CN-122021438-A - Wake flow assessment method and device for offshore wind power plant group

CN122021438ACN 122021438 ACN122021438 ACN 122021438ACN-122021438-A

Abstract

The invention provides a wake flow assessment method and a wake flow assessment device of an offshore wind power plant group, which relate to the technical field of wind power generation engineering and computational fluid dynamics, and the method comprises the steps of acquiring and fusing marine atmospheric multi-source data, identifying and extracting the position, the scale and the strength of a thermodynamic vortex structure by adopting continuous wavelet transformation, constructing a topological network by taking the vortex structure as a node and taking the gradient communication relation between the vortices as edges, establishing a quantitative physical model by combining a background flow field, generating a dynamic gradient guiding vector field, mapping the positions of each unit to the network, and obtaining the space-time evolution track of the centroid of each unit wake flow by solving a motion control equation under the joint driving of the gradient guiding vector field and the background flow field; and finally, calculating a dynamic wake velocity loss field through a physical mapping function according to the vortex structure of the track passing position, and obtaining an integral wake field through space-time superposition, thereby evaluating the wind speed loss of the downstream target position and generating a report.

Inventors

CAO QINGWEI
HUANG HAOCHENG
YAO ZHONGYUAN
ZHAO JIANJIAN
ZHOU FENGFENG
XU CHANG
XUE FEIFEI
HUO ZHIHONG
GU YICHENG
Xin Xinyao

Assignees

盛东如东海上风力发电有限责任公司
华能国际电力江苏能源开发有限公司清洁能源分公司
河海大学

Dates

Publication Date: 20260512
Application Date: 20260202

Claims (8)

1. A wake assessment method for an offshore wind farm group, comprising the specific steps of: Step 1, acquiring ocean surface state parameter field and atmospheric boundary layer vertical section parameter data of a target sea area, performing space-time fusion and registration to form an ocean atmospheric composite field, performing multi-scale calculation analysis on the composite field by adopting continuous wavelet transformation, identifying vortex structures in the composite field, and extracting the central position, the scale and the strength of each vortex structure; Step 2, constructing a vortex topological network by taking each vortex structure as a node and taking a gradient communication relation between adjacent vortex structures as an edge, and constructing a quantitative relation model between gradients and flow directions by combining a background horizontal flow field of a target sea area based on the vortex topological network through fluid dynamics so as to construct a gradient guiding vector field covering the whole target sea area; Mapping the geographic position of each wind generating set in a wind power plant group to a vortex topology network, taking the position of the set at the beginning of evaluation as a wake centroid initial point in a set evaluation interval, and solving a wake centroid motion control equation through a numerical value under the common driving of a gradient guiding vector field and a background horizontal flow field to obtain a space-time evolution motion track; And 4, calculating wake velocity loss distribution according to the scale and the intensity of a vortex structure of each space-time evolution motion track and through a pre-established physical mapping function so as to generate a dynamic wake velocity loss field, performing space-time superposition on all the dynamic wake velocity loss fields to obtain an integral wake field, calculating the wind speed loss of a downstream target position according to the integral wake field, and generating a wake evaluation report.
2. A wake assessment method of an offshore wind farm group according to claim 1, wherein the space-time fusion and registration process comprises the steps of: The method comprises the steps of acquiring a sea surface temperature field and a sea surface height field which cover a target sea area from a satellite remote sensing data source to serve as sea surface state parameter fields, extracting vertical profile data of air temperature, humidity and wind speed on the target sea area from an atmospheric analysis database, a buoy and ship observation data to serve as atmospheric boundary layer vertical profile parameter data, setting a unified time reference and a space grid coordinate system, and resampling the sea surface state parameter fields and the atmospheric boundary layer vertical profile parameter data to a set of space-time frame grid points with unified time step, unified horizontal grid resolution and unified vertical layering height by adopting a three-dimensional space-time interpolation algorithm, so that space-time alignment and fusion of multi-source data are completed, and a sea atmosphere composite field is formed; The method comprises the steps of selecting a preset two-dimensional mother wavelet function, performing two-dimensional continuous wavelet transformation on a sea surface temperature field in a marine atmosphere composite field to obtain a complex form wavelet coefficient field under multiple scales, obtaining a corresponding wavelet coefficient mode value field by calculating the mode value of each wavelet coefficient in the wavelet coefficient field, identifying a local maximum value point of the wavelet coefficient mode value field under each scale, determining the central position of a vortex structure, determining the scale of the vortex structure by a characteristic scale generating maximum mode response, and obtaining the strength of the vortex structure by normalizing the wavelet coefficient mode value at the central position of the vortex structure.
3. The method for estimating wake flow of an offshore wind farm group according to claim 2, wherein each vortex structure is used as a node, and gradient communication relation between adjacent vortex structures is used as an edge, and a vortex topology network is constructed, specifically comprising: Defining each identified vortex structure as a node, calculating Euclidean distance between the center positions of any two nodes, calculating the scale product of the two nodes, and carrying out normalization processing on the Euclidean distance by utilizing the product to obtain normalized distance measurement; the method comprises the steps of respectively extracting temperature gradient vectors at the central positions of the two temperature gradient vectors from a marine atmosphere composite field, calculating the dot product absolute values of the two temperature gradient vectors to obtain a direction consistency measurement for representing the direction consistency of the two gradients; Setting a gradient communication intensity threshold, when the gradient communication intensity value between any two nodes is larger than the threshold, establishing a connecting edge between the two nodes, wherein the weight of the connecting edge is the gradient communication intensity value, otherwise, not establishing the connecting edge between the two nodes, so as to form a weighted undirected graph, namely a vortex topological network.
4. The method for estimating wake flow of an offshore wind farm group according to claim 3, wherein a continuous networked gradient field covering the whole target sea area is generated by a spatial weighted interpolation algorithm based on the temperature gradient vector of each node in the vortex topology network and by utilizing the weight information of the connecting edges between each node and all adjacent nodes, and the contribution weight of each node to any interpolation point in space in the spatial weighted interpolation algorithm is determined by the spatial geometric distance of the node to the interpolation point and the sum of the weights of all connecting edges of the node in the vortex topology network; Establishing a quantitative relation model between gradient and flow direction, wherein the gradient guiding vector field defined by the model is formed by weighted synthesis of three vector components and corresponding physical coupling coefficients, wherein the first vector component is a networked gradient field and represents a pure gradient driving effect directly driven by thermodynamic non-uniformity; the method comprises the steps of obtaining historical high-fidelity simulation data, determining an optimal value of a physical coupling coefficient through an optimization algorithm based on a fluid dynamics principle, calibrating the physical coupling coefficient of a first vector and a second vector as time-varying parameters, setting the physical coupling coefficient of a third vector to be a fixed value, and applying a calibrated quantitative relation model, namely weighting and synthesizing a networked gradient field, a background horizontal flow field and a calculated deflection effect field at each spatial position according to the respective calibrated coupling coefficients, so as to construct a gradient guiding vector field covering the whole target sea area.
5. A wake assessment method for an offshore wind farm cluster according to claim 4, wherein mapping the geographical location of each wind park in the wind farm cluster to a vortex topology network comprises: Aiming at each wind generating set, calculating the Euclidean distance from the geographic coordinate of the hub center of each wind generating set to the center position of each node in the vortex topology network, and associating the wind generating set to the node with the nearest Euclidean distance according to the nearest neighbor principle, so as to determine the local dominant vortex environment of the wind generating set; aiming at each wind generating set with nodes related in the mode, the wake centroid motion control equation is solved to obtain the time-space evolution motion trail of the wind generating set, which comprises the following steps: For each set of related nodes, taking the geographical coordinates of the hub center of the set at the estimated starting moment as the initial position of the wake centroid, and sequentially executing the following operations on each calculated time step from the starting moment to the ending moment to advance the wake centroid position, namely acquiring a vector of a background horizontal flow field where the current wake centroid position is located and a vector of a gradient guide vector field, and calculating the wake centroid movement speed of the current time step, wherein the speed is synthesized by three vectors: The method comprises the steps of obtaining a wake mass center, obtaining a dynamic coupling coefficient, obtaining a random diffusion speed vector, obtaining a local turbulence intensity modulation of the current position of the wake mass center, obtaining a dynamic coupling coefficient, obtaining a local turbulence intensity modulation of the current position of the wake mass center, obtaining a local turbulence intensity modulation of the dynamic coupling coefficient, obtaining a local turbulence intensity modulation of the local turbulence intensity vector, obtaining a local turbulence intensity modulation of the local turbulence vector, and obtaining a local turbulence intensity by adopting a dynamic integration method to obtain a local turbulence motion trajectory.
6. The wake evaluation method of the offshore wind farm group according to claim 5, wherein the wake evaluation method is characterized by calculating wake speed deficit distribution through a pre-established physical mapping function according to the scale and strength of a vortex structure passed by a space-time evolution motion track, and specifically comprises the steps of setting a proximity distance threshold value for each point on the track according to the space-time evolution motion track of each unit wake centroid, determining local environment attributes for calculation, namely defining a physical influence area of the vortex structure by taking the center position of the vortex structure represented by each node in a vortex topology network as the center of a circle, taking K times of the scale as the radius, judging whether a current track point is positioned in the physical influence area of any vortex structure, if the current track point is positioned in the physical influence area of a certain vortex structure, directly adopting the scale and strength of the vortex structure as local environment attributes of the current track point, and taking the distance from the current track point to the center position of all structures as the interpolation points, and taking the distance from the current track point as the interpolation point of a weighted set of the adjacent vortex structure, and taking the distance from the current track point to the adjacent structure as the center of the vortex structure as the interpolation point of the interpolation point, and taking the adjacent structure as the interpolation point of the adjacent structure as the interpolation point of the vortex structure of the adjacent structure; the method comprises the steps of inputting the determined local environment attribute into a pre-established physical mapping function based on machine learning, outputting two core parameters of a speed loss core amplitude and a wake characteristic radius at corresponding track points by the physical mapping function, calculating and generating wake speed loss distribution on a two-dimensional plane which takes the track points as the center and is perpendicular to the track direction by utilizing the output core parameters and combining a Gaussian distribution model, and executing the process on each track point of each time step in an evaluation interval, thereby obtaining a dynamic wake speed loss field.
7. The method for estimating the wake flow of the offshore wind farm group according to claim 6 is characterized by comprising the steps of carrying out space-time superposition on all dynamic wake flow velocity deficiency fields to obtain an overall wake flow field, specifically comprising the steps of obtaining free incoming flow wind speeds at each space grid point and each time step in a target sea area, collecting velocity deficiency core amplitudes caused by all upstream unit wake flows at the space grid points, calculating the momentum deficiency caused by each unit wake flow at the point based on a momentum deficiency conservation principle, wherein the value is the difference between the square of the free incoming flow wind speeds and the square of the wind speeds after the influence of the unit wake flows, carrying out square sum superposition on the momentum deficiency caused by all unit wake flows to obtain total momentum deficiency, carrying out squaring operation on the total momentum deficiency to obtain equivalent wind speed deficiency quantity caused by a synthetic effect, subtracting the equivalent wind speed deficiency quantity from the free incoming flow wind speeds to obtain the synthetic wind speeds after the comprehensive influence of the wind farm wake flows, and carrying out the calculation on all the time steps in the evaluation space and the evaluation interval in the target sea area to obtain the overall wind speed; the wake flow evaluation report is generated by extracting wind speed time sequence data of one or more specified downstream target positions in an evaluation period based on the integral wake flow field, comparing the wind speed time sequence with corresponding free incoming flow wind speed time sequences, calculating to obtain wind speed loss rates of the downstream target positions, and generating a quantized wake flow evaluation report based on the calculated wind speed loss rates.
8. A wake assessment device of an offshore wind farm group, characterized in that the assessment device is adapted to perform a wake assessment method of an offshore wind farm group according to any of claims 1-7, comprising: The vortex computing module is used for acquiring ocean surface state parameter fields of a target sea area and vertical profile parameter data of an atmospheric boundary layer, carrying out space-time fusion and registration to form an ocean atmosphere composite field, carrying out multi-scale computing analysis on the composite field by adopting continuous wavelet transformation, identifying vortex structures in the composite field, and extracting the central position, the scale and the strength of each vortex structure; The vector construction module is used for constructing a vortex topological network by taking each vortex structure as a node and taking a gradient communication relation between adjacent vortex structures as an edge, and establishing a quantitative relation model between gradients and flow directions through fluid dynamics by combining a background horizontal flow field of a target sea area based on the vortex topological network so as to construct a gradient guiding vector field covering the whole target sea area; The mass center mapping module is used for mapping the geographic position of each wind generating set in the wind power plant group to a vortex topology network, taking the position of the set at the evaluation starting moment as a wake mass center initial point in a set evaluation interval, and solving a wake mass center motion control equation through a numerical value under the common driving of the gradient guiding vector field and the background horizontal flow field to obtain a space-time evolution motion track; The report evaluation module is used for calculating wake velocity loss distribution according to the scale and the intensity of the vortex structure of each space-time evolution motion track to generate dynamic wake velocity loss fields, performing space-time superposition on all the dynamic wake velocity loss fields to obtain an integral wake field, calculating the wind speed loss of a downstream target position according to the integral wake field, and generating a wake evaluation report.

Description

Wake flow assessment method and device for offshore wind power plant group Technical Field The invention relates to the technical field of wind power generation engineering and computational fluid dynamics, in particular to a wake flow assessment method and device of an offshore wind power generation field group. Background The offshore wind power generation is taken as an important component of clean energy, the large-scale development is usually carried out in the form of wind power station groups, in the mode, the wake generated by the operation of an upstream wind generating set can obviously change the wind resource condition of a downstream area, so that the inflow wind speed of the downstream unit is reduced, the turbulence intensity is increased, the generated energy loss and the mechanical load are further increased, the accurate wake evaluation of the offshore wind power station groups is a key premise for optimizing the layout of the wind power station groups, improving the overall power generation efficiency of the groups and guaranteeing the safe and stable operation of equipment, and particularly in the complex marine atmospheric environment, the dynamic interaction of a sea-air interface makes the wind resource distribution and wake evolution mechanism particularly complex, and the traditional evaluation method faces serious challenges. Currently, the estimation of wake effects of wind power plants mainly depends on two types of technical routes, one type is a numerical simulation method based on computational fluid dynamics, for example, an actuating disc or an actuating line model is used for combining large vortex simulation, so as to analyze a wake field with high fidelity by solving a fluid control equation, the other type is an engineering wake model based on historical data or a simplified physical model and a data driving method, wake speed loss is rapidly predicted by a parameterized formula or a machine learning model, and the methods generally consider an atmospheric boundary layer and a sea surface as uniform or parameterized background conditions, and focus on disturbance generated by a simulation unit as momentum sink on an average flow field and propagation of the disturbance in a downwind direction; However, the prior art schemes have obvious limitations when processing the interaction of the peculiar complex environment at sea, firstly, the secondary circulation structures which are led by heterogeneous thermal processes in the marine atmosphere interface, such as the fundamental influence of thermal vortex on a background flow field and a wake transport path, are generally not fully considered, so that physical mechanism is lost when the wake space deflection, diffusion and recovery processes are evaluated, secondly, the prior art methods mostly adopt a simulation model of unidirectional action, namely, the unit wake evolves in a preset static or simple dynamic background field, the feedback action which can be possibly generated by taking the wake as momentum sink to the local microenvironment is not effectively simulated, and how the dynamic interaction further influences the evolution of the downstream wake, which is not consistent with the bidirectional coupling action mechanism between the actual offshore wind farm group and the atmospheric boundary layer and the dynamic sea surface, and finally, the fast prediction can be realized based on the black box model driven by pure data, but the result often lacks definite physical interpretability, the intrinsic mechanism of the wake influence is difficult to be revealed, so that the application value of the unit is limited in high decision-making steps such as cooperative control and layout depth optimization of the unit. The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art. Disclosure of Invention The invention aims to provide a wake flow assessment method and device for an offshore wind farm group, which are used for solving the problems in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions: a wake flow assessment method of an offshore wind power plant group comprises the following specific steps: Step 1, acquiring ocean surface state parameter field and atmospheric boundary layer vertical section parameter data of a target sea area, performing space-time fusion and registration to form an ocean atmospheric composite field, performing multi-scale calculation analysis on the composite field by adopting continuous wavelet transformation, identifying vortex structures in the composite field, and extracting the central position, the scale and the strength of each vortex structure; Step 2, constructing a vortex topological network by