Wind power plant micro-siting evaluation method based on spatial analysis and climate adaptability

Abstract

The invention discloses a wind power plant micro-siting assessment method based on spatial analysis and climate adaptability, relates to the technical field of wind power plant siting, and solves the problems that in the prior art, spatial analysis depends on static topographic data, climate adaptability assessment uses dynamic meteorological data, and the static topographic data and the dynamic meteorological data lack a space-time synchronization mechanism; and a prediction result has a large deviation. The method comprises the following steps: firstly, acquiring multi-source data and fusing to realize time-space synchronization of the data; then, constructing and solving an objective function to obtain a candidate point set, and screening out an optimized point distribution scheme of which the prevailing wind direction coverage is greater than a set threshold value; and finally, inputting prediction data of various extreme climates into the cost-per-kilowatt-hour model, introducing extreme climate toughness parameters into calculation to carry out climate adaptability data correction, and calculating return on investment cycles under different carbon emission scenes. Through the steps, the limitation that only static wind resources are concerned on traditional site selection is broken through, and combined evaluation of the regional climate adaptability and the complex terrain adaptability is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind farm site selection, and in particular to a wind farm micro-site selection evaluation method based on spatial analysis and climate adaptability. Background Art

[0002] As a core part of wind power project development, the technological evolution of wind farm micro-site selection has always revolved around the two dimensions of spatial analysis and climate adaptability. The traditional method uses geographic information systems (GIS) for terrain modeling and combines it with computational fluid dynamics (CFD) to simulate wind field distribution, and has formed a software tool chain represented by WASP and WindFarmer. These tools can handle wind resource assessments under complex terrains and achieve multi-objective optimization of wind turbine layouts through optimization algorithms (such as genetic algorithms and particle swarm algorithms). In terms of climate adaptability, the industry has gradually introduced parameters such as turbulence intensity, wind shear, and 50-year extreme wind speeds, and combined them with the "Safety Requirements for Wind Turbine Generators" (GB 1845.1-2001) to establish a safety threshold model, forming a technical framework with wind energy resource assessment-safety level classification-power generation calculation as the core.

[0003] In existing technologies, spatial analysis relies on static terrain data, while climate adaptability assessment uses dynamic meteorological data. The two lack a spatiotemporal synchronization mechanism. For example, the wind tower layout scheme used for wind farm micro-site selection does not take into account the differences in microclimate changes in complex terrain. The wind speed assessment error can reach 30%, which has a significant bias in the final wind farm micro-site selection results.

[0004] In view of this, a wind farm micro-site selection assessment method based on spatial analysis and climate adaptability is needed. Summary of the Invention

[0005] In order to address the problem that existing technologies rely on static terrain data for spatial analysis and dynamic meteorological data for climate adaptability assessment, and that the two lack a spatiotemporal synchronization mechanism, resulting in large deviations in prediction results, this invention provides a wind farm micro-site selection and assessment method based on spatial analysis and climate adaptability, which can improve the accuracy of site selection assessment and the value of the project throughout its life cycle. The specific technical solution is as follows:

[0006] The wind farm micro-site selection assessment method based on spatial analysis and climate adaptability includes the following steps:

[0007] Acquire multi-source data through various monitoring and collection methods, and fuse the multi-source data to achieve temporal and spatial synchronization of data;

[0008] The target function is constructed and solved based on at least the dominant wind direction coverage and the turbulence avoidance rate, candidate point sets are obtained, and the wind farm state of the site selection place is simulated by generating virtual wind data at the candidate points to screen out an optimal layout scheme with a dominant wind direction coverage greater than a set threshold;

[0009] The wind field is divided into partitions, each partition independently calibrates the wind tower data weight coefficient, and in the iteration process of the optimization algorithm, the fitness function gradient is adjusted according to the partition weight to guide the population to converge to the high weight area.

[0010] The prediction data of various extreme climates are input into the degree electricity cost model, the extreme climate resilience parameter is introduced into the calculation to correct the climate adaptability data, and the investment return period under different carbon emission scenarios is calculated.

[0011] Preferably, the multi-source data obtained through various monitoring and collection means at least includes: air temperature, air pressure, humidity collected in real time through meteorological satellites, unmanned aerial vehicle weather stations and ground Doppler radars.

[0012] Preferably, the multi-source data is fused to realize the specific process of space-time synchronization of data as follows:

[0013] The WRF-Chem module is embedded in the CFD model, the real-time meteorological data and terrain data are spatio-temporally interpolated, and a three-dimensional wind field model updated at a set time period is established.

[0014] Preferably, after the multi-source data is fused, the following operations are further performed:

[0015] A dynamic air density correction module is set, a variable density term is introduced into the RANS equation, the Boussinesq approximation and the compressible fluid model switching strategy are used, when the temperature difference is greater than a preset value, the compressible model is automatically switched, and the data calculation deviation value caused by the dynamic change of air density is corrected.

[0016] Preferably, the target function is represented as follows:

[0017] F = w1·S 风速代表性 +w2·S 湍流规避率 +w3·S 经济性

[0018] In the formula, w1, w2, and w3 are weight coefficients, S 风速代表性 is the coverage of the wind tower to the dominant wind direction, S 湍流规避率 is the proportion of avoiding strong turbulence area, and S 经济性 is an economic index, which is the sum of initial construction cost, annual operation and maintenance cost and data transmission fee.

[0019] Preferably, the calculation of degree electricity cost is as follows:

[0020]

[0021] In the formula, C cap is the initial construction cost, C om is the annual operation and maintenance cost benchmark value, y is the operation and maintenance cost amplification coefficient caused by extreme weather, and the value is greater than 1, C clim is the additional investment for climate adaptability reconstruction, δ is the equipment residual value rate, E annual is the benchmark annual power generation capacity without considering climate correction, ψ is a product item of multi-dimensional climate correction factors, and the value is as follows:

[0022] ψ = ψ typhoon · ψ ice

[0023] In the formula, ψ typhoon is a typhoon reduction coefficient factor, ψ ice is an icing correction factor.

[0024] Preferably, the following steps are further included:

[0025] The modified degree of electricity cost is added to the calculation of the economic index of the objective function in at least an additive manner, and a viscous vortex wake model is further used to simulate the wake superposition effect between wind turbines, the spacing between wind turbines is optimized by a genetic algorithm, including the spacing in the main wind direction and the spacing in the crosswind direction, so as to reduce the wake loss; in a complex terrain area, such as a valley or a ridge, an asymmetric layout strategy is used, and the row / column spacing of the wind turbine layout in the selected site is adjusted according to the local wind acceleration effect.

[0026] A computer readable storage medium, comprising a stored program, wherein when the program runs, the device where the computer readable storage medium is located is controlled to execute the wind farm micro-siting evaluation method based on spatial analysis and climate adaptability as described above.

[0027] A processor for running a program, wherein when the program runs, the wind farm micro-siting evaluation method based on spatial analysis and climate adaptability as described above is executed.

[0028] Compared with the prior art, the present application has the following beneficial effects:

[0029] The application firstly obtains multi-source data through various monitoring and collecting means, and fuses the multi-source data to realize the space-time synchronization of data; then a target function is constructed and solved based on the dominant wind direction coverage and the turbulence avoidance rate, the candidate point set is obtained, and the wind farm state of the site selection place is simulated by generating virtual wind data at the candidate points to screen out the optimization layout scheme with the dominant wind direction coverage greater than the set threshold; the wind field is processed by partition, each partition independently calibrates the wind tower data weight coefficient, in the optimization algorithm iteration process, the fitness function gradient is adjusted according to the partition weight to guide the population to converge to the high weight area; finally, the prediction data of various extreme climates is input into the degree of electricity cost model, the extreme climate resilience parameter is introduced into the calculation to correct the climate adaptability data, and the investment return period under different carbon emission scenarios is calculated. Through the above steps, the application breaks through the limitation of traditional site selection which only focuses on static wind resources, combines long-term climate prediction and extreme event probability with spatial terrain analysis into the decision framework, realizes the combined evaluation of regional climate adaptability and complex terrain adaptability, and has the function of dynamic economic evaluation, which evaluates the risk and expected return of wind farm investment and construction from multiple aspects. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the specific embodiments or prior art of the present application, the drawings needed in the specific embodiments or prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual proportion.

[0031] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0033] It should be understood that when used in the present specification and the appended claims, the terms "comprise" and "include" indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or sets thereof.

[0034] It should also be understood that the terms used in the specification of the present application are used merely for the purpose of describing particular embodiments and are not intended to limit the present application. As used in the specification and the appended claims of the present application, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0035] It should be further understood that the term "and / or" used in the specification and the appended claims of the present application means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0036] In one embodiment of the present application, a wind farm micro-siting evaluation method based on spatial analysis and climate adaptability is provided, as shown in Figure 1 The method specifically comprises the following steps:

[0037] Step 1: Multi-source data fusion and dynamic time-space synchronization mechanism

[0038] 1.1 High-precision time-space data acquisition

[0039] 1.1.1 Establish a multi-dimensional meteorological monitoring network: deploy a monitoring system of meteorological satellites, unmanned aerial vehicle meteorological stations and ground Doppler radars in linkage, collect dynamic parameters such as air temperature, air pressure and humidity in real time, construct an air density field combining ERA5 reanalysis data, obtain different meteorological data at different point positions in a specified region space, and build a density field matrix data according to the formula: ρ = P / (R·T) and combining spatial coordinate analysis, wherein ρ is air density, R is gas constant, T is temperature, P is pressure of the measuring point, and temperature correction coefficient is integrated to correct the deviation value influence of temperature change of complex terrain during measurement.

[0040] The temperature correction coefficient is defined as follows:

[0041]

[0042] In the formula, α is the temperature correction coefficient, ΔT terr is the temperature deviation amount caused by the terrain, T meas is the instrument measured temperature value, and (x, y, z) is the space three-dimensional coordinate.

[0043] ERA5 reanalysis data is the fifth generation of global climate and weather reanalysis data set developed by the European Center for Medium-Range Weather Forecasts (ECMWF), aiming to build a spatiotemporal continuous global climate database by integrating historical observation data and modern numerical models. The principle is to conduct multi-scale climate simulation based on the integrated forecasting system, coupled with atmospheric, land surface and wave models. In data-scarce areas, reasonable estimates are generated by model physical laws extrapolation. It has an hourly update of analysis fields such as surface pressure and temperature, a mechanism for updating terrain roughness and vegetation data every quarter, and real-time data streaming access.

[0044] 1.1.2 Establish a dynamic terrain update system: Use synthetic aperture radar (SAR) and LiDAR technology to update terrain roughness and vegetation cover data every quarter, implement terrain gridding dynamic data management through GIS platform, obtain dynamic update data of terrain spatial changes, and mark the corresponding terrain time and space based on the terrain spatial change data, finally obtain dynamic terrain data with time and space characteristics.

[0045] 1.2 Build a spatiotemporal coupling modeling framework

[0046] 1.2.1 Synchronize the data obtained, embed the WRF-Chem module in the CFD model, perform spatiotemporal interpolation of real-time meteorological data and terrain data, and establish a three-dimensional wind field model updated every hour. For example, for mountainous areas, use unstructured grids, set the minimum grid size according to the required three-dimensional model accuracy, and based on historical data analysis to capture turbulent vortex structures and simulate and evolve them in the three-dimensional wind field model, dynamically couple and match meteorological data and terrain data, achieve data spatiotemporal synchronization. WRF-Chem module is an extended version of the Weather Research and Forecasting (WRF) model, designed specifically for simulating the evolution of atmospheric chemical components, including gaseous pollutants, aerosols and chemical processes. WRF-Chem is designed for online coupling, enabling simultaneous simulation of meteorological processes and transport, transformation and deposition of atmospheric pollutants, achieving real-time two-way interaction between meteorological and chemical fields, significantly improving simulation accuracy and reliability.

[0047] 1.2.2 Set up dynamic air density correction module, introduce variable density term in RANS equation, use Boussinesq approximation and compressible fluid model switching strategy, when the temperature difference is greater than a certain preset value (for example, greater than 15℃), automatically switch to the compressible model, correct the data calculation deviation value caused by the dynamic change of air density, and reduce the prediction error of power generation. Boussinesq approximation is a fluid mechanics method for simplifying the control equation of buoyancy convection. Its core is to consider the density change caused by temperature in the buoyancy term only, and other physical parameters are considered constant, so as to ensure the accuracy of the results and ensure the efficiency.

[0048] Step 2: Microclimate perception wind tower optimization layout

[0049] 2.1 Reinforcement learning assisted site selection algorithm

[0050] Construct a multi-objective optimization model to form a multi-objective optimization function:

[0051] F = w1-wind speed representative + w2-turbulence avoidance rate + w3-economy;

[0052] Where w1, w2, and w3 are weight coefficients, wind speed representative is the coverage rate of the wind tower to the dominant wind direction, turbulence avoidance rate is the proportion of the site selection avoiding strong turbulence area, and NSGA-III algorithm is used to generate a candidate point set in the three-dimensional terrain model.

[0053] In the optimization layout of wind towers, economy can be evaluated by quantifying the total cost composed of initial construction cost, annual operation and maintenance cost, and data transmission fee. The weight can be determined by AHP or entropy weight method, and dynamically adjusted according to the project life cycle stage and terrain complexity. At the same time, the economy index is directly related to the levelized cost of electricity (LCOE), the cost of wind towers is one of the sub-items of LCOE, and the optimization layout can reduce the power generation deviation cost caused by wind resource assessment error, thereby indirectly reducing LCOE, and achieving the balance between economy and technology.

[0054] NSGA-III algorithm is an evolutionary algorithm for solving multi-objective optimization problems, which is suitable for handling optimization problems with three or more objective functions. Compared with traditional multi-objective algorithms, it performs better in handling high-dimensional target space.

[0055] Virtual wind tower simulation: According to the terrain data distribution in the site selection evaluation area, a plurality of candidate points are generated, based on virtual wind tower technology, virtual wind data is generated at the candidate points, cross validation is carried out with LiDAR radar scanning results, and the site selection scheme with dominant wind direction coverage > 85% is selected as the optimal scheme.

[0056] Generate a candidate point set through the objective function → select multiple candidate points from the virtual wind tower simulation → verify the selected candidate points to screen out a better point arrangement scheme

[0057] 2.2 Microclimate zoning correction strategy

[0058] 2.2.1 Terrain-climate coupling zoning: The wind field is divided into sub-zones with similar turbulence intensity and wind shear index (such as windward slope, leeward slope, valley unit), and the data weight coefficient of each sub-zone is calibrated independently. Specifically, a method based on terrain-climate coupling zoning and dynamic weight calibration is used.

[0059] In complex terrain wind farms, accurate division of microclimate zones is a key step to optimize wind resource assessment. The invention uses a terrain-meteorological dual constraint model to construct a multi-scale climate zoning system. First, based on model analysis data, the terrain slope, slope direction and dominant wind direction parameters are extracted, combined with turbulence intensity, wind shear index and other wind resource feature vectors to form a multi-dimensional data matrix. Through a double distance clustering algorithm (weighted spatial distance and meteorological parameter distance), the wind field is divided into multiple typical microclimate units such as windward slope, leeward slope and valley. To achieve dynamic calibration of zoned data, a variable weight coefficient mechanism is introduced: fixed wind tower data is given a higher weight (0.7-0.9) in stable dominant wind direction areas (such as windward slope areas), while the weight is dynamically adjusted (0.4-0.6) in transition areas (such as valley and pass junctions). This zoning strategy can significantly reduce the overall wind speed assessment error in wind farm assessment applications.

[0060] The final effect of the above assigned weights is as follows:

[0061] Data fusion layer: When calculating the representative wind speed, different zoned wind tower data is weighted and averaged differently. For example, windward slope area data is involved in dominant wind direction coverage calculation with a weight of 0.7-0.9, and transition area data is involved with a dynamically adjusted weight of 0.4-0.6, forming a zoned weighted coverage index.

[0062] Model training layer: When building a wind speed prediction model, the zoned weight is introduced as a regularization term into the loss function, so that the model training focuses more on the data matching degree in high weight areas.

[0063] Error backpropagation layer: During the optimization algorithm iteration process, the fitness function gradient is adjusted according to the zoned weight to guide the population to converge to high weight areas.

[0064] 2.2.2 Mobile wind measurement device deployment: Deploy vehicle-mounted SODAR equipment at the boundaries of the zoning to monitor the wind speed gradient changes in the transition area in real time and correct the extrapolation error of the fixed wind tower.

[0065] Specifically, SODAR is deployed in the transition zone of the partition, and multi-beam scanning technology is used to capture wind speed gradient and turbulence vortex evolution in real time. Through data assimilation algorithm, mobile observation data is combined with CFD coupling model to establish three-dimensional wind field dynamic correction matrix. The process includes: ① Based on the wind speed gradient of the partition boundary, a Gaussian process regression model is constructed; ② Kalman filter is used to fuse SODAR data and fixed wind tower data; ③ The extrapolation model parameters are updated through iterative optimization. In addition, the deployment strategy is set to dynamic self-adaptation, increasing the SODAR monitoring frequency to once an hour during the monsoon transition period (such as spring and summer), and adjusting to three times a day during the stable climate period, to realize the optimization of data accuracy and operation cost.

[0066] SODAR (Sonic Detection And Ranging) is an active remote sensing device for atmospheric detection using sound waves. It emits sound waves and receives their backscattered signals, and combines the Doppler effect to invert atmospheric parameters.

[0067] Step 3: Dynamic flow field simulation and data bidirectional correction

[0068] 3.1 Multi-scale nested simulation technology

[0069] 3.1.1 Medium-micro coupling model: Based on the architecture of embedding WRF in CFD model, the outer layer uses 3 km mesoscale grid to simulate atmospheric boundary layer, and the inner layer realizes bidirectional coupling of wind turbine wake and terrain through actuator disk model, solving the problem of underestimating tail loss in traditional model.

[0070] Specifically, the multi-scale coupling model based on the weather research and forecast model→computational fluid dynamics nested architecture realizes the coordinated simulation of atmospheric boundary layer and wind turbine wake through hierarchical grid division and physical field transmission mechanism. The outer layer uses 3 km mesoscale grid of weather research and forecast model to simulate the thermal-dynamic process of atmospheric boundary layer based on non-hydrostatic equation, including temperature gradient, pressure field and boundary layer turbulence parameterization, and through four-dimensional data assimilation technology, the weather reanalysis data is dynamically injected into the model boundary condition. The inner layer computational fluid dynamics model realizes the bidirectional coupling of wind turbine and terrain through actuator disk model, which simplifies the wind wheel as a porous medium disc, embeds volume force source term in Navier-Stokes equation to represent blade thrust and tangential force distribution, and dynamically corrects the momentum loss of wake through momentum conservation equation.

[0071] 3.1.2 Transient turbulence simulation: For extreme weather events such as typhoon, large eddy simulation mode is enabled, combining GPU parallel computing to compress the simulation step to 0.1 second level, capturing the characteristics of instantaneous wind speed mutation.

[0072] High-precision transient simulation of extreme weather events adopts a large eddy simulation model combined with a graphics processor parallel computing framework, breaking through the time and space resolution limits of traditional steady-state simulation. The large eddy simulation model is based on filtered Navier-Stokes equations, analytically resolving inertial sub-range vortices through a dynamic sub-grid stress model, and capturing the anisotropic characteristics of near-ground turbulence using a nonlinear return-to-isotropy model and an anisotropic sub-grid model.

[0073] To reduce computational resource consumption, a non-uniform vertical grid encryption technique is used to set the grid resolution to ≤2 meters in the near-ground area (0-200 meters) and gradually relax it to more than 10 meters in the high-altitude area, combined with a graphics processor heterogeneous computing architecture to achieve large-scale parallelization. Combined with an actuator disk dynamic thrust coefficient correction module, the strength of the volume force source term is adjusted according to the real-time incoming flow wind speed, effectively simulating the wake oscillation and vortex shedding phenomena caused by the sudden change of tip speed ratio during typhoon passage, achieving simulation of extreme weather, pre-simulating and evaluating the extreme weather pressure that the wind farm can bear in this terrain, and verifying the safety risks of this site.

[0074] 3.2 Data closed-loop correction system

[0075] 3.2.1 Online error feedback mechanism: compare the simulated wind speed with the actual output data in real time, and trigger model parameter self-correction when the deviation is >8%.

[0076] The online error feedback mechanism achieves high-frequency collaborative calibration of the simulation model and the real wind field through real-time data flow closed-loop control. The system collects wind turbine output data such as power curves and yaw angles in real time based on edge computing nodes, and meteorological station measured wind speed, uses a sliding time window to statistically construct a dynamic error matrix, and calculates the normalized absolute deviation of simulated wind speed and actual output data. When the deviation threshold breaks through 8%, a three-level response mechanism is triggered: ① use Kalman filter algorithm to separate system error and random noise; ② dynamically correct key physical quantities such as surface roughness coefficient and atmospheric stability parameter through gradient descent method; ③ use online learning module to update the thrust coefficient mapping relationship of the actuator disk model to realize real-time compensation of wake momentum loss.

[0077] 3.2.2 Digital twin verification platform: build a digital twin containing a historical climate pattern library, and evaluate the power generation fluctuation range under different climate scenarios through Monte Carlo simulation.

[0078] The digital twin verification platform takes a multi-modal historical climate database as the core, integrates reanalysis data, climate scenario simulation, and regional downscaling models, and constructs a climate pattern feature library. The specific process includes: ① establish a nonlinear response surface model of climate variables (wind speed, temperature, pressure) and power generation; ② introduce Copula function to describe the asymmetric correlation of multivariate joint distribution; ③ use resampling method to calculate the power generation quantile of 95% confidence interval.

[0079] Copula function is a mathematical tool that connects the multivariate joint distribution function with its marginal distribution function. Its core value lies in separating the randomness and coupling of variables, that is, the randomness of individual variables is described by the marginal distribution, while the dependence structure between variables is described by the Copula function.

[0080] In summary, step 3 realizes high-precision simulation and real-time optimization of the flow field of the wind farm through dynamic flow field simulation and data bidirectional correction technology. Specifically, the multi-scale nested simulation technology combines the mesoscale WRF model and the microscale CFD model. The outer 3 km grid captures the thermal-dynamic process of the atmospheric boundary layer, and the inner actuator disk model dynamically couples the wind turbine wake and the terrain, significantly improving the simulation accuracy of the wake effect. Especially in extreme weather, large eddy simulation combined with GPU parallel computing can capture transient turbulence at 0.1 seconds, accurately restoring the wake oscillation and vortex shedding phenomena in extreme events such as typhoons. At the same time, the data closed-loop correction system establishes a bidirectional feedback mechanism between simulation and measurement: the online error feedback module compares the simulated wind speed and wind turbine output data in real time. When the deviation exceeds 8%, it automatically triggers the Kalman filter to separate errors, corrects key parameters such as surface roughness using the gradient descent method, and updates the actuator disk thrust coefficient. The digital twin platform integrates a historical climate model library and uses Copula function to build a nonlinear response model of wind speed and power generation. Through Monte Carlo simulation, it quantifies the power generation fluctuation range (95% confidence interval) under different climate scenarios. This step not only reduces the wind speed evaluation error of the entire field, but also reduces the power generation prediction deviation under extreme climate, providing key technical support for the safe operation and maximum power generation of wind farms.

[0081] Step 4: Climate adaptability evaluation and dynamic optimization

[0082] 4.1 Climate resilience index system

[0083] For a few extreme climate changes, set the extreme climate resilience parameter, for example, define the typhoon reduction coefficient ψ typhoon = v / v 额定 , where v is the 10-minute average maximum wind speed of 50-year return period, v 额定 is the rated wind speed of typhoon, and α>1.3 is required; for example, introduce the icing condition turbulence intensity correction factor ψ ice =1+0.15T ice , where T ice is the annual icing days.

[0084] The predicted data of various extreme weather conditions are input into the degree electricity cost model, the extreme weather resilience parameter is introduced into the calculation to correct the climate adaptability data, the investment return period under different carbon emission scenarios is calculated, and the indicators are optimized through fan selection and terrain layout adjustment.

[0085] The calculation of degree electricity cost is as follows:

[0086]

[0087] In the formula, C cap is the initial construction cost, C om is the annual operation and maintenance cost benchmark value, y is the operation and maintenance cost amplification coefficient caused by extreme weather, which is greater than 1, C clim is the additional investment for climate adaptability modification, δ is the equipment residual value rate, E annual is the baseline annual power generation without considering climate correction, ψ is the product of multi-dimensional climate correction factors, and the value is as follows:

[0088] ψ = ψ typhoon · ψ ice

[0089] In the formula, ψ typhoon is the typhoon reduction coefficient factor, ψ ice is the icing correction factor.

[0090] 4.2 Wake effect and micro layout optimization

[0091] The viscous vortex wake model is used to simulate the superposition effect of wind turbine wake, and the genetic algorithm is used to optimize the spacing between wind turbines, including the main wind direction spacing and the crosswind direction spacing, so as to reduce the wake loss. In complex terrain areas such as valleys and ridges, an asymmetric layout strategy is adopted, and the row / column spacing of wind turbine layout in the selected site is adjusted according to the local wind acceleration effect.

[0092] The viscous vortex wake model is a mathematical method for simulating turbulent phenomena. It describes the behavior of fluid by simulating the vortex structure in turbulent flow and considering stress transmission. The model assumes that there are infinite vortices in the flow field. Under the interaction of these vortices, a series of forcing and dissipation actions occur between the vortices, which macroscopically presents the phenomenon of turbulent flow. In wind turbine layout optimization, the viscous vortex wake model is used to accurately predict the wind turbine wake effect and evaluate the wake loss under different layout schemes.

[0093] First, the viscous vortex wake model is used to simulate the wake effect under different fan layout schemes, and the wake loss is calculated. Then, the wake loss is input into the genetic algorithm as part of the fitness function. The genetic algorithm searches for the fan layout scheme with the smallest wake loss and the highest power generation efficiency through iterative optimization. Specifically, the input parameters of the genetic algorithm include fan parameters (such as rotor diameter, rated power, etc.), terrain data, initial layout scheme, etc.; the output parameters are the optimized inter-machine spacing (including the main wind direction spacing and the side wind direction spacing); other configuration parameters include population size, crossover rate, mutation rate, etc.

[0094] Local wind acceleration effect refers to the phenomenon of local wind speed increase caused by terrain changes. In complex terrain, such as ridges, valleys, etc., wind speed will change significantly due to terrain undulations. For example, at the ridge, wind speed may increase significantly due to the terrain lifting effect; while in the valley, wind speed may decrease due to the terrain shielding effect.

[0095] Asymmetric layout strategy adjusts the row / column spacing of wind turbine arrangement based on local wind acceleration effect. In wind acceleration area, increase the spacing between wind turbines to avoid wake superposition effect; in areas with lower wind speed, appropriately reduce the spacing between wind turbines to improve wind energy utilization. The specific adjustment strategy is as follows:

[0096] Terrain analysis: Determine the wind speed variation area caused by terrain through CFD simulation or wind tunnel test.

[0097] Spacing adjustment: Adjust the row / column spacing of wind turbine arrangement according to the wind speed variation area. In the wind acceleration area, the spacing between wind turbines can be increased to 5-7 times the rotor diameter; in the area with lower wind speed, the spacing between wind turbines can be reduced to 3-5 times the rotor diameter.

[0098] Asymmetric layout: In complex terrain areas, such as valleys, ridges, etc., adopt asymmetric layout strategy to adjust the row / column spacing of wind turbine arrangement based on local wind acceleration effect, so as to fully utilize wind energy resources and reduce wake loss.

[0099] The role of this step is reflected in the following aspects:

[0100] Reduce wake loss: By optimizing the inter-machine spacing, reduce the wake superposition effect between wind turbines, reduce the wake loss, and improve the overall power generation efficiency of the wind farm.

[0101] Increase power generation: By reasonably arranging wind turbines, fully utilize wind energy resources, and increase the annual power generation of the wind farm.

[0102] Reduce cost: By optimizing the inter-machine spacing, reducing unnecessary number of wind turbines and operation and maintenance cost, reducing the levelized cost of wind power.

[0103] Through the climate-space double driving data analysis, the application breaks through the limitation of traditional site selection only focusing on static wind resources, combines long-term climate prediction and extreme event probability with spatial terrain analysis into the decision framework, realizes the combined evaluation of regional climate adaptability and complex terrain adaptability, and has the function of dynamic economic evaluation, so that the risks and expected benefits of wind power plant investment and construction are evaluated from multiple aspects.

[0104] Those skilled in the art can appreciate that the units of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both, and in order to clearly illustrate the interchangeability of hardware and software, the components of the examples have been described in the above description in general terms. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0105] In the embodiments provided by the application, it should be understood that the division of units is only a logical functional division, and there can be another division manner in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0106] In addition, each functional unit in the various embodiments of the application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0107] When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the description of the present application.

Claims

1. A wind farm micro-site selection and assessment method based on spatial analysis and climate adaptability, characterized by: The following steps are involved: Acquire multi-source data through various monitoring and collection methods, and fuse the multi-source data to achieve temporal and spatial synchronization of data; At least based on the dominant wind direction coverage and turbulence avoidance rate, an objective function is constructed and solved to obtain a set of candidate points. Virtual wind measurement data is generated at the candidate points to simulate the state of the wind farm at the site, and an optimized site layout scheme with a dominant wind direction coverage greater than a set threshold is selected; The wind field is partitioned and the wind tower data weight coefficient is independently calibrated for each partition. During the iterative process of the optimization algorithm, the fitness function gradient is adjusted according to the partition weight to guide the population to converge to the high-weight area. The predicted data of various extreme climates are input into the cost per kilowatt-hour model, and the extreme climate resilience parameters are introduced into the calculation to correct the climate adaptability data and calculate the investment return cycle under different carbon emission scenarios.

2. The wind farm micro-site selection and evaluation method based on spatial analysis and climate adaptability according to claim 1 is characterized in that: The multi-source data obtained through various monitoring and collection methods include at least: temperature, air pressure and humidity collected in real time by meteorological satellites, UAV weather stations and ground Doppler radars.

3. The wind farm micro-site selection and evaluation method based on spatial analysis and climate adaptability according to claim 1 is characterized in that: The specific process of fusing multi-source data and achieving spatiotemporal synchronization of data is as follows: The WRF-Chem module is embedded in the CFD model to perform spatiotemporal interpolation between real-time meteorological data and terrain data to establish a three-dimensional wind field model that is updated at a set time period.

4. The wind farm micro-site selection and evaluation method based on spatial analysis and climate adaptability according to claim 3 is characterized in that: After fusing multi-source data, the following operations are performed: A dynamic air density correction module is set up, and a variable density term is introduced into the RANS equation. The Boussinesq approximation and compressible fluid model switching strategy are used. When the temperature difference is greater than the preset value, the compressible model is automatically switched to correct the data calculation deviation caused by the dynamic change of air density.

5. The wind farm micro-site selection and evaluation method based on spatial analysis and climate adaptability according to claim 1 is characterized in that: The objective function is expressed as follows: F=w1·S 风速代表性 +w2·S 湍流规避率 +w3·S 经济性 Where w1, w2, w3 are weight coefficients, S 风速代表性 is the coverage rate of the wind tower to the dominant wind direction, S 湍流规避率 is the ratio of points avoiding strong turbulence areas, S 经济性 It is an economic indicator, which is the sum of initial construction cost, annual operation and maintenance cost and data transmission cost.

6. The wind farm micro-site selection and evaluation method based on spatial analysis and climate adaptability according to claim 1, characterized in that: The cost per kilowatt-hour is calculated as follows: Where C cap is the initial construction cost, C om is the annual operation and maintenance cost benchmark value, y is the operation and maintenance cost magnification coefficient caused by extreme climate, and its value is greater than 1, C clim Additional investment for climate adaptation transformation, δ is the equipment residual value rate, E annual is the base year power generation without considering climate correction, ψ is the product term of multidimensional climate correction factor, and its value is as follows: ψ=ψ typhoon ·ψ ice Where, ψ typhoon is the typhoon reduction factor, ψ ice is the icing correction factor.

7. The wind farm micro-site selection and evaluation method based on spatial analysis and climate adaptability according to claim 1, characterized in that: The following steps are also included: At least add the corrected LCOE to the economic performance indicator calculation of the objective function in an additive manner, and further use a viscous vortex wake model to simulate the wake superposition effect between wind turbines. Optimize the turbine spacing, including the main wind direction spacing and the crosswind direction spacing, through a genetic algorithm to reduce wake losses. In complex terrain areas, such as valleys and ridges, an asymmetric layout strategy is adopted to adjust the row / column spacing of wind turbines at the selected site based on the local wind acceleration effect.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to execute the wind farm micro-site selection assessment method based on spatial analysis and climate adaptability according to any one of claims 1 to 7.

9. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the wind farm micro-site selection and assessment method based on spatial analysis and climate adaptability according to any one of claims 1 to 7.

Images