Abstract: A system and method of generating a plurality of actionable insights is disclosed herein. A computing system retrieves data corresponding to a work procedure. Each work procedure includes a plurality of steps. The computing system generates a predictive model for each actionable insight using a plurality of machine learning models by generating an input training based on the retrieved work procedure data and learning, by the plurality machine learning models, a metric corresponding to each actionable insight based on each respective input training set. The input data set for each actionable insight includes actionable insight specific information. The computing system receives a request to generate a plurality of actionable insights for a current work procedure. The computing system generates, via the predictive models, a plurality of metrics for a plurality of actionable insights based on data corresponding to the current work procedure.

Abstract: A wind turbine includes a tower, an aerodynamic rotor operable at a variable rotor speed and having a plurality of rotor blades each having an adjustable rotor blade setting angle and a generator for generating an electrical output power. An operating characteristic curve is prespecified for operating the wind turbine. The operating characteristic curve indicates a relationship between the rotor speed and the output power. A controller is provided, which sets the output power in accordance with the operating characteristic curve depending on the rotor speed. The the operating characteristic curve has a starting rotation speed to which the rotor speed increases as soon as the wind turbine starts when a sufficient wind speed is reached. The starting rotation speed is defined depending on a tower natural frequency of the wind turbine and/or depending on a detected turbulence measure of the prevailing wind.

Abstract: A two-part or multi-part rotor blade and also to a method which is associated with it. The rotor blade is split into at least one rotor blade component which is close to the hub and one rotor blade component which is remote from the hub at a separation point in the longitudinal direction, wherein the rotor blade component which is close to the hub and the rotor blade component which is remote from the hub can be connected at the separation point for operation of the wind turbine. A ratio of profile thickness to profile depth, called relative thickness, at the separation point lies within a range of from 0.4 to 0.5. An improved two-part or multi-part rotor blade in spite of the unexpectedly high relative thicknesses.

Abstract: A system and method of generating a plurality of actionable insights is disclosed herein. A computing system retrieves data corresponding to a work procedure. Each work procedure includes a plurality of steps. The computing system generates a predictive model for each actionable insight using a plurality of machine learning models by generating an input training based on the retrieved work procedure data and learning, by the plurality machine learning models, a metric corresponding to each actionable insight based on each respective input training set. The input data set for each actionable insight includes actionable insight specific information. The computing system receives a request to generate a plurality of actionable insights for a current work procedure. The computing system generates, via the predictive models, a plurality of metrics for a plurality of actionable insights based on data corresponding to the current work procedure.

Abstract: A method for designing and operating a wind power plant for generating electrical power from wind, wherein the wind power plant has an aerodynamic rotor with rotor blades of which the blade setting angle can be adjusted, wherein the rotor blades are populated with a plurality of vortex generators between the rotor blade root and the rotor blade tip, characterized in that the population with the vortex generators in the longitudinal direction of the respective rotor blade is carried out up to a radius position which is determined depending on the air density at a site of the wind power plant. A rotor blade of a wind power plant, to an associated wind power plant and to a wind farm.

Abstract: A method of controlling a wind power plant for generating electrical power from wind is provided. The plant comprises a rotor having rotor blades with adjustable blade angles and the rotor can be operated at a variable rotational speed. The method includes controlling the plant in a partial load mode when wind speed is below a nominal speed and, controlling the plant in a storm mode when the wind speed is above a storm commencement speed. An output power of the plant in the partial load mode and storm mode is adjusted according to an operating characteristic curve that determines a relationship between the rotational speed and the output power. A partial load characteristic curve is used as the operating characteristic curve for controlling the power plant in partial load mode, and a storm mode characteristic curve is used as the operating characteristic curve for controlling the plant in storm mode.

Abstract: A rotor blade for a wind turbine, a wind turbine, a wind park and a method for configuring a rotor blade which is divided in two.

Abstract: A rotor blade for a wind turbine, a wind turbine, a wind park and a method for configuring a rotor blade which is divided in two.

Abstract: A method of controlling a wind power plant for generating electrical power from wind is provided. The plant comprises a rotor having rotor blades with adjustable blade angles and the rotor can be operated at a variable rotational speed. The method includes controlling the plant in a partial load mode when wind speed is below a nominal speed and, controlling the plant in a storm mode when the wind speed is above a storm commencement speed. An output power of the plant in the partial load mode and storm mode is adjusted according to an operating characteristic curve that determines a relationship between the rotational speed and the output power. A partial load characteristic curve is used as the operating characteristic curve for controlling the power plant in partial load mode, and a storm mode characteristic curve is used as the operating characteristic curve for controlling the plant in storm mode.

Abstract: A wind turbine includes a tower, an aerodynamic rotor operable at a variable rotor speed and having a plurality of rotor blades each having an adjustable rotor blade setting angle and a generator for generating an electrical output power. An operating characteristic curve is prespecified for operating the wind turbine. The operating characteristic curve indicates a relationship between the rotor speed and the output power. A controller is provided, which sets the output power in accordance with the operating characteristic curve depending on the rotor speed. The the operating characteristic curve has a starting rotation speed to which the rotor speed increases as soon as the wind turbine starts when a sufficient wind speed is reached. The starting rotation speed is defined depending on a tower natural frequency of the wind turbine and/or depending on a detected turbulence measure of the prevailing wind.

Abstract: A system and method of generating a plurality of actionable insights is disclosed herein. A computing system retrieves data corresponding to a work procedure. Each work procedure includes a plurality of steps. The computing system generates a predictive model for each actionable insight using a plurality of machine learning models by generating an input training based on the retrieved work procedure data and learning, by the plurality machine learning models, a metric corresponding to each actionable insight based on each respective input training set. The input data set for each actionable insight includes actionable insight specific information. The computing system receives a request to generate a plurality of actionable insights for a current work procedure. The computing system generates, via the predictive models, a plurality of metrics for a plurality of actionable insights based on data corresponding to the current work procedure.

Abstract: A two-part or multi-part rotor blade and also to a method which is associated with it. The rotor blade is split into at least one rotor blade component which is close to the hub and one rotor blade component which is remote from the hub at a separation point in the longitudinal direction, wherein the rotor blade component which is close to the hub and the rotor blade component which is remote from the hub can be connected at the separation point for operation of the wind turbine. A ratio of profile thickness to profile depth, called relative thickness, at the separation point lies within a range of from 0.4 to 0.5. An improved two-part or multi-part rotor blade in spite of the unexpectedly high relative thicknesses.

Abstract: A system and method determine formation permeability and/or at least one property indicative of formation permeability of a subsurface geological reservoir having radial-flow. Pressure data is obtained with an observation probe during a formation test, wherein the observation probe is located at a setting position within an open hole wellbore formed within the reservoir. The system and method measure radial-flow response of the reservoir at or adjacent to the setting position of the observation probe by analyzing the collected pressure data. The formation permeability and/or at least one property indicative of the permeability of the reservoir is determined based on the measured radial-flow response of the reservoir at or adjacent to the observation probe.

Abstract: A computing system provides distributed training of a neural network model. Explore phase options, exploit phase options, a subset of a training dataset, and a validation dataset are distributed to a plurality of computing devices. (a) Execution of the model by the computing devices is requested using the subset stored at each computing device. (b) A first result of the execution is received from a computing device. (c) Next configuration data for the neural network model is selected based on the first result and distributed to the computing device. (a) to (c) is repeated until an exploration phase is complete. (d) Execution of the neural network model is requested. (e) A second result is received. (f) Next configuration data is computed based on the second result and distributed to the computing device. (d) to (f) is repeated until an exploitation phase is complete. The next configuration data defines the model.

Abstract: Convolutional neural networks can be visualized. For example, a graphical user interface (GUI) can include a matrix of symbols indicating feature-map values that represent a likelihood of a particular feature being present or absent in an input to a convolutional neural network. The GUI can also include a node-link diagram representing a feed forward neural network that forms part of the convolutional neural network. The node-link diagram can include a first row of symbols representing an input layer to the feed forward neural network, a second row of symbols representing a hidden layer of the feed forward neural network, and a third row of symbols representing an output layer of the feed forward neural network. Lines between the rows of symbols can represent connections between nodes in the input layer, the hidden layer, and the output layer of the feed forward neural network.

Abstract: A computing system provides distributed training of a neural network model. Explore phase options, exploit phase options, a subset of a training dataset, and a validation dataset are distributed to a plurality of computing devices. (a) Execution of the model by the computing devices is requested using the subset stored at each computing device. (b) A first result of the execution is received from a computing device. (c) Next configuration data for the neural network model is selected based on the first result and distributed to the computing device. (a) to (c) is repeated until an exploration phase is complete. (d) Execution of the neural network model is requested. (e) A second result is received. (f) Next configuration data is computed based on the second result and distributed to the computing device. (d) to (f) is repeated until an exploitation phase is complete. The next configuration data defines the model.

Abstract: Interactive visualizations of a convolutional neural network are provided. For example, a graphical user interface (GUI) can include a matrix having symbols indicating feature-map values that represent likelihoods of particular features being present or absent at various locations in an input to a convolutional neural network. Each column in the matrix can have feature-map values generated by convolving the input to the convolutional neural network with a respective filter for identifying a particular feature in the input. The GUI can detect, via an input device, an interaction indicating that that the columns in the matrix are to be combined into a particular number of groups. Based on the interaction, the columns can be clustered into the particular number of groups using a clustering method. The matrix in the GUI can then be updated to visually represent each respective group of columns as a single column of symbols within the matrix.

Abstract: Interactive visualizations of a convolutional neural network are provided. For example, a graphical user interface (GUI) can include a matrix having symbols indicating feature-map values that represent likelihoods of particular features being present or absent at various locations in an input to a convolutional neural network. Each column in the matrix can have feature-map values generated by convolving the input to the convolutional neural network with a respective filter for identifying a particular feature in the input. The GUI can detect, via an input device, an interaction indicating that that the columns in the matrix are to be combined into a particular number of groups. Based on the interaction, the columns can be clustered into the particular number of groups using a clustering method. The matrix in the GUI can then be updated to visually represent each respective group of columns as a single column of symbols within the matrix.

Abstract: Convolutional neural networks can be visualized. For example, a graphical user interface (GUI) can include a matrix of symbols indicating feature-map values that represent a likelihood of a particular feature being present or absent in an input to a convolutional neural network. The GUI can also include a node-link diagram representing a feed forward neural network that forms part of the convolutional neural network. The node-link diagram can include a first row of symbols representing an input layer to the feed forward neural network, a second row of symbols representing a hidden layer of the feed forward neural network, and a third row of symbols representing an output layer of the feed forward neural network. Lines between the rows of symbols can represent connections between nodes in the input layer, the hidden layer, and the output layer of the feed forward neural network.

Abstract: A method for determining rock and fluid properties of a fluid-containing subsurface geological formation is provided. First, a low resolution model of the geological formation is initially created from a lumped average parameter estimation derived from at least fluid pressure transient data obtained along a linear wellbore that traverses the formation. Next, the model parameters are updated using grid-based parameter estimation in which the low resolution pressure transient data are combined with data from at least one of seismic data, formation logs, and basic geological structural information surrounding the linear wellbore. Depending on the data available, this process may be carried out in a sequential manner by obtaining and combining additional dynamic data at selected areas. Through this process, multiple realizations of the properties of the geological formation (within the geological structural model) may be created based from the pressure-data conditioned geostatistics i.e.