Abstract: Systems and methods provide techniques for performing predictive maintenance data analysis based on telemetry data. In one embodiments, a method includes at least operations configured to obtain a training telemetry data object, determine a training input data object based on the training telemetry data object, obtain a maintenance data object, and generate a trained predictive maintenance convolutional neural network based on the training input data object and the maintenance data object. The trained predictive maintenance convolutional neural network can be utilized to generate maintenance predictions for a monitored system. The maintenance prediction can include data objects with visual explanatory capabilities, such as data objects that describe heatmaps over input telemetry data.

Abstract: Various embodiments described herein relate to automated setpoint generation for assets via cloud-based supervisory control. In this regard, a request to perform supervisory control with respect to an asset is received. The request comprises an asset identifier indicating an identity of the asset. In response to the request, one or more setpoints for the asset is determined based on the asset identifier. Also in response to the request, comfort constraint data indicative of one or more comfort constraints for an environment associated with the asset is determined based on the asset identifier. Furthermore, the one or more setpoints for the asset is adjusted based on the comfort constraint data.

Abstract: Various embodiments described herein relate to automated asset strategy selection for assets via cloud-based supervisory control. In this regard, one or more baselines for asset strategy data related to an asset is estimated based on a regression analysis process associated with the asset. Additionally, a maximum uncertainty level for the one or more baselines is determined, one or more conditions related to energy optimization for an environment associated with the asset is estimated for a future period of time, and a predicted uncertainty level for the one or more conditions related to the energy optimization is determined. Based on a comparison between the maximum uncertainty level and the predicted uncertainty level, a predetermined asset strategy or an energy optimization asset strategy for the future period of time is then selected.

Abstract: Method, apparatus and computer program product for generating a supply stream temperature set-point for a particular distribution channel element associated with a heating, ventilation, and air-conditioning (HVAC) system. In one example, a method includes identifying a hierarchical position of the particular distribution channel element; identifying potential set point configuration actions associated with the particular distribution channel element, wherein each potential set point configuration action is expected to cause transition of the particular distribution channel element from a current state to a future state; determining an overall cost measure for each potential set point configuration action based at least in part on the hierarchical position of the particular distribution channel element, and generating the supply stream temperature set-point based on each overall cost measure associated with a potential set point configuration action.

Abstract: Systems and methods provide techniques for performing predictive maintenance data analysis based on telemetry data. In one embodiments, a method includes at least operations configured to obtain a training telemetry data object, determine a training input data object based on the training telemetry data object, obtain a maintenance data object, and generate a trained predictive maintenance convolutional neural network based on the training input data object and the maintenance data object. The trained predictive maintenance convolutional neural network can be utilized to generate maintenance predictions for a monitored system. The maintenance prediction can include data objects with visual explanatory capabilities, such as data objects that describe heatmaps over input telemetry data.

Abstract: Systems and methods provide techniques for generating natural language data based on telemetry data. In one embodiments, a method includes at least operations configured to receive telemetry data, wherein the telemetry data includes temporal telemetry vectors; process the temporal telemetry vectors using an encoder model in order to generate a feature vector for the telemetry data; and process the feature vector using a decoder model in order to generate the natural language data.

Abstract: Method, apparatus and computer program product for generating a supply stream temperature set-point for a particular distribution channel element associated with a heating, ventilation, and air-conditioning (HVAC) system. In one example, a method includes identifying a hierarchical position of the particular distribution channel element; identifying potential set point configuration actions associated with the particular distribution channel element, wherein each potential set point configuration action is expected to cause transition of the particular distribution channel element from a current state to a future state; determining an overall cost measure for each potential set point configuration action based at least in part on the hierarchical position of the particular distribution channel element, and generating the supply stream temperature set-point based on each overall cost measure associated with a potential set point configuration action.