quality metric is generated for the predictive model and is used to determine whether the predictive model is a qualified predicative model. If so, a control system controls a subsystem of the work machine, using the qualified predictive model, and a position of the work machine, to perform the operation.