A method for emergency safety control of wind turbine generator set

Abstract

The present invention discloses an emergency safety control method for a wind turbine generator set, which belongs to the technical field of wind power control, and specifically comprises: obtaining the predicted output power of the wind turbine generator set, and determining the target output power of the wind turbine generator set according to the predicted output power; presetting a monitoring period, drawing a curve of the output power of the wind turbine generator set changing with time, obtaining all coordinate points in the change curve where the power generation power is the target output power and marking them as reference points; dividing the change curve into a plurality of sub-curves with any adjacent reference point as a dividing point, and calculating the risk coefficient of the wind turbine generator set according to the plurality of sub-curves; obtaining the deviation value of the wind turbine generator set, determining the grid-connected wind turbine generator set according to the risk coefficient and the deviation value, and performing three-phase division on all grid-connected wind turbine generator sets. The present invention reduces the imbalance of the three-phase load of the power grid when the wind turbine generator set is connected to the grid, and improves the operation stability and operation efficiency of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power control, and in particular to an emergency safety control method for a wind power generator set. Background Art

[0002] As traditional fossil energy sources such as coal and oil are becoming increasingly depleted, and environmental pollution is becoming increasingly serious, and as the global demand for renewable energy continues to increase, wind power generation, as an important form of renewable energy, has become an important energy supply method. However, since the power generated by wind turbines is usually affected by meteorological factors and has the characteristics of randomness, intermittency and volatility, the energy generated is uncertain, which brings challenges to the safe and stable operation of the power grid.

[0003] With the gradual increase in the number of wind turbine generators, on the one hand, when multiple wind turbines are connected to the grid at the same time, an instantaneous voltage rise is generated, causing an instantaneous voltage shock to the large power grid, posing a serious safety hazard. On the other hand, sudden changes in wind speed may cause unstable wind turbine speeds, affecting the continuity and reliability of power generation. The power output of wind turbines will fluctuate greatly. The above reasons lead to the uncertainty and time-varying nature of the three-phase connection of wind turbines to the grid, resulting in unbalanced three-phase voltage of the grid. The unbalanced voltage will increase the energy loss of the grid system, thereby causing fluctuations in grid voltage, reducing the operating efficiency of the grid, and increasing line losses and equipment losses. Summary of the invention

[0004] The purpose of the present invention is to provide an emergency safety control method for a wind turbine generator set to solve the following technical problems:

[0005] When multiple wind turbines are connected to the grid or the fan speed changes suddenly, the wind power output connected to the grid will fluctuate greatly, causing an impact on the grid, leading to an imbalance in the three-phase voltage of the grid and increasing line losses and equipment losses.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An emergency safety control method for a wind turbine generator set comprises the following steps:

[0008] Obtaining predicted output power of all wind turbines in the region based on historical data, and determining target output power P' of the wind turbines according to the predicted output power;

[0009] A monitoring period H is preset, and the output power P of any wind turbine is obtained during the monitoring period, and a curve f(t) showing the change of the output power P of the wind turbine with time is drawn, and all coordinate points with a power generation of P' in the change curve f(t) are obtained and marked as reference points D1, D2, ..., Dn; D = (t i , P'), where ti is the time corresponding to the i-th reference point;

[0010] The change curve f(t) is divided into several sub-curves using any adjacent reference point as a dividing point, and any sub-curve is selected and the dividing value C=P'-max(f(t)) of the sub-curve is calculated, t∈[t i , t i+1 ], where t i and t i+1 Respectively represent the starting time point and the ending time point of the sub-curve segment. If the segmentation value C is greater than 0, the sub-curve is calibrated as a positive curve. If the segmentation value C is less than 0, the sub-curve is calibrated as a negative curve. The risk coefficient K and the deviation value Pc of the wind turbine are calculated according to the positive curve and the negative curve.

[0011] The grid-connected powers of all grid-connected wind turbines are determined according to the risk coefficient K and the deviation value Pc, and three-phase grid connection is divided for all grid-connected wind turbines according to the grid-connected powers.

[0012] As a further solution of the present invention: the specific calculation process of the risk coefficient K is:

[0013] ;

[0014] Among them, Tz is the time length corresponding to any of the positive curves, Tf is the time length corresponding to any of the negative curves; M is the number of all positive curves, m is the number of all negative curves; Czq is the evaluation score corresponding to any positive curve; Cfu is the evaluation score corresponding to any negative curve.

[0015] As a further solution of the present invention: the specific calculation process of the deviation value is:

[0016] Pc = Czq-Cfu;

[0017] Among them, the calculation formulas of Czq and Cfu are:

[0018] ;

[0019] .

[0020] As a further solution of the present invention, it also includes determining the grid-connected wind turbine generator set, and the specific process of determining the grid-connected wind turbine generator set is:

[0021] If Pc is less than 0, the wind turbine is marked as a wind turbine to be connected to the grid; if Pc is greater than 0, according to the calculation formula The preferred score is calculated. If the preferred score is greater than or equal to the preset threshold, the wind turbine is calibrated as a grid-connected wind turbine; if the preferred score is less than the preset threshold, the wind turbine is calibrated as a wind turbine to be connected to the grid, where K' is the preset standard risk coefficient and α is the preset correction coefficient.

[0022] As a further solution of the present invention: the calculation process of the grid-connected power is:

[0023] With the deviation value as the horizontal axis and the risk coefficient as the vertical axis, an evaluation coordinate system is constructed to generate feature points corresponding to all grid-connected wind turbines, calculate the Euclidean distance I between all feature points, generate the Euclidean distance set U, set the clustering control radius R, take any feature point as the center, calculate the feature point density ρ within the control radius R, obtain the feature point density mean within the control radius R of all feature points, and mark the feature point density mean as Minρ; for any feature point, if the corresponding feature point density ρ is greater than Minρ, a category cluster is generated with the feature point as the core point, and all adjacent feature points within the radius R are classified into the category cluster to obtain several category clusters; if there is a feature point density ρ within the control radius R of a non-core point feature set that is also greater than Minρ, the category cluster generated by the non-core point is merged with the original category cluster to generate several category clusters;

[0024] With the core point in each category cluster as the center and R as the radius, several subclusters are generated. The degree of aggregation of each subcluster is calculated and marked as DP. The calculation formula of DP is:

[0025] ;

[0026] Where z represents the number of feature points in the cluster, I represents the Euclidean distance, z0 represents the core point of the cluster, and v0 represents any feature point in the cluster;

[0027] In each category cluster, the subcluster with the smallest DP value is selected as the representative cluster of the category cluster, and a representative set of each category cluster is generated. The mean value Pc' of any horizontal axis coordinate of the corresponding feature point of any representative set is obtained. According to the calculation formula P 并 =Pc' / H is used to calculate the grid-connected power of the wind turbines corresponding to this category cluster.

[0028] As a further solution of the present invention: the process of setting the cluster control radius R is:

[0029] Taking any feature point as the center, calculate the Euclidean distance between the feature point and any other feature point, sum each Euclidean distance to get u, and get the control radius R based on the value u. The calculation formula is as follows:

[0030] ;

[0031] ;

[0032] Among them, u is the sum of the Euclidean distance data values ​​of all feature points, and I is the Euclidean distance between any feature points.

[0033] As a further solution of the present invention: the total grid-connected power W per unit time of the corresponding wind turbines in all category clusters is obtained, the phase standard value W / 3 is set, the phase sequence combination of several category clusters is performed by exhaustive method, the deviation value of each phase load in any phase sequence combination and the standard load value W / 3 is calculated, the evaluation score of each phase sequence combination is calculated according to the deviation value, and the phase sequence combination with the smallest evaluation score is selected as the phase sequence allocation of the current category cluster;

[0034] Evaluation score = (X1w-W / 3)²+(X2w-W / 3)²+(X3w-W / 3)²;

[0035] Among them, X1w is the grid-connected power of phase X1 corresponding to any phase sequence combination, X2w is the grid-connected power of phase X2 corresponding to the phase sequence combination, and X3w is the grid-connected power of phase X3 corresponding to the phase sequence combination.

[0036] Beneficial effects of the present invention:

[0037] The present invention first obtains the predicted output power of the wind turbine generator set, and determines the target output power of the wind turbine generator set according to the predicted output power, which is the basis for subsequent analysis. By presetting the monitoring period and drawing the output power change curve according to the target output power, the reference point is determined by the target output power, which provides a benchmark point for further analysis and can reflect the actual output power trend of the wind turbine generator set; then the segmentation value is calculated and the positive curve and the negative curve are determined; the preferred score is calculated according to the positive curve, and the difference score is calculated according to the negative curve. It should be noted that the larger the preferred score, the more the wind turbine generator set meets the grid connection requirements, and the larger the difference score, the poorer the output effect of the actual wind turbine generator set, and the less it meets the grid connection requirements; it is worth noting that the longer the duration corresponding to the positive curve, the more time the output effect of the wind turbine generator set is greater than the target level in the actual process, which can reflect that the wind turbine generator set meets the grid connection requirements; the longer the duration corresponding to the negative curve, the more time the wind turbine generator set outputs more than the target level in the actual process, which can reflect that the wind turbine generator set meets the grid connection requirements; the longer the duration corresponding to the negative curve, the more time the wind turbine generator set outputs more than the target level in the actual process, which can reflect that the wind turbine generator set meets the grid connection requirements; The more time the output effect is less than the target level, the less the wind turbine unit meets the grid connection requirements. Finally, the risk coefficient and the deviation value are calculated comprehensively, and the grid-connected wind turbine units are screened out according to the risk coefficient and the deviation value. Since the three phases connected to the microgrid are uncertain and time-varying, the three-phase voltage of the grid is unbalanced. The unbalanced voltage will increase the energy loss of the microgrid system, thereby causing fluctuations in the grid voltage, reducing the operating efficiency of the grid, and increasing line losses and equipment losses. Therefore, the present invention uses the deviation value as the horizontal axis and the risk coefficient as the vertical axis to construct an evaluation coordinate system, generate characteristic points corresponding to all grid-connected wind turbine units, calculate the Euclidean distance between the characteristic points, and set the clustering control radius according to the Euclidean distance, so that all wind turbine units can be divided into several representative category clusters. By dividing the three-phase phase sequence of the representative category clusters and determining the grid-connected phase of each wind turbine unit according to the division results, the imbalance of the three-phase load of the grid during grid connection is reduced, and the stability of grid operation is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention will be further described below in conjunction with the accompanying drawings.

[0039] Figure 1 It is a flow chart of an emergency safety control method for a wind turbine generator set according to the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] See also Figure 1As shown, the present invention is an emergency safety control method for a wind turbine generator set, comprising the following steps:

[0042] Based on historical data, the predicted output power of all wind turbines in the area is obtained, and the target output power P' of the wind turbine is determined based on the predicted output power. This step involves the collection and analysis of historical data, and the use of statistical methods or machine learning algorithms to predict future wind power generation. Once the target output power P' is determined, the next step is to set a monitoring period H, during which the actual output power P of each wind turbine is continuously tracked, and a curve f(t) is drawn showing its change over time.

[0043] After obtaining the variation curve f(t), we need to find all the time points when the generated power is equal to P'. These points are called reference points D1, D2, ..., Dn. Each reference point can be expressed in coordinate form as D=(t i , P'), where t i Represents the time corresponding to the i-th reference point. Using these reference points as the basis for segmentation, the entire change curve f(t) is divided into multiple sub-intervals or sub-curve segments.

[0044] For each segmented sub-curve segment, the maximum value max(f(t)) between the current reference point and the next reference point is calculated (here it is assumed that such a maximum value exists), and then the target output power P' is subtracted from this maximum value to obtain the so-called "segmentation value" C=P'-max(f(t)). If C>0, it means that the actual power generated during this time period exceeds the expected level, so this section of the curve is regarded as a positive curve; on the contrary, if C<0, it means that the actual output is lower than expected, and this section is marked as a negative curve.

[0045] Next, the risk factor K and deviation value Pc of each wind turbine are calculated based on the number and distribution of positive and negative curves. The risk factor reflects the stability and reliability of the equipment's operating status, while the deviation value provides information about the difference between actual performance and the theoretical model. Specifically, quantitative evaluation can be performed by counting factors such as the ratio and length of positive and negative curves.

[0046] The last step is to adjust the grid-connected power settings of each wind turbine according to the calculated risk factor K and deviation value Pc. At the same time, considering the overall load demand of the power grid, it is also necessary to reasonably distribute the three phases of all wind turbines connected to the system to ensure the balance and safety of power supply. The power stations that meet the grid connection requirements are combined to make the grid-connected voltage of each phase sequence in a balanced state. This process aims to maximize the use of renewable energy while ensuring power supply quality and system stability.

[0047] In another preferred embodiment of the present invention, the specific calculation process of the risk coefficient K is:

[0048] ;

[0049] Among them, Tz is the time length corresponding to any of the positive curves, Tf is the time length corresponding to any of the negative curves; M is the number of all positive curves, m is the number of all negative curves; Czq is the evaluation score corresponding to any positive curve; Cfu is the evaluation score corresponding to any negative curve.

[0050] In another preferred embodiment of the present invention, the specific calculation process of the deviation value is:

[0051] Pc = Czq-Cfu;

[0052] Among them, the calculation formulas of Czq and Cfu are:

[0053] ;

[0054] .

[0055] In another preferred embodiment of the present invention, it further includes determining a grid-connected wind turbine generator set, and the specific process of determining the grid-connected wind turbine generator set is:

[0056] If Pc is less than 0, the wind turbine is marked as a wind turbine to be connected to the grid; if Pc is greater than 0, according to the calculation formula The preferred score is calculated. If the preferred score is greater than or equal to the preset threshold, the wind turbine is calibrated as a grid-connected wind turbine; if the preferred score is less than the preset threshold, the wind turbine is calibrated as a wind turbine to be connected to the grid, where K' is the preset standard risk coefficient and α is the preset correction coefficient.

[0057] In another preferred embodiment of the present invention, the calculation process of the grid-connected power is:

[0058] First, an evaluation coordinate system is constructed with the deviation value as the horizontal axis and the risk coefficient as the vertical axis. In this coordinate system, the feature points corresponding to all grid-connected wind turbines are generated. Then, the Euclidean distance I between all feature points is calculated to generate the Euclidean distance set U.

[0059] Next, set the cluster control radius R. Take any feature point as the center and calculate the feature point density ρ within the control radius R. Obtain the mean feature point density within the control radius R of all feature points and mark the mean feature point density as Minρ.

[0060] For any feature point, if the corresponding feature point density ρ is greater than Minρ, a category cluster is generated with the feature point as the core point, and all adjacent feature points within the radius R are classified into the category cluster to obtain several category clusters. If there is a feature point density ρ within the control radius R of a non-core point feature set that is also greater than Minrho, the category cluster generated by the non-core point is merged with the original category cluster to generate several category clusters.

[0061] Then, several subclusters are generated with the core point in each category cluster as the center and R as the radius. The degree of aggregation of each subcluster is calculated and marked as DP. The calculation formula of DP is:

[0062] ;

[0063] Among them, z represents the number of feature points in the cluster, I is the Euclidean distance, z0 represents the core point of the cluster, and v0 represents any feature point in the cluster.

[0064] In each category cluster, the subcluster with the smallest DP value is selected as the representative cluster of the category cluster to generate a representative set of each category cluster. Obtain the mean value Pc' of any horizontal axis coordinate of the corresponding feature point of any representative set, and calculate it according to the calculation formula P 并 =Pc' / H is used to calculate the grid-connected power of the wind turbines corresponding to this category cluster.

[0065] This method classifies wind turbines by constructing an evaluation coordinate system and calculating the Euclidean distance between feature points, and determines the grid-connected power based on the degree of aggregation of the category cluster. This can more accurately reflect the actual operation of wind turbines and improve the calculation accuracy of grid-connected power. At the same time, by processing the feature set of non-core points, it can avoid affecting the accuracy of the overall results due to individual abnormal data.

[0066] In a preferred case of this embodiment, the process of setting the cluster control radius R is:

[0067] First, taking any feature point as the center, calculate the Euclidean distance between the feature point and all other feature points. The Euclidean distance is the straight-line distance between two points in multidimensional space. Sum each calculated Euclidean distance I to get a sum u, which reflects the average closeness between all feature points.

[0068] Based on the value u, we can determine a suitable control radius R. This process may involve some statistical methods or empirical rules to ensure that R can reasonably reflect the distribution characteristics of the feature points. The calculation formula is:

[0069] ;

[0070] ;

[0071] Once the control radius R is determined, it can be applied to the clustering algorithm. In this scenario, the control radius R is used to determine which feature points should be classified into the same cluster. Specifically, if the distance between two feature points is less than or equal to R, they are considered adjacent and can be classified into the same class. In practical applications, it may be necessary to adjust the control radius R according to the quality of the clustering results. If the clustering results are too scattered or too concentrated, it may be necessary to increase or decrease the value of R to achieve better clustering results.

[0072] In another preferred embodiment of the present invention, the total grid-connected power W per unit time of the corresponding wind turbines in all category clusters is obtained, the phase standard value W / 3 is set, and the phase sequence combination of several category clusters is performed by exhaustive method, and the deviation value of each phase load in any phase sequence combination from the standard load value W / 3 is calculated, and the evaluation score of each phase sequence combination is calculated according to the deviation value, and the phase sequence combination with the smallest evaluation score is selected as the phase sequence allocation of the current category cluster;

[0073] Evaluation score = (X1w-W / 3)²+(X2w-W / 3)²+(X3w-W / 3)²;

[0074] Among them, X1w is the grid-connected power of phase X1 corresponding to any phase sequence combination, X2w is the grid-connected power of phase X2 corresponding to the phase sequence combination, and X3w is the grid-connected power of phase X3 corresponding to the phase sequence combination.

[0075] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An emergency safety control method for a wind turbine generator set, characterized in that: The following steps are involved: Obtaining predicted output power of all wind turbines in the region based on historical data, and determining target output power P' of the wind turbines according to the predicted output power; A monitoring period H is preset, and the output power P of any wind turbine is obtained during the monitoring period, and a curve f(t) showing the change of the output power P of the wind turbine with time is drawn, and all coordinate points with a power generation of P' in the change curve f(t) are obtained and marked as reference points D1, D2, ..., Dn; D = (t i , P'), where t i is the time corresponding to the i-th reference point; The change curve f(t) is divided into several sub-curves using any adjacent reference point as a dividing point, and any sub-curve is selected and the dividing value C=P'-max(f(t)) of the sub-curve is calculated, t∈[t i , t i+1 ], where t i and t i+1 Respectively represent the starting time point and the ending time point of the sub-curve segment. If the segmentation value C is greater than 0, the sub-curve is calibrated as a positive curve. If the segmentation value C is less than 0, the sub-curve is calibrated as a negative curve. The risk coefficient K and the deviation value Pc of the wind turbine are calculated according to the positive curve and the negative curve. Determine the grid-connected power of all grid-connected wind turbines according to the risk coefficient K and the deviation value Pc, and divide all grid-connected wind turbines into three-phase grid-connected groups according to the grid-connected power; The specific calculation process of risk factor K is: ； Wherein, Tz is the time length corresponding to any of the positive curves, Tf is the time length corresponding to any of the negative curves; M is the number of all positive curves, m is the number of all negative curves; Czq is the evaluation score corresponding to any positive curve; Cfu is the evaluation score corresponding to any negative curve; The specific calculation process of the deviation value is: Pc = Czq-Cfu; Among them, the calculation formulas of Czq and Cfu are: ； 。 2. The emergency safety control method of a wind turbine generator set according to claim 1, characterized in that: It also includes determining the wind turbines connected to the grid. The specific process of determining the wind turbines connected to the grid is as follows: If Pc is less than 0, the wind turbine is marked as a wind turbine to be connected to the grid; if Pc is greater than 0, according to the calculation formula The preferred score is calculated. If the preferred score is greater than or equal to the preset threshold, the wind turbine is calibrated as a grid-connected wind turbine; if the preferred score is less than the preset threshold, the wind turbine is calibrated as a wind turbine to be connected to the grid, where K' is the preset standard risk coefficient and α is the preset correction coefficient.

3. The emergency safety control method of a wind turbine generator set according to claim 1, characterized in that: The calculation process of grid-connected power is: With the deviation value as the horizontal axis and the risk coefficient as the vertical axis, an evaluation coordinate system is constructed to generate feature points corresponding to all grid-connected wind turbines, calculate the Euclidean distance I between all feature points, generate the Euclidean distance set U, set the clustering control radius R, take any feature point as the center, calculate the feature point density ρ within the control radius R, obtain the feature point density mean within the control radius R of all feature points, and mark the feature point density mean as Minρ; for any feature point, if the corresponding feature point density ρ is greater than Minρ, a category cluster is generated with the feature point as the core point, and all adjacent feature points within the radius R are classified into the category cluster to obtain several category clusters; if there is a feature point density ρ within the control radius R of a non-core point feature set that is also greater than Minρ, the category cluster generated by the non-core point is merged with the original category cluster to generate several category clusters; With the core point in each category cluster as the center and R as the radius, several subclusters are generated. The degree of aggregation of each subcluster is calculated and marked as DP. The calculation formula of DP is: ； Where z represents the number of feature points in the cluster, I represents the Euclidean distance, z0 represents the core point of the cluster, and v0 represents any feature point in the cluster; In each category cluster, the subcluster with the smallest DP value is selected as the representative cluster of the category cluster, and a representative set of each category cluster is generated. The mean value Pc' of any horizontal axis coordinate of the corresponding feature point of any representative set is obtained. According to the calculation formula P 并 =Pc' / H to calculate the grid-connected power of the wind turbines corresponding to this category cluster.

4. The emergency safety control method of a wind turbine generator set according to claim 3, characterized in that: The process of setting the cluster control radius R is: Taking any feature point as the center, calculate the Euclidean distance between the feature point and any other feature point, sum each Euclidean distance to get u, and get the control radius R based on the value u. The calculation formula is as follows: ； ； Among them, u is the sum of the Euclidean distance data values ​​of all feature points, and I is the Euclidean distance between any feature points.

5. The emergency safety control method of a wind turbine generator set according to claim 3, characterized in that: Obtain the total grid-connected power W per unit time of the corresponding wind turbines in all category clusters, set the phase standard value W / 3, use the exhaustive method to combine the phase sequences of several category clusters, calculate the deviation value of each phase load in any phase sequence combination from the standard load value W / 3, calculate the evaluation score of each phase sequence combination based on the deviation value, and select the phase sequence combination with the smallest evaluation score as the phase sequence allocation of the current category cluster; Evaluation score = (X1w-W / 3)²+(X2w-W / 3)²+(X3w-W / 3)²; Among them, X1w is the grid-connected power of phase X1 corresponding to any phase sequence combination, X2w is the grid-connected power of phase X2 corresponding to the phase sequence combination, and X3w is the grid-connected power of phase X3 corresponding to the phase sequence combination.

Images