Abstract: A computer-implemented method, computer-implemented tool and power plant control device for energy balancing solar power plants and a solar power plant system is provided.

Abstract: Deep learning is used to train a neural network for end-to-end prediction of short term (e.g., 20 minutes or less) solar irradiation based on camera images and metadata. The architecture of the neural network includes a recurrent network for temporal considerations. The images and metadata are input at different locations in the neural network. The resulting machine-learned neural network predicts solar irradiation based on camera images and metadata so that a solar plant and back-up power source may be controlled to minimize output power variation.

Abstract: In order to estimate states of health of batteries storing electrical energy significantly indicated by battery capacity and battery internal resistance, by which the estimation of states of health of batteries, deteriorated over time usage, is automated and carried out without a need of any specific test such as a capacity test and the usage of data out of normal operation, it is proposed to collect data from normal operation of a battery storing electrical energy relating to battery-internal physical properties such as a terminal direct current IDC, a terminal direct voltage UDV and a battery cell temperature T due to battery measurements and the determination or estimation, based on this data and a battery model, of a model parameter by solving an optimization-/model parameter estimation-problem and minimizing a difference between the battery model and the battery measurements.

Abstract: Systems, methods, and computer-readable media are disclosed for autonomous restoration of power systems after natural disasters. An operating mode of a circuit breaker that is in a power system grid and that is coupled to other circuit breakers in the power system grid via communication channels is determined. Based on determining that the operating mode is a normal operating mode, data to parameterize a plurality of power grid restoration scenarios is collected at the circuit breaker. Based on determining that the operating mode is an exception mode, measurement data that includes current and voltage data is received at the circuit breaker, and one of the power system grid restoration scenarios is initiated at the circuit breaker. The power system grid error restoration scenario is selected based at least in part on a status of the plurality of communication channels coupled to the circuit breaker and on the measurement data.

Abstract: A system and method for correcting short term irradiance prediction for control of a photovoltaic power station includes an irradiance data manager that generates an observed irradiance curve based on irradiance measurements received from an irradiation sensor, and a reference curve generation engine that generates a reference curve representative of irradiance values for a clear sky day. An irradiance prediction engine generates an irradiance prediction curve based on image segmentation and motion filtering of cloud pixels in sky images. A prediction correction engine corrects the irradiance prediction curve within a short term future time interval based on probabilistic analysis of time segmented intervals of the observed irradiance curve, the reference curve, and the irradiance prediction curve.

Abstract: A method and device which operates to record and evaluate an output of electrical energy of a hybrid power plant, wherein, in accordance with time and/or load, at least one expected generable and usable energy contribution resulting from the utilization of renewable energy sources and one expected energy contribution resulting from the utilization of convention energy carriers are recorded with different tariffs.

Abstract: Systems and methods are provided for adapting images from different cameras so that a single trained classifier or an analyzer may be used. The classifier or analyzer operates on images that include a particular color distribution or characteristic. A generative network is used to adapt images from other cameras to have a similar color distribution or characteristic for use by the classifier or analyzer. A generative adversarial process is used to train the generative network.

Abstract: In order to detect power production degradation of a solar power plant (SPP) with at least one photovoltaic module (PVM), by which the detection of power production degradation, affected for instance by long term aging or soiling of the photovoltaic module of the solar power plant, is automated and carried out sensorless, it is proposed to determine the degradation by calibrating local measurements of data related to module/plant-internal parameters (PT) such as irradiance I and temperature T or electric power P and temperature as measurement data (MDI,ti, MDT,ti, MDP,ti against irradiance reference data (RDI,ti) for instance satellite data in the form of weather data for the same time and then analyze a variation of the calibration.

Abstract: In order to estimate states of health of batteries storing electrical energy significantly indicated by battery capacity and battery internal resistance, by which the estimation of states of health of batteries, deteriorated over time usage, is automated and carried out without a need of any specific test such as a capacity test and the usage of data out of normal operation, it is proposed to collect data from normal operation of a battery storing electrical energy relating to battery-internal physical properties such as a terminal direct current IDC, a terminal direct voltage UDV and a battery cell temperature T due to battery measurements and the determination or estimation, based on this data and a battery model, of a model parameter by solving an optimization-/model parameter estimation-problem and minimizing a difference between the battery model and the battery measurements.

Abstract: To control a power network having a photovoltaic system, a time curve of a light radiation of the photovoltaic system is determined, wherein an increase of the time curve in relation to a reference curve of the light radiation is detected. As a result of the detection of the increase, a preparatory measure is then introduced to prepare the power network for an upcoming power drop of the photovoltaic system.

Abstract: Pixels are classified within a time series of first and second images for the first image, a first probability map is provided with a first probability for a cloud for each first pixel and, for the second image, a second probability map with a second probability for a cloud for each second pixel; first and second mean intensity values are calculated for the pixels; local zero mean images are calculated by subtracting the mean intensity value from the intensity value of the respective pixel; a maximum difference map is generated by calculating, for spatially corresponding pixels, an absolute difference value between a first and second zero mean value; a weighting map is produced by multiplying each absolute difference value with a non-linear function; and a classifying map is computed based on the first probability map, the second probability map, and the weighting map.

Abstract: A method of classifying a near sun sky image includes at least one of the following steps: providing a recurrent neural network in the shape of a long short-term memory cell, the memory cell having at least an input gate, a neuron with a self-recurrent connection, a forget gate, and an output gate, and using a convolutional neural network, which includes, in the cited order, at least an input layer, one or more convolutional layers, an average pooling layer, and an output layer.

Abstract: A method of predicting the intensity of sun light irradiating the ground. At least two input images are provided of a time series of images captured from the sky; a plurality of image features are extracted from the at least two input images; a set of meta data associated with the at least two input images are determined; the image features and the meta data are supplied as input data to a neural network; and neural network operations predict the future intensity of the sun light as a function of the input data. Further, a data processing unit and a computer program for controlling or carrying out the described method are described, as well as an electric power system with such a data processing unit.

Abstract: Deep learning is used to train a neural network for end-to-end prediction of short term (e.g., 20 minutes or less) solar irradiation based on camera images and metadata. The architecture of the neural network includes a recurrent network for temporal considerations. The images and metadata are input at different locations in the neural network. The resulting machine-learned neural network predicts solar irradiation based on camera images and metadata so that a solar plant and back-up power source may be controlled to minimize output power variation.

Abstract: A system and method for correcting short term irradiance prediction for control of a photovoltaic power station includes an irradiance data manager that generates an observed irradiance curve based on irradiance measurements received from an irradiation sensor, and a reference curve generation engine that generates a reference curve representative of irradiance values for a clear sky day. An irradiance prediction engine generates an irradiance prediction curve based on image segmentation and motion filtering of cloud pixels in sky images. A prediction correction engine corrects the irradiance prediction curve within a short term future time interval based on probabilistic analysis of time segmented intervals of the observed irradiance curve, the reference curve, and the irradiance prediction curve.

Abstract: Examples of techniques for configuring and parameterizing an energy control system are disclosed. In one example implementation according to aspects of the present disclosure, a computer-implemented method includes receiving, by a processing device, a plurality of digital twins. Each digital twin of the plurality of digital twins is associated with a component of a plurality of components. The method further includes computing, by the processing device, configuration and parameterization information for each of the plurality of components based at least in part on the plurality of digital twins. The method further includes transmitting, by the processing device, the configuration and parameterization information to respective components of the plurality of components to enable the plurality of components to implement the configuration and parameterization information.

Abstract: Systems, methods, and computer-readable media are described for addressing the unit commitment problem in a power grid by providing a communication-free control framework according to which each power generating unit in the power grid determines its own operating schedule for turning on or off based solely on local measurements.

Abstract: Provided is a method for charging or discharging an electrochemical energy store, the ageing behavior of which energy store is modelled by describing its depth of discharge at which a depth of discharge value is associated with each time period between two adjacent local maxima and minima in a time-dependent state of charge profile, which depth of discharge value corresponds to the charging or discharging stroke between the state of charge values of the two adjacent local maxima and minima.

Abstract: Systems, methods, and computer-readable media are disclosed for autonomous restoration of power systems after natural disasters. An operating mode of a circuit breaker that is in a power system grid and that is coupled to other circuit breakers in the power system grid via communication channels is determined. Based on determining that the operating mode is a normal operating mode, data to parameterize a plurality of power grid restoration scenarios is collected at the circuit breaker. Based on determining that the operating mode is an exception mode, measurement data that includes current and voltage data is received at the circuit breaker, and one of the power system grid restoration scenarios is initiated at the circuit breaker. The power system grid error restoration scenario is selected based at least in part on a status of the plurality of communication channels coupled to the circuit breaker and on the measurement data.

Abstract: Systems, methods, and computer-readable media are described for addressing the unit commitment problem in a power grid by providing a communication-free control framework according to which each power generating unit in the power grid determines its own operating schedule for turning on or off based solely on local measurements.