Systems, methods, apparatuses, and computer program products for a greenhouse indoor environment controller based on model predictive control (MPC), which can be integrated into existing greenhouse regulatory systems to optimally maintain critical climatic variables, including artificial lighting levels, CO.sub.2, indoor temperature, and humidity levels within acceptable limits. The objectives of the MPC may be to maximize the rate of crop photosynthesis while optimizing the use of the available water and energy resources, taking into account the unpredictability and intermittent nature of renewable energies and external atmospheric conditions. Accordingly, certain embodiments may facilitate the management of greenhouses by anticipating control actions for a better quality of production. For that, mathematical formulations of the optimal control problem may be described, and the numerical results related to the application of the MPC to case studies are described integrating the effects of greenhouse structural considerations and the influence of climate data on its operation.

A control system for a central energy facility with distributed energy storage includes a high level coordinator, a low level airside controller, a central plant controller, and a battery controller. The high level coordinator is configured to perform a high level optimization to generate an airside load profile for an airside system, a subplant load profile for a central plant, and a battery power profile for a battery. The low level airside controller is configured to use the airside load profile to operate airside HVAC equipment of the airside subsystem. The central plant controller is configured to use the subplant load profile to operate central plant equipment of the central plant. The battery controller is configured to use the battery power profile to control an amount of electric energy stored in the battery or discharged from the battery at each of a plurality of time steps in an optimization period.

A method includes receiving a first time-dependent data characterizing measurement by a first sensor operatively coupled to an oil and gas industrial machine; determining a first anomaly score associated with a first portion of the first time-dependent data over a time period, the determination is based on a first value of an operating characteristic over the time period and a second value of the operating characteristic over the time period, wherein the first value of the operating characteristic is calculated from the first time-dependent data and the second value of the operating characteristic is detected at the oil and gas industrial machine; and rendering, in a graphical user interface display space, a visual representation indicative of the first anomaly score. Related apparatus, systems, articles, and techniques are also described.

A building management system includes a processing circuit configured to perform a system identification process to identify one or more parameters of a system model that predicts a behavior of a building system. The one or more parameters include one or more model parameters and one or more Kalman gain parameters. The system identification process includes identifying the one or more model parameters, generating an initial guess of the one or more Kalman gain parameters based on the training data and results of a simulation that uses the one or more model parameters, and identifying the one or more Kalman gain parameters by initializing a prediction error minimization problem with the initial guess. The building management system also includes a controller configured to control building equipment to affect the behavior of the building system based on predictions of the system model.

A computer-implemented method of training a model for making time-series predictions of a computer-controlled system. The model uses a stochastic differential equation (SDE) comprising a drift component and a diffusion component. The drift component has a predefined part representing domain knowledge, that is received as an input to the training; and a trainable part. When training the model, values of the set of SDE variables at a current time point are predicted based on their values at a previous time point, and based on this, the model is refined. In order to predict the values of the set of SDE variables, the predefined part of the drift component is evaluated to get a first drift, and the first drift is combined with a second drift obtained by evaluating the trainable part of the drift component.

This disclosure includes systems and methods for optimizing aeration in wastewater treatment. The techniques described herein include receiving data for a wastewater treatment plant, the data being descriptive of water quality over a period of time. The techniques further include developing a predictive model for future water quality based on the received data. The techniques also include determining, based on the predictive model, a plurality of DO setpoints and airflow rates for the wastewater treatment plant. The techniques further include controlling an aeration system for the wastewater treatment plant using the plurality of DO setpoints and the airflow rates.

A state estimation method for a heat supply network in a steady state based on a bilateral equivalent model is provided. The method includes: establishing the bilateral equivalent model based on a mass flow rate in each supply branch of the heating network, a mass flow rate in each return branch of the heating network, a mass flow rate in each connecting branch of the heating network, a pressure and a temperature of each node in the heating network, wherein each heat source is configured as a connecting branch and each heat load is configured as a connecting branch; and repeatedly performing a state estimation on the heating network based on the bilateral equivalent model, until a coverage state estimation result is acquired.

A computing system is provided for training one or more machine learning models to perform at least a portion of a robotic task of a physical robotic system by monitoring a model-based control algorithm associated with the physical robotic system perform at least a portion of the robotic task. One or more robotic task predictions may be defined, via the one or more machine learning models, based upon, at least in part, the training of the one or more machine learning models. The one or more robotic task predictions may be provided to the model-based control algorithm associated with the physical robotic system. The robotic task may be performed, via the model-based control algorithm associated with the robotic system, on the physical robotic system based upon, at least in part, the one or more robotic task predictions defined by the one or more machine learning models.

Parameters are efficiently calculated. An information processing apparatus (1) includes a corresponding data calculation unit (2) configured to determine importance of each sample in accordance with a difference between a plurality of pieces of observation information observed when an input is given to an observation target and data of a second type generated by a simulator that simulates the observation target based on a sample of a parameter with respect to the plurality of samples and data of a first type indicating the input, and a contribution degree of each of the pieces of observation information in the plurality of pieces of observation information, and calculate data that corresponds to distribution of the parameters; and a new parameter sample generation unit (3) configured to generate a new sample of the parameters in accordance with predetermined processing using data that corresponds to distribution of the parameters.

The embodiments described herein include one embodiment that provides a control method that includes connecting a first controller to a control system; receiving control system configuration data from a database, in which the configuration data comprises holistic state data of a second controller in the control system; and configuring operation of the first controller based at least in part on the configuration data received.