A dynamic clustering equivalent modeling method for wind turbines considering unit uncertainty
By taking into account the uncertainty of wind turbines, the problem of insufficient modeling accuracy of wind turbines and insufficient wind energy utilization level analysis in the prior art is solved, and higher-precision equal-value modeling of wind turbines and more accurate wind energy utilization analysis is achieved.
Patent Information
- Application Number
- CN202411771466.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-04
- Publication Date
- 2025-04-01
AI Technical Summary
The prior art ignores the interaction between units in the dynamic equivalent modeling of wind turbines, resulting in insufficient modeling accuracy, unable to effectively reflect the interaction between wind turbines and the power grid, and in-depth analysis of the dynamic characteristics and wind energy utilization level of wind turbines.
A method of dynamic group group group group group group group group group group uncertainty is proposed. By extracting output data and meteorological data of wind turbine group, normalizing processing and Pearson correlation analysis, determining the best correlation parameters, establishing a dynamic grouping index set, and using the hierarchical clustering algorithm with the minimum distance for grouping, finally establishing a high-precision equivalent model.
The equivalent modeling accuracy of wind turbines is improved, and the dynamic characteristics and wind energy utilization level of wind turbines can be more accurately reflected, which enhances the safe and stable operation of wind turbines in the power grid.
Smart Images
Figure CN119740468A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of renewable energy modeling and simulation, and in particular to the field of a dynamic grouping and equivalent modeling method for wind turbines. Background Art
[0002] With the rapid development of wind power, the number of wind farms has increased rapidly. Compared with traditional energy sources such as thermal power, the large increase in wind turbines has a huge impact on the safe and stable operation of the power grid. Due to the large scale of the power grid, the complex structure of the power grid, and the diverse access methods, the stability analysis of the wind power access system has become more difficult. In order to further improve the safety and stability of wind power access to the power grid, it is necessary to conduct in-depth theoretical research on its operating characteristics. Wind turbine modeling is the prerequisite for realizing the analysis of its safe and stable operation characteristics in the power grid.
[0003] A reasonable and efficient wind turbine model is the basis for the stable operation analysis of the wind power generation system. If each component and each wind turbine in the wind power generation system is to be accurately modeled, a more accurate wind power generation system needs to be built. However, this will lead to problems such as large modeling volume and high modeling dimension, and it cannot be used for deeper theoretical research, and it will also reduce the calculation speed in the time domain.
[0004] In the existing analysis of wind turbine grid-connected stability, the interaction between units is often ignored, and the units are often equivalent to single or multiple units. On this basis, the stable operation of wind turbines is modeled. The traditional single-unit equivalent method is to equate the wind turbine to a single unit, which cannot reflect the interaction between the wind turbine and the power grid. In terms of dynamic analysis, the requirements for wind turbine models are higher and faster. Therefore, studying a high-precision dynamic equivalent model of wind turbine units is the key to ensuring the stable and reliable operation of wind turbine systems.
[0005] At present, in view of the randomness and uncertainty of wind energy, a dynamic clustering equivalent modeling method for wind turbines considering the uncertainty of wind turbines is proposed. The actual operating conditions and actual operating status of wind turbines are considered, and the optimal relevant parameters related to the maximum output of wind turbines are analyzed. A dynamic clustering method for wind turbines considering the uncertainty of wind turbines and suitable for multiple scenarios is established.
[0006] However, due to the large differences in the ideal output of wind turbines under different working conditions, an ideal output index is constructed based on the output data of wind turbines in continuous time periods to reflect their operating status. In addition, the inability of the power grid to fully absorb the power generated by wind turbines has led to the widespread phenomenon of "wind abandonment". In the dynamic equivalent modeling of wind turbines, it is necessary to pay attention to the significant differences in wind energy utilization levels caused by "wind abandonment". In order to further characterize the differences in wind energy utilization levels caused by wind abandonment, a wind turbine continuous output level index is constructed to characterize the continuity of wind turbine output levels in continuous time windows. Although the two indicators of ideal output and continuous output level can provide a basis for preliminary judgment of the output status of wind turbines, in actual applications, the evaluation based on measured data will be interfered by extreme weather or abnormal conditions, and even cover up the performance degradation of wind turbines caused by failures. Therefore, based on the two existing judgment indicators of ideal output and continuous output level, high and low output capacity indicators are further defined to more accurately reflect the dynamic performance of wind turbines under different system states, and to judge the high and low output capacity of wind turbines by the maximum value of the difference between the actual output value and the ideal value of the wind turbine at a certain moment. Summary of the invention
[0007] The purpose of the present invention application is to establish a set of unit grouping indicators including ideal output, continuous output level, unit state output capacity and optimal related parameters, to carry out dynamic grouping of wind turbines, and to establish an equivalent model of wind turbines based on the grouping results. The established equivalent model can further improve the equivalent modeling accuracy of wind turbines.
[0008] In order to achieve the above invention objectives, the present invention adopts the following technical solutions:
[0009] A method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines is provided. The method dynamically groups wind turbines and performs equivalent modeling according to the uncertainty of wind turbines. The specific steps are as follows:
[0010] S1. Extract the data of wind turbines in different wind farms in the region to characterize the output and uncertainty of wind turbines. The extracted data include but are not limited to: the nameplate data of wind turbines, such as rated voltage, rated current, rated power, and rated capacity; real-time data such as voltage, current, active power, and reactive power during the actual operation of wind turbines; meteorological data such as temperature, humidity, wind speed, and irradiance of the environment where the wind turbines are located; and the longitude and latitude that can reflect the location of the wind turbines;
[0011] S2. Normalize all extracted parameters to reduce the differences in the order of magnitude of each parameter. Normalize the extracted feature parameters. The normalization process is as follows:
[0012]
[0013] In the above formula represents the normalized value of the parameter, x represents the measured value of the extracted parameter, and x min represents the minimum value of the extracted parameter of this type, x max Indicates the measured maximum value of the extracted parameter of this type.
[0014] S3. Use the Pearson correlation coefficient to analyze the correlation between various factors. The Pearson correlation coefficient calculation formula is as follows:
[0015]
[0016] Where, X m represents the mth data of various parameters extracted, and P represents the historical output of the corresponding wind turbine; The parameter that is most correlated with the output is defined as the best correlation parameter, and is used as a clustering index set together with the wind turbine output index defined later. According to the above calculation structure, the best correlation parameter is selected and determined as one of the clustering index set parameters;
[0017] S4. Establish a dynamic grouping index set considering the uncertainty of wind turbines; the index set mainly includes: ideal output index, continuous output level index, unit state output capacity index and optimal related parameters;
[0018] S5. The wind turbine grouping is determined by the minimum distance hierarchical clustering algorithm, and the morphological similarity distance of Manhattan distance is used as the calculation basis of the clustering distance of wind turbines:
[0019] S6. According to the dynamic grouping results of wind turbines, equivalent wind turbines are used to replace the wind turbines in the group, and the equivalent parameters of each link of the equivalent wind turbines are determined according to the Pearson correlation analysis;
[0020] The equivalent parameters of each link of the equivalent wind turbine generator set are calculated based on the Pearson correlation weight, and the Pearson correlation coefficient is determined as the Pearson correlation weight of the equivalent unit.
[0021] Furthermore, the ideal output index is based on the output data of the wind turbine in a continuous period of time to evaluate the operating status of the wind turbine; based on the operating conditions of the wind turbine in 12 consecutive sliding time windows, the difference between the actual output of the wind turbine and the ideal output of the wind turbine is evaluated. The ideal output represents the ideal output of the wind turbine. The ideal output of wind turbines in different working conditions varies greatly. The calculation formula is as follows:
[0022]
[0023] In the formula, ε t represents the ideal output of the wind turbine in the six sliding time windows at and after time t, P t act and are the actual values of active power and reactive power of wind turbines at time t; P t id and are respectively the active power and reactive power of the wind turbine under ideal conditions at time t.
[0024] Furthermore, the continuous output level index is constructed to characterize the difference in wind energy utilization level caused by wind abandonment, and is used to characterize the continuity of the wind turbine output level of the wind turbine within a continuous time window.
[0025] The continuous output level of the wind turbine generator set is calculated using the following formula:
[0026]
[0027] In the formula, γ t Indicates the continuous output level of wind turbines, C P It represents the wind energy utilization coefficient of the wind turbine, P t act represents the actual output level of the wind turbine at time t, P t id It represents the ideal maximum output of the wind turbine at time t;
[0028] The continuous output level γ t The larger the value, the stronger the output continuity of the wind turbine, the higher the wind energy utilization level of the wind turbine, and the less wind abandonment phenomenon of the wind turbine.
[0029] Furthermore, the wind abandonment phenomenon is caused by the high uncertainty and randomness of wind power generation, which results in the inability of the power grid to fully absorb the power generated by wind turbines, resulting in the "wind abandonment" phenomenon; in view of the universality of the "wind abandonment" phenomenon, in the dynamic equivalent modeling of wind turbines, it is necessary to pay attention to the significant differences in wind energy utilization levels caused by "wind abandonment";
[0030] Furthermore, the high and low output capacity index is a measure of the unit's state output capacity, and is a further defined index, in order to more comprehensively evaluate the dynamic response capability of the wind turbine unit, more accurately reflect the dynamic performance capability of the wind turbine unit under different system states, and solve the problem that the evaluation of measured data based on the existing two judgment indicators of ideal output and continuous output level will be disturbed by extreme weather or abnormal conditions, and even cover up the performance degradation of the wind turbine unit due to faults;
[0031] The calculation formula of the wind turbine output capacity at time t is as follows:
[0032]
[0033] In the formula, ψ t Indicates the output capacity of the wind turbine at time t. They represent the actual active power output and actual reactive power output of the wind turbine at time t+1, It represents the ideal active power output and ideal reactive power output of the wind turbine at time t+i.
[0034] Furthermore, the wind turbine group dynamic grouping index set is constructed based on the ideal output, continuous output level, unit state output capability and optimal related parameters. The wind turbine group dynamic grouping index set adopts the formula Representation.
[0035] Furthermore, the Manhattan distance is calculated using the following formula:
[0036]
[0037] In the formula, the Manhattan distance expression between the index parameters of any two units i and j is: dis(Y i ,Y j ), where Y ik ,Y jk are the index parameters of any two wind turbines;
[0038] The morphological similarity distance expression for calculating the Manhattan distance between the multidimensional parameters is:
[0039]
[0040] In the formula, y i and j are the elements in the wind turbine dynamic clustering index set, where y ik ,y jk It is the corresponding clustering index parameter of the two wind turbines in the morphological similarity distance expression, which represents the morphological similarity distance between any two wind turbines calculated based on the Manhattan distance.
[0041] Furthermore, the shortest distance of morphological similarity is the basis for classifying wind turbines. The morphological similarity distance is used as the calculation basis of the hierarchical clustering algorithm. The shortest distance of morphological similarity is expressed as:
[0042] D ij =mind(y i ,y j )
[0043] Initialize and set each wind turbine group as an independent wind turbine group cluster, calculate the shortest distance between any two wind turbine group clusters with similar shapes, find the two wind turbine group clusters with the closest distance, and merge these two wind turbine group clusters into a new wind turbine group.
[0044] By cyclically calculating the shortest distance between wind turbines with similar shapes, the wind turbine grouping results under any number of wind turbine groupings can be obtained. The number of wind turbine groupings can be determined according to the specific number of actual calculation examples.
[0045] Furthermore, the equivalent parameters of each link of the equivalent wind turbine set are calculated based on the Pearson correlation weight, and the Pearson correlation coefficient is determined as the Pearson correlation weight of the equivalent set:
[0046] Among them, the Pearson correlation weight of each wind turbine is used as the wind speed weight in the wind speed equalization process, and the equivalent wind speed model is expressed as:
[0047]
[0048] Where: is the calculation result of wind speed correlation of each wind turbine; v i represents the wind speed of each wind turbine in the wind power cluster replaced by the equivalent unit; v eq is the alternative wind speed of the equivalent wind turbine.
[0049] Furthermore, the equivalent electrical parameters of each link of the equivalent wind turbine are determined, and the Pearson coefficient is used as the weight value of the modeling of each unit in the grouped equivalent module unit, as shown in the following formula:
[0050]
[0051] In the formula, S eq , P eq and Q eq are the equivalent capacity, equivalent active power and equivalent reactive power of equivalent wind turbine generator sets respectively, and are the sum of the corresponding parameters of wind turbine generator sets contained in each equivalent wind turbine generator set; represents the Pearson correlation parameter value of the unit, which is used as the equal weight parameter within the equal unit cluster; H eq Denotes the equivalent axis inertia time constant, K eq Denotes the equivalent shaft stiffness coefficient, D eq Represents the equivalent shaft damping coefficient.
[0052] How it works
[0053] The invention provides a method for dynamic clustering equivalent modeling of wind turbines considering the uncertainty of wind turbines. The method dynamically clusters wind turbines according to the uncertainty of wind turbines and carries out equivalent modeling, extracts output data, historical meteorological data, external characteristic data such as voltage and current of wind turbines in different wind farms in nearby areas, performs normalization processing, and adopts Pearson correlation analysis to analyze the correlation between various parameters and select the best relevant parameters; according to the result of Pearson correlation analysis, a dynamic clustering index set considering the uncertainty of wind turbines is established, and the index set mainly includes: ideal output, continuous output level, unit state output capacity and best relevant parameters; a multi-dimensional parameter morphological similarity distance considering Manhattan distance is used as the clustering distance algorithm basis of wind turbines, and a hierarchical clustering algorithm with minimum distance is adopted to determine the clustering results of wind turbines; an equivalent principle and a calculation method of equivalent parameters in the process of dynamic equivalent modeling of wind power generation system are proposed, and an equivalent model of wind turbines is established for wind turbines with obtained clustering results according to the equivalent method of weight parameters determined by Pearson correlation analysis.
[0054] Beneficial Effects
[0055] The present invention proposes a method for dynamic clustering equivalent modeling of wind turbines considering the operating status of wind turbines, which fully considers the parameters of each link in the actual operating status of the wind turbines and the operating status of the units, determines the optimal correlation parameters through Pearson correlation analysis, establishes a set of dynamic clustering indicators of wind turbines considering the operating status of the wind turbines, defines and calculates the morphological similarity distance based on Manhattan distance as the clustering distance of the wind turbines, and adopts a hierarchical clustering algorithm with a minimum distance to determine the clustering results of the wind turbines; according to the Pearson correlation coefficient, an equivalent parameter calculation method for the Pearson correlation weight in the process of dynamic equivalent modeling of the wind power generation system is established, the wind turbines are clustered according to the clustering results and an equivalent model is established, which can maintain the consistency of the dynamic characteristics of the wind turbines and the accuracy of the dynamic response of the wind turbines to the greatest extent on the basis of ensuring the equivalence of the wind turbine modeling. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 This is a schematic diagram of wind turbine grouping;
[0057] Figure 2 This is a flow chart of the equivalent modeling method for dynamic grouping of wind turbines considering unit uncertainty;
[0058] Figure 3 This is a diagram of the idea of equivalent modeling of wind turbines based on uncertainty characterization. DETAILED DESCRIPTION
[0059] In order to make those skilled in the art better understand the scheme of the present invention, the following will be combined with the attached embodiment of the present invention. Figure 1-2, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0060] Example 1 is as shown in the attached Figure 1 In the process shown:
[0061] S1. Extract the data of wind turbines from different wind farms in the region to characterize the output and uncertainty of wind turbines.
[0062] The extracted data include but are not limited to: nameplate data such as rated voltage, rated current, rated power, and rated capacity of the wind turbine; real-time data such as voltage, current, active power, and reactive power during the actual operation of the wind turbine; meteorological data such as temperature, humidity, wind speed, and irradiance of the environment where the wind turbine is located; and longitude and latitude that can reflect the location of the wind turbine.
[0063] S2. Normalize all extracted parameters to reduce the differences in the order of magnitude of each parameter. Normalize the extracted feature parameters. The normalization process is as follows:
[0064]
[0065] In the above formula represents the normalized value of the parameter, x represents the measured value of the extracted parameter, and x min represents the minimum value of the extracted parameter of this type, x max Indicates the measured maximum value of the extracted parameter of this type.
[0066] S3. Use the Pearson correlation coefficient to analyze the correlation between various factors and determine the optimal output-related parameters.
[0067] S31. The calculation formula of Pearson correlation coefficient is as follows:
[0068]
[0069] Where, X m It represents the mth data among the extracted parameters, and P represents the historical output of the corresponding wind turbine.
[0070] S32. Select and determine the best correlation parameter through the calculation results, which is defined as It represents the parameter related to the maximum output, and together with the wind turbine output index defined later, it serves as the clustering index set.
[0071] S4. Establish a dynamic grouping index set that takes into account the uncertainty of wind turbines; the index set mainly includes: ideal output index, continuous output level index, unit state output capacity index and optimal related parameters.
[0072] S41. Further, the ideal output index is based on the output data of the wind turbine in a continuous period of time to evaluate the operating status of the wind turbine; based on the operating conditions of the wind turbine in 12 consecutive sliding time windows, the difference between the actual output of the wind turbine and the ideal output of the wind turbine is evaluated. The ideal output represents the ideal output of the wind turbine. The ideal output of wind turbines in different working conditions varies greatly. The calculation formula is as follows:
[0073]
[0074] In the formula, ε t represents the ideal output of the wind turbine in the six sliding time windows at and after time t, P t act and are the actual values of active power and reactive power of wind turbines at time t; P t id and are respectively the active power and reactive power of the wind turbine under ideal conditions at time t.
[0075] S42. In order to characterize the difference in wind energy utilization level caused by wind abandonment, a wind turbine continuous output level index is constructed to characterize the continuity of the wind turbine output level within a continuous time window.
[0076] Furthermore, the wind abandonment phenomenon is due to the high uncertainty and randomness of wind power generation, which causes the power grid to be unable to fully absorb the power generated by wind turbines, resulting in the "wind abandonment" phenomenon; in view of the universality of the "wind abandonment" phenomenon, in the dynamic equivalent modeling of wind turbines, it is necessary to pay attention to the significant differences in wind energy utilization levels caused by "wind abandonment";
[0077] The continuous output level of the wind turbine generator set is calculated using the following formula:
[0078]
[0079] In the formula, γ t Indicates the continuous output level of wind turbines, C P It represents the wind energy utilization coefficient of the wind turbine, P t act represents the actual output level of the wind turbine at time t, P t id It represents the ideal maximum output of the wind turbine at time t;
[0080] The continuous output level γ t The larger the value, the stronger the output continuity of the wind turbine, the higher the wind energy utilization level of the wind turbine, and the less wind abandonment phenomenon of the wind turbine.
[0081] S43. High and low output capacity index is a further defined index of the unit state output capacity measurement standard, in order to more comprehensively evaluate the dynamic response capability of the wind turbine unit, more accurately reflect the dynamic performance capability of the wind turbine unit under different system states, and solve the problem that the evaluation of measured data based on the existing two judgment indicators of ideal output and continuous output level will be disturbed by extreme weather or abnormal conditions, and even cover up the performance degradation of the wind turbine unit due to faults;
[0082] The calculation formula of the output capacity of the wind turbine at time t is as follows:
[0083]
[0084] In the formula, ψ t Indicates the output capacity of the wind turbine at time t. They represent the actual active power output and actual reactive power output of the wind turbine at time t+1, It represents the ideal active power output and ideal reactive power output of the wind turbine at time t+i.
[0085] S44, the index set is constructed based on the ideal output, continuous output level, unit state output capacity and optimal related parameters. The wind turbine dynamic grouping index set adopts the formula Representation.
[0086] S5. The wind turbine grouping is determined by the minimum distance hierarchical clustering algorithm, and the morphological similarity distance of Manhattan distance is used as the calculation basis of the clustering distance of wind turbines:
[0087] S51. Further, the Manhattan distance is calculated using the following formula:
[0088]
[0089] In the formula, the Manhattan distance expression between the index parameters of any two wind turbines i and j is: dis(Y i ,Y j ), where Y ik ,Y jk is the index parameter of any two wind turbines.
[0090] S52, the morphological similarity distance expression of the calculation considering the Manhattan distance between the multi-dimensional parameters is:
[0091]
[0092] In the formula, y i and j are the elements in the wind turbine dynamic clustering index set, where y ik ,y jk It is the corresponding clustering index parameter of the two wind turbines in the morphological similarity distance expression, which represents the morphological similarity distance between any two wind turbines calculated based on the Manhattan distance.
[0093] S53. Further, the morphological similarity shortest distance is the basis for classifying wind turbines. The morphological similarity distance is used as the calculation basis of the hierarchical clustering algorithm. The morphological similarity shortest distance is expressed as:
[0094] D ij =mind(y i ,y j )
[0095] S54, initialize and set each wind turbine group as an independent wind turbine group cluster, calculate the shortest distance between any two wind turbine group clusters with similar shapes, find out the two wind turbine group clusters with the closest distance, and merge these two wind turbine group clusters into a new wind turbine group.
[0096] By cyclically calculating the shortest distance between wind turbines with similar shapes, the wind turbine grouping results under any number of wind turbine groupings can be obtained. The number of wind turbine groupings can be determined according to the specific number of actual calculation examples.
[0097] S6. Based on the dynamic grouping results of wind turbines, equivalent units are used to replace the wind turbines in the group, and the equivalent parameters of each link of the equivalent wind turbine are determined based on the Pearson correlation analysis.
[0098] The equivalent parameters of each link of the equivalent wind turbine generator set are calculated based on the Pearson correlation weight, and the Pearson correlation coefficient is determined as the Pearson correlation weight of the equivalent unit.
[0099] S61. Further, the equivalent parameters of each link of the equivalent wind turbine set are calculated by Pearson correlation weight, and the Pearson correlation coefficient is determined as the Pearson correlation weight of the equivalent set:
[0100] Among them, the Pearson correlation weight of each wind turbine is used as the wind speed weight in the wind speed equalization process, and the equivalent wind speed model is expressed as:
[0101]
[0102] Where: is the calculation result of wind speed correlation of each wind turbine; vi represents the wind speed of each wind turbine in the wind power cluster replaced by the equivalent unit; v eq is the alternative wind speed of the equivalent wind turbine.
[0103] S62. Further, the equivalent electrical parameters of each link of the equivalent wind turbine generator set are determined, and the Pearson coefficient is used as the weight value of the modeling of each unit in the grouped equivalent module unit, as shown in the following formula:
[0104]
[0105] In the formula, S eq , P eq and Q eq are the equivalent capacity, equivalent active power and equivalent reactive power of equivalent wind turbine generator sets respectively, and are the sum of the corresponding parameters of wind turbine generator sets contained in each equivalent wind turbine generator set; represents the Pearson correlation parameter value of the unit, which is used as the equal weight parameter within the equal unit cluster; H eq Denotes the equivalent axis inertia time constant, K eq Denotes the equivalent shaft stiffness coefficient, D eq Represents the equivalent shaft damping coefficient.
Claims
1. A method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines. The method dynamically groups wind turbines and performs equivalent modeling according to the uncertainty of wind turbines, and is characterized in that: The specific implementation steps are as follows: S1. Extract the data of wind turbines in different wind farms in the region to characterize the output and uncertainty of wind turbines. The extracted data include: the rated voltage, rated current, rated power, rated capacity and other nameplate data of wind turbines; the real-time data of voltage, current, active power and reactive power during the actual operation of wind turbines; the meteorological data of temperature, humidity, wind speed and irradiance of the environment where the wind turbines are located; and the longitude and latitude that can reflect the location of the wind turbines; S2. Normalize all extracted parameters to reduce the differences in the order of magnitude of each parameter. Normalize the extracted feature parameters. The normalization process is as follows: In the above formula represents the normalized value of the parameter, x represents the measured value of the extracted parameter, and x min represents the minimum value of the extracted parameter of this type, x max Indicates the measured maximum value of the extracted parameter of this type; S3. Use the Pearson correlation coefficient to analyze the correlation between various factors. The calculation formula of the Pearson correlation coefficient is as follows: Where, X m represents the mth data of various parameters extracted, and P represents the historical output of the corresponding wind turbine; The parameter that is most correlated with the output is defined as the best correlation parameter, and the output index of the wind turbine set defined later is used as a clustering index set; according to the above calculation structure, the best correlation parameter is selected and determined as one of the parameters of the clustering index set; S4. Establish a dynamic grouping index set that takes into account the uncertainty of wind turbines; the index set includes: ideal output index, continuous output level index, unit state output capacity index and optimal related parameters; S5. The wind turbine groups are clustered using a minimum distance hierarchical clustering algorithm, and the morphological similarity distance between multidimensional parameters calculated based on Manhattan distance is used as the calculation basis for the clustering distance of wind turbines; S6. Based on the dynamic grouping results of wind turbines, equivalent units are used to replace the wind turbines in the group, and the equivalent parameters of each link of the equivalent wind turbine are determined based on the Pearson correlation analysis.
2. A method for dynamic grouping and equivalent modeling of wind turbines considering unit uncertainty according to claim 1, characterized in that: The ideal output index is based on the output data of the wind turbine in a continuous period of time to evaluate the operating status of the wind turbine; based on the operating conditions of the wind turbine in 12 consecutive sliding time windows, the difference between the actual output of the wind turbine and the ideal output of the wind turbine is evaluated. The ideal output represents the ideal output of the wind turbine. The ideal output of wind turbines in different working conditions varies greatly. The calculation formula is as follows: In the formula, ε t represents the ideal output of the wind turbine in the six sliding time windows at and after time t, P t act and are the actual values of active power and reactive power of wind turbines at time t; P t id and are respectively the active power and reactive power of the wind turbine under ideal conditions at time t.
3. The method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines according to claim 1 is characterized in that: The continuous output level index is used to characterize the difference in wind energy utilization level caused by wind abandonment phenomenon, and the continuous output level index of the wind turbine is constructed to characterize the continuity of the wind turbine output level of the wind turbine within a continuous time window; The continuous output level of the wind turbine generator set is calculated using the following formula: In the formula, γ t Indicates the continuous output level of wind turbines, C P It represents the wind energy utilization coefficient of the wind turbine, P t act represents the actual output level of the wind turbine at time t, P t id It represents the ideal maximum output of the wind turbine at time t; The continuous output level γ t The larger the value, the stronger the output continuity of the wind turbine, the higher the wind energy utilization level of the wind turbine, and the less wind abandonment phenomenon of the wind turbine.
4. The method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines according to claim 3 is characterized in that: The wind abandonment phenomenon is a phenomenon in which the power grid cannot fully absorb the power generated by the wind turbines due to the high uncertainty and randomness of wind power generation.
5. The method for dynamic grouping and equivalent modeling of wind turbines considering unit uncertainty according to claim 1 is characterized in that: The calculation formula of the wind turbine output capacity at time t is as follows: In the formula, ψ t Indicates the output capacity of the wind turbine at time t. They represent the actual active power output and actual reactive power output of the wind turbine at time t+1, It represents the ideal active power output and ideal reactive power output of the wind turbine at time t+i.
6. The method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines according to claim 1 is characterized in that: The wind turbine group dynamic grouping index set is constructed based on the ideal output, continuous output level, unit state output capacity and optimal related parameters. The wind turbine group dynamic grouping index set adopts the formula Representation.
7. The method for dynamic grouping and equivalent modeling of wind turbines considering unit uncertainty according to claim 1 is characterized in that: The Manhattan distance is calculated using the following formula: In the formula, the Manhattan distance expression between the index parameters of any two units i and j is: dis(Y i ,Y j ), where Y ik ,Y jk are the index parameters of any two wind turbines; The morphological similarity distance expression of the Manhattan distance calculation between the multidimensional parameters is: In the formula, y i and j are the elements in the wind turbine dynamic clustering index set, where y ik ,y jk It is the corresponding clustering index parameter of the two wind turbines in the morphological similarity distance expression, which represents the morphological similarity distance between any two wind turbines calculated based on the Manhattan distance.
8. The method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines according to claim 1 is characterized in that: The shortest distance of morphological similarity is expressed as: D ij =mind(y i ,y j ) Initialize and set each wind turbine group as an independent wind turbine group cluster, calculate the shortest distance between any two wind turbine group clusters with similar shapes, find the two wind turbine group clusters with the closest distance, and merge these two wind turbine group clusters into a new wind turbine group; By cyclically calculating the shortest distance between wind turbines with similar shapes, the wind turbine grouping results under any number of wind turbine groupings can be obtained. The number of wind turbine groupings can be determined according to the specific number of actual calculation examples.
9. The method for dynamic grouping and equivalent modeling of wind turbines considering unit uncertainty according to claim 1 is characterized in that: The equivalent parameters of each link of the equivalent wind turbine set are calculated by Pearson correlation weight, and the Pearson correlation coefficient is determined as the Pearson correlation weight of the equivalent set; Among them, the Pearson correlation weight of each wind turbine is used as the wind speed weight in the wind speed equalization process, and the equivalent wind speed model is expressed as: Where: is the calculation result of wind speed correlation of each wind turbine; v i represents the wind speed of each wind turbine in the wind power cluster replaced by the equivalent unit; v eq is the alternative wind speed of the equivalent wind turbine.
10. The method for dynamic grouping and equivalent modeling of wind turbines considering the uncertainty of wind turbines according to claim 1, characterized in that: The equivalent electrical parameters of the equivalent model are determined by using the Pearson correlation weight as the weight value of the modeling of each unit in the grouped equivalent module unit, as shown in the following formula: In the formula, S eq , P eq and Q eq are the equivalent capacity, equivalent active power and equivalent reactive power of equivalent wind turbine generator sets respectively, and are the sum of the corresponding parameters of wind turbine generator sets contained in each equivalent wind turbine generator set; represents the Pearson correlation parameter value of the unit, which is used as the equal weight parameter within the equal unit cluster; H eq Denotes the equivalent axis inertia time constant, K eq Denotes the equivalent shaft stiffness coefficient, D eq Represents the equivalent shaft damping coefficient.