Abstract: Embodiments the present invention set forth techniques for generating power output predictions of power generation devices. In some embodiments, the techniques include receiving one or more data samples associated with at least one power generation device, each data sample including a plurality of features; processing a first portion of the plurality of features by a first portion of a machine learning model to generate one or more latent space variables; and processing the one or more latent space variables and a second portion of the plurality of features by a second portion of the machine learning model to generate a predicted power output of the at least one power generation device. The one or more latent space variables and the second portion of the plurality of features are based on physics relationships of the at least one power generation device.

Abstract: Embodiments the present invention set forth techniques for generating training data sets for power output detection. In some embodiments, the techniques include receiving a set of data samples of features of at least one power generation device, determining, for each data sample, a distance between the features of the data sample and features of other data samples, identifying at least one outlier data sample of the data sample set, the identifying being based on the distance determined for each data sample, and generating a training data set for a machine learning model, wherein the training data set includes the set of data samples excluding at least one of the at least one outlier data sample.

Abstract: Systems and methods for controlling Battery Energy Storage Systems (BESSs), including determining historical minimum state of charge (SOC) for peak shaving of a previous day based on historical photovoltaic (PV)/load profiles, historical demand charge thresholds (DCT), and battery capacity of the BESSs. A minimum SOC for successful peak shaving of a next day is estimated by generating a weighted average based on the historical minimum SOC, and optimal charging/discharging profiles for predetermined intervals are generated based on estimated PV/load profiles for a next selected time period and grid feed-in limitations. Continuous optimal charging/discharging functions are provided for the one or more BESSs using a real-time controller configured for overriding the optimal charging/discharging profiles when at least one of a high excess PV generation, a peak shaving event, or a feed-in limit violation is detected.

Abstract: Systems and methods for minimizing demand charges includes generating historical demand profiles having a plurality of demand charge threshold (DCT) values over a period of time, filtering the DCT values using weighting factors to provide filtered DCT values such that DCT underestimations and overestimations are removed from optimization, determining an optimal DCT output value from the filtered DCT values such that the optimal DCT output value is based on the historical demand profiles and the weighting factors, and distributing energy in one or more batteries in accordance with the optimal DCT output value and current energy demand to minimize energy and demand charges.

Abstract: Aspects of the present disclosure describe a single battery degradation model and methods that considers both CYCLING and CALENDAR aging and useful in both energy management and battery management systems that may employ any of a variety of known battery technologies.

Abstract: Computer-implemented methods and, a system are provided. A method includes constructing by an Energy Management System (EMS), one or more optimization-based techniques for resilient battery charging based on an optimization problem having an EMS cost-based objective function. The one or more optimization-based techniques are constructed to include a battery degradation metric in the optimization problem. The method further includes charging, by the EMS, one or more batteries in a power system in accordance with the one or more optimization-based techniques.

Abstract: A system and method for management of one or more grid-scale Energy Storage Systems (GSSs), including generating an optimal GSS schedule in the presence of frequency regulation uncertainties. The GSS scheduling includes determining optimal capacity deployment factors to minimize penalties for failing to provide scheduled energy and frequency regulation up/down services subject to risk constraints; generating a schedule for a GSS unit by performing co-optimization using the optimal capacity deployment factors, the co-optimization including tracking upper and/or lower bounds on a state of charge (SoC) and including the bounds as a hard constraints; and calculating risk indices based on the optimal scheduling for the GSS unit, and outputting an optimal GSS schedule if risk constraints are satisfied. A controller charges and/or discharges energy from GSS units based on the generated optimal GSS schedule.

Abstract: Systems and methods for controlling Battery Energy Storage Systems (BESSs), including determining historical minimum state of charge (SOC) for peak shaving of a previous day based on historical photovoltaic (PV)/load profiles, historical demand charge thresholds (DCT), and battery capacity of the BESSs. A minimum SOC for successful peak shaving of a next day is estimated by generating a weighted average based on the historical minimum SOC, and optimal charging/discharging profiles for predetermined intervals are generated based on estimated PV/load profiles for a next selected time period and grid feed-in limitations. Continuous optimal charging/discharging functions are provided for the one or more BESSs using a real-time controller configured for overriding the optimal charging/discharging profiles when at least one of a high excess PV generation, a peak shaving event, or a feed-in limit violation is detected.

Abstract: Systems and methods for energy distribution for one or more grid-scale Energy Storage Systems (ESSs), including generating one or more time series models to provide forecasted pricing data for one or more markets, determining battery life and degradation costs for one or more batteries in or more ESSs to provide battery life and degradation costs, optimizing bids for the one or more markets to generate optimal bids based on at least one of the forecasted pricing data or the battery life and degradation costs, and distributing energy to or from the one or more ESSs based on the optimal bids generated.

Abstract: A computer-implemented method for controlling a distributed energy storage system (ESS) communicating with one or more microgrids is presented. The method includes assigning, via the processor, a weight to a first objective function pertaining to minimizing demand charge (DC) cost, assigning, via the processor, a weight to a second function pertaining to maximizing photovoltaic (PV) utilization, receiving historical demand profiles including demand data and historical PV profiles including PV data, and determining ESS power and capacity. The method further includes employing a multi-objective DC cost and PV utilization optimization module to obtain a plurality of optimal solutions by concurrently processing the assigned weights of the first and second objective functions, the historical demand and PV profiles, and the ESS power and capacity.

Abstract: A computer-implemented method executed on a processor for outputting a smoothed photovoltaic (PV) power output from a battery of a power control system communicating with one or more microgrids is presented. The method includes curtailing an input signal received from a plurality of sensors, smoothing the input signal by employing a fuzzy logic based low pass filter having an adaptive window to generate a power reference command, applying a hard ramp rate limit to the power reference command, adjusting battery power output of the battery to satisfy battery constraints and a no-power-from-grid constraint, and distributing energy from the battery based on the adjusted battery power output.

Abstract: A system, method, and computer-program product are provided for controlling a distributed energy storage system (ESS) operatively coupled to one or more microgrids. The system includes a memory for storing program code. The system further includes a processor for running the program code to respectively assign a first, a second, and a third set of weights to a first, a second, and a third objective function formulated to minimize an ESS cost, minimize a Demand Charge (DC) cost, and maximize a Photovoltaic utilization, respectively. The processor further runs the program code to execute a multi-objective ESS optimizing engine to obtain a set of different monthly-based optimal solutions for controlling the ESS by concurrently processing the first, the second, and the third objective functions using different ones of the weights from the first, the second, and the third set of weights.

Abstract: Systems and methods for minimizing demand charges includes generating historical demand profiles having a plurality of demand charge threshold (DCT) values over a period of time, filtering the DCT values using weighting factors to provide filtered DCT values such that DCT underestimations and overestimations are removed from optimization, determining an optimal DCT output value from the filtered DCT values such that the optimal DCT output value is based on the historical demand profiles and the weighting factors, and distributing energy in one or more batteries in accordance with the optimal DCT output value and current energy demand to minimize energy and demand charges.

Abstract: Systems and methods for minimizing demand charges, including determining one or more optimal monthly demand charge thresholds based on historical load data, time of use charges, demand charges, and energy storage unit size for one or more end users. A grid power dispatch setpoint is calculated for a particular time step based on a daily load forecast and a daily economic dispatch solution based on the determined optimal monthly demand charge thresholds. A grid power dispatch setpoint for a subsequent time step is determined by iteratively solving the daily energy dispatch for the subsequent time step to determine an optimal grid power dispatch setpoint. Energy and demand charges are minimized by controlling charging and discharging operations for the energy storage unit in real-time based on the determined optimal grid power dispatch setpoint.

Abstract: Computer-implemented methods and, a system are provided. A method includes constructing by an Energy Management System (EMS), one or more optimization-based techniques for resilient battery charging based on an optimization problem having an EMS cost-based objective function. The one or more optimization-based techniques are constructed to include a battery degradation metric in the optimization problem. The method further includes charging, by the EMS, one or more batteries in a power system in accordance with the one or more optimization-based techniques.

Abstract: Aspects of the present disclosure describe a single battery degradation model and methods that considers both CYCLING and CALENDAR aging and useful in both energy management and battery management systems that may employ any of a variety of known battery technologies.

Abstract: Systems and methods are disclosed for multi-objective energy management of micro-grids. A two-layer control method is used. In the first layer which is the advisory layer, a Model Predictive Control (MPC) method is used as a long term scheduler. The result of this layer will be used as optimality constraints in the second layer. In the second layer, a real-time controller guarantees a second-by-second balance between supply and demand subject to the constraints provided by the advisory layer.

Abstract: A system and method for management of one or more grid-scale Energy Storage Systems (GSSs), including generating an optimal GSS schedule in the presence of frequency regulation uncertainties. The GSS scheduling includes determining optimal capacity deployment factors to minimize penalties for failing to provide scheduled energy and frequency regulation up/down services subject to risk constraints; generating a schedule for a GSS unit by performing co-optimization using the optimal capacity deployment factors, the co-optimization including tracking upper and/or lower bounds on a state of charge (SoC) and including the bounds as a hard constraints; and calculating risk indices based on the optimal scheduling for the GSS unit, and outputting an optimal GSS schedule if risk constraints are satisfied. A controller charges and/or discharges energy from GSS units based on the generated optimal GSS schedule.

Abstract: The invention is directed to a method or management system which 1) dispatches high efficiency generators first, 2) charges/discharges energy storage units in a way to enhance efficiency of generators in the system or avoid the necessity of dispatching generators during their low efficiency operations at all. In this way the method or management system utilizes its knowledge about the efficiency characteristics of generators in the system and its ability to change the net demand seen by the generators through charge and discharge of energy storage units to increase the overall efficiency of the energy system.

Abstract: A management framework is disclosed that achieves maximum energy storage device lifetime based on energy storage device life estimation and the price of energy. The management framework includes a battery life estimation from a supercycle model, for a time window between two consecutive full charges of a battery, which allows assessing a worst case scenario impact of all partial cycles within a supercycle on the battery life. The battery life estimation then considers each supercycle as a single discharge unit instead of treating each individual discharge period separately.