TotalEnergies OneTech

The present invention concerns a control method (100) for controlling a wind farm comprising a plurality of wind turbines.The method comprises a learning phase (1100) and an exploitation phase (1200).During the learning phase receiving (1110) a model architecture is received. The model architecture comprises an attention comprising at least on multi-head self-attention neural network. The learning phase further comprises training (1120) the model architecture according to a reinforcement learning technic implying tuning the tunable parameters of the model architecture to obtain a trained model optimizing a performance of the wind farm. The performance of the wind farm depends on an estimated power output by the wind farm, and losses due to wake effect,The exploitation phase comprises receiving (1210) estimated wind velocities and directions. The exploitation phase further comprises applying (1230) the trained model to derive rotations commands, and sending (1240) the rotation commands to wind turbines.

Le procédé l'injection, à des instants de mesure, dans la cellule de mesure portée par un drone d'un faisceau laser, à une longueur d'onde de mesure caractéristique du gaz traceur à détecter.La signature spectrale du gaz traceur ou de l'ensemble de gaz est plus large qu'une largeur d'accordabilité de la source laser de mesure. Le procédé comprend l'injection, à des instants de référence, dans la cellule de mesure d'un faisceau laser de référence à une longueur d'onde de référence située à l'écart de ladite signature spectrale.Le procédé comporte la transmission de données représentatives de teneur en gaz traceur obtenues à partir du signal de mesure et de données représentatives de référence obtenues à partir du signal de référence à un système de calcul pour calculer une teneur en gaz à partir des données représentatives de teneur en gaz traceur et des données représentatives de référence.

The invention relates to a method for determining surfaces of waterbodies (W) in a target area, the method comprising: - a phase of building a training database (TDB) comprising for each elementary portion (E) of a georeferenced representation of a training area: an aerial image (IM<M>), water index image(s) (IM<W>) derived from the aerial image (IM<M>) using a water index formula and a ground truth (G) binary mask, each elementary portion (E) representing a zone of the training area where at least a part of a waterbody surface (W) is present, - a phase of training a determination model using the training database (TDB) so as to obtain a trained determination model, and - a phase of inferencing the trained determination model.

The assembly comprises an elongated porous dry electrode line (20) comprising an elongated core electrode (30) configured to be electrically connected to an electrical power source or to a utility.The line (20) comprises an outer sheath (32) made at least partially of an ion permeable and gas impervious membrane applied on the electrode (30), the electrode (30) defining inside the outer sheath (32) at least a longitudinal gas circulation canal (34) emerging at one gas collection outlet or gas supply inlet.The assembly comprises a gas capacity to collect gas from the electrode line (20) or to supply gas to electrode line (20), the gas collection outlet or the gas supply inlet of the electrode line (20) being fluidly connected to the gas capacity.

A control method (100) for controlling a wind farm comprising a plurality of wind turbines, based on an estimated wind velocity and an estimated wind direction of the wind on the wind farm. The method comprises receiving a wind direction and a wind velocity. The method further comprises determining (120) a group of yaw angles sets and evaluating (130), for each yaw angle set, a power generated by the wind farm. The method further comprises generating (150) new yaw angle sets and modifying (160) the value of each yaw angle according to a probability law. The method further comprises adjusting (170). The method further comprises selecting (190), among the adjusted yaw angle sets, the set for which the power generated is the highest, the yaw angles of such selection set being the desired yaw angles, and sending (200), to each wind turbine, the respective desired yaw angle.

The invention relates to a control method (100) for controlling a wind farm comprising a plurality of wind turbines, based on wind velocity direction.The method (100) comprising a learning phase (1100) and an exploitation phase (1200).The learning phase comprises obtaining (1110) a wind direction, a wind velocity, sets of yaw angle values and a parameterized model configured to determine a rotation command. The learning phase further comprises determining (1130) an ordered list of wind turbines, and training (1140) of the parametrized model according to a reinforcement learning algorithm.The reinforcement learning algorithm comprising several iterations determining rotations command based on the parameterized model, performed sequentially on the wind turbines according to the ordered list of wind turbines.The exploitation phase comprising applying (1230) the trained model to each wind turbine to obtain rotation commands, and sending (1240) them to the wind turbines.

A control method (100) for controlling a wind farm comprising a plurality of wind turbines, based on a plurality of estimated wind velocities and a plurality of estimated wind directions of the wind on the wind farm. The method comprises determining (120), for each wind turbine, a respective rotation command by solving an optimization problem. The optimization problem aims at maximizing an objective function, comprising a first component and a second component. The first component is representative of power generated by the wind farm for said set of successive time instants. The second component is representative of the fatigue of each wind turbine of the wind farm, surrogated over the set of successive time instants. Said second component depends on the fatigue induced by mechanical moments of wind forces on each wind turbine, at each time instant.

The present invention refers to a photovoltaic module carrying structure (11) comprising a metallic frame providing:- a holding part (12) comprising a first vertical post (11a) comprising a first slot (11b) configured for receiving a first side edge of a photovoltaic module (13) and a second vertical post (11a) comprising a second slot (11b) configured for facing the first slot (11b) for receiving a second side edge of the photovoltaic module (13),- a supporting part (14) comprising a first surface (16a) configured for extending from the bottom of the holding part (14) on a first side of the holding part (12) with an angle higher than 90°, with respect to the orientation of the first vertical post (11a) and the second vertical post (11a) and a second surface (16b) configured for extending from the bottom of the holding part (12) on a second side of the holding part (12) opposite to the first side with an angle higher than 90° with respect to the orientation of the first vertical post (11a) and the second vertical post (11a).

The present invention refers to a photovoltaic equipment (10) comprising:- a carrying structure (11),- at least one photovoltaic module (13) arranged sensibly vertically on the carrying structure (11),- at least one reflective surface (17) with a top side arranged on the carrying structure (11) below the, at least one, photovoltaic module (13),wherein the angle between the reflective surface (17) and the photovoltaic module (13) is higher than 90°.

It is disclosed a method for determining the locations of flow entities for operating a reservoir, comprising:- Receiving a geological description of the reservoir,- building a geographical graph comprising nodes and edges between nodes,wherein each node represents a flow entity location and is assigned geographic coordinates, and each edge represents relationships between adjacent nodes and is assigned a value of at least one edge property derived from the geological description of the reservoir,- implementing a deformation process on the geological graph, comprising:∘ defining a geological quantity over the geological graph, being function of the spacing between the nodes and of the edge property assigned to the edges of the graph,∘ starting from initial coordinates of the nodes, adjusting the locations of the nodes based on the defined geological function, and- outputting the adjusted flow entities locations as determined locations for operating the reservoir.