CN115034159A - Power prediction method, device, storage medium and system for offshore wind farm - Google Patents
Power prediction method, device, storage medium and system for offshore wind farmDownload PDFInfo
- Publication number
- CN115034159A CN115034159ACN202210718925.4ACN202210718925ACN115034159ACN 115034159 ACN115034159 ACN 115034159ACN 202210718925 ACN202210718925 ACN 202210718925ACN 115034159 ACN115034159 ACN 115034159A
- Authority
- CN
- China
- Prior art keywords
- wind
- power
- prediction
- wind farm
- real
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Geometry (AREA)
- Marketing (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Power Engineering (AREA)
- Quality & Reliability (AREA)
- Algebra (AREA)
- Fluid Mechanics (AREA)
Abstract
Description
Translated fromChinese技术领域technical field
本发明涉及海上风电场的功率预测技术领域，尤其涉及一种海上风电场的功率预测方法、装置、计算机可读存储介质及系统。The present invention relates to the technical field of power prediction of offshore wind farms, and in particular, to a power forecasting method, device, computer-readable storage medium and system for offshore wind farms.
背景技术Background technique
相比于陆上风电，海上风电的预测难度有较大的增加。首先，陆上风电场通常分布面积较广，风功率的波动随风场规模的增大趋于缓和；而相同装机容量的海上风电场具有“高集中”的发展趋势，使得功率波动可能达到非常显著的水平；再加之海上气温、气压、风速等气象要素容易突变，加剧了风场出力在短时间内大幅度波动，发生爬坡事件，这些以风力发电产生严重变化为特征的事件与风电的低可预测性直接相关，严重影响风电预测统计模型的整体性能。Compared with onshore wind power, offshore wind power is more difficult to forecast. First of all, onshore wind farms usually have a wide distribution area, and the fluctuation of wind power tends to ease with the increase of wind farm scale; while offshore wind farms with the same installed capacity have a development trend of "high concentration", so that power fluctuations may reach very high levels. In addition, meteorological elements such as sea temperature, air pressure, and wind speed are prone to sudden changes, aggravating the large fluctuation of wind farm output in a short period of time, and the occurrence of slope climbing events. These events characterized by serious changes in wind power generation are related to wind power generation. Low predictability is directly related and seriously affects the overall performance of wind power forecasting statistical models.
在现有技术中，通常通过对海上风电环境进行获取，并根据环境参数和环境模型对海上风电场进行物理建模，从而预测海上风电场的功率。In the prior art, the power of the offshore wind farm is usually predicted by acquiring the offshore wind power environment, and physically modeling the offshore wind farm according to environmental parameters and an environmental model.
但是，现有技术仍存在如下缺陷：海水的高比热容、海上的风流热效应以及放大的尾流效应，使得海上风电场的物理建模运算非常繁琐；且由于地理环境的差异，物理建模的灵活泛化性存在一定缺陷，并不适用于海上的短期功率预测；在此基础上，不同于陆上风电，我国用于海上风电场的NWP(Numerical Weather Prediction，数值天气预报)信息开发晚、精度不高，使预测误差增大，因此，不准确的数值天气预报信息也是预测误差的主要来源之一。However, the existing technology still has the following defects: the high specific heat capacity of seawater, the thermal effect of wind flow at sea, and the amplified wake effect make the physical modeling operation of offshore wind farms very complicated; and due to differences in geographical environment, the flexibility of physical modeling There are certain defects in generalization, and it is not suitable for short-term power prediction in the sea; on this basis, unlike onshore wind power, my country's NWP (Numerical Weather Prediction, Numerical Weather Prediction) information for offshore wind farms is developed late and accurate. If it is not high, the prediction error will increase. Therefore, inaccurate numerical weather forecast information is also one of the main sources of prediction error.
因此，当前需要一种海上风电场的功率预测方法、装置、计算机可读存储介质以及系统，从而克服现有技术中存在的上述缺陷。Therefore, there is currently a need for a power prediction method, device, computer-readable storage medium and system for an offshore wind farm, so as to overcome the above-mentioned defects in the prior art.
发明内容SUMMARY OF THE INVENTION
本发明提供一种海上风电场的功率预测方法、装置、计算机可读存储介质以及系统，以提升海上风电场的功率预测的准确性。The present invention provides a power prediction method, device, computer-readable storage medium and system for an offshore wind farm, so as to improve the accuracy of the power prediction of the offshore wind farm.
本发明一实施例提供一种海上风电场的功率预测方法，所述功率预测方法包括：An embodiment of the present invention provides a power prediction method for an offshore wind farm, where the power prediction method includes:
获取海上风电场的环境数据组、发电量数据、实时测风塔数据以及数值天气预报数据，并根据所述环境数据组、预设的致动线风电机组模型以及预设的大涡湍流模型，建立并计算所述海上风电场的第一流场模型结果；Obtain the environmental data set, power generation data, real-time wind tower data and numerical weather forecast data of the offshore wind farm, and according to the environmental data set, the preset actuation line wind turbine model and the preset large eddy turbulence model, establishing and calculating a first flow field model result for the offshore wind farm;
根据预设的尾流条件组、所述实时测风塔数据以及所述第一流场模型结果，推算实时风资源模拟观测数据组以及风机点位风速组，并根据所述数值天气预报数据，拟合所述环境数据组以及所述发电量数据之间的第一统计关系；According to the preset wake condition group, the real-time wind measurement tower data and the results of the first flow field model, the real-time wind resource simulation observation data group and the fan point wind speed group are calculated, and according to the numerical weather forecast data, fitting a first statistical relationship between the environmental data set and the power generation data;
根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，对所述海上风电场进行功率预测以获取预测结果。According to the real-time wind resource simulation observation data set, the wind speed set at the point of the wind turbine, and the first statistical relationship, power prediction is performed on the offshore wind farm to obtain a prediction result.
作为上述方案的改进，根据所述环境数据组、预设的致动线风电机组模型以及预设的大涡湍流模型，建立并计算所述海上风电场的第一流场模型结果，具体包括：As an improvement of the above scheme, according to the environmental data set, the preset actuation line wind turbine model and the preset large eddy turbulence model, the results of the first flow field model of the offshore wind farm are established and calculated, specifically including:
根据预设的致动线风电机组模型以及预设的大涡湍流模型，建立所述海上风电场的第一流场模型；establishing a first flow field model of the offshore wind farm according to a preset actuation line wind turbine model and a preset large eddy turbulence model;
根据所述环境数据组以及所述第一流场模型，计算所述海上风电场的第一流场模型结果。A first flow field model result of the offshore wind farm is calculated according to the environmental data set and the first flow field model.
作为上述方案的改进，根据预设的尾流条件组、所述实时测风塔数据以及所述第一流场模型结果，推算实时风资源模拟观测数据组以及风机点位风速组，具体包括：As an improvement of the above scheme, according to the preset wake condition group, the real-time wind measurement tower data and the results of the first flow field model, the real-time wind resource simulation observation data group and the fan point wind speed group are calculated, specifically including:
根据所述第一流场模型结果以及所述实时测风塔数据，推算海上风电场中风机的各个点位的各个高度的风资源模拟观测数据组；According to the result of the first flow field model and the real-time wind tower data, calculate the wind resource simulation observation data set of each height of each point of the wind turbine in the offshore wind farm;
根据所述风资源模拟观测数据组以及预设的尾流条件组，计算输出在所述尾流条件组下的风机的各个点位的计算风速，并将所述风速存储入风机点位风速组。According to the wind resource simulation observation data set and the preset wake condition set, calculate and output the calculated wind speed of each point of the fan under the wake condition set, and store the wind speed into the fan point wind speed set .
作为上述方案的改进，根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，对所述海上风电场进行功率预测以获取预测结果，具体包括：As an improvement of the above scheme, according to the real-time wind resource simulation observation data group, the wind speed group at the wind turbine point and the first statistical relationship, perform power prediction on the offshore wind farm to obtain a prediction result, which specifically includes:
根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，基于CFD动力降尺度，对所述海上风电场进行风电功率短期预测以获取短期预测结果；According to the real-time wind resource simulation observation data set, the wind speed set at the point of the wind turbine, and the first statistical relationship, based on CFD dynamic downscaling, perform short-term wind power prediction on the offshore wind farm to obtain a short-term prediction result;
根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，根据多种预测方法，对所述海上风电场进行风电功率集合预测以获取集合预测结果；According to the real-time wind resource simulation observation data set, the wind speed set at the wind turbine location, and the first statistical relationship, and according to a variety of prediction methods, the wind power set prediction is performed on the offshore wind farm to obtain the set prediction result;
将所述短期预测结果以及所述集合预测结果作为预测结果进行输出。The short-term prediction result and the aggregated prediction result are output as prediction results.
作为上述方案的改进，根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，基于CFD动力降尺度，对所述海上风电场进行风电功率短期预测以获取短期预测结果，具体包括：As an improvement of the above solution, according to the real-time wind resource simulation observation data set, the wind speed set at the point of the wind turbine, and the first statistical relationship, based on CFD dynamic downscaling, short-term prediction of wind power for the offshore wind farm is performed to obtain Get short-term forecast results, including:
根据所述实时风资源模拟观测数据组以及所述风机点位风速组，通过预设的中尺度风速预报订正方法，对所述数值天气预报数据进行订正；According to the real-time wind resource simulation observation data set and the wind speed set at the fan point, correct the numerical weather forecast data by using a preset mesoscale wind speed forecast and correction method;
根据订正后的数值天气预报数据，在预设的微尺度风场基础数据库中选出对应的风电场微尺度风场分布，分析计算各个风机的自由来流风速；According to the revised numerical weather forecast data, select the corresponding wind farm micro-scale wind field distribution in the preset micro-scale wind farm basic database, and analyze and calculate the free flow wind speed of each fan;
根据所述第一统计关系、预设的风电机组模拟发电功率订正方法以及所述自由来流风速，计算得到各个风机的模拟发电功率；According to the first statistical relationship, the preset correction method for the simulated generated power of the wind turbine, and the free flow wind speed, the simulated generated power of each wind turbine is obtained by calculating;
根据预设的风电场运行维护计划，去除不运行的机组和机组不运行的时段，得到风电场全场发电功率预测。According to the preset operation and maintenance plan of the wind farm, the non-operational units and the non-operational periods of the units are removed to obtain the wind farm's full-scale power generation forecast.
作为上述方案的改进，根据所述风资源模拟观测数据组以及预设的尾流条件组，计算输出在所述尾流条件组下的风机的各个点位的计算风速，具体包括：As an improvement of the above solution, according to the wind resource simulation observation data group and the preset wake condition group, calculate and output the calculated wind speed of each point of the fan under the wake condition group, which specifically includes:
根据预设的Larsen尾流计算模型以及风速数据组，计算风机的各个点位的计算风速；所述风资源模拟观测数据组包括风速数据组。According to the preset Larsen wake calculation model and the wind speed data set, the calculated wind speed of each point of the wind turbine is calculated; the wind resource simulation observation data set includes the wind speed data set.
作为上述方案的改进，所述多种预测方法包括：基于禁忌算法的神经网络预测方法、功率时间序列预测方法、粒子群算法以及卡尔曼滤波修正的功率预测方法。As an improvement of the above solution, the multiple prediction methods include: a neural network prediction method based on the tabu algorithm, a power time series prediction method, a particle swarm algorithm, and a power prediction method modified by Kalman filter.
本发明另一实施例对应提供了一种海上风电场的功率预测装置，所述功率预测装置包括数据获取单元、计算拟合单元以及功率预测单元，其中，Another embodiment of the present invention correspondingly provides a power prediction device for an offshore wind farm, the power prediction device includes a data acquisition unit, a calculation fitting unit, and a power prediction unit, wherein,
所述数据获取单元用于获取海上风电场的环境数据组、发电量数据、实时测风塔数据以及数值天气预报数据，并根据所述环境数据组、预设的致动线风电机组模型以及预设的大涡湍流模型，建立并计算所述海上风电场的第一流场模型结果；The data acquisition unit is used to acquire the environmental data set, power generation data, real-time wind tower data and numerical weather forecast data of the offshore wind farm, and according to the environmental data set, the preset actuation line wind turbine model and the forecast The established large eddy turbulence model, establish and calculate the results of the first flow field model of the offshore wind farm;
所述计算拟合单元用于根据预设的尾流条件组、所述实时测风塔数据以及所述第一流场模型结果，推算实时风资源模拟观测数据组以及风机点位风速组，并根据所述数值天气预报数据，拟合所述环境数据组以及所述发电量数据之间的第一统计关系；The calculation and fitting unit is configured to calculate the real-time wind resource simulation observation data set and the fan point wind speed set according to the preset wake condition set, the real-time wind measurement tower data and the first flow field model result, and Fitting a first statistical relationship between the environmental data set and the power generation data according to the numerical weather forecast data;
所述功率预测单元用于根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，对所述海上风电场进行功率预测以获取预测结果。The power prediction unit is configured to perform power prediction on the offshore wind farm to obtain a prediction result according to the real-time wind resource simulation observation data group, the wind speed group at the point of the wind turbine, and the first statistical relationship.
作为上述方案的改进，所述数据获取单元还用于：As an improvement of the above solution, the data acquisition unit is also used for:
根据预设的致动线风电机组模型以及预设的大涡湍流模型，建立所述海上风电场的第一流场模型；establishing a first flow field model of the offshore wind farm according to a preset actuation line wind turbine model and a preset large eddy turbulence model;
根据所述环境数据组以及所述第一流场模型，计算所述海上风电场的第一流场模型结果。A first flow field model result of the offshore wind farm is calculated according to the environmental data set and the first flow field model.
作为上述方案的改进，所述计算拟合单元还用于：As an improvement of the above scheme, the calculation and fitting unit is also used for:
根据所述第一流场模型结果以及所述实时测风塔数据，推算海上风电场中风机的各个点位的各个高度的风资源模拟观测数据组；According to the result of the first flow field model and the real-time wind tower data, calculate the wind resource simulation observation data set of each height of each point of the wind turbine in the offshore wind farm;
根据所述风资源模拟观测数据组以及预设的尾流条件组，计算输出在所述尾流条件组下的风机的各个点位的计算风速，并将所述风速存储入风机点位风速组。According to the wind resource simulation observation data set and the preset wake condition set, calculate and output the calculated wind speed of each point of the fan under the wake condition set, and store the wind speed into the fan point wind speed set .
作为上述方案的改进，所述功率预测单元还用于：As an improvement of the above solution, the power prediction unit is also used for:
根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，基于CFD动力降尺度，对所述海上风电场进行风电功率短期预测以获取短期预测结果；According to the real-time wind resource simulation observation data set, the wind speed set at the point of the wind turbine, and the first statistical relationship, based on CFD dynamic downscaling, perform short-term wind power prediction on the offshore wind farm to obtain a short-term prediction result;
根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，根据多种预测方法，对所述海上风电场进行风电功率集合预测以获取集合预测结果；According to the real-time wind resource simulation observation data set, the wind speed set at the wind turbine location, and the first statistical relationship, and according to a variety of prediction methods, the wind power set prediction is performed on the offshore wind farm to obtain the set prediction result;
将所述短期预测结果以及所述集合预测结果作为预测结果进行输出。The short-term prediction result and the aggregated prediction result are output as prediction results.
作为上述方案的改进，所述功率预测单元还用于：As an improvement of the above solution, the power prediction unit is also used for:
根据所述实时风资源模拟观测数据组以及所述风机点位风速组，通过预设的中尺度风速预报订正方法，对所述数值天气预报数据进行订正；According to the real-time wind resource simulation observation data set and the wind speed set at the fan point, correct the numerical weather forecast data by using a preset mesoscale wind speed forecast and correction method;
根据订正后的数值天气预报数据，在预设的微尺度风场基础数据库中选出对应的风电场微尺度风场分布，分析计算各个风机的自由来流风速；According to the revised numerical weather forecast data, select the corresponding wind farm micro-scale wind field distribution in the preset micro-scale wind farm basic database, and analyze and calculate the free flow wind speed of each fan;
根据所述第一统计关系、预设的风电机组模拟发电功率订正方法以及所述自由来流风速，计算得到各个风机的模拟发电功率；According to the first statistical relationship, the preset correction method for the simulated generated power of the wind turbine, and the free flow wind speed, the simulated generated power of each wind turbine is obtained by calculating;
根据预设的风电场运行维护计划，去除不运行的机组和机组不运行的时段，得到风电场全场发电功率预测。According to the preset operation and maintenance plan of the wind farm, the non-operational units and the non-operational periods of the units are removed to obtain the wind farm's full-scale power generation forecast.
作为上述方案的改进，所述计算拟合单元还用于：As an improvement of the above scheme, the calculation and fitting unit is also used for:
根据预设的Larsen尾流计算模型以及风速数据组，计算风机的各个点位的计算风速；所述风资源模拟观测数据组包括风速数据组。According to the preset Larsen wake calculation model and the wind speed data set, the calculated wind speed of each point of the wind turbine is calculated; the wind resource simulation observation data set includes the wind speed data set.
本发明另一实施例提供了一种计算机可读存储介质，所述计算机可读存储介质包括存储的计算机程序，其中，在所述计算机程序运行时控制所述计算机可读存储介质所在设备执行如前所述的海上风电场的功率预测方法。Another embodiment of the present invention provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, wherein, when the computer program runs, the device where the computer-readable storage medium is located is controlled to execute the following The aforementioned power prediction method for offshore wind farms.
本发明另一实施例提供了一种海上风电场的功率预测系统，所述功率预测系统包括处理器、存储器以及存储在所述存储器中且被配置为由所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现如前所述的海上风电场的功率预测方法。Another embodiment of the present invention provides a power prediction system for an offshore wind farm, the power prediction system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the When the processor executes the computer program, the aforementioned method for predicting power of an offshore wind farm is implemented.
与现有技术相比，本技术方案存在如下有益效果：Compared with the prior art, the technical solution has the following beneficial effects:
本发明提供了一种海上风电场的功率预测方法、装置、计算机可读存储介质以及系统，通过数值天气预报和第一流场模拟技术以进行联合建模，综合了实时测风塔测风数据和尾流计算，该功率预测方法、装置、计算机可读存储介质以及系统提升了功率预测的准确性。The present invention provides a power prediction method, device, computer-readable storage medium and system for an offshore wind farm. The numerical weather forecast and the first flow field simulation technology are used for joint modeling, and the real-time wind measurement tower wind measurement data is synthesized. and wake calculation, the power prediction method, apparatus, computer readable storage medium and system improve the accuracy of power prediction.
进一步地，本发明提供的一种海上风电场的功率预测方法、装置、计算机可读存储介质以及系统还通过对数值天气预报数据进行修正，从而进一步提升功率预测的准确性。Further, the power prediction method, device, computer-readable storage medium and system of an offshore wind farm provided by the present invention further improve the accuracy of power prediction by correcting numerical weather forecast data.
附图说明Description of drawings
图1是本发明一实施例提供的一种海上风电场的功率预测方法的流程示意图；1 is a schematic flowchart of a power prediction method for an offshore wind farm provided by an embodiment of the present invention;
图2是本发明一实施例提供的一种海上风电场的功率预测装置的结构示意图。FIG. 2 is a schematic structural diagram of a power prediction device for an offshore wind farm according to an embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
具体实施例一Specific embodiment one
本发明实施例首先描述了一种海上风电场的功率预测方法。图1是本发明一实施例提供的一种海上风电场的功率预测方法的流程示意图。The embodiments of the present invention first describe a power prediction method for an offshore wind farm. FIG. 1 is a schematic flowchart of a power prediction method for an offshore wind farm according to an embodiment of the present invention.
如图1所示，所述功率预测方法包括：As shown in Figure 1, the power prediction method includes:
S1:获取海上风电场的环境数据组、发电量数据、实时测风塔数据以及数值天气预报数据，并根据所述环境数据组、预设的致动线风电机组模型以及预设的大涡湍流模型，建立并计算所述海上风电场的第一流场模型结果。S1: Acquire the environmental data set, power generation data, real-time wind tower data and numerical weather forecast data of the offshore wind farm, and according to the environmental data set, the preset actuation line wind turbine model and the preset large eddy turbulence model, establish and calculate the first flow field model results of the offshore wind farm.
由于每个风电场的地理位置、地形和气象条件都不相同，每个风电场都具有相对独特性。精确的风电场功率预报既依赖于准确的中尺度数据，也依赖于对风电场地形、粗糙度、尾流等特性的准确模拟。要实现这种准确模拟，基于概率统计模型、人工神经网络(ANN)以及数值天气预报(NWP)的结果，将风速等气象参数传递至测风塔点位，再基于CFD((Computational Fluid Dynamics，计算机流体动力学)模型和尾流模型将这些结果从空间上迁移到各个风机点位。Each wind farm is relatively unique due to its unique geographic location, topography, and meteorological conditions. Accurate wind farm power forecasting relies not only on accurate mesoscale data, but also on accurate simulation of wind farm terrain, roughness, wake and other characteristics. To achieve this accurate simulation, based on the results of probabilistic and statistical models, artificial neural networks (ANN) and numerical weather forecasting (NWP), meteorological parameters such as wind speed are transferred to the wind measuring tower points, and then based on CFD (Computational Fluid Dynamics, Computational fluid dynamics) models and wake models transfer these results spatially to individual turbine locations.
由于地形和粗糙度信息对结果的准确性有相当的影响，因此高质量的地形高程和粗糙度数据对于进行准确的推算风机点位的风速是非常必要的。在本发明实施例中，首先生成针对特定地形、粗糙度和气象条件下的风电场CFD模型从而使得模型最大化地贴近客观情况，提升预测准确率。CFD模型为采用计算流体力学仿真方法，利用优化后的致动线风电机组模型替代传统的实际叶片几何复杂模型，并结合大涡湍流模型，从而带来如下技术优势：(1)采用的双方程湍流模型精度更高，对复杂地形的模拟更精确；(2)提供孤立求解器、耦合求解器和并行求解器，在功能上更完备，求解速度和收敛性有保证；(3)能够在考虑IEC(国际电工委员会)风电机组规范的前提下自动获得最优的风电机组布置。Since the terrain and roughness information have considerable influence on the accuracy of the results, high-quality terrain elevation and roughness data are very necessary to accurately estimate the wind speed at the fan point. In the embodiment of the present invention, a CFD model of the wind farm under specific terrain, roughness and meteorological conditions is first generated, so that the model is as close to the objective situation as possible, and the prediction accuracy is improved. The CFD model adopts the computational fluid dynamics simulation method, uses the optimized actuating line wind turbine model to replace the traditional geometric complex model of the actual blade, and combines the large eddy turbulence model, which brings the following technical advantages: (1) The two equations used The turbulence model has higher accuracy and more accurate simulation of complex terrain; (2) provides isolated solvers, coupled solvers and parallel solvers, which are more complete in function, and the solution speed and convergence are guaranteed; (3) can be considered The optimal wind turbine layout is automatically obtained under the premise of IEC (International Electrotechnical Commission) wind turbine specifications.
风电场CFD模型主要用于模拟CFD流场，计算域的出口边界设定为压力边界条件，下表面的边界条件设定为无滑移壁面边界条件，将给定高度上的风向和风速作为入口边界条件，设定为风速轮廓线。通过求解雷诺平均的Navier-Stokes方程来确定风电场模型，其湍流闭合方案采用双方程的湍流模型，对于相对复杂的海上风电场的流畅计算而言，根据工程实际以及不同湍流模型的适用性，选用RNGk-e湍流模型进行流场模拟，由于流动方程是非线性的，通过迭代计算进行模型求解，由假设的初始条件开始，逐步迭代导最终收敛结果。The wind farm CFD model is mainly used to simulate the CFD flow field. The outlet boundary of the computational domain is set as the pressure boundary condition, the boundary condition on the lower surface is set as the no-slip wall boundary condition, and the wind direction and wind speed at a given height are used as the inlet. Boundary condition, set to wind speed profile. The wind farm model is determined by solving the Reynolds-averaged Navier-Stokes equation. The turbulent closure scheme adopts a two-equation turbulence model. For the smooth calculation of relatively complex offshore wind farms, according to engineering practice and the applicability of different turbulence models, The RNGk-e turbulence model is used to simulate the flow field. Since the flow equation is nonlinear, the model is solved by iterative calculation. Starting from the assumed initial conditions, iteratively leads to the final convergence result.
在本步骤中，建立进行地形模型和风场CFD计算具体包括：获取风机信息和风电场的布局信息后，时间序列文件从参考测风点位传递至各个风机点位，从中尺度数据开始，将中尺度数据通过统计模型推算至风电场内的实时测风塔点位，然后再由实时测风塔点位推算至各风机点位；其中风机信息和风电场的布局信息包括：风机的坐标、轮毂高度、以及该场区的一个实时测风塔的点位。In this step, establishing a terrain model and wind farm CFD calculation specifically includes: after obtaining wind turbine information and wind farm layout information, the time series file is transferred from the reference wind measurement point to each wind turbine point, starting from the mesoscale data, and transferring the middle The scale data is calculated to the real-time wind tower points in the wind farm through statistical models, and then from the real-time wind tower points to the wind turbine points; the wind turbine information and the layout information of the wind farm include: the coordinates of the wind turbine, the hub height, and the location of a real-time wind tower in the field.
如果海上风电场周围有屏障，可考虑加入屏障模型。风场模块中通常采用16个扇区，对于具体风电场也可采用36个扇区以提高准确性，确保在各个方向上都能获得准确的模拟结果。If there is a barrier around the offshore wind farm, consider adding a barrier model. 16 sectors are usually used in the wind farm module, and 36 sectors can also be used for specific wind farms to improve accuracy and ensure accurate simulation results in all directions.
在一个实施例中，根据所述环境数据组、预设的致动线风电机组模型以及预设的大涡湍流模型，建立并计算所述海上风电场的第一流场模型结果，具体包括：根据预设的致动线风电机组模型以及预设的大涡湍流模型，建立所述海上风电场的第一流场模型；根据所述环境数据组以及所述第一流场模型，计算所述海上风电场的第一流场模型结果。In one embodiment, establishing and calculating a first flow field model result of the offshore wind farm according to the environmental data set, the preset actuating line wind turbine model and the preset large eddy turbulence model, specifically including: According to the preset actuation line wind turbine model and the preset large eddy turbulence model, a first flow field model of the offshore wind farm is established; according to the environmental data set and the first flow field model, the First flow field model results for an offshore wind farm.
S2:根据预设的尾流条件组、所述实时测风塔数据以及所述第一流场模型结果，推算实时风资源模拟观测数据组以及风机点位风速组，并根据所述数值天气预报数据，拟合所述环境数据组以及所述发电量数据之间的第一统计关系。S2: According to the preset wake condition group, the real-time wind measurement tower data and the first flow field model result, calculate the real-time wind resource simulation observation data group and the fan point wind speed group, and according to the numerical weather forecast data, and fitting a first statistical relationship between the environmental data set and the power generation data.
统计方法避免了中间复杂的物理建模步骤，它将高精度数值模拟技术NWP预报等信息作为输入数据，通过大量分析寻找气象信息与风电场的发电量之间的潜在统计关系。因此采用统计方法NWP和CFD方法联合进行建模。并且，针对以上的海上风电多爬坡事件、NWP预测不准的问题考虑对NWP数据进行修正之后进行建模。The statistical method avoids the complex physical modeling steps in the middle. It uses information such as high-precision numerical simulation technology NWP forecast as input data, and searches for the potential statistical relationship between meteorological information and wind farm power generation through extensive analysis. Therefore, the statistical method NWP and the CFD method are combined for modeling. In addition, in view of the above problems of multiple climbing events of offshore wind power and inaccurate NWP prediction, the NWP data should be revised and then modeled.
在一个实施例中，根据预设的尾流条件组、所述实时测风塔数据以及所述第一流场模型结果，推算实时风资源模拟观测数据组以及风机点位风速组，具体包括：根据所述第一流场模型结果以及所述实时测风塔数据，推算海上风电场中风机的各个点位的各个高度的风资源模拟观测数据组；根据所述风资源模拟观测数据组以及预设的尾流条件组，计算输出在所述尾流条件组下的风机的各个点位的计算风速，并将所述风速存储入风机点位风速组。In one embodiment, according to the preset wake condition group, the real-time wind measurement tower data and the first flow field model result, the real-time wind resource simulation observation data group and the fan point wind speed group are calculated, specifically including: According to the results of the first flow field model and the real-time wind measurement tower data, calculate the wind resource simulation observation data set at each height of each point of the wind turbine in the offshore wind farm; according to the wind resource simulation observation data set and forecast Set the wake condition group, calculate and output the calculated wind speed of each point of the fan under the wake condition group, and store the wind speed into the fan point wind speed group.
在一个实施例中，根据所述风资源模拟观测数据组以及预设的尾流条件组，计算输出在所述尾流条件组下的风机的各个点位的计算风速，具体包括：根据预设的Larsen尾流计算模型以及风速数据组，计算风机的各个点位的计算风速；所述风资源模拟观测数据组包括风速数据组。In one embodiment, according to the wind resource simulation observation data group and a preset wake condition group, calculating and outputting the calculated wind speed of each point of the wind turbine under the wake condition group, specifically includes: according to a preset The Larsen wake calculation model and the wind speed data set are used to calculate the calculated wind speed of each point of the fan; the wind resource simulation observation data set includes the wind speed data set.
S3:根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，对所述海上风电场进行功率预测以获取预测结果。S3: Perform power prediction on the offshore wind farm to obtain a prediction result according to the real-time wind resource simulation observation data group, the wind speed group at the point of the wind turbine, and the first statistical relationship.
考虑到风电场区的风场分布是天气系统的运动在局地地形驱动下的结果，天气系统的运动可采用中尺度数值模式进行预报，由于局地地形是稳定不动的，可以采用CFD技术事先模拟出各种天气背景条件下的风场分布，建立微尺度风场基础数据库，然后直接根据中尺度数值天气预报，通过选择对应的风电场的风场分布，给出每台风电机组的风电功率预测。Considering that the distribution of the wind field in the wind farm area is the result of the movement of the weather system driven by the local terrain, the movement of the weather system can be predicted using a mesoscale numerical model. Since the local terrain is stable, CFD technology can be used. The wind field distribution under various weather background conditions is simulated in advance, the basic database of micro-scale wind field is established, and then directly according to the mesoscale numerical weather forecast, the wind field distribution of the corresponding wind farm is selected to give the wind power of each wind turbine. Power forecast.
与此同时，由于每种预测模型和方法利用的数据都不完全相同,组合预测方法能够最大限度地利用各种单一预测方法的有用信息，在一定程度上增加系统的预测准确性。具体地，将不同的预测模型和方法结合，从而根据多种预测方法所提供的信息，以适当的加权平均方式得出多元化的模型结构，形成组合预测模型能够进一步提升预测准确率。该方法一般采用绝对误差作为准则来计算组合预测方法的权系数。At the same time, since the data used by each forecasting model and method are not exactly the same, the combined forecasting method can maximize the useful information of various single forecasting methods and increase the forecasting accuracy of the system to a certain extent. Specifically, combining different forecasting models and methods to obtain a diversified model structure in an appropriate weighted average manner based on the information provided by multiple forecasting methods, and forming a combined forecasting model can further improve the forecasting accuracy. This method generally uses the absolute error as a criterion to calculate the weight coefficient of the combined prediction method.
在一个实施例中，根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，对所述海上风电场进行功率预测以获取预测结果，具体包括：根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，基于CFD动力降尺度，对所述海上风电场进行风电功率短期预测以获取短期预测结果；根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，根据多种预测方法，对所述海上风电场进行风电功率集合预测以获取集合预测结果；将所述短期预测结果以及所述集合预测结果作为预测结果进行输出。In one embodiment, performing power prediction on the offshore wind farm to obtain a prediction result according to the real-time wind resource simulation observation data group, the wind speed group at the point of the wind turbine, and the first statistical relationship, specifically including: according to The real-time wind resource simulation observation data set, the wind speed set at the wind turbine location, and the first statistical relationship, based on CFD dynamic downscaling, perform short-term wind power prediction on the offshore wind farm to obtain a short-term prediction result; According to the real-time wind resource simulation observation data set, the wind speed set of the wind turbine points and the first statistical relationship, according to a variety of prediction methods, the wind power set prediction is performed on the offshore wind farm to obtain the set prediction result; The short-term prediction results and the aggregated prediction results are output as prediction results.
在一个实施例中，根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，基于CFD动力降尺度，对所述海上风电场进行风电功率短期预测以获取短期预测结果，具体包括：根据所述实时风资源模拟观测数据组以及所述风机点位风速组，通过预设的中尺度风速预报订正方法，对所述数值天气预报数据进行订正；根据订正后的数值天气预报数据，在预设的微尺度风场基础数据库中选出对应的风电场微尺度风场分布，分析计算各个风机的自由来流风速；根据所述第一统计关系、预设的风电机组模拟发电功率订正方法以及所述自由来流风速，计算得到各个风机的模拟发电功率；根据预设的风电场运行维护计划，去除不运行的机组和机组不运行的时段，得到风电场全场发电功率预测。In one embodiment, according to the real-time wind resource simulation observation data set, the wind speed set at the point of the wind turbine, and the first statistical relationship, based on CFD dynamic downscaling, short-term wind power prediction is performed on the offshore wind farm to obtain Obtaining a short-term forecast result specifically includes: correcting the numerical weather forecast data by using a preset mesoscale wind speed forecast correction method according to the real-time wind resource simulation observation data group and the wind speed group at the fan point; After obtaining the numerical weather forecast data, select the corresponding micro-scale wind field distribution of the wind farm in the preset micro-scale wind farm basic database, and analyze and calculate the free flow wind speed of each fan; according to the first statistical relationship, the preset According to the method for correcting the simulated generating power of wind turbines and the free flow wind speed, the simulated generating power of each wind turbine can be obtained by calculating; according to the preset operation and maintenance plan of the wind farm, the non-operating units and the time periods when the units are not running are removed to obtain the wind farm. Full-scale power generation forecast.
在一个实施例中，所述多种预测方法包括：基于禁忌算法的神经网络预测方法、功率时间序列预测方法、粒子群算法以及卡尔曼滤波修正的功率预测方法。In one embodiment, the multiple prediction methods include: a neural network prediction method based on tabu algorithm, a power time series prediction method, a particle swarm algorithm, and a power prediction method modified by Kalman filter.
其中，基于禁忌算法的神经网络预测方法：Among them, the neural network prediction method based on the tabu algorithm:
禁忌搜索(Tabu Search或Taboo Search，简称TS)是对局部邻域搜索的一种扩展，是一种全局逐步寻优算法，是对人类智力过程的模拟。TS算法通过引入一个灵活的存储结构和相应的禁忌准则来避免迂回搜索，并通过藐视准则来赦免一些被禁忌的优良状态，进而保证多样化的有效探索以最终实现全局优化。基本思想是：假设给出一个解邻域，首先在邻域中找出一个初始局部解x作为当前解，并令当前解为最优解，然后以这个当前解x作为起点，在解邻域中搜索最优解x'，为了避免x'与x相同造成迂回搜索，设置一个记忆近期操作的禁忌表，如果当前的搜索操作是记录在禁忌表中的操作，那么这一搜索操作就被禁止，否则以x'代替x作为当前解。Tabu Search (Tabu Search or Taboo Search, TS for short) is an extension of local neighborhood search, a global step-by-step optimization algorithm, and a simulation of the human intelligence process. The TS algorithm avoids roundabout searches by introducing a flexible storage structure and corresponding taboo criteria, and forgives some taboo good states by defying the criteria, thereby ensuring diversified and effective exploration to finally achieve global optimization. The basic idea is: Assuming that a solution neighborhood is given, first find an initial local solution x in the neighborhood as the current solution, and make the current solution the optimal solution, and then use the current solution x as the starting point, in the solution neighborhood In order to avoid the roundabout search caused by the same x' and x, set up a taboo table to memorize recent operations. If the current search operation is an operation recorded in the taboo table, then this search operation is prohibited. , otherwise replace x with x' as the current solution.
BP神经网络是一种按照误差反向传播算法训练的多层前馈网络，由两个过程组成：信息的正向传播过程以及误差的反向传播过程。信息的正向传播过程，即输入信息从输入层输入网络，并传递到中间层，中间层对信息进行变换处理，然后将处理的结果传给输出层，在输出层得到网络的实际输出信息。误差的反向传播过程，即按误差梯度下降的方式通过输出层，修正各层的权值，按照中间层、输入层的顺序逐一反传。两个过程周而复始的进行，直到网络输出与期望输出之间的误差减少到可以接受的程度。在误差反向传播过程中用到的主要是“负梯度下降”理论。BP neural network is a multi-layer feedforward network trained according to the error back propagation algorithm, which consists of two processes: the forward propagation process of information and the back propagation process of error. The forward propagation process of information, that is, the input information is input into the network from the input layer and transmitted to the intermediate layer, the intermediate layer transforms the information, and then transmits the processing result to the output layer, and the actual output information of the network is obtained at the output layer. The back-propagation process of the error, that is, through the output layer in the way of error gradient descent, correcting the weights of each layer, and back-propagating one by one according to the order of the intermediate layer and the input layer. The two processes are repeated until the error between the network output and the expected output is reduced to an acceptable level. The "negative gradient descent" theory is mainly used in the process of error back propagation.
神经网络有很强的非线性拟合能力，可映射任意复杂的非线性关系，而且学习规则简单，便于计算机实现，具有很强的鲁棒性、记忆能力、非线性映射能力以及强大的自学习能力，但是在学习过程中容易陷入“过学习”。基于禁忌搜索算法的人工神经网络应用的基本原理是用具有记忆功能的禁忌搜索算法对神经网络的权值进行优化学习，利用禁忌搜索算法的全局搜索能力来获取全局最优解，从而避免训练陷入局部极小。The neural network has strong nonlinear fitting ability, can map any complex nonlinear relationship, and the learning rules are simple, easy to implement by computer, with strong robustness, memory ability, nonlinear mapping ability and powerful self-learning Ability, but it is easy to fall into "over-learning" in the learning process. The basic principle of the artificial neural network application based on the tabu search algorithm is to use the tabu search algorithm with memory function to optimize the learning of the weights of the neural network, and use the global search ability of the tabu search algorithm to obtain the global optimal solution, so as to avoid training trapped in Locally minimal.
为进一步描述，将举例以对禁忌搜索算法优化神经网络的过程进行描述：For further description, an example will be used to describe the process of the tabu search algorithm to optimize the neural network:
假设某一BP神经网路的误差函数是f＝f(Wh,Wo,θh,θo)，其中，Wh、Wo、θh、θo分别表示输入层和隐含层的连接权重、隐含层和输出层的连接权重、隐含层的阈值和输出层的阈值。对网络的优化就是求minf(Wh,Wo,θh,θo)的过程。为叙述方便，此处用Δ来表示(Wh,Wo,θh,θo)。Suppose the error function of a BP neural network is f=f(Wh , Wo , θh , θo ), where Wh , Wo , θh , θo represent the input layer and the hidden layer, respectively. Connection weights, connection weights for hidden and output layers, thresholds for hidden layers, and thresholds for output layers. The optimization of the network is the process of finding minf(Wh , Wo , θh , θo ). For the convenience of description, Δ is used here to represent (Wh , Wo , θh , θo ).
首先，初始化Δ(具体方法是为Δ的每个分量赋一个较小的随机数，记为Δinitial)；接着，以Δbest表示到目前为止搜索导的最优解，Δnow表示当前解，令Δbest＝Δinitial，Δnow＝Δinitial，并将Δnow存储在禁忌表里；然后，产生Δinitial的一个邻域解Δnew，计算f(Δnew)和f(Δbest)；若f(Δnew)连续多次(事先设定好的次数)没有改变，则终止算法并输出结果，否则继续下面的步骤；若f(Δnew)<f(Δbest)，则Δbest＝Δnew，Δnow＝Δnew，Δnew进入禁忌表，表内存储的点顺序后移，即禁忌表被更新；若f(Δnew)≥f(Δbest)，判断Δnew是否被禁忌，如果Δnew在禁忌表中某存储点的给定邻域里，表示Δnew被禁忌，则重新产生一个邻域解向量Δnew；如果Δnew没有被禁忌，则Δnow＝Δnew，同时更新禁忌表；最后，产生Δnow的一个邻域解Δnew，转到第四步；当训练结束时，得到训练好的权重向量Wh、Wo以及阈值向量θh、θo。First, initialize Δ (the specific method is to assign a small random number to each component of Δ, denoted as Δinitial ); then, Δbest represents the optimal solution of the search derivative so far, Δnow represents the current solution, Let Δbest = Δinitial , Δnow = Δinitial , and store Δnow in the taboo table; then, generate a neighborhood solution Δnew of Δinitial , and calculate f(Δnew ) and f(Δbest ); if If f(Δnew ) does not change for many consecutive times (pre-set times), then terminate the algorithm and output the result, otherwise continue the following steps; if f(Δnew )<f(Δbest ), then Δbest =Δnew , Δnow = Δnew , Δnew enters the taboo table, the points stored in the table are moved backward, that is, the taboo table is updated; if f(Δnew )≥f(Δbest ), judge whether Δnew is taboo, if Δnew is in the given neighborhood of a certain storage point in the taboo table, indicating that Δnew is forbidden, then a neighborhood solution vector Δnew is regenerated; if Δnew is not forbidden, then Δnow = Δnew , and the taboo is updated at the same time Finally, generate a neighborhood solutionΔnew ofΔnow , and go to the fourth step; when the training ends, get trained weight vectors Wh , Wo and threshold vectors θh , θo .
时间序列分析是处理动态数据的一种行之有效的参数化时域分析方法。差分自回归滑动平均模型(ARIMA--Autoregressive Integrated Moving Average)通过分析时间序列资料来进行预测和控制。本发明采用功率时间序列预测技术，根据功率的历史记录资料，建立一个时间序列数学模型，用这个数学模型一方面来描述功率这个随机变量变化过程的统计规律性，另一方面在该数学模型的基础上再确立功率预测的数学表达式，对未来的功率进行预测。Time series analysis is an effective parametric time domain analysis method for processing dynamic data. The differential autoregressive moving average model (ARIMA--Autoregressive Integrated Moving Average) performs prediction and control by analyzing time series data. The invention adopts the power time series prediction technology, establishes a time series mathematical model according to the historical record data of power, and uses this mathematical model to describe the statistical regularity of the change process of the random variable of power on the one hand, and on the other hand, in the mathematical model. On the basis, the mathematical expression of power prediction is established, and the future power is predicted.
ARIMA由三部分组成：自回归项autoregressive(AR)、差分项integrated(I)和滑动平均项moving average(MA)。ARIMA是在ARMA的基础上提出来的，ARMA的数学表达式是：ARIMA consists of three parts: the autoregressive term autoregressive (AR), the difference term integrated (I) and the moving average term moving average (MA). ARIMA is proposed on the basis of ARMA. The mathematical expression of ARMA is:
其中，Xt表示时间序列，即功率序列；
在功率预测中，假设功率X(t)的样本函数为x(t)，采样值(功率的历史资料)为x1,x2,…,xt,…，在具体实施时，假设有限个采样值构成的有限序列为x1,x2,…,xN。时间序列预测就是根据这个有限序列去推断产生这个有限序列的原来序列的性质。要确切地找到原来的序列是很困难的，不过可以找到一个基本上与有限序列相符合的预测模型来替代原始序列。这个过程需要进行模式识别和参数估计。模型识别就是判定要求的预测模型是AR、MA、ARMA中的哪一类或不属于这三类。参数估计，就是在模型识别之后，根据适当的方法计算模型中的未知参数。模型识别和参数估计都是根据有限序列式去推断原始序列式的性质来完成的，这种推断不可能完全准确，因而对所确定的随机模型是否合适还需要进行检验。In power prediction, it is assumed that the sample function of power X(t) is x(t), and the sampled values (historical data of power) are x1 , x2 ,…,xt ,…. In the specific implementation, it is assumed that a finite number of The finite sequence of sampled values is x1 ,x2 ,…,xN . Time series prediction is to infer the properties of the original sequence that generated the finite sequence based on this finite sequence. It is difficult to find the original sequence exactly, but a predictive model that basically fits the finite sequence can be found to replace the original sequence. This process requires pattern recognition and parameter estimation. Model identification is to determine whether the required prediction model is AR, MA, ARMA or does not belong to these three categories. Parameter estimation is to calculate the unknown parameters in the model according to the appropriate method after the model is identified. Model identification and parameter estimation are done based on the finite sequence formula to infer the properties of the original sequence formula. This inference cannot be completely accurate. Therefore, it is necessary to test whether the determined stochastic model is suitable.
模型识别的判别依据是序列的自相关函数和偏相关函数。经过模型识别后，模型的类别、结构和阶次都己初步确定，然后需要估计模型中的未知参数。参数估计的方法很多，应用最多的方法主要是矩估计和最小二乘估计。经过模型识别和参数估计后，接下来的问题就是要确定模型是否恰当，如果经过检验，模型符合要求，就可以进行预测工作了；反之，如果存在拟合严重不当的证据，那么就需要修改模型或重新识别模型，直至满足要求。如果经过模型识别和参数估计初步确定的随机模型没通过检验，即经检验确定该模型不合适时，可作如下处理：利用参数估计的标准差评价参数估计值的统计显著性。一般规定，参数点估计的绝对值小于二倍标准差者为统计显著性差或称其不显著，对于这样的参数做系数的项，应当从模型中删除掉。与此同时，若需要在模型中增加一个自回归项或滑动平均项，就加进这个初步确定的模型中去，但需要检验增加后的新模型应比原初步确定的模型更有适用性，否则增加参数项无意义；根据在残差分析残差所提供的信息，对模型进行适当的修改；如果不对初步确定的模型进行修改，也可以重新识别模型；经模型检验认为初步确定的模型不合适时，经过模型修改或重新识别得到新的模型后，还必须进行模型检验，以确定其合理性。The discriminant basis of model identification is the autocorrelation function and partial correlation function of the sequence. After model identification, the type, structure and order of the model have been preliminarily determined, and then the unknown parameters in the model need to be estimated. There are many methods for parameter estimation, and the most widely used methods are mainly moment estimation and least squares estimation. After model identification and parameter estimation, the next problem is to determine whether the model is appropriate. If the model meets the requirements after inspection, the prediction work can be carried out; on the contrary, if there is evidence of serious improper fitting, then the model needs to be modified. Or re-identify the model until the requirements are met. If the stochastic model preliminarily determined by model identification and parameter estimation fails the test, that is, when it is determined that the model is not suitable after the test, the following processing can be done: use the standard deviation of parameter estimates to evaluate the statistical significance of parameter estimates. It is generally stipulated that if the absolute value of the parameter point estimate is less than two times the standard deviation, the statistical significance is poor or it is said to be insignificant, and the coefficient term for such a parameter should be deleted from the model. At the same time, if an autoregressive term or a moving average term needs to be added to the model, it should be added to the preliminarily determined model, but it is necessary to check that the added new model should be more applicable than the original preliminarily determined model. Otherwise, it is meaningless to add parameter items; according to the information provided by the residual analysis residuals, make appropriate modifications to the model; if you do not modify the initially determined model, you can also re-identify the model; after model inspection, it is considered that the initially determined model is not When appropriate, after a new model is obtained through model modification or re-identification, model checking must be performed to determine its rationality.
为进一步描述，将举例以对粒子群算法进行描述。For further description, an example will be used to describe the particle swarm algorithm.
粒子群算法是一种有效的全局寻优算法，开始模拟的是鸟群觅食的过程，是一种利用群体智能来优化的算法，在群体中，通过个体间的合作与竞争实现群体智能优化搜索。理论中把每一个优化问题都看作成在空中觅食的鸟群，所谓的“粒子”就是在空中飞行的每一只觅食“鸟”，也即优化问题的一个解，而最优解就是鸟群最终寻找到的“食物”。Particle swarm optimization is an effective global optimization algorithm. It starts to simulate the process of bird flock foraging. It is an algorithm that uses swarm intelligence to optimize. In the group, swarm intelligence optimization is realized through cooperation and competition among individuals. search. In the theory, each optimization problem is regarded as a flock of birds foraging in the air. The so-called "particles" are every foraging "bird" flying in the air, that is, a solution to the optimization problem, and the optimal solution is The "food" that the flock eventually finds.
在粒子群算法中，算法首先生成初始解，即在D维可行解空间中随机初始化N个粒子组成种群z＝{z1,z2,…,zN}，每个粒子所对应两个向量，位置和速度，即zi＝{zi1,zi2,…,ziD}和vi＝{vi1,vi2,…,viD}，然后根据目标函数计算粒子的适应度值，在S维解空间中进行迭代搜索。在这个过程中，粒子通过个体本身搜索到的最优解，即“个体极值”Pid和群体索到的最优解，即“极值”Pgd，来更新自己的位置和速度。每个粒子根据下式来更新自己的位置和速度：In particle swarm optimization, the algorithm first generates an initial solution, that is, randomly initializes N particles in the D-dimensional feasible solution space to form a population z={z1 , z2 ,...,zN }, and each particle corresponds to two vectors , position and velocity, namely zi ={zi1 ,zi2 ,...,ziD } and vi ={vi1 ,vi2 ,...,viD }, and then calculate the fitness value of the particle according to the objective function, in Iterative search is performed in the S-dimensional solution space. In this process, the particle updates its position and velocity through the optimal solution searched by the individual itself, namely the "individual extreme value" Pid and the optimal solution obtained by the group, namely the "extreme value" Pgd . Each particle updates its position and velocity according to:
vid(t+1)＝ωvid(t)+c1r1[pid-zid(t)]+c2r2[pgd-zid(t)]；vid (t+1)=ωvid (t)+c1 r1 [pid -zid (t)]+c2 r2 [pgd -zid (t)];
zid(t+1)＝zid(t)+vid(t+1)；zid (t+1)=zid (t)+vid (t+1);
式中，vid(t+1)表示第i个粒子在t+1次迭代中第d维上的速度，ω为惯性权重，c1,c2为加速常数，r1，r2为0～1之间的随机数。In the formula, vid (t+1) represents the velocity of the i-th particle on the d-th dimension in the t+1 iteration, ω is the inertia weight, c1 , c2 are acceleration constants, r1 , r2 are 0 A random number between ~1.
惯性权重ω表示对当前速率的保持，ω调节的是粒子的全局和局部搜索能力。较大的惯性权重使粒子具有较强的全局能力搜索能力，较小的惯性权重使粒子具有较强的局部搜索能力。学习因子c1，c2分别代表粒子自我学习能力和社会学习的能力，即调节粒子“飞向”个体最优与群体最优的最大步长，决定粒子的种群经验和个体经验对粒子自身的影响，反映了粒子个体与个体，个体与种群之间信息的交流。The inertia weight ω represents the maintenance of the current rate, and ω adjusts the global and local search ability of the particle. The larger inertia weight makes the particle have stronger global search ability, and the smaller inertia weight makes the particle have stronger local search ability. The learning factors c1 and c2 represent the particle's self-learning ability and social learning ability respectively, that is, it adjusts the maximum step size of the particle "flying" to the individual optimum and the group optimum, and determines the effect of the particle's population experience and individual experience on the particle itself. Influence reflects the exchange of information between particle individuals and individuals, and between individuals and populations.
为进一步描述，将举例以对卡尔曼滤波修正的功率预测方法进行描述。For further description, the power prediction method modified by Kalman filter will be described as an example.
在应用统计方法进行风电场功率预测时，NWP是进行预报的数据基础，因此NWP提供的数据精度将直接决定着最终功率预测结果的可靠性。一般情况下，NWP处理次网格现象的能力不足以弥补物理参数初始化存在缺陷，从而导致了气象模式输出存在着一定程度的系统误差，此误差引入风电功率的计模型之中必然影响最终的预测精度。When applying statistical methods for wind farm power forecasting, NWP is the data basis for forecasting, so the data accuracy provided by NWP will directly determine the reliability of the final power forecasting result. In general, the ability of NWP to deal with sub-grid phenomena is not enough to compensate for the defects in the initialization of physical parameters, which leads to a certain degree of systematic error in the output of the meteorological model. The introduction of this error into the wind power meter model will inevitably affect the final prediction. precision.
从MOS中演化而来的多种统计方法可以用来消除此类系统误差。然而，MOS方程的建立需要大量的历史资料，而积累历史资料是很困难的，动态更新方程的参数也是MOS所不能及的。卡尔曼滤波算法则不同，它不需要大量的历史资料，仅通过误差反馈便可动态修正预测方程系数，这对于提高气象模式输出的精度具有重要的现实意义。Various statistical methods evolved from MOS can be used to remove such systematic errors. However, the establishment of the MOS equation requires a large amount of historical data, and it is very difficult to accumulate historical data, and the parameters of the dynamic update equation are beyond the reach of MOS. The Kalman filter algorithm is different. It does not require a large amount of historical data, and can dynamically correct the coefficients of the prediction equation only through error feedback, which is of great practical significance for improving the accuracy of the meteorological model output.
本发明实施例还采用了卡尔曼滤波算法修正后的风电场功率预测模型。NWP提供的历史气象数据经卡尔曼滤波算法修正后形成BP神经网络的训练集合，从风力机组和监视控制系统(SCADA)采集得到的各个风机的功率序列将作为BP神经网络的目标集合，经充分训练之后得到气象数据与功率输出之间的非线性映射关系即为BP网络预测模型。同理，经卡尔曼滤波算法修正后的未来气象数据经过神经网络训练后得到最终的预测功率输出。The embodiment of the present invention also adopts the wind farm power prediction model modified by the Kalman filter algorithm. The historical meteorological data provided by NWP is corrected by the Kalman filter algorithm to form the training set of BP neural network. The power sequence of each wind turbine collected from the wind turbine and the supervisory control system (SCADA) will be used as the target set of the BP neural network. After training, the nonlinear mapping relationship between meteorological data and power output is obtained as the BP network prediction model. Similarly, the future meteorological data corrected by the Kalman filter algorithm is trained by the neural network to obtain the final predicted power output.
用卡尔曼滤波器对NWP输出的风速时间序列进行修正时，将风速的预测误差作为NWP风速输出数据的函数，并对此误差进行估计。假定vt是NWP模型在t时刻的输出风速，yt是t时刻的预测误差，yt可用一个关于vt的三阶多项式来描述，即：When the Kalman filter is used to correct the wind speed time series output by NWP, the prediction error of wind speed is regarded as a function of NWP wind speed output data, and the error is estimated. Suppose vt is the output wind speed of the NWP model at time t, yt is the prediction error at time t, and yt can be described by a third-order polynomial about vt , namely:
yt＝x0,t+x1,tvt+x2,tvt2+x3,tvt3+qt；yt =x0,t +x1,t vt +x2,t vt2 +x3,t vt3 +qt ;
式中，xi,t(i＝0,1,2,3)是采用卡尔曼滤波器进行估计的系数；qt为上一步中生成的高斯非线性系统误差，采用待估计的系数矩阵作为状态向量，即xt＝[x0,t x1,t x2,t x3,t]T，用yt作为观测向量，观测矩阵Ht＝[1 vt vt2 vt3]，系统矩阵取单位矩阵，基于此，系统方程及量测方程为：In the formula, xi,t (i=0,1,2,3) is the coefficient estimated by Kalman filter; qt is the Gaussian nonlinear system error generated in the previous step, and the coefficient matrix to be estimated is used as State vector, i.e. xt =[x0,t x1,t x2,t x3,t ]T , using yt as observation vector, observation matrix Ht =[1 vt vt2 vt3 ] , the system matrix takes the unit matrix, based on this, the system equation and the measurement equation are:
xt＝xt-1+wt；xt =xt-1 +wt ;
yt＝Htxt+qt。yt =Ht xt +qt .
在一个实施例中，所述功率预测方法还包括：以预设的时间周期对风电场CFD模型优化。具体地，在区域内的基准风电场的风电场CFD模型建立完成之后，需要运行一段时间，再根据一定时间段内的风电场的全面观测资料，进行风电场CFD模型的优化。In one embodiment, the power prediction method further includes: optimizing the wind farm CFD model in a preset time period. Specifically, after the wind farm CFD model of the benchmark wind farm in the region is established, it needs to run for a period of time, and then optimize the wind farm CFD model according to the comprehensive observation data of the wind farm within a certain period of time.
本发明实施例描述了一种海上风电场的功率预测方法，通过数值天气预报和第一流场模拟技术以进行联合建模，综合了实时测风塔测风数据和尾流计算，该功率预测方法、装置、计算机可读存储介质以及系统提升了功率预测的准确性；进一步地，本发明实施例描述的一种海上风电场的功率预测方法还通过对数值天气预报数据进行修正，从而进一步提升功率预测的准确性。The embodiment of the present invention describes a power prediction method for an offshore wind farm. The numerical weather forecast and the first flow field simulation technology are used for joint modeling, and the wind measurement data of the real-time wind tower and the wake calculation are synthesized. The method, the device, the computer-readable storage medium and the system improve the accuracy of power prediction; further, the power prediction method for an offshore wind farm described in the embodiment of the present invention further improves the numerical weather forecast data by revising the numerical weather forecast data. Accuracy of power predictions.
具体实施例二Specific embodiment two
除上述方法外，本发明实施例还公开了一种海上风电场的功率预测装置。图2是本发明一实施例提供的一种海上风电场的功率预测装置的结构示意图。In addition to the above method, the embodiment of the present invention also discloses a power prediction device for an offshore wind farm. FIG. 2 is a schematic structural diagram of a power prediction device for an offshore wind farm according to an embodiment of the present invention.
如图2所示，所述功率预测装置包括数据获取单元11、计算拟合单元12以及功率预测单元13。As shown in FIG. 2 , the power prediction apparatus includes a
数据获取单元11用于获取海上风电场的环境数据组、发电量数据、实时测风塔数据以及数值天气预报数据，并根据所述环境数据组、预设的致动线风电机组模型以及预设的大涡湍流模型，建立并计算所述海上风电场的第一流场模型结果。The
计算拟合单元12用于根据预设的尾流条件组、所述实时测风塔数据以及所述第一流场模型结果，推算实时风资源模拟观测数据组以及风机点位风速组，并根据所述数值天气预报数据，拟合所述环境数据组以及所述发电量数据之间的第一统计关系。The calculation and
功率预测单元13用于根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，对所述海上风电场进行功率预测以获取预测结果。The
在一个实施例中，数据获取单元11还用于：根据预设的致动线风电机组模型以及预设的大涡湍流模型，建立所述海上风电场的第一流场模型；根据所述环境数据组以及所述第一流场模型，计算所述海上风电场的第一流场模型结果。In one embodiment, the
在一个实施例中，计算拟合单元12还用于：根据所述第一流场模型结果以及所述实时测风塔数据，推算海上风电场中风机的各个点位的各个高度的风资源模拟观测数据组；根据所述风资源模拟观测数据组以及预设的尾流条件组，计算输出在所述尾流条件组下的风机的各个点位的计算风速，并将所述风速存储入风机点位风速组。In one embodiment, the calculation and
在一个实施例中，功率预测单元13还用于：根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，基于CFD动力降尺度，对所述海上风电场进行风电功率短期预测以获取短期预测结果；根据所述实时风资源模拟观测数据组、所述风机点位风速组以及所述第一统计关系，根据多种预测方法，对所述海上风电场进行风电功率集合预测以获取集合预测结果；将所述短期预测结果以及所述集合预测结果作为预测结果进行输出。In one embodiment, the
在一个实施例中，功率预测单元13还用于：根据所述实时风资源模拟观测数据组以及所述风机点位风速组，通过预设的中尺度风速预报订正方法，对所述数值天气预报数据进行订正；根据订正后的数值天气预报数据，在预设的微尺度风场基础数据库中选出对应的风电场微尺度风场分布，分析计算各个风机的自由来流风速；根据所述第一统计关系、预设的风电机组模拟发电功率订正方法以及所述自由来流风速，计算得到各个风机的模拟发电功率；根据预设的风电场运行维护计划，去除不运行的机组和机组不运行的时段，得到风电场全场发电功率预测。In one embodiment, the
在一个实施例中，计算拟合单元12还用于：根据预设的Larsen尾流计算模型以及风速数据组，计算风机的各个点位的计算风速；所述风资源模拟观测数据组包括风速数据组。In one embodiment, the calculation and
其中，所述功率预测装置集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。所述计算机可读存储介质包括存储的计算机程序，其中，在所述计算机程序运行时控制所述计算机可读存储介质所在设备执行如前所述的海上风电场的功率预测方法。Wherein, if the integrated unit of the power prediction device is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device on which the computer-readable storage medium is located is controlled to execute the aforementioned method for predicting power of an offshore wind farm.
基于这样的理解，本发明实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是，所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括电载波信号和电信信号。Based on this understanding, the present invention can implement all or part of the processes in the methods of the above embodiments, and can also be completed by instructing relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When executed by a processor, the steps of each of the above method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, and the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer-readable media Electric carrier signals and telecommunication signals are not included.
需说明的是，以上所描述的装置实施例仅仅是示意性的，其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。另外，本发明提供的装置实施例附图中，单元之间的连接关系表示它们之间具有通信连接，具体可以实现为一条或多条通信总线或信号线。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。It should be noted that the device embodiments described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical unit, that is, it can be located in one place, or it can be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. In addition, in the drawings of the apparatus embodiments provided by the present invention, the connection relationship between the units indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative effort.
本发明实施例描述了一种海上风电场的功率预测装置及计算机可读存储介质，通过数值天气预报和第一流场模拟技术以进行联合建模，综合了实时测风塔测风数据和尾流计算，该功率预测装置及计算机可读存储介质提升了功率预测的准确性；进一步地，本发明实施例描述的一种海上风电场的功率预测装置及计算机可读存储介质还通过对数值天气预报数据进行修正，从而进一步提升功率预测的准确性。The embodiments of the present invention describe a power prediction device and a computer-readable storage medium for an offshore wind farm, which perform joint modeling through numerical weather forecasting and the first flow field simulation technology, synthesizing the real-time wind measurement tower wind measurement data and tail wind data. Flow calculation, the power prediction device and the computer-readable storage medium improve the accuracy of power prediction; further, the power prediction device and the computer-readable storage medium for an offshore wind farm described in the embodiment of the present invention also use logarithmic weather The forecast data is revised to further improve the accuracy of the power forecast.
具体实施例三Specific embodiment three
除上述方法和装置外，本发明实施例还描述了一种海上风电场的功率预测系统。In addition to the above method and device, the embodiments of the present invention also describe a power prediction system for an offshore wind farm.
该功率预测系统包括处理器、存储器以及存储在所述存储器中且被配置为由所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现如前所述的海上风电场的功率预测方法。The power prediction system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor executing the computer program to implement the aforementioned offshore wind farm's Power prediction method.
所称处理器可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，所述处理器是所述装置的控制中心，利用各种接口和线路连接整个装置的各个部分。The processor may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf processors Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The processor is the control center of the device, and uses various interfaces and lines to connect various parts of the entire device.
所述存储器可用于存储所述计算机程序和/或模块，所述处理器通过运行或执行存储在所述存储器内的计算机程序和/或模块，以及调用存储在存储器内的数据，实现所述装置的各种功能。所述存储器可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等；存储数据区可存储根据手机的使用所创建的数据(比如音频数据、电话本等)等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(FlashCard)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。The memory can be used to store the computer program and/or module, and the processor implements the apparatus by running or executing the computer program and/or module stored in the memory and calling the data stored in the memory various functions. The memory may mainly include a stored program area and a stored data area, wherein the stored program area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the storage data area may store Data (such as audio data, phonebook, etc.) created according to the usage of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory such as hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card , a flash memory card (FlashCard), at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
本发明实施例描述了一种海上风电场的功率预测系统，通过数值天气预报和第一流场模拟技术以进行联合建模，综合了实时测风塔测风数据和尾流计算，该功率预测方法、装置、计算机可读存储介质以及系统提升了功率预测的准确性；进一步地，本发明实施例描述的一种海上风电场的功率预测系统还通过对数值天气预报数据进行修正，从而进一步提升功率预测的准确性。The embodiment of the present invention describes a power prediction system for an offshore wind farm, which performs joint modeling through numerical weather forecasting and the first flow field simulation technology, and integrates real-time wind measurement tower wind measurement data and wake calculation. The method, the device, the computer-readable storage medium and the system improve the accuracy of power prediction; further, the power prediction system for an offshore wind farm described in the embodiment of the present invention further improves the numerical weather forecast data by revising the numerical weather forecast data. Accuracy of power predictions.
以上所述是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也视为本发明的保护范围。The above are the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made, and these improvements and modifications may also be regarded as It is the protection scope of the present invention.
Claims (10)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210718925.4ACN115034159A (en) | 2022-06-23 | 2022-06-23 | Power prediction method, device, storage medium and system for offshore wind farm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210718925.4ACN115034159A (en) | 2022-06-23 | 2022-06-23 | Power prediction method, device, storage medium and system for offshore wind farm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115034159Atrue CN115034159A (en) | 2022-09-09 |
Family
ID=83126880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210718925.4APendingCN115034159A (en) | 2022-06-23 | 2022-06-23 | Power prediction method, device, storage medium and system for offshore wind farm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115034159A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115693666A (en)* | 2022-12-30 | 2023-02-03 | 中国华能集团清洁能源技术研究院有限公司 | Offshore wind farm generated energy determination method and system based on satellite inversion |
| WO2025016116A1 (en)* | 2023-07-18 | 2025-01-23 | 三峡国际能源投资集团有限公司 | Offshore wind resource prediction method and apparatus, and computer device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104699936A (en)* | 2014-08-18 | 2015-06-10 | 沈阳工业大学 | Sector management method based on CFD short-term wind speed forecasting wind power plant |
| CN106505631A (en)* | 2016-10-29 | 2017-03-15 | 塞壬智能科技（北京）有限公司 | Intelligent wind power wind power prediction system |
| CN114552570A (en)* | 2022-02-24 | 2022-05-27 | 广东电网有限责任公司 | Offshore wind power prediction management system |
- 2022
- 2022-06-23CNCN202210718925.4Apatent/CN115034159A/enactivePending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104699936A (en)* | 2014-08-18 | 2015-06-10 | 沈阳工业大学 | Sector management method based on CFD short-term wind speed forecasting wind power plant |
| CN106505631A (en)* | 2016-10-29 | 2017-03-15 | 塞壬智能科技（北京）有限公司 | Intelligent wind power wind power prediction system |
| CN114552570A (en)* | 2022-02-24 | 2022-05-27 | 广东电网有限责任公司 | Offshore wind power prediction management system |
Non-Patent Citations (1)
| Title |
|---|
| 胡号朋等: "基于致动线模型的风电机组CFD 仿真研究及验证", 船舶工程, 15 November 2020 (2020-11-15), pages 196 - 200* |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115693666A (en)* | 2022-12-30 | 2023-02-03 | 中国华能集团清洁能源技术研究院有限公司 | Offshore wind farm generated energy determination method and system based on satellite inversion |
| WO2025016116A1 (en)* | 2023-07-18 | 2025-01-23 | 三峡国际能源投资集团有限公司 | Offshore wind resource prediction method and apparatus, and computer device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126704B (en) | Multi-region precipitation prediction model construction method based on multi-graph convolution and memory network | |
| Velo et al. | Wind speed estimation using multilayer perceptron | |
| CN113283588B (en) | Near-shore single-point wave height forecasting method based on deep learning | |
| CN115310648A (en) | A mid- and long-term wind power combined forecasting method based on multi-meteorological variable model identification | |
| CN111723523B (en) | Estuary surplus water level prediction method based on cascade neural network | |
| CN113592144B (en) | Medium-long term runoff probability forecasting method and system | |
| CN117526274A (en) | New energy power prediction method, electronic equipment and storage medium in extreme climate | |
| CN116454875A (en) | Method and system for medium-term power probability prediction method and system of regional wind farms based on cluster division | |
| CN115034159A (en) | Power prediction method, device, storage medium and system for offshore wind farm | |
| CN117408171B (en) | Hydrologic set forecasting method of Copula multi-model condition processor | |
| CN107274030A (en) | Runoff Forecast method and system based on hydrology variable year border and monthly variation characteristic | |
| CN116706992A (en) | Self-adaptive power prediction method, device and equipment for distributed photovoltaic cluster | |
| CN119228070B (en) | A decision-making method for reservoir group operation considering multi-source uncertainty propagation and evolution tracking | |
| CN115329930A (en) | Flood process probability forecasting method based on mixed deep learning model | |
| CN106526710A (en) | Haze prediction method and device | |
| CN117874989A (en) | Method and system for simulating and predicting runoff in non-data area based on space-time characteristics | |
| CN111325376B (en) | Wind speed prediction method and device | |
| CN118690656A (en) | A multi-model integrated runoff simulation method and system | |
| CN117233866A (en) | CNOP-based weather set forecasting system, method and storage medium | |
| CN117147396A (en) | A pollen concentration distribution prediction method and system | |
| Huang et al. | Hybrid neural network models for hydrologic time series forecasting based on genetic algorithm | |
| CN116303786A (en) | Block chain financial big data management system based on multidimensional data fusion algorithm | |
| CN114741952A (en) | Short-term load prediction method based on long-term and short-term memory network | |
| CN114066250A (en) | A method, device, equipment and storage medium for calculating repair cost of power transmission project | |
| US20240320549A1 (en) | Information processing apparatus, information processing method, and recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |