Abstract: A method includes operating equipment to affect a variable state or condition of a space and training weights of a neural network by outputting control dispatches from the neural network based on simulated inputs, simulating costs associated with operating the equipment in accordance with control dispatches, and adjusting the weights to reduce the costs. The method also includes generating a control dispatch for the equipment by applying an actual measurement relating to the space as an input to the neural network, and controlling the equipment in accordance with the control dispatch.

Abstract: A heating ventilating or air conditioning (HVAC) system includes a unit configured for use in a heating or a cooling operation for an environment. The unit includes an interface configured to receive electric energy from an alternative energy source and a controller configured to perform an optimization of an objective function for the unit to determine a first amount of electric energy to receive from an energy grid and a second amount of the electric energy from the alternative energy source for use in the heating or cooling operation at each time step of an optimization period in response to a cost characteristic of the electric energy received from the energy grid. The unit is configured to use the first amount and the second amount for the heating or cooling operation.

Abstract: A method includes determining control setpoints for equipment based on a time-varying availability of green energy and revenue from an incentive program of an energy provider. The method also includes controlling the equipment using the control setpoints.

Abstract: A method for correcting faults in a building management system (BMS) includes receiving time series data characterizing an operating performance of one or more BMS devices, generating a first fault detection result by processing the time series data using a first fault detection technique, generating a second fault detection result that conflicts with the first fault detection result by processing the time series data using a second fault detection technique different than the first fault detection technique, resolving a conflict between the first fault detection result and the second fault detection result by applying both the first and second fault detection results as inputs to a neural network configured to output an indication of whether a fault condition is occurring in the BMS, and initiating an action to resolve the fault condition in response to the indication indicating that the fault condition is occurring in the BMS.

Abstract: A predictive controller for HVAC equipment obtains a control objective or a constraint that defines an amount of energy consumed by both waterside equipment and airside equipment during each time step of a time period as a summation of multiple equipment-specific energy components including, for each time step, a waterside energy component indicating an amount of energy consumed by the waterside equipment during the time step and an airside energy component indicating an amount of energy consumed by the airside equipment during the time step. The predictive controller performs a predictive control process using the control objective or the constraint to determine values of the equipment-specific energy components during each time step of the time period as outputs of the predictive control process. The predictive controller operates the waterside equipment and the airside equipment using the values of the equipment-specific energy components.

Abstract: A refrigerant charge controller for heating, ventilation, or air conditioning (HVAC) equipment includes a processing circuit configured to analyze usage data for the HVAC equipment using a machine learning model to estimate an amount of refrigerant used by the HVAC equipment, identify a refrigerant deficiency based on the amount of refrigerant, and initiate a corrective action in response to identifying the refrigerant deficiency.

Abstract: A method for detecting faults in a building management system (BMS) is shown. The method includes receiving time series data characterizing an operating performance of one or more BMS devices. The method further includes processing the time series data using multiple different fault detection methods to generate multiple fault detection results. The method includes providing the multiple fault detection results as outputs from the multiple different fault detection methods. The method includes applying the multiple fault detection results as inputs to a neural network that determines whether the multiple fault detection results are indicative of a fault condition in the BMS.

Abstract: A system includes a plurality of thermostats corresponding to a plurality of HVAC systems that serve a plurality of spaces and a computing system communicable with the plurality of thermostats via a network. The computing system is configured to, for each space of the plurality of spaces, obtain a set of training data relating to thermal behavior of the space, identify a model of thermal behavior of the space based on the set of training data, perform a model predictive control process using the model of thermal behavior of the space to obtain a temperature setpoint for the space, and provide the temperature setpoint to the thermostat corresponding to the HVAC system serving the space. The plurality of thermostats are configured to control the plurality of HVAC systems in accordance with the temperature setpoints.

Abstract: A refrigerant charge controller for heating, ventilation, or air conditioning (HVAC) equipment includes a processing circuit configured to analyze usage data for the HVAC equipment using a machine learning model to estimate an amount of refrigerant used by the HVAC equipment, identify a refrigerant deficiency based on the amount of refrigerant, and initiate a corrective action in response to identifying the refrigerant deficiency.

Abstract: A model predictive maintenance (MPM) system for building equipment includes an operational cost predictor configured to predict a cost of operating the building equipment over a duration of an optimization period, a maintenance cost predictor configured to predict a cost of performing maintenance on the building equipment over the duration of the optimization period, and an objective function optimizer configured to optimize an objective function to predict a total cost associated with the building equipment over the duration of the optimization period. The objective function includes the predicted cost of operating the building equipment and the predicted cost of performing maintenance on the building equipment. The MPM system includes an equipment controller configured to operate the building equipment to affect a variable state or condition in a building in accordance with values of one or more decision variables obtained by optimizing the objective function.

Abstract: A predictive controller for a building energy system obtains a function that defines an amount of energy consumed by both waterside equipment and airside equipment during a time period as a summation of multiple equipment-specific energy components including a waterside energy component indicating an amount of energy consumed by the waterside equipment during the time period and an airside energy component indicating an amount of energy consumed by the airside equipment during the time period. The predictive controller performs a predictive control process using the function to determine values of the equipment-specific energy components during the time period. The predictive control process includes predicting the amount of energy consumed by both the waterside equipment and the airside equipment during the time period based on the values of the equipment-specific energy components.

Abstract: A method for controlling HVAC equipment includes using a dynamic thermal model that describes heat transfer characteristics of a building to predict a temperature of the building as a function of control inputs to HVAC equipment that operate to affect the temperature of the building. The method includes obtaining time-varying utility prices and a predicted heating or cooling load for the building and performing a predictive control process using the dynamic thermal model, the time-varying utility prices, and the predicted heating or cooling load to generate the control inputs to the HVAC equipment subject to constraints on the temperature of the building.

Abstract: An air conditioning system for a building includes an air conditioning unit including one or more powered air conditioning components configured to apply heating or cooling to a refrigerant and an air conditioning controller. The air conditioning controller is configured to cause a determined amount of power to be received at the one or more powered air conditioning components for use in powering the air conditioning unit and control the one or more powered air conditioning components by determining a refrigerant temperature setpoint for the refrigerant or a supply air temperature setpoint for supply air output by the air conditioning system based on the determined amount of power.

Abstract: A model predictive maintenance (MPM) system generates operating decisions and maintenance decisions for building equipment by performing a MPM process. The MPM process determines a specific type of maintenance activity to be performed at a service time from a set of multiple different types of maintenance activities based on (1) first costs of operating the building equipment predicted to result from the multiple different types of the maintenance activities and (2) second costs of performing maintenance on the building equipment predicted to result from the multiple different types of the maintenance activities. The MPM system causes the specific type of maintenance activity to be performed on the building equipment at the service time and controls the building equipment by generating electronic control signals based on the operating decisions and causing the building equipment to affect a variable state or condition in a building using the electronic control signals.

Abstract: A thermostat for a building zone includes at least one of a model predictive controller and an equipment controller. The model predictive controller is configured to obtain a cost function that accounts for a cost of operating HVAC equipment during each of a plurality of time steps, use a predictive model to predict a temperature of the building zone during each of the plurality of time steps, and generate temperature setpoints for the building zone for each of the plurality of time steps by optimizing the cost function subject to a constraint on the predicted temperature. The equipment controller is configured to receive the temperature setpoints generated by the model predictive controller and drive the temperature of the building zone toward the temperature setpoints during each of the plurality of time steps by operating the HVAC equipment to provide heating or cooling to the building zone.

Abstract: A model predictive maintenance (MPM) system for building equipment includes one or more processing circuits having one or more processors and memory. The memory store instructions that, when executed by the one or more processors, cause the one or more processors to perform operations including estimating a degradation state of the building equipment, using a degradation impact model to predict an amount of one or more input resources consumed by the building equipment to produce one or more output resources based on the degradation state of the building equipment, generating a maintenance schedule for the building equipment based on the amount of the one or more input resources predicted by the degradation impact model, and initiating a maintenance activity for the building equipment in accordance with the maintenance schedule.

Abstract: Systems and methods for reducing health risks with respect to an infectious disease in buildings. Health data for an infectious diseases is used to determine a health risk level for building spaces and individuals in the building. An air handling action or a disinfection action is performed based on the health risk level.

Abstract: A variable refrigerant flow (VRF) system for a building includes an outdoor VRF unit, a plurality of indoor VRF units, a battery, and a predictive VRF controller. The outdoor VRF unit includes powered VRF components configured to apply heating or cooling to a refrigerant. The indoor VRF units are configured to use the heated or cooled refrigerant to provide heating or cooling to a plurality of building zones. The battery is configured to store electric energy and discharge the stored electric energy for use in powering the powered VRF components. The predictive VRF controller is configured to optimize a predictive cost function to determine an optimal amount of electric energy to purchase from an energy grid and an optimal amount of electric energy to store in the battery or discharge from the battery for use in powering the powered VRF components at each time step of an optimization period.

Abstract: A model predictive maintenance (MPM) system for building equipment includes one or more processing circuits including one or more processors and memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations include obtaining one or more performance indicators for the building equipment and determining whether a trigger condition has been satisfied based on the one or more performance indicators. The operations include triggering a model predictive maintenance process to generate a maintenance schedule for the building equipment in response to determining that the trigger condition has been satisfied. The operations include initiating a maintenance activity for the building equipment in accordance with the maintenance schedule.

Abstract: A building optimization system includes charging and discharging a battery of a battery power vehicle. The building optimization system includes a charging system configured to cause the battery of the battery powered vehicle to charge or discharge. The building optimization system also includes an optimization controller including a processing circuit. The processing circuit is configured to receive charging constraints for the battery powered vehicle, determine whether to charge discharge the battery of the battery powered vehicle based on the charging constraints, and cause the charging system to charge or discharge the battery of the battery powered vehicle based on the optimization.