A method and system for constructing a set of wind farm operation scenarios

By constructing a set of wind farm operation scenarios, the problems of cumbersome procedures and low efficiency in existing technologies have been solved. This has enabled the rapid construction of wind farm operation scenario sets and data correlation, thereby improving the operational stability of wind farms and the power grid.

CN117436272BActive Publication Date: 2026-11-10GUANGDONG POWER GRID CO LTD +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202311507515.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-13
Publication Date
2026-11-10
Estimated Expiration
2043-11-13

AI Technical Summary

Technical Problem

Existing methods for generating wind farm operation scenarios are cumbersome and inefficient, resulting in insufficient training of artificial intelligence decision-making technology, which makes it difficult to meet the production and operation needs of wind farms and power grids.

Method used

By collecting meteorological data from wind farms, we construct an original set of wind farm operation scenarios, model the boundary conditions for wind farm operation, calculate wind farm operation evaluation indicators, classify them using principal component analysis, and construct a target set of wind farm operation scenarios to ensure diversity and separability.

Benefits of technology

It enables the rapid construction of wind farm operation scenario sets, with strong correlations between data, improving the efficiency and accuracy of scenario generation and providing better data support for artificial intelligence decision-making.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117436272B_ABST
    Figure CN117436272B_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for constructing a wind farm operation scenario set. By collecting meteorological condition data of the wind farm, an original wind farm operation scenario set is constructed based on the meteorological condition data. Then, the wind farm operation boundary conditions are modeled based on the randomness of wind power output and wind speed to obtain the wind farm output range. Wind farm operation evaluation indicators are calculated using the wind farm output range. The original wind farm operation scenario set is then classified using the wind farm operation evaluation indicators to obtain a target wind farm operation scenario set. This method constructs the target wind farm operation scenario set through wind farm operation boundary conditions and various wind farm operation evaluation indicators, resulting in a target wind farm operation scenario set with good diversity and separability, and strong correlation between data points. This achieves rapid construction of the wind farm operation scenario set and solves the problems of cumbersome steps and low efficiency in existing wind farm operation scenario generation methods.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of power system operation and control technology, and in particular to a method and system for constructing a wind farm operation scenario set. Background Technology

[0002] With the continuous increase in energy demand, new energy power generation has developed rapidly, leading to significant changes in the energy structure. Numerous new energy power generation systems, such as photovoltaic and wind power, are connected to the power grid, impacting its security and stability. Taking wind power as an example, due to the dynamic and time-varying characteristics of weather, wind farms involve nonlinear energy conversion processes with strong uncertainties and complex spatiotemporal correlations. These issues all affect the safe and stable operation of the grid.

[0003] With the improvement of power grid automation and the development of artificial intelligence, AI-based wind farm stability analysis and decision-making technologies are expected to solve the above problems. However, the application and development of AI-based decision-making technologies in wind farms are currently limited by insufficient and imbalanced data samples, resulting in inadequate training of AI decision-making methods, overfitting, and other phenomena, making it difficult to meet the production and operation needs of wind farms and power grids.

[0004] Existing scene generation technologies mainly focus on the clustering or reduction of scene sets. Clustering methods primarily include K-means clustering, hierarchical clustering, density clustering, and fuzzy C-means clustering, while reduction methods mainly include forward scene reduction and backward scene reduction. Both emphasize the classification and screening of power grid and meteorological data, have high requirements for the selection of initial cluster centers and the number of clusters, involve relatively cumbersome steps, may result in unstable clustering effects, and have low efficiency in constructing and generating scene sets. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and apparatus for constructing a wind farm operation scenario set. By constructing a target wind farm operation scenario set through wind farm operation boundary conditions and various wind farm operation evaluation indicators, the constructed target wind farm operation scenario set has good diversity and separability characteristics, and the data have strong correlations. This solves the problems of cumbersome steps and low generation efficiency in existing wind farm operation scenario generation methods.

[0006] A first aspect of this invention provides a method for constructing a wind farm operation scenario set, the method comprising:

[0007] Collect meteorological data from wind farms and construct an original set of wind farm operation scenarios based on the meteorological data;

[0008] Based on the meteorological data, the boundary conditions for wind farm operation are modeled for wind power processing and wind speed randomness to obtain the power output range of the wind farm.

[0009] Based on the output range of the wind farm, wind farm operation evaluation indicators are calculated. Based on the wind farm operation evaluation indicators, the original wind farm operation scenario set is classified and processed to obtain the target wind farm operation scenario set.

[0010] This embodiment collects meteorological condition data from wind farms, constructs an original set of wind farm operation scenarios based on the meteorological condition data, and then models the wind farm operation boundary conditions based on the randomness of wind power output and wind speed to obtain the wind farm output range. The wind farm operation evaluation index is then calculated using the wind farm output range, and the original wind farm operation scenario set is classified using the wind farm operation evaluation index to obtain the target wind farm operation scenario set. This method constructs the target wind farm operation scenario set through wind farm operation boundary conditions and various wind farm operation evaluation indices, resulting in a target wind farm operation scenario set with good diversity and separability, and strong correlation between data. This achieves rapid construction of wind farm operation scenario sets and solves the problems of cumbersome steps and low efficiency in existing wind farm operation scenario generation methods.

[0011] In one possible implementation of the first aspect, the wind farm operation boundary conditions are modeled based on meteorological data to account for wind power processing and wind speed randomness, thereby obtaining the wind farm output range, specifically:

[0012] Based on meteorological data, a wind speed stochasticity model is constructed to obtain a wind speed probability model, which is as follows:

[0013]

[0014] In the formula, V t Let be the wind speed at time t, k be the shape parameter, and c be the scale parameter;

[0015] Based on the wind speed probability model, a wind speed prediction output is obtained. This wind speed prediction output is then used to construct a wind farm output model, which is as follows:

[0016]

[0017] In the formula, P w To provide power for the wind turbine, y t Let P be the wind speed at time t. r V is the rated power of the fan. ci V is the cut-in wind speed for the fan. r V is the rated wind speed of the fan. co The cutoff wind speed of the fan is given by A, B, and C, which are all fitting coefficients.

[0018] Based on the wind farm output model, a prediction output is obtained, and a conditional probability model of the wind farm output prediction error is constructed based on the prediction output. The conditional probability model of the wind farm output prediction error is as follows:

[0019]

[0020]

[0021]

[0022]

[0023] Among them, f XY (x, y) is [X T Y T ] T The joint probability density function, where M is the total number of Gaussian components, ω m N represents the weighting coefficient. m (•) represents a multidimensional normal distribution, W is the number of wind farms, det(•) is the matrix determinant, and μ m and σ m Let X be the mean vector and covariance matrix of the m-th Gaussian component, respectively; X is the historical actual wind speed result; and Y is the wind speed prediction output result.

[0024] Based on the conditional probability model, the power output range of the wind farm is obtained.

[0025] In one possible implementation of the first aspect, the wind farm output is predicted based on the wind farm output model to obtain the wind farm output prediction output result. A conditional probability model for the wind farm output prediction error is then constructed based on the wind farm output prediction output result, specifically as follows:

[0026] The wind speed is predicted and output based on the wind speed probability model to obtain the wind speed values ​​at different times. The wind speed values ​​at different times are then input into the wind farm output model to obtain the wind speed prediction output results for each wind speed value.

[0027] Based on the wind speed prediction output and the historical actual wind speed for each wind speed value, the joint probability distribution of the wind farm output prediction and actual value is obtained. Based on the joint probability distribution, a conditional probability model of the wind farm output prediction error is constructed.

[0028] In one possible implementation of the first aspect, the power output range of the wind farm is obtained according to the conditional probability model, specifically as follows:

[0029] The wind speed prediction output for each wind speed value is updated based on the conditional probability model to obtain the power output range of the wind farm.

[0030] In one possible implementation of the first aspect, the wind farm operation evaluation index is calculated based on the wind farm's output range, specifically as follows:

[0031] Based on the power output range of the wind farm, the voltage and phase angle of the system nodes are obtained through power flow calculations;

[0032] Calculate the wind farm operation evaluation indicators based on the voltage and phase angle of the system nodes.

[0033] In one possible implementation of the first aspect, the voltage and phase angle of the system nodes are obtained through power flow calculations based on the output range of the wind farm, specifically:

[0034] After determining the power output range of the wind farm, the active and reactive power at the equilibrium node are obtained using power flow calculations. The power flow calculation formula is as follows:

[0035]

[0036] Where the subscripts i and j represent the system node numbers, n represents the number of nodes, and ΔP i Let P be the active power imbalance at node i. Gi P is the injected active power at node i. Di U represents the outflow active power of node i. i U is the effective value of the voltage at node i. j G represents the effective voltage value at node j. ij Let θ be the conductance between node i and node j. i Let θ be the voltage phase angle at node i. j Let B be the voltage phase angle at node j. ij Let ΔQ be the susceptance between node i and node j. i Let Q be the reactive power imbalance at node i. Gi The injected reactive power at node i, Q Di Let i be the reactive power flowing out of node i;

[0037] The active and reactive power of nodes other than the slack node are determined by the output range of the wind farm.

[0038] Based on the active and reactive power of the slack node and other nodes, the voltage and phase angle of each node are obtained. The formulas for voltage and phase angle are as follows:

[0039] (U i θ i )=f PF (P i Q i )

[0040] In the formula, U i θ represents the effective voltage value at system node i.i P represents the phase angle of system node i. i Q represents the active power of system node i. i f represents the reactive power of system node i. PF It represents a function that calculates node voltage and phase angle based on active and reactive power.

[0041] In one possible implementation of the first aspect, the original wind farm operation scenario set is classified according to wind farm operation evaluation indicators to obtain the target wind farm operation scenario set, specifically as follows:

[0042] The original wind farm operation scenario set was classified according to the wind farm operation evaluation indicators to obtain the scenario classification results;

[0043] The scene classification results were clustered using a visualization-based dimensionality reduction method based on principal component analysis to obtain the target wind farm operation scene set.

[0044] A second aspect of this invention provides a wind farm operation scenario set construction system, the system comprising:

[0045] The data acquisition module is used to collect meteorological data from the wind farm and construct an original set of wind farm operation scenarios based on the meteorological data.

[0046] The wind farm output range calculation module is used to model the wind farm operation boundary conditions based on meteorological data, wind power processing and wind speed randomness, and obtain the wind farm output range.

[0047] The classification module is used to calculate the wind farm operation evaluation index based on the wind farm's output range, and to classify the original wind farm operation scenario set based on the wind farm operation evaluation index to obtain the target wind farm operation scenario set.

[0048] In one possible implementation of the second aspect, the wind farm output range calculation module includes a first calculation unit, a second calculation unit, a third calculation unit, and a fourth calculation unit.

[0049] The first calculation unit is used to model the randomness of wind speed based on meteorological data, and obtain a wind speed probability model, wherein the wind speed probability model is:

[0050]

[0051] In the formula, V t Let be the wind speed at time t, k be the shape parameter, and c be the scale parameter;

[0052] The second calculation unit is used to predict the wind speed output based on the wind speed probability model, obtain the wind speed prediction output result, and construct the wind farm output model using the wind speed prediction output result. The wind farm output model is as follows:

[0053]

[0054] In the formula, P w To power the wind turbine, V t Let P be the wind speed at time t. r V is the rated power of the fan. ci V is the cut-in wind speed for the fan. r V is the rated wind speed of the fan. co The cutoff wind speed of the fan is given by A, B, and C, which are all fitting coefficients.

[0055] The third calculation unit is used to predict the output based on the wind farm output model, obtain the wind farm output prediction output result, and construct a conditional probability model of the wind farm output prediction error based on the wind farm output prediction output result. The conditional probability model of the wind farm output prediction error is as follows:

[0056]

[0057]

[0058]

[0059]

[0060] Among them, f XY (x, y) is [X T Y T ] T The joint probability density function, where M is the total number of Gaussian components, ω m N represents the weighting coefficient. m (·) represents a multidimensional normal distribution, W is the number of wind farms, det(·) is the matrix determinant, and μ m and σ m Let X be the mean vector and covariance matrix of the m-th Gaussian component, respectively; X is the historical actual wind speed result; and Y is the wind speed prediction output result.

[0061] The fourth calculation unit is used to obtain the power output range of the wind farm based on the conditional probability model.

[0062] In one possible implementation of the second aspect, the third computational unit includes a wind speed prediction subunit and a conditional probability model unit.

[0063] Among them, the wind speed prediction subunit is used to make prediction output based on the wind speed probability model, obtain the wind speed value at different times, and input the wind speed value at different times into the wind farm output model to obtain the wind speed prediction output result for each wind speed value.

[0064] The conditional probability model unit is used to obtain the joint probability distribution of the wind farm output prediction value and the actual value based on the wind speed prediction output of each wind speed value and the historical actual wind speed value. Based on the joint probability distribution, a conditional probability model of the wind farm output prediction error is constructed. Attached Figure Description

[0065] Figure 1 : A flowchart illustrating an embodiment of the wind farm operation scenario set construction method provided by the present invention;

[0066] Figure 2 This is a schematic diagram of the wind power output variation curve of one embodiment of the wind farm operation scenario set construction method provided by the present invention.

[0067] Figure 3 : A schematic diagram illustrating the correlation of wind farm operation evaluation indicators in an embodiment of the wind farm operation scenario set construction method provided by the present invention;

[0068] Figure 4 : A schematic diagram of the spatial distribution of the target wind farm operation scenario set in one embodiment of the wind farm operation scenario set construction method provided by the present invention;

[0069] Figure 5 : A schematic diagram of the system structure of another embodiment of the wind farm operation scenario set construction method provided by the present invention;

[0070] Figure 6 : A schematic diagram of an electronic device structure for another embodiment of the wind farm operation scenario set construction method provided by the present invention; Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Example 1

[0073] Please refer to Figure 1 This is a flowchart illustrating an embodiment of the wind farm operation scenario set construction method provided by the present invention, including steps S11 to S13, each step of which is as follows:

[0074] S11. Collect meteorological data from the wind farm and construct an original set of wind farm operation scenarios based on the meteorological data.

[0075] In this embodiment, since the uncertainty of wind farm operation is mainly caused by meteorological factors, which can directly affect the wind power output in the wind power generation system, it is necessary to first obtain meteorological condition data, which comes from real datasets.

[0076] After acquiring meteorological data, a data sample can be generated for each day. For each data sample, a time resolution can be set, such as 5 minutes. In this case, a data sample contains 288 moments, each corresponding to a system state. The collection of these system states constitutes the original wind farm operation scenario set.

[0077] It should be noted that meteorological data includes, but is not limited to, wind speed, ambient temperature, air density, and geographical location for the current date, as well as historical data such as wind speed, ambient temperature, air density, and geographical location.

[0078] S12. Based on meteorological data, model the boundary conditions for wind farm operation based on wind power processing and wind speed randomness to obtain the output range of the wind farm.

[0079] In a preferred embodiment, the wind farm operation boundary conditions are modeled based on meteorological data to account for wind power processing and wind speed randomness, thereby obtaining the wind farm output range, specifically:

[0080] Based on meteorological data, a wind speed stochasticity model is constructed to obtain a wind speed probability model, which is as follows:

[0081]

[0082] In the formula, V t Let be the wind speed at time t, k be the shape parameter, and c be the scale parameter;

[0083] Based on the wind speed probability model, a wind speed prediction output is obtained. This wind speed prediction output is then used to construct a wind farm output model, which is as follows:

[0084]

[0085] In the formula, P w To power the wind turbine, V t Let P be the wind speed at time t. r V is the rated power of the fan. ci V is the cut-in wind speed for the fan. r V is the rated wind speed of the fan. co The cutoff wind speed of the fan is given by A, B, and C, which are all fitting coefficients.

[0086] Based on the wind farm output model, a prediction output is obtained, and a conditional probability model of the wind farm output prediction error is constructed based on the prediction output. The conditional probability model of the wind farm output prediction error is as follows:

[0087]

[0088]

[0089]

[0090]

[0091] Among them, f XY (x, y) is [X T Y T ] T The joint probability density function, where M is the total number of Gaussian components, ω m N represents the weighting coefficient. m (·) represents a multidimensional normal distribution, W is the number of wind farms, det(·) is the matrix determinant, and μ m and σ m Let X be the mean vector and covariance matrix of the m-th Gaussian component, respectively; let X be the historical actual wind speed result; and let Y be the wind speed prediction output result.

[0092] Based on the conditional probability model, the power output range of the wind farm is obtained.

[0093] In a preferred embodiment, a wind farm output prediction output is obtained by predicting the output based on the wind farm output model. A conditional probability model for the wind farm output prediction error is then constructed based on this prediction output, specifically:

[0094] The wind speed is predicted and output based on the wind speed probability model to obtain the wind speed values ​​at different times. The wind speed values ​​at different times are then input into the wind farm output model to obtain the wind speed prediction output results for each wind speed value.

[0095] Based on the wind speed prediction output and the historical actual wind speed for each wind speed value, the joint probability distribution of the wind farm output prediction and actual value is obtained. Based on the joint probability distribution, a conditional probability model of the wind farm output prediction error is constructed.

[0096] In a preferred embodiment, the power output range of the wind farm is obtained according to a conditional probability model, specifically as follows:

[0097] The wind speed prediction output for each wind speed value is updated based on the conditional probability model to obtain the power output range of the wind farm.

[0098] In this embodiment, based on the acquired meteorological data, boundary condition modeling for wind farm operation was performed to address the randomness of wind power output and wind speed, resulting in a relatively precise range for wind farm output. Since the uncertainty in wind farm operation is primarily caused by meteorological factors, which directly affect wind power output in the wind power generation system, this embodiment mainly focuses on boundary condition modeling for wind farm operation, addressing the randomness of wind power output and wind speed. Wind farm operation boundary condition modeling includes wind farm output modeling, wind speed randomness modeling, and conditional probability modeling of wind farm output prediction errors.

[0099] First, a random model of wind speed is constructed based on meteorological data to obtain a wind speed probability model, which is expressed as follows:

[0100]

[0101] Among them, V t Let t be the wind speed, k be the shape parameter, and c be the scale parameter.

[0102] Then, based on the predicted output of the wind speed probability model, a wind farm output model is constructed. The wind farm output model (the relationship between wind power output and wind speed) is expressed as follows:

[0103]

[0104] Among them, P w To power the wind turbine, V t Let P be the wind speed at time t. r V is the rated power of the fan. ci V is the cut-in wind speed for the fan. r V is the rated wind speed of the fan. co The cutoff wind speed of the fan is represented by A, B, and C, which are all fitting coefficients.

[0105] The purpose of constructing the wind speed probability model is to introduce randomness. Based on the wind speed probability model, wind speed values ​​at different times are obtained. Then, by substituting multiple different wind speed values ​​into the wind farm output model, the predicted wind power output value corresponding to each wind speed value can be obtained.

[0106] Then, based on the predicted and historical data of wind farm output, the joint probability distribution of the actual and predicted wind farm output is obtained. Finally, a conditional probability model of the wind farm output prediction error is constructed to improve the wind farm operation boundary condition model and obtain a more accurate range of wind farm output. The joint probability distribution of the actual and predicted wind farm output is as follows:

[0107]

[0108]

[0109]

[0110]

[0111] Among them, f XY (x,y) is [X T Y T ] T The joint probability density function, where M is the total number of Gaussian components, ω m N represents the weighting coefficient. m (·) represents a multidimensional normal distribution, W is the number of wind farms, det(·) is the matrix determinant, and μ m and σ m Let X be the mean vector and covariance matrix of the m-th Gaussian component, respectively; let Y be the actual output of the wind farm and Y be the predicted output of the wind farm.

[0112] The wind farm forecasting error is the difference between the actual wind farm output and the predicted wind farm output, i.e.:

[0113] Z = XY

[0114] Understandably, based on multiple wind power output values ​​predicted by the wind farm output model, a Gaussian Mixture Model (GMM) is used to characterize the joint probability distribution of the actual and predicted wind farm output values. This constructs a conditional probability model of the wind farm output prediction error. Then, based on this conditional probability model, the predicted wind farm output values ​​are updated to obtain an accurate range for the wind farm output. The formula is as follows:

[0115] Y*=Y+Z

[0116] Where Y* is the predicted output value of the wind farm considering the conditional probability of the wind farm output prediction error.

[0117] S13. Calculate the wind farm operation evaluation index based on the wind farm output range, and classify the original wind farm operation scenario set according to the wind farm operation evaluation index to obtain the target wind farm operation scenario set.

[0118] In a preferred embodiment, the wind farm operation evaluation index is calculated based on the wind farm's output range, specifically as follows:

[0119] Based on the power output range of the wind farm, the voltage and phase angle of the system nodes are obtained through power flow calculations;

[0120] Calculate the wind farm operation evaluation indicators based on the voltage and phase angle of the system nodes.

[0121] In a preferred embodiment, the voltage and phase angle of the system nodes are obtained through power flow calculation based on the power output range of the wind farm, specifically as follows:

[0122] After determining the power output range of the wind farm, the active and reactive power at the equilibrium node are obtained using power flow calculations. The power flow calculation formula is as follows:

[0123]

[0124] Where the subscripts i and j represent the system node numbers, n represents the number of nodes, and ΔP i Let P be the active power imbalance at node i. Gi P is the injected active power at node i. Di U represents the outflow active power of node i. i U is the effective value of the voltage at node i. j G represents the effective voltage value at node j. ij Let θ be the conductance between node i and node j. i Let θ be the voltage phase angle at node i. j Let B be the voltage phase angle at node j. ij Let ΔQ be the susceptance between node i and node j. i Let Q be the reactive power imbalance at node i. Gi The injected reactive power at node i, Q Di Let i be the reactive power flowing out of node i;

[0125] The active and reactive power of nodes other than the slack node are determined by the output range of the wind farm.

[0126] Based on the active and reactive power of the slack node and other nodes, the voltage and phase angle of each node are obtained. The formulas for voltage and phase angle are as follows:

[0127] (U i θ i )=f PF (P i Q i )

[0128] In the formula, U i θ represents the effective voltage value at system node i. i P represents the phase angle of system node i. i Q represents the active power of system node i. i f represents the reactive power of system node i. PF It represents a function that calculates node voltage and phase angle based on active and reactive power.

[0129] In a preferred embodiment, the original wind farm operation scenario set is classified according to wind farm operation evaluation indicators to obtain the target wind farm operation scenario set, specifically:

[0130] The original wind farm operation scenario set was classified according to the wind farm operation evaluation indicators to obtain the scenario classification results;

[0131] The scene classification results were clustered using a visualization-based dimensionality reduction method based on principal component analysis to obtain the target wind farm operation scene set.

[0132] In this embodiment, the system node state variables are obtained through power flow calculation based on the power output range of the wind farm; and the wind farm operation evaluation index is calculated based on the system node state variables.

[0133] Understandably, for a wind farm system with a defined topology, the state of the system can be considered determined once the power output range of the wind farm is determined. System node state quantities characterizing the state of a wind farm system include active power, reactive power, voltage, and phase angle at each node.

[0134] Except for the slack node, whose active and reactive power can be directly determined by the wind farm output, the active and reactive power of the slack node, as well as the voltage and phase angle of the other nodes, can be calculated by power flow.

[0135]

[0136] Where the subscripts i and j represent the system node numbers, n represents the number of nodes, and ΔP i Let P be the active power imbalance at node i. Gi P is the injected active power at node i. Di U represents the outflow active power of node i. i U is the effective value of the voltage at node i. j G represents the effective voltage value at node j. ij Let θ be the conductance between node i and node j. i Let θ be the voltage phase angle at node i. j Let B be the voltage phase angle at node j. ij Let ΔQ be the susceptance between node i and node j. i Let Q be the reactive power imbalance at node i. Gi The injected reactive power at node i, Q Di Let be the outflow reactive power of node i.

[0137] For a system in its normal operating range, i.e., when the power flow calculation converges, for a given set of active and reactive power, there is a unique set of node voltages and phase angles, as follows:

[0138] (Ui θ i )=f PF (P i Q i )

[0139] In the formula, U i θ represents the effective voltage value at system node i. i P represents the phase angle of system node i. i Q represents the active power of system node i. i f represents the reactive power of system node i. PF It represents a function that calculates node voltage and phase angle based on active and reactive power.

[0140] After obtaining these system node state variables, wind farm operation evaluation indicators are obtained through simulation calculations. Based on the system node state variables, wind farm operation evaluation indicators are calculated, including but not limited to voltage indicators, harmonic value indicators, three-phase voltage imbalance indicators, and power indicators.

[0141] Among them, voltage indicators include voltage stability margin, voltage deviation rate, voltage fluctuation, and flicker interference value.

[0142] 1) Regarding voltage stability margin

[0143] When the voltage in a system exceeds the static voltage stability limit, a voltage collapse accident occurs, jeopardizing system safety. In practical calculations, methods such as singular value decomposition, sensitivity method, continuous power flow extension method, collapse point method, and nonlinear programming method can be used to determine the voltage level.

[0144] 2) Regarding voltage offset rate

[0145] The formula for calculating the voltage offset rate is as follows:

[0146]

[0147] Where n is the number of system nodes, V i V is the actual voltage at system node i. ri This is the rated voltage of system node i.

[0148] This formula represents the average voltage offset rate of all nodes in the system.

[0149] 3) Regarding voltage fluctuations.

[0150] Voltage fluctuations can be described by the root mean square (RMS) curve of voltage, with the voltage variation d and the frequency r measuring the magnitude and rate of voltage fluctuation. The specific calculation formulas are as follows:

[0151]

[0152] Where ΔU is the difference between two adjacent extreme voltages on the root mean square voltage curve, U N This is the system's nominal voltage.

[0153] 4) Regarding flicker interference values.

[0154] Flicker is the cumulative effect of voltage fluctuations over a period of time, and is composed of a short-term flicker value P. st and long-term flicker value P lt To measure.

[0155] Short-time flicker value P st The long-term flicker value P can be calculated through a process involving a square demodulator, bandpass weighted filter, and square first-order low-pass filter. lt The short-time flicker value P contained within the measurement time period st The calculation is obtained using the following formula:

[0156]

[0157] In the formula, n is the number of short-time flicker values ​​included in the long-time flicker value measurement time, and P sti Let be the value of the i-th short-time flicker.

[0158] Harmonic values ​​include harmonic voltage content, harmonic current content, harmonic voltage content, harmonic current content, total harmonic distortion of voltage, and total harmonic distortion of current.

[0159] By performing Fourier series decomposition on the voltage and current at the grid connection point of the wind farm, the harmonic components of the voltage and current can be obtained.

[0160] 1) Regarding the harmonic voltage content.

[0161] It can be calculated using the following formula:

[0162]

[0163] In the formula, U h U1 is the root mean square value of the h-th harmonic voltage, and U2 is the root mean square value of the fundamental voltage.

[0164] 2) Regarding the h-th harmonic current content

[0165] It can be calculated using the following formula:

[0166]

[0167] In the formula, I h Ih is the root mean square value of the h-th harmonic current, and I1 is the root mean square value of the fundamental current.

[0168] 3) Regarding harmonic voltage content

[0169] The harmonic voltage content is calculated using the following formula:

[0170]

[0171] 4) Regarding harmonic current content.

[0172] It can be calculated using the following formula:

[0173]

[0174] 5) Regarding the total harmonic distortion rate of voltage.

[0175] It can be calculated using the following formula:

[0176]

[0177] 6) Regarding the total harmonic distortion rate of current.

[0178] It can be calculated using the following formula:

[0179]

[0180] The three-phase voltage unbalance index includes the unbalance degree.

[0181] By decomposing the three-phase voltage of the system using the symmetrical component method, we can obtain its positive-sequence, negative-sequence, and zero-sequence components in the symmetrical system.

[0182] The degree of imbalance can be calculated using the following formula:

[0183]

[0184] Where, ε U2 ε represents the negative sequence unbalance of three-phase voltage. U0 U1 represents the root mean square value of the positive sequence component of the three-phase voltage, U2 represents the root mean square value of the negative sequence component of the three-phase voltage, and U0 represents the root mean square value of the zero sequence component of the three-phase voltage.

[0185] Power indicators include, but are not limited to, the line loss rate, which is the ratio of the difference between the total power generation and the total power consumption of the wind farm in the system to the total power generation. The line loss rate can be calculated using the following formula:

[0186]

[0187] Among them, P Gi P represents the injected active power at system node i. Di This represents the outflow of active power from system node i.

[0188] Take a real wind farm grid connection system as an example.

[0189] First, we obtained nearly a year's worth of meteorological data from real datasets, including wind speed, ambient temperature, and air density.

[0190] Using one day as a data sample with a time resolution of 5 minutes, each data sample contains 288 time points, each corresponding to a system state. Boundary conditions for wind farm operation are modeled based on wind speed probability models, wind farm output models, and conditional probability models to determine the wind power output of each node in the wind farm's grid-connected system at each time point.

[0191] Specifically, Figure 2 The diagram shows the wind power output variation curve provided by this invention. Figure 2 It can be seen that the wind power output obtained by modeling the boundary conditions of wind farm operation using the above method can meet good temporal characteristics, and the wind power output of different nodes in the system has obvious correlation characteristics, indicating that the target wind farm operation scenario set constructed on this basis has diversity and separability. Then, after obtaining the system state variables, the wind farm operation evaluation index is calculated.

[0192] Taking voltage as an example, Figure 3 A correlation diagram of the wind farm operation evaluation indicators provided by this invention is shown. According to... Figure 3 It can be seen that there is a strong linear correlation among the various indicators.

[0193] Finally, the scene samples of the wind farm were analyzed and clustered. After reducing the various indicators to two dimensions using principal component analysis, it was found that 97% of the variance information was retained, indicating that the dimensionality reduction method was effective.

[0194] Specifically, the spatial distribution of scene samples is determined using the dimensionality-reduced data. Figure 4 The spatial distribution map of the target wind farm operation scenario set provided by the present invention is shown.

[0195] exist Figure 4 In the diagram, each point represents a scene sample, and points of different colors represent samples of corresponding categories. From c1 to c5, the voltage level increases progressively. According to... Figure 4 It can also be seen that the various types of samples have clear boundaries in the dimensionality-reduced space, indicating that the dimensionality reduction and clustering methods are effective. In addition, the scene samples have certain distribution characteristics in space, indicating that the samples have diversity.

[0196] In this embodiment, system node state variables are obtained through power flow calculation based on the wind farm's output range. Based on these system node state variables, wind farm operation evaluation indicators are calculated. Then, the original wind farm operation scenario set is classified according to these indicators to obtain the target wind farm operation scenario set. The original wind farm operation scenario set is constructed based on meteorological condition data. This method constructs the target wind farm operation scenario set through wind farm operation boundary conditions and various wind farm operation evaluation indicators. This results in a target wind farm operation scenario set with good diversity and separability, and strong correlation between data points. It overcomes the shortcomings of existing wind farm operation scenario generation methods, which are cumbersome and inefficient. This method enables rapid construction of wind farm operation scenario sets, providing better data support for complex grid connection situation simulations of wind farms and for training AI-based perception and diagnostic engines.

[0197] It should be noted that the wind farm operation scenario set construction method provided by the present invention constructs a target wind farm operation scenario set through wind farm operation boundary conditions and various operation evaluation indicators. The constructed scenario set has good diversity and separability characteristics, strong correlation between data, and more efficient scenario clustering method.

[0198] Example 2

[0199] Accordingly, see Figure 5 , Figure 5 This invention provides a wind farm operation scenario set construction system, as shown in the figure. The wind farm operation scenario set construction system includes:

[0200] The data acquisition module 501 is used to acquire meteorological data from the wind farm and construct an original set of wind farm operation scenarios based on the meteorological data.

[0201] The wind farm output range calculation module 502 is used to model the wind farm operation boundary conditions based on the meteorological data, wind power processing and wind speed randomness, and obtain the wind farm output range.

[0202] The classification module 503 is used to calculate the wind farm operation evaluation index based on the wind farm output range, and to classify the original wind farm operation scenario set based on the wind farm operation evaluation index to obtain the target wind farm operation scenario set.

[0203] In a preferred embodiment, the wind farm output range calculation module includes a first calculation unit, a second calculation unit, a third calculation unit, and a fourth calculation unit.

[0204] The first calculation unit is used to model the randomness of wind speed based on meteorological data, and obtain a wind speed probability model, wherein the wind speed probability model is:

[0205]

[0206] In the formula, V t Let be the wind speed at time t, k be the shape parameter, and c be the scale parameter;

[0207] The second calculation unit is used to predict the wind speed output based on the wind speed probability model, obtain the wind speed prediction output result, and construct the wind farm output model using the wind speed prediction output result. The wind farm output model is as follows:

[0208]

[0209] In the formula, P w To power the wind turbine, V t Let P be the wind speed at time t. r V is the rated power of the fan. ci V is the cut-in wind speed for the fan. r V is the rated wind speed of the fan. co The cutoff wind speed of the fan is given by A, B, and C, which are all fitting coefficients.

[0210] The third calculation unit is used to predict the output based on the wind farm output model, obtain the wind farm output prediction output result, and construct a conditional probability model of the wind farm output prediction error based on the wind farm output prediction output result. The conditional probability model of the wind farm output prediction error is as follows:

[0211]

[0212]

[0213]

[0214]

[0215] Among them, f XY (x, y) is [X T Y T ] T The joint probability density function, where M is the total number of Gaussian components, ω m N represents the weighting coefficient. m (·) represents a multidimensional normal distribution, W is the number of wind farms, det(·) is the matrix determinant, and μ m and σ m Let X be the mean vector and covariance matrix of the m-th Gaussian component, respectively; X is the historical actual wind speed result; and Y is the wind speed prediction output result.

[0216] The fourth calculation unit is used to obtain the power output range of the wind farm based on the conditional probability model.

[0217] In a preferred embodiment, the third calculation unit includes a wind speed prediction subunit and a conditional probability model unit.

[0218] Among them, the wind speed prediction subunit is used to make prediction output based on the wind speed probability model, obtain the wind speed value at different times, and input the wind speed value at different times into the wind farm output model to obtain the wind speed prediction output result for each wind speed value.

[0219] The conditional probability model unit is used to obtain the joint probability distribution of the wind farm output prediction value and the actual value based on the wind speed prediction output of each wind speed value and the historical actual wind speed value. Based on the joint probability distribution, a conditional probability model of the wind farm output prediction error is constructed.

[0220] In a preferred embodiment, the fourth calculation unit includes a wind farm output range determination subunit.

[0221] The wind farm output range determination subunit is used to update the wind speed prediction output results of each wind speed value according to the conditional probability model to obtain the wind farm output range.

[0222] In a preferred embodiment, the classification module 503 includes a power flow calculation unit and an evaluation index determination unit.

[0223] The power flow calculation unit is used to obtain the voltage and phase angle of the system nodes through power flow calculation based on the output range of the wind farm.

[0224] The evaluation index determination unit is used to calculate the wind farm operation evaluation index based on the voltage and phase angle of the system nodes.

[0225] In a preferred embodiment, the power flow calculation unit includes a power calculation subunit for the slack node, a calculation subunit for other nodes, and a voltage phase angle determination subunit.

[0226] The slack node power calculation subunit is used to determine the output range of the wind farm and then use power flow calculation to obtain the active and reactive power of the slack node. The power flow calculation formula is as follows:

[0227]

[0228] Where the subscripts i and j represent the system node numbers, n represents the number of nodes, and ΔP i Let P be the active power imbalance at node i. Gi P is the injected active power at node i. Di U represents the outflow active power of node i. i U is the effective value of the voltage at node i. j G represents the effective voltage value at node j. ijLet θ be the conductance between node i and node j. i Let θ be the voltage phase angle at node i. j Let B be the voltage phase angle at node j. ij Let ΔQ be the susceptance between node i and node j. i Let Q be the reactive power imbalance at node i. Gi The injected reactive power at node i, Q Di Let i be the reactive power flowing out of node i;

[0229] Other node calculation subunits are used to determine the active and reactive power of nodes other than the balancing node based on the output range of the wind farm.

[0230] The voltage phase angle determination subunit is used to obtain the voltage and phase angle of each node based on the active power and reactive power of the slack node and the other nodes, wherein the formulas for the voltage and the phase angle are:

[0231] (U i ,θ i )=f PF (P i Q i )

[0232] In the formula, U i θ represents the effective voltage value at system node i. i P represents the phase angle of system node i. i Q represents the active power of system node i. i f represents the reactive power of system node i. PF It represents a function that calculates node voltage and phase angle based on active and reactive power.

[0233] In a preferred embodiment, the classification module 503 further includes a scene classification unit and a clustering unit.

[0234] The classification unit is used to classify the original wind farm operation scenario set according to the wind farm operation evaluation index to obtain the scenario classification result;

[0235] The clustering unit is used to cluster the scene classification results using a visualization dimensionality reduction method based on principal component analysis to obtain a set of target wind farm operation scenes.

[0236] Accordingly, see Figure 6 , Figure 6This invention provides an electronic device, as shown in the figure. The electronic device includes a processor 610, a communication interface 620, a memory 630, and a communication bus 640. The processor 610, communication interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute a method for constructing a wind farm operation scenario set. This method includes: collecting meteorological data from the wind farm; constructing an original wind farm operation scenario set based on the meteorological data; modeling the wind farm operation boundary conditions based on the meteorological data for wind power processing and wind speed randomness to obtain the wind farm output range; calculating wind farm operation evaluation indicators based on the wind farm output range; and classifying the original wind farm operation scenario set based on the wind farm operation evaluation indicators to obtain a target wind farm operation scenario set.

[0237] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0238] Accordingly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the wind farm operation scenario set construction method provided by the above methods.

[0239] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0240] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0241] For a more detailed explanation of the working principle and procedures of this embodiment, please refer to the relevant description in Embodiment 1.

[0242] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0243] By collecting meteorological data of wind farms, and constructing an initial set of wind farm operation scenarios based on this data, the method then models the wind farm operation boundary conditions for the randomness of wind power output and wind speed, obtaining the wind farm output range. Using this output range, wind farm operation evaluation indicators are calculated. These indicators are then used to classify the initial set of wind farm operation scenarios, resulting in a target set of wind farm operation scenarios. This method constructs the target set of wind farm operation scenarios through wind farm operation boundary conditions and various wind farm operation evaluation indicators. This results in a target set of wind farm operation scenarios with good diversity and separability, and strong correlations between data points. It achieves rapid construction of wind farm operation scenario sets, solving the problems of cumbersome steps and low efficiency in existing wind farm operation scenario generation methods.

[0244] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for constructing a wind farm operation scenario set, characterized in that, include: Collect meteorological data from wind farms and construct an original set of wind farm operation scenarios based on the meteorological data; Based on the meteorological data, wind power processing and wind speed randomness are modeled to determine the wind farm operation boundary conditions, thereby obtaining the wind farm output range. The process involves modeling the wind farm's operational boundary conditions based on the meteorological data, considering wind power processing and wind speed randomness, to obtain the wind farm's output range. Specifically: Based on the meteorological data, a wind speed stochasticity model is performed to obtain a wind speed probability model, wherein the wind speed probability model is as follows: In the formula, for Wind speed at any moment For shape parameters, For scale parameters; Based on the wind speed probability model, a wind speed prediction output is obtained. The wind farm output model is then constructed using this wind speed prediction output. The wind farm output model is as follows: In the formula, To provide power for the wind turbine, for Wind speed at any moment This refers to the rated power of the fan. The cut-in wind speed for the fan. The rated wind speed of the fan. To cut off the wind speed for the fan. , , All are fitting coefficients; Based on the wind farm output model, a prediction output is obtained, yielding a wind farm output prediction result. A conditional probability model for the wind farm output prediction error is then constructed based on this prediction result. The conditional probability model for the wind farm output prediction error is as follows: , in, for The joint probability density function, The total number of Gaussian components. These are the weighting coefficients. It follows a multidimensional normal distribution. The number of wind farms. It is the determinant of a matrix. and The first The mean vector and covariance matrix of each Gaussian component, X is the historical actual wind speed result, and Y is the wind speed prediction output result; The power output range of the wind farm is obtained based on the conditional probability model. Based on the output range of the wind farm, wind farm operation evaluation indicators are calculated. Based on the wind farm operation evaluation indicators, the original wind farm operation scenario set is classified and processed to obtain the target wind farm operation scenario set.

2. The method for constructing a wind farm operation scenario set as described in claim 1, characterized in that, The process involves predicting output based on the wind farm output model to obtain a wind farm output prediction result. A conditional probability model for the wind farm output prediction error is then constructed based on this prediction result. Specifically: Based on the wind speed probability model, the wind speed values ​​at different times are predicted and output. The wind speed values ​​at different times are then input into the wind farm output model to obtain the wind speed prediction output results for each wind speed value. Based on the wind speed prediction output results of each wind speed value and the historical actual wind speed results of each wind speed value, a joint probability distribution of the wind farm output prediction value and the actual value is obtained. Based on the joint probability distribution, a conditional probability model of the wind farm output prediction error is constructed.

3. The method for constructing a wind farm operation scenario set as described in claim 2, characterized in that, The wind farm output range is obtained based on the conditional probability model, specifically as follows: The wind speed prediction output for each wind speed value is updated based on the conditional probability model to obtain the power output range of the wind farm.

4. The method for constructing a wind farm operation scenario set as described in claim 1, characterized in that, The calculation of wind farm operation evaluation indicators based on the wind farm's output range specifically includes: Based on the power output range of the wind farm, the voltage and phase angle of the system nodes are obtained through power flow calculation; The wind farm operation evaluation index is calculated based on the voltage and phase angle of the system nodes.

5. The method for constructing a wind farm operation scenario set as described in claim 4, characterized in that, The voltage and phase angle of the system nodes are obtained through power flow calculation based on the output range of the wind farm, specifically as follows: After determining the power output range of the wind farm, the active and reactive power of the equilibrium node are obtained using power flow calculations. The power flow calculation formula is as follows: Among them, subscript and This represents the system node number, and n represents the number of system nodes. For system nodes The active power imbalance, For system nodes The injected active power, For system nodes The outflow of active power, For system nodes The effective value of the voltage, For system nodes The effective value of the voltage, For system nodes and system nodes The electrical conductivity between them For system nodes voltage phase angle, For system nodes voltage phase angle, For system nodes and system nodes The susceptance between them For system nodes The reactive power imbalance, For system nodes Injected reactive power, For system nodes The outflow of reactive power; The active and reactive power of nodes other than the balancing node are determined by the output range of the wind farm. Based on the active power and reactive power of the balancing node and the other nodes, the voltage and phase angle of each system node are obtained, wherein the formulas for the voltage and the phase angle are: In the formula, Represents system nodes The effective value of the voltage, Represents system nodes voltage phase angle, Represents system nodes i active power, Represents system nodes i reactive power, It represents a function that calculates the node voltages and phase angles of a system based on active and reactive power.

6. The method for constructing a wind farm operation scenario set as described in claim 1, characterized in that, The step of classifying the original wind farm operation scenario set according to the wind farm operation evaluation indicators to obtain the target wind farm operation scenario set is as follows: The original set of wind farm operation scenarios is classified according to the wind farm operation evaluation indicators to obtain scenario classification results; The scenario classification results were clustered using a visualization-based dimensionality reduction method based on principal component analysis to obtain a set of target wind farm operation scenarios.

7. A system for constructing a set of wind farm operation scenarios, characterized in that, A method for constructing a wind farm operation scenario set as described in any one of claims 1 to 6 includes: The data acquisition module is used to collect meteorological data from the wind farm and construct an original set of wind farm operation scenarios based on the meteorological data. The wind farm output range calculation module is used to model the wind farm operation boundary conditions based on the meteorological data, wind power processing and wind speed randomness, and obtain the wind farm output range. The classification module is used to calculate the wind farm operation evaluation index based on the wind farm's output range, and to classify the original wind farm operation scenario set based on the wind farm operation evaluation index to obtain the target wind farm operation scenario set.

8. The wind farm operation scenario set construction system as described in claim 7, characterized in that, The wind farm output range calculation module includes a first calculation unit, a second calculation unit, a third calculation unit, and a fourth calculation unit. The first calculation unit is used to perform wind speed stochasticity modeling based on the meteorological data to obtain a wind speed probability model, wherein the wind speed probability model is: In the formula, for Wind speed at any moment For shape parameters, For scale parameters; The second calculation unit is used to make a prediction output based on the wind speed probability model, obtain the wind speed prediction output result, and construct a wind farm output model using the wind speed prediction output result, wherein the wind farm output model is: In the formula, To provide power for the wind turbine, for Wind speed at any moment This refers to the rated power of the fan. The cut-in wind speed for the fan. The rated wind speed of the fan. To cut off the wind speed for the fan. , , All are fitting coefficients; The third calculation unit is used to perform prediction output based on the wind farm output model to obtain the wind farm output prediction output result, and to construct a conditional probability model of the wind farm output prediction error based on the wind farm output prediction output result, wherein the conditional probability model of the wind farm output prediction error is: , in, for The joint probability density function, The total number of Gaussian components. These are the weighting coefficients. It follows a multidimensional normal distribution. The number of wind farms. It is the determinant of a matrix. and The first m The mean vector and covariance matrix of each Gaussian component, X is the historical actual wind speed result, and Y is the wind speed prediction output result; The fourth calculation unit is used to obtain the power output range of the wind farm based on the conditional probability model.

9. The wind farm operation scenario set construction system as described in claim 8, characterized in that, The third calculation unit includes a wind speed prediction subunit and a conditional probability model unit. The wind speed prediction subunit is used to make prediction outputs based on the wind speed probability model to obtain wind speed values ​​at different times, and input the wind speed values ​​at different times into the wind farm output model to obtain wind speed prediction output results for each wind speed value. The conditional probability model unit is used to obtain the joint probability distribution of the predicted and actual wind farm output values ​​based on the wind speed prediction output results and the historical actual wind speed results of each wind speed value, and to construct a conditional probability model of the wind farm output prediction error based on the joint probability distribution.

Citation Information

Patent Citations

  • Static voltage stability margin-based new energy network-related influence quantitative calculation method and system, electronic equipment and storage medium

    CN115470866A