A wind resource assessment system for repowering a wind farm from large to small

Abstract

The application discloses a kind of wind resource evaluation system for wind farm to big change small transformation, comprising: original data acquisition module: for obtaining original wind farm information, original wind farm historical operation data and measured wind data, generate complete wind resource data;Wind farm information simulation module: for constructing multidimensional wind farm model M and expected wind farm model M3 based on original wind farm information;Evaluation index determination module: for calculating the index value of wind resource index according to the factors considered in big change small transformation, factors include positive correlation factors and negative correlation factors;Transformation evaluation module: for priority calculation according to index weight value;Risk assessment module: for risk assessment to specified wind turbine, according to management and maintenance difficulty, management level and technical decision scheme are provided.According to the above technical scheme, the running environment of the transformed wind farm can be simulated, the priority level of wind turbine transformation is selected, and the power generation after transformation is predicted;Predict various risk factors during and after the operation of wind farm transformation.

Description

A wind resource assessment system for wind farm retrofitting. Technical Field

[0001] This invention relates to the field of wind power development, and more specifically, to a wind resource assessment system for the large-to-small conversion of wind farms. Background Technology

[0002] In recent years, my country has been carrying out technological upgrades on aging wind farms. These upgrades can be broadly categorized into two types: "technical upgrading for efficiency improvement" and "replacing large turbines with smaller ones." "Technical upgrading for efficiency improvement" refers to upgrading existing wind turbine foundations, basic conditions, and the original height and condition of the towers to improve power generation efficiency. "Replacing large turbines with smaller ones" involves completely dismantling the old turbines and rebuilding new, high-efficiency wind turbines to enhance the overall efficiency of the wind farm. Notably, the wind resources in the selected areas when wind farms were built many years ago are generally higher than those in the current areas. Upgrading aging wind farms through "replacing large turbines with smaller ones" aims to fully utilize the existing wind resources and improve the efficiency of wind farms, serving as an important means and guarantee for the steady development of my country's new energy industry.

[0003] When implementing "replacing large with small" technological upgrades, there are situations where the existing wind measurement towers' heights no longer meet the current technological requirements. If new wind measurement towers are built only after the wind resource assessment requirements (at least 12 consecutive months of measurement) are met, the work will be costly, time-consuming, and unable to effectively utilize measurement data from years of wind turbine operation. This leads to insufficient accuracy and scientific rigor in predicting the upgrades and hinders the wind resource assessment work required for future large-scale wind farm upgrades. Furthermore, current wind turbine design and manufacturing technologies are more advanced than before, significantly reducing the number of turbine sites required for the same installed capacity. Therefore, based on current policies and investors' actual needs, it is not necessary to upgrade all existing wind farm turbine sites. Thus, before formal upgrades and development, it is necessary to assess the upgrade costs of existing turbine sites in the wind farm: evaluating the upgrade scale, single-unit capacity, and economic costs for each site. On the other hand, considering the overall environment of the wind farm, an assessment should be conducted to determine the potential geological and climatic risks to the turbines after the upgrade, as well as the management and maintenance difficulties.

[0004] In this situation, a technical solution is needed to assess the cost of wind farm site renovation and to propose scheduling plans based on the current operation of the wind farm, so as to avoid the loss of wind farm power generation and damage to wind farm equipment. Summary of the Invention

[0005] To achieve the above objectives, this application provides a wind resource assessment system for wind farm retrofitting, comprising:

[0006] Raw data acquisition module: used to acquire raw wind farm information, historical operation data and measured wind data of the raw wind farm, process data for data unification and validity, delete and replace outliers and improve missing values, and generate complete wind resource data; among which, raw wind farm information includes: wind farm environmental information, basic information of wind turbine units in the wind farm and raw wind measurement tower information.

[0007] Wind farm information simulation module: used to construct a multi-dimensional wind farm model M and a expected wind farm model M3 based on the original wind farm information; the multi-dimensional wind farm model M includes: historical wind farm model Mq, theoretical wind farm model M1, and actual wind farm model M2; the expected wind farm model M3 is generated by constructing the multi-dimensional wind farm model M.

[0008] Evaluation index determination module: used to calculate the index value of wind resource index based on the factors to be considered in the large-to-small wind turbine renovation, including positive and negative correlation factors; the renovation index is a factor that affects the price level of wind turbine renovation, including: the expected power generation revenue of the wind turbine, the cost of renovating and expanding the access road to the wind turbine, the cost of renovating and expanding the wind turbine site, and the cost of renovating the wind turbine collection line.

[0009] Modification assessment module: used to calculate priority based on the indicator values;

[0010] Risk assessment module: Used to conduct risk assessments on a specified wind turbine unit and provide decision-making solutions based on environmental risks and management and maintenance difficulties.

[0011] The wind farm environmental information includes: wind farm topographic information, obtained through the original design drawings and topographic maps of the wind farm; wherein, the design drawings contain the basic information of the original wind farm turbine units, including the foundation design type of the wind turbine, the foundation design elevation of the turbine location, the site leveling range, the tower base and tower height of the overhead power collection line, as well as the relevant information of the buried cable, and the wind farm turbine units of the surrounding existing and under-construction wind farms;

[0012] The basic information of the original wind farm's wind turbine units consists of the wind turbine information β that makes up the units. j constitute;

[0013] The original meteorological tower information includes: the original meteorological tower location. The coordinates of the aircraft position, the original height of the meteorological tower, the height of the original meteorological tower sensor's wind measurement layer, and the measured wind data obtained from the original meteorological tower: wind direction, wind speed, temperature, and air pressure data;

[0014] The historical operating data includes wind speed, wind direction, temperature, and air pressure data for each wind turbine during its operation over the years, as well as operational information during the operation period;

[0015] The measured wind data can be obtained from the original wind measurement tower before the wind farm is built, or from the wind turbines or power prediction tower after the wind farm is built.

[0016] In the wind farm information simulation module, when constructing a multi-dimensional wind farm model M, three-dimensional data modeling is performed, relevant wind parameters are obtained, the wind parameters are classified, and a multi-dimensional wind farm model M is established during the classification process, generating a historical wind farm model Mq, a theoretical wind farm model M1, and an actual wind farm model M2.

[0017] Furthermore, the results of classifying the wind parameters include the original set of parameters for the wind measurement tower and the set of parameters for each wind turbine;

[0018] Among them, the theoretical wind farm model M1 is a theoretical wind farm model simulated and constructed based on the original set of meteorological tower parameters; the actual wind farm model M2 is a wind farm model constructed based on the set of parameters of each wind turbine.

[0019] A multidimensional wind farm model M is established during the classification process, including:

[0020] Wind measurement tower The wind measurement tower was built before the wind farm was built and was demolished after the wind farm was built: ignoring the wind measurement tower. Wind measurement data;

[0021] Wind measurement tower The wind measurement tower was built before the wind farm was constructed, and was added after the wind turbines were built. Power prediction tower not dismantled For the same tower, correlation is used to replace wind measurement data from three parties for both same-tower and different-tower data; for wind measurement towers The wind measurement data from continuous measurements at the same location are split into segments, including: wind measurement towers before the construction of the wind farm. The first part uses wind measurement data to construct the historical wind farm model Mq; the second part uses wind measurement data from the period after the wind farm was built and during the operation of the wind turbines to construct the theoretical wind farm model M1.

[0022] Furthermore, the construction of the expected wind farm model M3 includes the following steps:

[0023] Prepare forecast data and establish a virtual wind measurement tower data source;

[0024] Based on virtual meteorological tower data sources, predictive wind resource data is constructed, including: for wind farms with multiple meteorological towers, calculating the functional relationships between the various meteorological layers of the same tower, the functional relationships between any two towers, and the relationship between the meteorological tower and its corresponding virtual meteorological tower γ. t The functional relationships between different wind turbines and between different wind turbines; and based on the correlation magnitude, prioritizing relationships between different layers within the same tower, between different wind turbines, and between wind measurement towers and virtual wind measurement towers. tThe order in which these data are used to construct the predicted wind resource data;

[0025] By comparing the wind resource parameters of the theoretical wind farm model M1 and the actual wind farm model M2, the deviation between the actual and theoretical values ​​is verified. Based on the deviation, deviation diagnosis is performed. The expected wind farm model M3 is constructed by combining the predicted wind resource data calculated by the virtual wind measurement tower and the deviation.

[0026] Furthermore, the priority calculation based on the aforementioned indicator value includes the following steps:

[0027] Assume there are n in the wind farm j Typhoon generators, and each generator has m i There are 10 evaluation indicators, where i is the evaluation indicator number, m is the number of evaluation indicators, j is the wind turbine number that can be modified, and n is the maximum number of wind turbines that can be modified.

[0028] Assess the weight of the i-th evaluation index for the j-th wind turbine in the wind farm, and generate the entropy value of each evaluation index for each wind turbine. ;

[0029] Based on the entropy values ​​of each evaluation index for each wind turbine Calculate the weighted ratio for each wind turbine. Determine the priority of wind turbine retrofitting.

[0030] Among them, the entropy values ​​of each evaluation index for each wind turbine are generated. Includes the following steps:

[0031] The original data matrix R′=(r′) is constructed using the entropy weight method. ij ) m×n The m evaluation indicators of the j-th wind turbine are normalized one by one, where r′ ij This is the evaluation value of the i-th evaluation indicator for the j-th wind turbine, after normalization. Represented as: ;

[0032] Calculate the entropy value of the i-th evaluation index for the j-th wind turbine. The calculation method is as follows: ,

[0033] in, Let be the weight of the i-th indicator of the j-th wind turbine, and: , .

[0034] Calculate the weighted ratio for each wind turbine. include:

[0035] The weight of the i-th evaluation indicator is defined as follows: ;

[0036] Calculate the weighted ratio for each wind turbine. The calculation method is as follows: ,

[0037] Where s represents the number of indicators that are positively correlated with the priority of wind power retrofitting, and g represents the number of indicators that are positively correlated with the priority of wind power retrofitting.

[0038] Furthermore, the risk assessment module includes a 3D modeling unit, a data processing unit, and an analysis unit;

[0039] The three-dimensional modeling unit is used to obtain the original wind farm information from the original data acquisition module and perform three-dimensional data modeling. After the three-dimensional model is constructed, environmental data under various conditions can be obtained, including: different flood frequencies and severity, altitude, slope, topography determined by surface vegetation, and return period of maximum wind speed.

[0040] The data processing unit obtains environmental data from the 3D modeling unit, loads a pre-trained third-party environmental risk prediction model, and obtains risk prediction results for each wind turbine under different environments after inputting the environmental data.

[0041] The analysis unit is used to combine risk prediction results to assess the impact of uncertainty factors in the operation of wind farms on different wind turbine operating scenarios, and to determine whether geological environmental factors within the wind farm will pose risks to wind turbine operation.

[0042] This invention can classify and apply historical data from the construction and operation of existing wind farms over many years. Data is segmented based on construction and application time points. Virtual wind measurement towers are used to predict wind measurement data after renovation, simulating the operating environment of the renovated wind farm and predicting post-renovation power generation. Various influencing factors are introduced to weight and evaluate each wind turbine, prioritizing the "replacement of large turbines with smaller ones" technology for the renovation of old wind farm locations. Furthermore, AI technology, combined with environmental information, is used to predict various risk factors during and after the wind farm renovation, reducing labor costs, lowering the difficulty of wind farm management, and improving the effectiveness and scientific rigor of wind farm renovation and management. Attached Figure Description

[0043] Figure 1 is a schematic diagram of the structure of a wind resource assessment system for wind farm renovation by replacing large-scale wind farms with smaller-scale wind farms, provided according to an embodiment of the present invention. Detailed Implementation

[0044] Currently, the "replacement of smaller turbines with larger ones" retrofitting of wind farms faces the following challenges: 1) Previously, individual turbine capacities were relatively small. When using the current larger turbine models for retrofitting, fewer turbine sites are needed for a wind farm of the same scale. Therefore, all turbine sites need to be prioritized and weighted to select the most suitable site. 2) Wind resource data measured during actual turbine operation is the most accurate. However, during turbine construction, unless otherwise specified by the investor, only relevant wind resource measurement sensors are installed at the hub height, which does not meet the height requirements for wind resource measurement in the retrofitted wind farm. This invention addresses these issues by proposing a wind resource assessment system that integrates AI technology. Based on wind farm construction data, historical operation data, real-time data, and expected data, the system corrects wind resource measurement data, achieving a more accurate assessment of turbine site retrofitting.

[0045] The specific implementation of the present invention will now be described in detail with reference to the accompanying drawings.

[0046] The wind resource assessment system for wind farm retrofitting provided by this invention is shown in Figure 1, and includes the following parts:

[0047] P100: Raw Data Acquisition Module: Used to acquire raw wind farm information, historical operating data of the raw wind farm, and measured wind data;

[0048] The raw data acquisition module acquires data through various methods, including manual collection of data on the operation and scheduling modes of various wind farms, the causes and scale of various geological disasters, topographic data of the wind farm interior and surrounding areas obtained through manual inspections and drone patrols, and operational data acquired by sensors of each unit during wind turbine operation.

[0049] The original wind farm information includes: wind farm environmental information, basic information of the original wind turbine units within the wind farm, and original meteorological tower information;

[0050] Specifically, wind farm environmental information includes: wind farm topographic information; this needs to be obtained through the original wind farm design drawings and topographic maps; the design drawings contain basic information about the original wind farm turbine units, such as the foundation design type of the turbine, the design elevation of the turbine foundation, the site leveling area, the tower foundation and tower height of the overhead power collection line, as well as the relevant information of the buried cable, the buried cable of the power collection line, and the wind farm turbine units of the surrounding existing and under-construction wind farms;

[0051] The basic information of the original wind turbine units in the wind farm consists of the wind turbine information β that makes up the unit. jThe array consists of (j = 1, ..., n), where n is the number of wind turbines. Wind turbine information includes: wind turbine center point coordinates, hub height, foundation elevation, model, manufacturer-provided theoretical power curve, impeller diameter, hub height, and swept area. The wind turbine center point coordinates are generated by creating a digital elevation model from a topographic map and overlaying it with the foundation elevation. The swept area is calculated from the impeller diameter.

[0052] Original meteorological tower information includes: original meteorological tower location. The coordinates of the positions of the original meteorological tower (x = 1, …, k), the height of the original meteorological tower, the height of the original meteorological tower sensor's wind measurement layer, and the measured wind data obtained from the original meteorological tower: wind direction, wind speed, temperature, and air pressure data; where k is the number of original meteorological tower positions.

[0053] The original historical operating data of the wind farm includes: wind speed, wind direction, temperature and air pressure data during the operation of each wind turbine over the years, as well as operation information during operation, including: SCADA (Supervisory Control and Data Acquisition) data in 10-minute increments, fault logs, wind speed cut-out time, actual operating power curves and other data, which can determine whether cut-out wind speeds have occurred within the wind farm area and the downtime of the units.

[0054] Actual wind measurement data can be obtained from the original wind measurement tower before the wind turbine is built or from the wind turbine after it is built. This includes information such as wind speed, wind direction, temperature, and air pressure. If a power prediction tower is built concurrently with the wind farm after its construction... (y = 0, ..., r), acquire data from the power prediction tower (compared with the original wind measurement tower). The information is consistent; as is customary, the original wind measurement towers built before the construction of older wind farms... The original wind measurement towers can be left undemolished and operated concurrently with the wind turbines in later stages. Therefore, the original wind measurement towers in older wind farms can be reused after construction. This is what we call a power prediction tower. , and The difference lies in the time scale of the data acquired; in this case, the measured wind data is obtained through a power prediction tower. (y = 0, …, r) is obtained; where r is the number of power prediction towers.

[0055] The raw data acquisition module processes the wind measurement data from the original wind measurement tower, the wind measurement data from each wind turbine, and the data provided by the power prediction tower, and performs data unification and validity processing, deleting and replacing outliers and improving missing values ​​to generate a systematic and complete wind resource data.

[0056] P110: Wind Farm Information Simulation Module: Used to construct a multi-dimensional wind farm model M and a prospective wind farm model M3 based on complete wind resource data;

[0057] The multidimensional wind farm model M includes: P111 historical wind farm model Mq, P111 theoretical wind farm model M1, P112 actual wind farm model M2; and P113 expected wind farm model M3, which is generated by constructing the multidimensional wind farm model M.

[0058] The wind farm information simulation module performs three-dimensional data modeling of the ground and air features of the wind farm in the wind farm environmental information; and based on the topographic data of the entire wind farm and its surroundings, it also incorporates various meteorological data of the entire wind farm into the model to obtain the spatial meteorological factor distribution at different heights of the wind farm.

[0059] Specifically, the wind farm information simulation module first processes the complete wind resource data generated by the raw data acquisition module to perform wind farm resource dynamics simulation calculations and obtain relevant wind parameters, including: wind speed distribution, wind direction frequency distribution, wake loss between each wind turbine, air density at hub height, energy density, wind shear index, and wind farm turbulence intensity.

[0060] Wind parameters are categorized, and a multi-dimensional wind farm model M is established during the categorization process. The categorization results include the original set of parameters from the wind measurement tower and the set of parameters for each wind turbine (including data from the power prediction tower). Using these two sets of parameters as reference points, a theoretical wind farm model M1 and an actual wind farm model M2 are generated.

[0061] Among them, the theoretical wind farm model M1 is a theoretical wind farm model simulated and constructed based on the original set of meteorological tower parameters; the actual wind farm model M2 is a wind farm model constructed based on the actual values ​​generated during the operation of each wind turbine, that is, the set of parameters of each wind turbine (including the power prediction tower).

[0062] The establishment of a multidimensional wind farm model M during the classification process also includes the following cases:

[0063] 1) Wind measuring tower These are the types of wind measurement towers that were built before the wind farm was built and were dismantled after the wind farm was built. Time scale of wind measurement data and power prediction tower , fan β j Inconsistent, therefore this type of wind measurement tower is ignored. Wind measurement data.

[0064] 2) After the wind turbine is built, the wind measurement tower Power prediction tower not dismantled For the same tower, three parties (originally the wind measurement tower) , fan β jand power prediction tower With consistent time scales, the wind measurement data from the three parties can be used to replace data from the same or different towers using correlation analysis; for wind measurement towers... The wind measurement data from continuous measurements at the same location are split into the following parts: Part 1: Wind measurement towers before wind farm construction The first part uses wind measurement data to construct the historical wind farm model Mq; the second part uses wind measurement data from the period after the wind farm was built and during the operation of the wind turbines to construct the theoretical wind farm model M1.

[0065] Same-tower replacement and different-tower replacement are methods to improve the data of devices with insufficient data integrity. In particular, same-tower replacement has a higher priority than different-tower replacement.

[0066] Among them, the same-tower replacement is mainly for wind measurement data of wind measurement towers with multiple wind speed layers. It is necessary to obtain the functional relationship between the data of each wind speed layer, and use the functional relationship to perform interpolation, replacement and extension of missing values ​​and outliers.

[0067] Among them, the method of replacing data with data from different towers is a data improvement method when the data integrity is still insufficient after replacing data from the same tower. For example, if the sub-units β1 and β2 only have the top wind speed and there is no possibility of replacing data from the same tower, a more reliable functional relationship between β1 and β2 can be established to repair and improve the data to meet the requirements. Alternatively, if all the data of each wind speed of the wind measuring tower is judged as invalid data for a certain period of time, it is also necessary to use a set of data from other towers with more complete data. The calculation principle and method are the same as described above.

[0068] 3) When carrying out the "replacement of large with small" transformation, since the hub height and performance of the replaced wind turbine equipment are better than the existing wind turbine, the expected wind farm model M3 is constructed based on the parameters of the replaced wind turbine equipment, the theoretical wind farm model Mq, the theoretical wind farm model M1 and the actual wind farm model M2.

[0069] First, prepare the forecast data by setting up a virtual wind measurement tower data source: typically, the original wind measurement tower... , fan β j and power prediction tower The maximum wind measurement height does not meet the requirements of the new model, and there are also cases of missing and abnormal data (even after the aforementioned same-tower replacement and different-tower replacement, there is still a possibility that the data will not meet the requirements); therefore, in β j as well as A virtual wind measurement tower γ is set up at the coordinate point. t(t = 1, …, k+r+n), the virtual wind measurement tower data source can be, but is not limited to, commonly used databases such as MERRA2 and ERA5 for assignment;

[0070] Secondly, based on the virtual meteorological tower data source, predictive wind resource data is constructed: for wind farms with multiple meteorological towers... For a wind farm with x = 1, …, k, calculate the functional relationships between the various meteorological layers of the same meteorological tower (x = 1, …, k), the functional relationships between any two different meteorological towers (when x = 1, only the functional relationships between layers of the same tower need to be calculated), and the relationship between the meteorological tower and its corresponding virtual meteorological tower γ. t The functional relationships between different wind turbines and between the wind measurement data of different wind turbines were analyzed, and the correlation was determined according to the magnitude of the correlation, prioritizing relationships between different layers of the same wind turbine, different wind turbines, and wind measurement towers and virtual wind measurement towers. t The order in which these parameters are used to construct the predicted wind resource data.

[0071] By using predicted wind resource data and a multidimensional wind farm model M, a projected wind farm model M3 (P113) is constructed. During the construction process, the wind resource parameters of the theoretical wind farm model M1 and the actual wind farm model M2 are compared to verify the deviation between the actual and theoretical values. Based on the deviation, a deviation diagnosis is performed, and the reasons for the deviation between the theoretical and actual values ​​are analyzed. Finally, the projected wind farm model M3 is constructed by combining the predicted wind resource data calculated by the virtual wind measurement tower with the deviation.

[0072] This includes: theoretical wind farm model M1 and historical wind farm model M q A comparison was made to obtain the difference between actual and theoretical wind parameters; based on the actual values, power prediction was performed on the wind farm model M2 for towers with many years of operating data. The data is combined with the actual operating power curves of the wind turbines over many years to construct a reference wind farm model M′2; by comparing and analyzing models M1, M2, M′2 and Mq, the wind farm model deviation is diagnosed and corrected according to the model parameters of the wind turbine to be replaced, and the expected wind farm model M3 is generated.

[0073] The wind resources of the wind farm after the "large-to-small" wind turbine renovation is calculated using the expected wind farm model M3 to obtain the theoretical power generation revenue.

[0074] P120: Evaluation index determination module: used to calculate the index value of wind resource index based on the factors to be considered in the large-to-small replacement renovation, including positive and negative correlation factors;

[0075] When calculating the expected wind resource indicators, the wind resource calculation module specifies the β value for each wind turbine. j Modification items, calculate β for each of the aforementioned wind turbines. jThe modification indicators corresponding to the modification items; the modification indicators include: power indicators and cost indicators.

[0076] Specifically, the upgrade indicators are factors that affect the price level of wind turbine upgrades, including: the expected power generation revenue of the wind turbine, the cost of upgrading and expanding the access road to the wind turbine, the cost of upgrading and expanding the wind turbine site, and the cost of upgrading the wind turbine collection line, etc.

[0077] P130: Modification Assessment Module: Used to calculate priority based on the evaluation value.

[0078] The calculation of priority includes the following steps:

[0079] 1) Assume there are n in the wind farm j Typhoon generators, and each generator has m i The evaluation indicators include: the need to use the new model M3 to obtain wind resource parameters (wind speed, wind direction, wind power density, turbulence intensity, etc.) to calculate the theoretical power generation and thus obtain the expected power generation revenue of the wind turbine; the cost of upgrading and expanding the access road to the j-th wind turbine in the original wind farm; the cost of upgrading and expanding the site of the j-th wind turbine; and the cost of upgrading the collection line of the j-th wind turbine. Among them, i is the evaluation indicator number, m is the number of evaluation indicators, j is the wind turbine number that can be upgraded, and n is the maximum number of wind turbines that can be upgraded. The maximum number of wind turbines that can be upgraded, n = P / Z, where P represents the maximum construction scale of the wind farm approved by the relevant department, Z represents the maximum capacity of the single wind turbine expected to be upgraded, and n is less than or equal to the number of wind turbines in the original old wind farm.

[0080] 2) Evaluate the weight of the i-th evaluation index of the j-th wind turbine in the wind farm and generate the evaluation value of each wind turbine;

[0081] First, the original data matrix R′=(r′) is constructed using the entropy weight method. ij ) m×n The m evaluation indicators of the j-th wind turbine are normalized one by one, where r′ ij This is the evaluation value of the i-th evaluation indicator for the j-th wind turbine, after normalization. Represented as:

[0082] ,

[0083] Next: Calculate the entropy value of the i-th evaluation index for the j-th wind turbine. ,

[0084] ,

[0085] ,in, The numerical weight of the i-th indicator for the j-th wind turbine;

[0086] , assuming hour, ;

[0087] 3) Calculate the weighted ratio for each wind turbine. Determine the priority of wind turbine retrofitting:

[0088] In this invention, the weights are divided into two types: the first weight δ1 indicates that the evaluation index is positively correlated with the priority of wind power retrofitting; the second weight δ2 indicates that the evaluation index is negatively correlated with the priority of wind power retrofitting.

[0089] The weight of the i-th evaluation index is expressed as: .

[0090] Calculate the weighted ratio for each wind turbine. , ;

[0091] Where s is the number of indicators that are positively correlated with the priority of wind power transformation, and g is the number of indicators that are positively correlated with the priority of wind power transformation; and s + g = m, s = 1, …, m, g = i = 1, …, m.

[0092] in For real numbers greater than 0, when 0 < When <1, it indicates that the wind turbine has no prospect of being upgraded. A value greater than 1 indicates that the wind turbine has potential for retrofitting.

[0093] By weighted ratio of each wind turbine It can be determined that there are ε (0 ≤ ε ≤ n) wind turbines in the wind farm that do not currently have the prospect of being upgraded, while there are n-ε (0 ≤ ε ≤ n) wind turbines that do have the prospect of being upgraded.

[0094] The n-ε (0 ≤ ε ≤ n) typhoons Sort the wind turbines in the wind farm from largest to smallest to determine their upgrade priority.

[0095] P140 Risk Assessment Module: This invention also provides a risk assessment module that can perform risk assessments on a specified wind turbine unit and provide decision-making solutions based on environmental risks and management and maintenance difficulties.

[0096] Specifically, the risk assessment module includes a 3D modeling unit, a data processing unit, and an analysis unit:

[0097] 1) 3D Modeling Unit: Obtains raw wind farm information from the raw data acquisition module for 3D data modeling, including: ground and air features of the wind farm; during 3D data modeling, acquires topographic data of the entire wind farm interior and surrounding areas, and incorporates various meteorological data of the entire wind farm into the model to obtain the spatial meteorological factor distribution at different heights of the wind farm; after the 3D model is built, environmental data under various conditions can be acquired, including: different flood frequencies and severity, altitude, slope, topography determined by surface vegetation, return period of maximum wind speed, etc.

[0098] 2) Data processing unit: The data processing unit obtains environmental data from the 3D modeling unit, loads the pre-trained environmental risk prediction model from a third party, and after inputting the environmental data, it can obtain the risk prediction results of each wind turbine under different environments.

[0099] 3) Analysis Unit: Based on the regional spatial characteristics in the original wind farm information, and combined with the risk prediction results, assess the impact of uncertainty factors in the wind farm operation process on different wind turbine operation scenarios. In particular, the model should conduct wind farm uncertainty analysis based on the corresponding changes in external conditions and the different scenarios set in the early stage, and determine whether geological environmental factors in the wind farm will bring risks to the operation of wind turbines.

[0100] The risk assessment module in this invention can also be connected to the control system of an operating wind farm, enabling real-time management while conducting risk assessments: upon discovering a destructive geological environment, it determines that geological factors within the wind farm will pose a risk to wind turbine operation and generates a real-time scheduling plan. This function is crucial for enabling emergency response in case of emergencies during wind farm operation, effectively avoiding the negative impacts of human factors.

[0101] For example, if the model calculations detect that the wind speed in and around the wind farm may temporarily exceed the design safety wind speed, the model can promptly prompt for remote shutdown. The model can also adjust the operation plan of each wind turbine unit in a timely manner based on the wind turbine layout, wind speed, and other wind parameter distribution factors in the wind farm, effectively reducing the impact of wake and other interference factors on the wind farm's power generation.

[0102] Furthermore, the risk prediction results and uncertainty analysis will be used to assess the risks of wind farms under various scenarios and make relevant decisions; the assessment results will be uploaded to the visualization window of the management terminal.

[0103] In response to the current lack of standardized practices and experience in the technological upgrading of aging wind farms in China, this invention combines wind resource assessment methods to rationally segment the data of various equipment in operating wind farms, introduces various influencing factors to conduct weighted evaluations of each wind turbine, and obtains the priority of "replacing large turbines with smaller ones" for the technological upgrading of aging wind farm locations; it also introduces an ecological environment AI model to predict various risk factors during and after the wind farm's upgrading process, thereby reducing labor costs, lowering the difficulty of wind farm management, and improving the effectiveness and scientific nature of wind farm upgrading and management.

[0104] The above-disclosed embodiments are merely a few specific examples of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A wind resource assessment system for wind farm retrofitting, characterized in that, include: The raw data acquisition module is used to acquire raw wind farm information, historical operational data, and measured wind data. It performs data unification and validity processing, deletes and replaces outliers, and corrects missing values ​​to generate complete wind resource data. Raw wind farm information includes: wind farm environmental information, basic information on wind turbine units within the wind farm, and information on the original wind measurement tower. The basic information on wind turbine units includes the foundation design type, foundation elevation, site leveling area, overhead power line tower foundations, tower height, and information on buried cables, as well as information on existing and under-construction wind turbine units in the surrounding area. The wind farm information simulation module is used to construct a multi-dimensional wind farm model M and a projected wind power model based on the raw wind farm information. The wind farm model M3 includes: historical wind farm model Mq, theoretical wind farm model M1, and actual wind farm model M2; the expected wind farm model M3 is generated from the multi-dimensional wind farm model M; the evaluation index determination module is used to calculate the index values ​​of wind resource indicators based on the factors to be considered in the large-to-small replacement retrofit, including power indicators and cost indicators; the retrofit evaluation module is used to calculate priorities based on the index values; the risk assessment module is used to conduct risk assessments on specified wind turbine units and provide decision-making solutions based on environmental risks and management and maintenance difficulties; the construction of the expected wind farm model M3 includes: establishing a virtual meteorological tower data source; the virtual meteorological tower is located at the intersection of the original meteorological tower, wind turbine, and power prediction tower. At the coordinate point; by referencing a virtual meteorological tower, the wind measurement data after the upgrade is predicted, simulating the operating environment of the upgraded wind farm, and predicting the power generation after the upgrade; based on the virtual meteorological tower data source, predicted wind resource data is constructed, including: for wind farms with multiple meteorological towers, calculating the functional relationships between different meteorological layers of the same tower, the functional relationships between different towers, the functional relationships between a meteorological tower and its corresponding virtual meteorological tower, and the functional relationships between the meteorological data of each wind turbine; and constructing predicted wind resource data according to the correlation magnitude in the following order: first, different layers of the same tower; then, between different wind turbines; and finally, between a meteorological tower and a virtual meteorological tower; comparing the wind resource parameters of the theoretical wind farm model M1 and the actual wind farm model M2. The process involves several steps: first, verifying the deviation between actual and theoretical values; second, diagnosing the deviation based on the deviation; and third, constructing a projected wind farm model M3 by combining predicted wind resource data calculated from a virtual wind measurement tower, the set of parameters from the virtual wind measurement tower, and the deviation. The theoretical wind farm model M1 and the historical wind farm model Mq are compared to obtain the differences between actual and theoretical wind parameters. A reference wind farm model M′2 is constructed by combining data from a power prediction tower with years of operational data from the actual wind farm model M2 with the actual operating power curves of the wind turbines over many years. Finally, by comparing and analyzing models M1, M2, M′2, and Mq, and based on the model parameters of the wind turbine to be replaced, the wind farm model deviation is diagnosed and corrected to generate the projected wind farm model M3.

2. The wind resource assessment system according to claim 1, characterized in that, The wind farm environmental information includes: wind farm topographic information, obtained through the original wind farm design drawings and topographic maps; wherein, the design drawings contain the basic information of the original wind farm turbine units; the basic information of the original wind farm turbine units consists of the turbine information that makes up the unit; the original meteorological tower information includes: the original meteorological tower location coordinates, the original meteorological tower height, the original meteorological tower sensor wind measurement layer height, and the measured wind data obtained from the original meteorological tower: wind direction, wind speed, temperature, and air pressure data; the historical operation data includes wind speed, wind direction, temperature, and air pressure data during the operation of each wind turbine over the years, as well as operation information during the operation period; the measured wind data is obtained from the original meteorological tower before the wind farm construction and from the wind turbines or power prediction towers after the construction.

3. The wind resource assessment system according to claim 2, characterized in that, When the wind farm information simulation module constructs a multi-dimensional wind farm model M, it performs three-dimensional data modeling, obtains relevant wind parameters, classifies the wind parameters, establishes the multi-dimensional wind farm model M during the classification process, and generates a historical wind farm model Mq, a theoretical wind farm model M1, and an actual wind farm model M2.

4. The wind resource assessment system according to claim 3, characterized in that, The classification of the wind parameters includes the original set of meteorological tower parameters and the set of parameters for each wind turbine. The theoretical wind farm model M1 is a theoretical wind farm model constructed based on the original set of meteorological tower parameters. The actual wind farm model M2 is a wind farm model constructed based on the set of parameters for each wind turbine.

5. The wind resource assessment system according to claim 3, characterized in that, The process of establishing a multi-dimensional wind farm model M during classification includes: If the wind measurement tower was built before the wind farm was built and then dismantled after the wind farm was built, the wind measurement data of the wind measurement tower is ignored; if the wind measurement tower was built before the wind farm was built, and the wind measurement tower was not dismantled after the wind turbine was built, and the power prediction tower is the same tower, then the wind measurement data of the original wind measurement tower, wind turbine, and power prediction tower are used to determine the same tower and different tower replacements based on correlation; the wind measurement data of continuous wind measurements taken by wind measurement towers at the same location are split, including: wind measurement data of wind measurement towers before the wind farm was built are used to construct a historical wind farm model Mq; wind measurement data from the wind farm after its construction and during the operation of the wind turbine are used to construct a theoretical wind farm model M1.

6. The wind resource assessment system according to claim 1, characterized in that, The risk assessment module includes a 3D modeling unit, a data processing unit, and an analysis unit. The 3D modeling unit acquires raw wind farm information from the raw data acquisition module to create a 3D data model. After the 3D model is built, environmental data under various conditions is acquired, including: different flood frequencies and severity, altitude, slope, topography determined by surface vegetation, and the return period of maximum wind speed. The data processing unit acquires environmental data from the 3D modeling unit, loads a pre-trained third-party environmental risk prediction model, and obtains risk prediction results for each wind turbine under different environments. The analysis unit combines the risk prediction results to assess the impact of uncertainties in wind farm operation on different wind turbine operating scenarios, determining whether geological environmental factors within the wind farm will pose risks to wind turbine operation.

Images