Abstract: This disclosure is directed to real-time electric vehicle charging detection using edge computing. An edge device can receive a time series of aggregated current waveform measurements at the site and determine, based on the measurements, a magnitude of one or more harmonics and a metric. The edge device can generate, based on a baseline for the site, a feature vector based on the magnitude of the one or more harmonics and the metric and determine, using a machine learning model, a probability that an EV is charged at the site during a time window of the measurements. The edge device can transmit, based on a comparison of the probability with a threshold, a notification to cause a remote data processing system to execute an operation related to distribution of electricity via the distribution grid.

Abstract: The solution provides a system remote from edge devices at sites of an electricity grid. The system can have a processor to trigger, based on a characteristic of electricity related to the electricity distribution grid, a control function. The processor can ping, using a natural language-based protocol compatible with a generative machine learning model, edge devices for data on electricity consumption at the grid and receive, from the edge devices, information having data structures generated via the generative machine learning model executed by the edge devices. The processor can generate, based on the information and a grid layout, responsive to the trigger of the control function, guidance to impact the electricity characteristic at one or more sites. The system can transmit, to the edge devices, the guidance to adjust the electricity characteristic.

Abstract: Systems and methods for solar voltaic disaggregation from metered net load is presented. A processor coupled with memory can be configured to identify, using a model trained for a site electrically coupled with an electricity distribution grid, a tilt, azimuth, and capacity factor for a photovoltaic system at the site. The processor can receive, from a data repository, irradiance data for a first time interval. The processor can determine an amount of power generated by the photovoltaic system during the first time interval based on the irradiance data, the tilt, the azimuth, and the capacity factor. The processor can execute an action related to power delivery to the site based on the amount of power generated by the photovoltaic system during the first time interval.

Abstract: High-resolution electricity load forecasting is provided. A data processing system, located at a site, can obtain a waveform data set comprising characteristics of power consumption measured for the site during a first time interval at a first sampling rate; determine power-system values based on the waveform data set; determine harmonic information based on a frequency transform performed on the waveform data set; construct, based on the power-system values and the harmonic information (as well as other exogeneous inputs such as weather), a processed waveform data set having a time series of values at a second sampling rate; predict, based on the processed waveform data set, using one or more models, a characteristic of load at the site during a second time interval; and execute an action related to power delivery for the site based on the characteristic of the load predicted during the second time interval.

Abstract: Technical solutions provide a system with a device having a processor coupled with memory to receive, via an interface, a query related to a characteristic of electricity and retrieve data for the grid edge device. The data can include any combination of contextual data, operational data and metering data related to the device or the site. The device can construct a prompt data structure using generative artificial intelligence based on the input query and the data for the device. The device can generate, using the generative artificial intelligence on the device, code responsive to the prompt data structure. The device can execute the processor-executable code to generate output responsive to the query and perform an action in accordance with the output.

Abstract: This disclosure is directed to the dispatch of distributed energy resources (DER). A DER control device can be located at a site connected with an electricity distribution grid. The DER control device receives, from a data processing system of the electricity distribution grid that is remote from the DER control device, an operational constraint. The DER control device generates, for a time interval, trajectories for characteristics of electricity at the site based on the operational constraint and bounds for local power quality at the site. The DER control device detects, in real-time during the time interval, a condition related to the characteristics of electricity. The DER control device controls, in real-time and responsive to the detection, during the time interval, delivery of power from one or more DERs based on i) one or more of the trajectories, ii) the detected condition, and iii) the bounds for local power quality.

Abstract: Systems and methods comprising a metering device located on an electricity distribution grid, the metering device comprising one or more processors and memory. The metering device can detect a drop in characteristic of electricity below a threshold indicating a fault on the electricity distribution grid. The metering device can generate, responsive to the drop in the characteristic of electricity below the threshold, a time series of a rate of change of the characteristic of electricity for a predetermined number of cycles subsequent to the detection of the drop. The metering device can determine, based on a comparison of the time series of the rate of change with a predetermined pattern, a location of the metering device on the electricity distribution grid relative to a location of the fault on the electricity distribution grid.

Abstract: Systems and methods for meter-to-transformer connectivity. The system can include a data processing system comprising one or more processors coupled with memory. The data processing system can receive a time-series data set of voltage measured by meters at a plurality of loads in an electricity distribution grid. The data processing system can construct, based on the time-series data set, a matrix with values that indicate similarities between pairs of meters of the meters. The data processing system can generate, via clustering techniques applied to the matrix, a silhouette score for each of the meters in the matrix to group the meters into transformer groups. The data processing system can provide, for output via a graphical user interface, a digital map with an indication of a subset of meters of the meters associated with a transformer of the transformer groups generated based on the clustering techniques applied to the matrix.

Abstract: Systems and methods for a smart grid distributed operations platform. The system can include a data processing system comprising one or more processors, coupled with memory. The data processing system can identify a network connection with edge devices located on an electricity distribution grid, wherein each of the edge devices comprise a processor and memory. The data processing system can identify, based on a configuration associated with the edge devices, an application configured to perform a function to manage delivery of electricity via the electricity distribution grid. The data processing system can select a model from models trained with machine learning based on a local environment attribute associated with a location on the electricity distribution grid at which the edge devices are located. The data processing system can provide, for execution on the edge devices, the application configured with the model to perform the function to manage delivery of electricity.

Abstract: Anomaly detection, classification, and prediction is provided. A system can include one or more processors coupled with memory. The system can identify voltage waveform data corresponding to electricity distributed over a utility grid and measured by a metering device. The system can detect, based on a comparison with baseline voltage waveform data, an anomaly in at least a portion of the voltage waveform data. The system can generate spectrogram data for the at least the portion of the voltage waveform data comprising the anomaly. The system can determine, via a model trained with machine learning, a type of the anomaly based on the spectrogram data. The system can provide an indication of the type of the anomaly to cause an action to be performed on the utility grid responsive to determination of the type of anomaly.

Abstract: Meter voltage fingerprint is provided. A system can include a metering system at a location downstream from a substation on a utility grid that distributes electricity. The metering system can receive data samples of a voltage waveform corresponding to the electricity distributed at the location. The metering system can determine first metrics for the voltage waveform over a time interval via a statistical technique. The metering system can determine second metrics for the voltage waveform over the time interval based on a difference between the voltage waveform and a model waveform. The metering system can construct a data structure comprising the first metrics, the second metrics, and an identifier for the location. The metering system can provide the data structure to a data processing system remote from the metering system to cause the data processing system to evaluate a performance of the utility grid.

Abstract: Systems and methods for real-time (EV) charging detection. The system can include a metering system. The metering system can detect, via a sensor, current associated with electricity consumed in the utility grid during a first time interval. The metering system can determine a current harmonic metric and a power metric associated with the electricity delivered over the utility grid in the first time interval. The metering system can input the current harmonic metric and the power metric into a model trained with machine learning and deployed on the metering system to determine a likelihood that at least a portion of the electricity delivered over the utility grid in the first time interval is used to charge an electric vehicle. The metering system can execute an action associated with performance of the utility grid responsive to the likelihood satisfying a threshold.

Abstract: Systems and methods comprising a metering device located on an electricity distribution grid, the metering device comprising one or more processors and memory. The metering device can detect a drop in characteristic of electricity below a threshold indicating a fault on the electricity distribution grid. The metering device can generate, responsive to the drop in the characteristic of electricity below the threshold, a time series of a rate of change of the characteristic of electricity for a predetermined number of cycles subsequent to the detection of the drop. The metering device can determine, based on a comparison of the time series of the rate of change with a predetermined pattern, a location of the metering device on the electricity distribution grid relative to a location of the fault on the electricity distribution grid.

Abstract: Systems and methods comprising a metering device located on an electricity distribution grid, the metering device comprising one or more processors and memory. The metering device can detect a drop in characteristic of electricity below a threshold indicating a fault on the electricity distribution grid. The metering device can generate, responsive to the drop in the characteristic of electricity below the threshold, a time series of a rate of change of the characteristic of electricity for a predetermined number of cycles subsequent to the detection of the drop. The metering device can determine, based on a comparison of the time series of the rate of change with a predetermined pattern, a location of the metering device on the electricity distribution grid relative to a location of the fault on the electricity distribution grid.

Abstract: Systems and methods are directed controlling components of a utility grid. The system can receive signals. The system can determine one or more statistical metrics based on the signals. The system can generate an input matrix. The system can input the input matrix into a machine learning model. The system can predict, based on the input matrix and via the machine learning model, the value for the signal of the utility grid at a time period for which the value is not provided in the input matrix. The system can provide a command to control a component of the utility grid responsive to the value for the signal of the utility grid predicted by the machine learning model.

Abstract: Systems and methods are directed to controlling components of a utility grid. The system can receive data samples including signals detected at one or more portions of a utility grid. The system can construct a matrix having a first dimension and a second dimension. The system can train a machine learning model based on the matrix to predict values for signals of the utility grid not provided in the matrix. The system can receive bounds for one or more input variables, constraints on one or more output variables, and a performance objective for the utility grid. The system can determine, based on the machine learning model and via an optimization technique, an adjustment to a component of the utility grid that satisfies the performance objective. The system can provide the adjustment to the component of the utility grid to satisfy the performance obj ective.

Abstract: Systems and methods comprising a metering device located on an electricity distribution grid, the metering device comprising one or more processors and memory. The metering device can detect a drop in characteristic of electricity below a threshold indicating a fault on the electricity distribution grid. The metering device can generate, responsive to the drop in the characteristic of electricity below the threshold, a time series of a rate of change of the characteristic of electricity for a predetermined number of cycles subsequent to the detection of the drop. The metering device can determine, based on a comparison of the time series of the rate of change with a predetermined pattern, a location of the metering device on the electricity distribution grid relative to a location of the fault on the electricity distribution grid.

Abstract: Systems and methods are directed to controlling components of a utility grid. The system can receive data samples including signals detected at one or more portions of a utility grid. The system can construct a matrix having a first dimension and a second dimension. The system can train a machine learning model based on the matrix to predict values for signals of the utility grid not provided in the matrix. The system can receive bounds for one or more input variables, constraints on one or more output variables, and a performance objective for the utility grid. The system can determine, based on the machine learning model and via an optimization technique, an adjustment to a component of the utility grid that satisfies the performance objective. The system can provide the adjustment to the component of the utility grid to satisfy the performance objective.

Abstract: Systems and methods comprising a metering device located on an electricity distribution grid, the metering device comprising one or more processors and memory. The metering device can detect a drop in characteristic of electricity below a threshold indicating a fault on the electricity distribution grid. The metering device can generate, responsive to the drop in the characteristic of electricity below the threshold, a time series of a rate of change of the characteristic of electricity for a predetermined number of cycles subsequent to the detection of the drop. The metering device can determine, based on a comparison of the time series of the rate of change with a predetermined pattern, a location of the metering device on the electricity distribution grid relative to a location of the fault on the electricity distribution grid.

Abstract: Systems and methods are directed to controlling components of a utility grid. The system can receive data samples including signals detected at one or more portions of a utility grid. The system can construct a matrix having a first dimension and a second dimension. The system can train a machine learning model based on the matrix to predict values for signals of the utility grid not provided in the matrix. The system can receive bounds for one or more input variables, constraints on one or more output variables, and a performance objective for the utility grid. The system can determine, based on the machine learning model and via an optimization technique, an adjustment to a component of the utility grid that satisfies the performance objective. The system can provide the adjustment to the component of the utility grid to satisfy the performance objective.