A method and system for reducing wind power output based on point-by-point correction of time series

Abstract

The invention discloses a wind power output reduction method and system based on time series point-by-point correction, comprising: obtaining wind turbine data and SCADA data of each wind turbine in an existing wind farm, and determining an actual output curve of the wind farm; modeling and CFD calculation of the wind farm according to the wind turbine data and SCADA data to obtain an annual theoretical output curve; determining a reduction coefficient at each moment based on the actual output curve of the wind farm and the annual theoretical output curve; grouping the reduction coefficients to determine the reduction coefficient corresponding to each wind speed segment; performing function fitting based on the reduction coefficient corresponding to each wind speed segment to determine a reduction probability model; performing distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine a reduction coefficient distribution model corresponding to each wind speed segment; obtaining a target annual theoretical output curve of a target wind farm and a wind speed at each moment, and correcting the target annual theoretical output curve based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy resource analysis, and more specifically, to a method and system for reducing wind power output based on point-by-point correction of time series. Background Art

[0002] With the proposal of the "dual carbon" goal, the energy structure is accelerating its transformation from fossil energy to clean energy. New energy sources such as wind and solar energy have become the main source of new installed capacity in the power system due to their pollution-free and renewable characteristics. However, the intermittent, volatile and random nature of renewable energy generation has brought unprecedented challenges to the traditional power system. In order to meet these challenges, a new power system is actively being built to achieve a high proportion of renewable energy utilization. The core goal of the new power system is to achieve clean, low-carbon, safe and efficient. The key lies in coordinating the balance of renewable energy generation, load demand and grid dispatch. In this system, not only is a high degree of flexibility required on the power supply side (such as wind power, photovoltaics, etc.), but the load side and energy storage facilities also need to work more closely together.

[0003] In the new power system, source-grid-load-storage integrated projects have gradually become an important means to solve the problem of new energy fluctuations. By systematically designing and co-optimizing the power generation end (source), transmission and distribution network (grid), user-side load (load) and energy storage system (storage), source-grid-load-storage integration can play an important role in peak shaving and valley filling, improving power quality and enhancing system safety. Such projects can not only achieve efficient utilization of new energy electricity, but also improve the operating economy and reliability of the power system. In source-grid-load-storage integrated projects, the prediction and evaluation of new energy output is crucial. In the planning and design stage of such projects, production simulation analysis of the annual power supply and demand is required to ensure the balance and economy of the project under various operating conditions.

[0004] In production simulation, the annual 8760-hour output curve of new energy is one of the core input data. This curve describes in detail the output of wind farms or photovoltaic power stations in each hour, and is the basis for supply and demand balance analysis, energy storage optimization configuration and scheduling strategy research.

[0005] The randomness and volatility of renewable energy output determine that the power system dispatch must take into account the hourly output changes. Especially in the source-grid-load-storage integrated project, the 8760-hour output curve of renewable energy directly affects the system's dispatch strategy and economic evaluation. Therefore, generating an 8760-hour output sequence that conforms to the actual situation is crucial for production simulation.

[0006] As an important component of the current installed capacity of new energy, wind power has a particularly significant volatility and uncertainty in output. Compared with photovoltaic output, which is strongly affected by the day and night cycle, wind power output is more complexly affected by meteorological conditions (such as wind speed, wind direction, temperature, etc.) and has a higher degree of randomness.

[0007] In actual projects, the wind power 8760-hour output curve is not only used for annual power generation evaluation, but can also be directly used in the following scenarios: load matching analysis: whether the wind power output can meet the load demand in each time period; energy storage configuration optimization: how the energy storage system balances the output between the wind power off-peak period and the peak period; grid dispatching research: the impact of wind power fluctuations on the stability of grid operation.

[0008] In actual operation, the total annual power generation of a wind farm is usually lower than the theoretical value. This is because factors such as equipment failure, changes in meteorological conditions, and scheduling strategies will affect the actual power generation. In order to compensate for this deviation, a reduction factor is introduced in engineering practice. Although this method is simple and effective, it fails to refine the hourly time series dynamic characteristics and ignores the changes in the reduction factor in different time periods, which has a certain impact on the accuracy of production simulation.

[0009] Although the traditional reduction coefficient method can correct the theoretical power generation well on a macro level, it lacks time resolution. The reduction coefficient is only used to correct the annual total power generation, ignoring the hourly dynamic characteristics of 8760 hours.

[0010] Therefore, a wind power output reduction method based on point-by-point correction of time series is needed to dynamically and finely reduce the output. Summary of the invention

[0011] The present invention proposes a wind power output reduction method and system based on time series point-by-point correction to solve the problem of how to correct the wind power output.

[0012] In order to solve the above problem, according to one aspect of the present invention, a method for reducing wind power output based on point-by-point correction of time series is provided, the method comprising:

[0013] Obtaining wind turbine data and SCADA data of each wind turbine in the existing wind farm, and determining an actual output curve of the wind farm based on the SCADA data;

[0014] Modeling and CFD calculation of the wind farm are performed according to the wind turbine data and SCADA data to obtain an annual theoretical output curve;

[0015] Based on the actual output curve and the annual theoretical output curve of the wind farm, calculations are performed at preset time intervals to determine the reduction coefficient at each moment;

[0016] The reduction coefficients are grouped according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment;

[0017] Perform function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model;

[0018] Based on the reduction coefficient corresponding to each wind speed segment, distribution fitting is performed to determine the reduction coefficient distribution model corresponding to each wind speed segment;

[0019] The target year theoretical output curve of the target wind farm and the wind speed at each moment are obtained, and the target year theoretical output curve is corrected based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment.

[0020] Preferably, the method further comprises:

[0021] The actual output curve of the wind farm is cleaned to remove abnormal values ​​and missing values.

[0022] Preferably, the method further comprises:

[0023] When modeling and CFD timing the wind farm according to the wind turbine data and SCADA data, if the wind measurement data of the wind tower within the preset range of the wind farm is obtained, the wind resources are analyzed and calculated based on the wind measurement data; if the wind measurement data of the wind tower within the preset range of the wind farm is not obtained, the wind resources are analyzed and processed based on the SCADA data of the first wind turbine in the main wind power direction of the wind farm.

[0024] Preferably, grouping the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment includes:

[0025] If there are at least two years of reduction factor data, the reduction factors are initially grouped by season, and then finally grouped by preset wind speed segments in each season to determine the reduction factor corresponding to each wind speed segment;

[0026] If there is only one year of reduction factor data, the data is grouped directly according to the preset wind speed segments to determine the reduction factor corresponding to each wind speed segment.

[0027] Preferably, performing function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model includes:

[0028] The probability of reduction in each wind speed segment is calculated, and a function is fitted based on the probability of reduction in all wind speed segments to determine the reduction probability model.

[0029] Preferably, performing distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution model corresponding to each wind speed segment includes:

[0030]

[0031] Among them, f R (x) is the density function describing the reduction factor x in a certain wind speed range; u R is the mean value, indicating the average reduction factor of this wind speed segment, σ R is the standard deviation, which indicates the dispersion of the reduction factor.

[0032] Preferably, the target year theoretical output curve is corrected based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment, including:

[0033] For any moment in the target year theoretical output curve, according to the wind speed corresponding to the any moment, the reduction probability model and the reduction coefficient distribution model are used to determine the target reduction probability and the target reduction coefficient distribution corresponding to the any moment;

[0034] According to the product of the theoretical output value corresponding to any moment, the target reduction probability and the target reduction coefficient distribution, the reduced output value corresponding to any moment is determined to correct the target year theoretical output curve.

[0035] According to another aspect of the present invention, a wind power output reduction system based on time series point-by-point correction is provided, the system comprising:

[0036] A data acquisition unit, used to acquire wind turbine data and SCADA data of each wind turbine in the existing wind farm, and determine an actual output curve of the wind farm based on the SCADA data;

[0037] An annual theoretical output curve acquisition unit is used to perform modeling and CFD calculation on the wind farm according to the wind turbine data and SCADA data to obtain an annual theoretical output curve;

[0038] A reduction coefficient determination unit, configured to determine the reduction coefficient at each moment by performing calculations at preset time intervals based on the actual output curve of the wind farm and the annual theoretical output curve;

[0039] A reduction coefficient grouping unit, used to group the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment;

[0040] A reduction probability model determination unit is used to perform function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model;

[0041] A reduction coefficient distribution model determination unit is used to perform distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution model corresponding to each wind speed segment;

[0042] The correction unit is used to obtain the target year theoretical output curve of the target wind farm and the wind speed at each moment, and correct the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment.

[0043] Preferably, the system further comprises:

[0044] The data cleaning unit is used to perform data cleaning on the actual output curve of the wind farm to remove abnormal values ​​and missing values.

[0045] Preferably, the annual theoretical output curve acquisition unit is further used for:

[0046] When modeling and CFD timing the wind farm according to the wind turbine data and SCADA data, if the wind measurement data of the wind tower within the preset range of the wind farm is obtained, the wind resources are analyzed and calculated based on the wind measurement data; if the wind measurement data of the wind tower within the preset range of the wind farm is not obtained, the wind resources are analyzed and processed based on the SCADA data of the first wind turbine in the main wind power direction of the wind farm.

[0047] Preferably, the reduction coefficient grouping unit groups the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment, including:

[0048] If there are at least two years of reduction factor data, the reduction factors are initially grouped by season, and then finally grouped by preset wind speed segments in each season to determine the reduction factor corresponding to each wind speed segment;

[0049] If there is only one year of reduction factor data, the data is grouped directly according to the preset wind speed segments to determine the reduction factor corresponding to each wind speed segment.

[0050] Preferably, the reduction probability model determination unit is used to perform function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model, including:

[0051] The probability of reduction in each wind speed segment is calculated, and a function is fitted based on the probability of reduction in all wind speed segments to determine the reduction probability model.

[0052] Preferably, the reduction coefficient distribution model determination unit performs distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution model corresponding to each wind speed segment, including:

[0053]

[0054] Among them, f R (x) is the density function describing the reduction factor x in a certain wind speed range; u R is the mean value, indicating the average reduction factor of this wind speed segment, σ R is the standard deviation, which indicates the dispersion of the reduction factor.

[0055] Preferably, the correction unit corrects the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment, including:

[0056] For any moment in the target year theoretical output curve, according to the wind speed corresponding to the any moment, the reduction probability model and the reduction coefficient distribution model are used to determine the target reduction probability and the target reduction coefficient distribution corresponding to the any moment;

[0057] According to the product of the theoretical output value corresponding to any moment, the target reduction probability and the target reduction coefficient distribution, the reduced output value corresponding to any moment is determined to correct the target year theoretical output curve.

[0058] The present invention provides a wind power output reduction method and system based on time series point-by-point correction, comprising: obtaining wind turbine data and SCADA data of each wind turbine in an existing wind farm, and determining an actual wind farm output curve based on the SCADA data; modeling and CFD calculation of the wind farm according to the wind turbine data and SCADA data to obtain an annual theoretical output curve; calculating according to a preset time interval based on the actual wind farm output curve and the annual theoretical output curve to determine a reduction coefficient at each moment; grouping the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment; performing function fitting based on the reduction coefficient corresponding to each wind speed segment to determine a reduction probability model; performing distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine a reduction coefficient distribution model corresponding to each wind speed segment; obtaining a target annual theoretical output curve of a target wind farm and a wind speed at each moment, and correcting the target annual theoretical output curve based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment. The present invention improves the accuracy of wind power output curves, solves the problems of traditional methods overestimating power generation and ignoring dynamic changes, and can provide more reliable data support for wind farm design and evaluation; meets the needs of source-grid-load-storage integrated projects for refined time-series output, and can provide high-precision basic data support for wind farm supply and demand balance analysis, energy storage configuration optimization and dynamic scheduling, improves the matching degree between wind farms and load centers, reduces wind abandonment rate, and optimizes overall project operation efficiency; provides a highly adaptable general method, which can be widely used in the planning and design of new wind farms and the performance evaluation of existing wind farms, and provides a general and efficient solution for new energy output prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] A more complete understanding of exemplary embodiments of the present invention may be obtained by referring to the following drawings:

[0060] Figure 1 Flow chart of a method 100 for reducing wind power output based on point-by-point correction of time series according to an embodiment of the present invention;

[0061] Figure 2 A flow chart generated by a wind power output reduction model according to an embodiment of the present invention;

[0062] Figure 3 Schematic diagram of the structure of a wind power output reduction system 300 based on time series point-by-point correction according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0064] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0065] The present invention provides a wind power output reduction method based on point-by-point refined correction of time series. By studying the reduction of theoretical output and actual output of existing wind farms, a reduction model based on different wind speed ranges is found. The purpose is to solve the shortcomings of existing wind power output prediction technology in dynamic adaptability, refined requirements, calculation efficiency and matching of actual operating conditions, and provide reliable technical support for the planning, design and production simulation of source-grid-load-storage integrated projects.

[0066] Figure 1 FIG. 1 is a flow chart of a method 100 for reducing wind power output based on point-by-point correction of a time series according to an embodiment of the present invention. Figure 1 As shown, the wind power output reduction method based on time series point-by-point correction provided by the embodiment of the present invention improves the accuracy of the wind power output curve, solves the problem of overestimation of power generation and neglect of dynamic changes in traditional methods, and can provide more reliable data support for wind farm design and evaluation; meets the demand for refined time series output of source-grid-load-storage integrated projects, and can provide high-precision basic data support for wind farm supply and demand balance analysis, energy storage configuration optimization and dynamic scheduling, improves the matching degree between wind farms and load centers, reduces wind abandonment rate, and optimizes the overall project operation efficiency; provides a general method with strong adaptability, which can be widely used in the planning and design of new wind farms and the performance evaluation of operating wind farms, and provides a general and efficient solution for new energy output prediction. The wind power output reduction method 100 based on time series point-by-point correction provided by the embodiment of the present invention starts from step 101. In step 101, the wind turbine data and SCADA data of each wind turbine in the existing wind farm are obtained, and the actual output curve of the wind farm is determined based on the SCADA data.

[0067] Preferably, the method further comprises:

[0068] The actual output curve of the wind farm is cleaned to remove abnormal values ​​and missing values.

[0069] In the present invention, the wind turbine data of each wind turbine in an existing wind farm is collected, including parameters such as wind turbine latitude and longitude coordinates, hub height, blade diameter, power curve, etc. At the same time, the SCADA data of wind turbines in operation in the wind farm are collected, including data such as wind speed, wind direction, output, etc., and the actual wind farm output curve P is formed. actual (t), then P actual (t) Perform data cleaning to remove outliers and missing values ​​(such as zero power due to shutdown for maintenance) to avoid affecting the calculation of the reduction factor.

[0070] In step 102, the wind farm is modeled and CFD calculated based on the wind turbine data and SCADA data to obtain an annual theoretical output curve.

[0071] Preferably, the method further comprises:

[0072] When modeling and CFD timing the wind farm according to the wind turbine data and SCADA data, if the wind measurement data of the wind tower within the preset range of the wind farm is obtained, the wind resources are analyzed and calculated based on the wind measurement data; if the wind measurement data of the wind tower within the preset range of the wind farm is not obtained, the wind resources are analyzed and processed based on the SCADA data of the first wind turbine in the main wind power direction of the wind farm.

[0073] In the present invention, if the wind measurement data of the synchronous wind measurement towers around the wind farm are collected, the wind measurement data are analyzed and processed according to the relevant wind resource evaluation standards. If the synchronous wind measurement data of the surrounding areas are not collected, the SCADA wind speed and wind direction data of the first wind turbine in the main wind energy direction of the wind farm are used to perform wind resource analysis and processing.

[0074] Then, the wind farm was modeled using MeteodynWT software. The wind farm site elevation data, surface roughness, longitude and latitude of each wind turbine, wind turbine hub height, wind turbine output model, wind tower location and wind measurement data were input to model the wind farm and perform CFD calculations to generate the annual theoretical output curve P. theoretical (t), which is the full-field theoretical output curve after the software considers the fan wake and air density reduction. In the present invention, it is also necessary to align P actual (t) and P theoretical (t) Time axis, ensure that the time axes of the two curves are consistent for subsequent comparison.

[0075] In step 103, based on the actual output curve of the wind farm and the annual theoretical output curve, calculations are performed at preset time intervals to determine the reduction coefficient at each moment.

[0076] In the present invention, the ratio of actual output power to theoretical output power per hour is calculated according to the actual output curve of the wind farm and the annual theoretical output curve, so as to obtain the reduction coefficient R(t) corresponding to each moment.

[0077] In step 104, the reduction coefficients are grouped according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment.

[0078] Preferably, grouping the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment includes:

[0079] If there are at least two years of reduction factor data, the reduction factors are initially grouped by season, and then finally grouped by preset wind speed segments in each season to determine the reduction factor corresponding to each wind speed segment;

[0080] If there is only one year of reduction factor data, the data is grouped directly according to the preset wind speed segments to determine the reduction factor corresponding to each wind speed segment.

[0081] Combination Figure 2 As shown, in the present invention, if the data is multi-year data (when the data volume is large), R(t) is divided into four groups according to the time period: spring, summer, autumn, and winter, and in each season, the wind speed is divided into multiple wind speed segments, such as 3-5m / s, 5-7m / s, 7-9m / s, 9-11m / s, etc.; if the data is one year's data (when the data volume is small), R(t) is directly divided into multiple wind speed segments, such as 3-5m / s, 5-7m / s, 7-9m / s, 9-11m / s, etc.

[0082] In step 105, a function fitting is performed based on the reduction coefficient corresponding to each wind speed segment to determine a reduction probability model.

[0083] Preferably, performing function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model includes:

[0084] The probability of reduction in each wind speed segment is calculated, and a function is fitted based on the probability of reduction in all wind speed segments to determine the reduction probability model.

[0085] In the present invention, the reduction probability model represents the relationship between the probability of reduction occurring in a certain wind speed range and the wind speed, and can be fitted into a function.

[0086] Specifically, it includes: for each wind speed segment, calculating the probability of reduction, the calculation formula is:

[0087]

[0088] Among them, N R(t)＜1is the number of times the reduction occurs in each wind speed segment, N total is the total data quantity of this wind speed segment.

[0089] And obtain the probability P of the reduction corresponding to each wind speed segment reduction (v) Then, polynomial fitting is used to obtain the reduced probability model:

[0090] P reduction (v) = a0 + a1v + a2v 2 +...,

[0091] Or perform exponential fitting to obtain the reduced probability model:

[0092] P reduction (v) = a·e -bv +c.

[0093] Among them, P reduction (v) is the reduced probability when the wind speed at the wind tower is v.

[0094] In step 106, distribution fitting is performed based on the reduction coefficient corresponding to each wind speed segment to determine a reduction coefficient distribution model corresponding to each wind speed segment.

[0095] Preferably, performing distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution model corresponding to each wind speed segment includes:

[0096]

[0097] Among them, f R (x) is the density function describing the reduction factor x in a certain wind speed range; u R is the mean value, indicating the average reduction factor of this wind speed segment, σ R is the standard deviation, which indicates the dispersion of the reduction factor.

[0098] In the present invention, distribution fitting is performed on the reduction value of each wind speed segment to obtain its probability distribution function. If the reduction coefficient R of each wind speed segment is normally distributed, then:

[0099]

[0100] Among them, f R (x) is the density function describing the reduction factor x in a certain wind speed range; u R is the mean value, indicating the average reduction factor of this wind speed segment, σ R is the standard deviation, which indicates the degree of dispersion of the reduction factor. This step can reflect the probability density of the reduction factor.

[0101] In step 107, the target year theoretical output curve of the target wind farm and the wind speed at each moment are obtained, and the target year theoretical output curve is corrected based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment.

[0102] Preferably, the target year theoretical output curve is corrected based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment, including:

[0103] For any moment in the target year theoretical output curve, according to the wind speed corresponding to the any moment, the reduction probability model and the reduction coefficient distribution model are used to determine the target reduction probability and the target reduction coefficient distribution corresponding to the any moment;

[0104] According to the product of the theoretical output value corresponding to any moment, the target reduction probability and the target reduction coefficient distribution, the reduced output value corresponding to any moment is determined to correct the target year theoretical output curve.

[0105] In the present invention, for a newly built wind farm, the conventional wind power engineering method can be used to obtain the annual theoretical power generation curve of the wind farm, and the annual power generation curve is finely reduced hour by hour with reference to the real-time wind speed of the wind tower of the project. Specifically, the theoretical output value P of the wind farm every hour theoretical Each of them corresponds to a wind speed value v of the wind tower. For any hour, based on the corresponding wind speed value of the wind tower, the wind speed reduction probability model can be used to obtain the P under the wind speed in any hour. theoretical The reduced probability P reduction ; Then, using the reduction coefficient distribution model, we can get P at this wind speed in any hour. theoretical The distribution of the reduction coefficient; then for any hour, the reduced output P actual For: P actual =P theoretica ·P reduction ·f R , thereby correcting the theoretical output value of each hour in the theoretical output curve of the target year.

[0106] In the present invention, if a seasonal discount model is used, it is necessary to use the seasonal discount model according to the season in which the data is located.

[0107] The present invention introduces a reduction method based on point-by-point refinement of time series, and solves the deficiencies of existing wind power output prediction technology in terms of dynamic adaptability, refinement requirements, data dependence and computational efficiency based on actual engineering needs. The specific technical effects produced by the present invention include:

[0108] (1) Improve the accuracy of wind power output curve

[0109] The present invention refines the traditional static annual total power generation reduction method into hourly dynamic adjustment by correcting the 8760-hour theoretical output curve of wind power point by point. By introducing the reduction probability model and reduction coefficient distribution model based on wind speed and season, combined with actual operating conditions and statistical laws, the actual output generated by the corrected theoretical curve is more in line with the actual operating conditions. Compared with the traditional empirical reduction method (for example, a unified reduction coefficient of 0.75), the output curve generated by the present invention can effectively reflect the multiple influences of wind speed, equipment maintenance plan, scheduling restrictions and meteorological conditions. The error of the wind power 8760-hour output curve is significantly reduced, and the fit with the actual operating data is improved. It solves the problem of overestimation of power generation and neglect of dynamic changes in traditional methods, and provides more reliable data support for wind farm design and evaluation.

[0110] (2) Meeting the demand for refined timing output of integrated power generation, grid-load and storage projects

[0111] The planning and design of integrated source-grid-load-storage projects requires an accurate assessment of the matching degree between the output of new energy and the load, while the traditional reduction method cannot meet the needs of hourly dynamic analysis. The dynamic reduction method of the present invention effectively solves this problem: the actual output curve corrected point by point can be directly used for the production simulation of the source-grid-load-storage project to meet the accuracy requirements of load demand and scheduling optimization. By introducing the reduction model of wind speed segments and seasonality (including reduction probability and reduction value), the output characteristics under different operating conditions are fully reflected. It can provide high-precision basic data support for wind farm supply and demand balance analysis, energy storage configuration optimization and dynamic scheduling. It improves the matching degree between wind farms and load centers, reduces the wind abandonment rate, and optimizes the overall project operation efficiency.

[0112] (3) Providing a general method with strong adaptability

[0113] The present invention avoids dependence on a large amount of historical data by establishing a reduction model based on statistical laws and actual data, while having both adaptability and robustness. In scenarios with limited historical data, a reliable reduction model can be generated through theoretical data and statistical analysis; in wind farms supported by existing operating data, dynamic correction can be further used to optimize the reduction factor to enhance the applicability of the model. The method of the present invention can be widely used in the planning and design of new wind farms and the performance evaluation of existing wind farms; it provides a general and efficient solution for the prediction of new energy output.

[0114] (4) Improve model transparency and explainability

[0115] Compared with black-box methods such as machine learning, this invention uses transparent mathematical models and physical principles to ensure the interpretability and engineering practicality of the results. The definition and calculation of the reduction factor are directly related to actual variables such as wind speed, equipment status, and scheduling strategy, making the model results easy to understand; users can adjust model parameters (such as wind speed ranges, seasonal characteristics, etc.) according to different scenarios and flexibly apply them to a variety of engineering scenarios. The usability and trustworthiness of model results in engineering practice are improved; engineers are supported to quickly verify and correct the forecast results of new energy output.

[0116] (5) Providing efficient computing solutions

[0117] The present invention effectively reduces the computational complexity through a point-by-point correction algorithm of dynamic reduction: no complex numerical weather forecast models and high-performance computing resources are required, and dynamic reduction can be achieved with only wind speed time series data and simple statistical analysis. Compared with traditional reduction methods, not only is the accuracy significantly improved, but the computational cost is still kept within a reasonable range. It supports fast calculations of large-scale wind farms or long-term prediction scenarios, and is suitable for real-time scheduling optimization; under the condition of limited hardware resources, high-precision output curves can still be generated.

[0118] (6) Comprehensive support for analysis and optimization of reduction characteristics

[0119] This invention further reveals the regularity of reduction behavior by constructing a reduction probability model for wind speed segments and seasonality: by analyzing the functional relationship between the reduction probability and the reduction size, the reduction characteristics under different wind speeds and seasons are quantified; a model based on probability distribution is provided to make the reduction analysis more systematic and scientific. It can provide data support for wind farm operation optimization and scheduling strategies, such as selecting the optimal maintenance period or designing load-side response plans; in the wind farm design stage, the reduction risk under different wind resource conditions can be evaluated in advance, and the wind farm layout and capacity configuration can be optimized.

[0120] Figure 3 FIG. 3 is a schematic diagram of a wind power output reduction system 300 based on time series point-by-point correction according to an embodiment of the present invention. Figure 3 As shown, the wind power output reduction system 300 based on time series point-by-point correction provided in an embodiment of the present invention includes: a data acquisition unit 301, an annual theoretical output curve acquisition unit 302, a reduction coefficient determination unit 303, a reduction coefficient grouping unit 304, a reduction probability model determination unit 305, a reduction coefficient distribution model determination unit 306 and a correction unit 307.

[0121] Preferably, the data acquisition unit 301 is used to acquire wind turbine data and SCADA data of each wind turbine in an existing wind farm, and determine an actual output curve of the wind farm based on the SCADA data.

[0122] Preferably, the system further comprises:

[0123] The data cleaning unit is used to perform data cleaning on the actual output curve of the wind farm to remove abnormal values ​​and missing values.

[0124] Preferably, the annual theoretical output curve acquisition unit 302 is used to perform modeling and CFD calculation on the wind farm according to the wind turbine data and SCADA data to acquire the annual theoretical output curve.

[0125] Preferably, the annual theoretical output curve acquisition unit 302 is further used for:

[0126] When modeling and CFD timing the wind farm according to the wind turbine data and SCADA data, if the wind measurement data of the wind tower within the preset range of the wind farm is obtained, the wind resources are analyzed and calculated based on the wind measurement data; if the wind measurement data of the wind tower within the preset range of the wind farm is not obtained, the wind resources are analyzed and processed based on the SCADA data of the first wind turbine in the main wind power direction of the wind farm.

[0127] Preferably, the reduction coefficient determination unit 303 is used to perform calculations at preset time intervals based on the actual output curve of the wind farm and the annual theoretical output curve to determine the reduction coefficient at each moment.

[0128] Preferably, the reduction coefficient grouping unit 304 is used to group the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment.

[0129] Preferably, the reduction coefficient grouping unit 304 groups the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment, including:

[0130] If there are at least two years of reduction factor data, the reduction factors are initially grouped by season, and then finally grouped by preset wind speed segments in each season to determine the reduction factor corresponding to each wind speed segment;

[0131] If there is only one year of reduction factor data, the data is grouped directly according to the preset wind speed segments to determine the reduction factor corresponding to each wind speed segment.

[0132] Preferably, the reduction probability model determining unit 305 is used to perform function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model.

[0133] Preferably, the reduction probability model determining unit 305 is used to perform function fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction probability model, including:

[0134] The probability of reduction in each wind speed segment is calculated, and a function is fitted based on the probability of reduction in all wind speed segments to determine the reduction probability model.

[0135] Preferably, the reduction coefficient distribution model determining unit 306 is used to perform distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution model corresponding to each wind speed segment.

[0136] Preferably, the reduction coefficient distribution model determining unit 306 performs distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution model corresponding to each wind speed segment, including:

[0137]

[0138] Among them, f R (x) is the density function describing the reduction factor x in a certain wind speed range; u R is the mean value, indicating the average reduction factor of this wind speed segment, σ R is the standard deviation, which indicates the dispersion of the reduction factor.

[0139] Preferably, the correction unit 307 is used to obtain the target year theoretical output curve of the target wind farm and the wind speed at each moment, and correct the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment.

[0140] Preferably, the correction unit 307 corrects the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution model and the wind speed corresponding to each moment, including:

[0141] For any moment in the target year theoretical output curve, according to the wind speed corresponding to the any moment, the reduction probability model and the reduction coefficient distribution model are used to determine the target reduction probability and the target reduction coefficient distribution corresponding to the any moment;

[0142] According to the product of the theoretical output value corresponding to any moment, the target reduction probability and the target reduction coefficient distribution, the reduced output value corresponding to any moment is determined to correct the target year theoretical output curve.

[0143] The wind power output reduction system 300 based on time series point-by-point correction of the embodiment of the present invention corresponds to the wind power output reduction method 100 based on time series point-by-point correction of another embodiment of the present invention, and will not be described in detail herein.

[0144] The present invention has been described with reference to a few embodiments. However, it is known to those skilled in the art that other embodiments than the one disclosed above are equally within the scope of the present invention.

[0145] Generally, all terms used in the present invention are interpreted according to their ordinary meaning in the technical field, unless otherwise explicitly defined therein. All references to "a / said / the [device, component, etc.]" are open to interpretation as at least one instance of the device, component, etc., unless otherwise explicitly stated. The steps of any method disclosed herein do not necessarily have to be performed in the exact order disclosed, unless explicitly stated.

[0146] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0147] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for reducing wind power output based on point-by-point correction of time series, characterized in that: The method comprises: Obtaining wind turbine data and SCADA data of each wind turbine in the existing wind farm, and determining an actual output curve of the wind farm based on the SCADA data; Modeling and CFD calculation of the wind farm are performed according to the wind turbine data and SCADA data to obtain an annual theoretical output curve; Based on the actual output curve and the annual theoretical output curve of the wind farm, calculations are performed at preset time intervals to determine the reduction coefficient at each moment; The reduction coefficients are grouped according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment; The reduction probability of each wind speed segment is calculated, and a function fitting is performed based on the reduction probabilities corresponding to all wind speed segments to determine the reduction probability model; Based on the reduction coefficient corresponding to each wind speed segment, distribution fitting is performed to determine the reduction coefficient distribution probability model corresponding to each wind speed segment; Obtaining a target year theoretical output curve of a target wind farm and the wind speed at each moment, and revising the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution probability model and the wind speed corresponding to each moment; The target year theoretical output curve is corrected based on the reduction probability model, the reduction coefficient distribution probability model and the wind speed corresponding to each moment, including: For any moment in the target year theoretical output curve, according to the wind speed corresponding to the any moment, the target reduction probability and the probability of the target reduction coefficient distribution corresponding to the any moment are determined by using the reduction probability model and the reduction coefficient distribution probability model respectively; The reduced output value corresponding to any moment is determined according to the product of the theoretical output value corresponding to any moment, the target reduction probability and the probability of the target reduction coefficient distribution, so as to correct the target year theoretical output curve.

2. The method according to claim 1, characterized in that The method further comprises: Data cleaning is performed on the actual output curve of the wind farm to remove abnormal values ​​and missing values.

3. The method according to claim 1, characterized in that The method further comprises: When modeling and CFD timing the wind farm according to the wind turbine data and SCADA data, if the wind measurement data of the wind tower within the preset range of the wind farm is obtained, the wind resources are analyzed and calculated based on the wind measurement data; if the wind measurement data of the wind tower within the preset range of the wind farm is not obtained, the wind resources are analyzed and processed based on the SCADA data of the first wind turbine in the main wind power direction of the wind farm.

4. The method according to claim 1, characterized in that: The reduction coefficients are grouped according to a preset grouping strategy to determine the reduction coefficients corresponding to each wind speed segment, including: If there are at least two years of reduction factor data, the reduction factors are initially grouped by season, and then finally grouped by preset wind speed segments in each season to determine the reduction factor corresponding to each wind speed segment; If there is only one year of reduction factor data, the data is directly grouped according to the preset wind speed segments to determine the reduction factor corresponding to each wind speed segment.

5. The method according to claim 1, characterized in that The reduction coefficient distribution probability model conforms to the normal distribution.

6. A wind power output reduction system based on time series point-by-point correction, characterized in that: The system comprises: A data acquisition unit, used to acquire wind turbine data and SCADA data of each wind turbine in the existing wind farm, and determine an actual output curve of the wind farm based on the SCADA data; An annual theoretical output curve acquisition unit is used to perform modeling and CFD calculation on the wind farm according to the wind turbine data and SCADA data to obtain an annual theoretical output curve; A reduction coefficient determination unit, configured to determine the reduction coefficient at each moment by performing calculations at preset time intervals based on the actual output curve of the wind farm and the annual theoretical output curve; A reduction coefficient grouping unit, used to group the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment; A reduction probability model determination unit is used to calculate the reduction probability of each wind speed segment, and to perform function fitting based on the reduction probabilities corresponding to all wind speed segments to determine the reduction probability model; A reduction coefficient distribution probability model determination unit is used to perform distribution fitting based on the reduction coefficient corresponding to each wind speed segment to determine the reduction coefficient distribution probability model corresponding to each wind speed segment; A correction unit, used to obtain a target year theoretical output curve of a target wind farm and a wind speed at each moment, and to correct the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution probability model and the wind speed corresponding to each moment; The correction unit corrects the target year theoretical output curve based on the reduction probability model, the reduction coefficient distribution probability model and the wind speed corresponding to each moment, including: For any moment in the target year theoretical output curve, according to the wind speed corresponding to the any moment, the target reduction probability and the probability of the target reduction coefficient distribution corresponding to the any moment are determined by using the reduction probability model and the reduction coefficient distribution probability model respectively; The reduced output value corresponding to any moment is determined according to the product of the theoretical output value corresponding to any moment, the target reduction probability and the probability of the target reduction coefficient distribution, so as to correct the target year theoretical output curve.

7. The system according to claim 6, characterized in that The system further comprises: The data cleaning unit is used to perform data cleaning on the actual output curve of the wind farm to remove abnormal values ​​and missing values.

8. The system according to claim 6, characterized in that The annual theoretical output curve acquisition unit is also used for: When modeling and CFD timing the wind farm according to the wind turbine data and SCADA data, if the wind measurement data of the wind tower within the preset range of the wind farm is obtained, the wind resources are analyzed and calculated based on the wind measurement data; if the wind measurement data of the wind tower within the preset range of the wind farm is not obtained, the wind resources are analyzed and processed based on the SCADA data of the first wind turbine in the main wind power direction of the wind farm.

9. The system according to claim 6, characterized in that The reduction coefficient grouping unit groups the reduction coefficients according to a preset grouping strategy to determine the reduction coefficient corresponding to each wind speed segment, including: If there are at least two years of reduction factor data, the reduction factors are initially grouped by season, and then finally grouped by preset wind speed segments in each season to determine the reduction factor corresponding to each wind speed segment; If there is only one year of reduction factor data, the data is directly grouped according to the preset wind speed segments to determine the reduction factor corresponding to each wind speed segment.

10. The system according to claim 6, characterized in that The reduction coefficient distribution probability model conforms to the normal distribution.

Images