Power optimization dispatching methods, devices and equipment for renewable energy grid connection scenarios

Abstract

The present application relates to the technical field of power data optimization processing, and specifically to a power optimization dispatching method, device and equipment in a new energy grid connection scenario, the method comprising: obtaining the power influence intensity of each wind turbine according to the numerical distribution and synchronous change of the electric power vector of each wind turbine in the frequency domain and its corresponding energy amplitude at all frequencies; obtaining the significant fluctuation weight of each wind turbine according to the comparison of the power influence intensity before and after the electric power vector of each wind turbine is combined with other wind turbines; obtaining the array power fluctuation degree according to the significant fluctuation weight and power influence intensity of all wind turbines; predicting the output power at a moment after a preset time period at the current moment according to the array power fluctuation degree and output power at the current moment, obtaining the output power prediction value; dispatching power equipment according to the output power prediction value. The present application can improve the dispatching accuracy of power equipment.

Description

Technical Field

[0001] The present application relates to the technical field of power data optimization processing, and specifically to power optimization scheduling methods, devices and equipment in a new energy grid-connected scenario. Background Art

[0002] Wind power generation is a common renewable energy generation technology. It has a large power generation capacity and clean energy, and is an excellent energy generation method. However, due to the large fluctuations in wind energy, the fluctuations in wind power generation are also large. When wind power generation accounts for a large proportion of the power grid, the high volatility of wind power will cause large fluctuations in the power grid, resulting in unstable power supply.

[0003] In the process of wind power grid connection, in order to ensure the stability of power supply to the grid, it is necessary to obtain the fluctuation characteristics of wind power through wind power prediction and optimize power dispatch. At present, wind power prediction mainly relies on neural network technology. However, the existing prediction method fails to fully consider the wind field characteristic information brought by the power output difference between different wind turbines in the same wind farm, resulting in a decrease in the accuracy of power prediction, thereby affecting the accuracy of power equipment dispatch. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method, device and equipment for optimizing power dispatching in a renewable energy grid-connected scenario. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for optimizing power dispatch in a new energy grid-connected scenario, the method comprising the following steps:

[0006] S1, collecting the electric power vector of each wind turbine in the wind power array;

[0007] S2, according to the numerical distribution and synchronous change of the electric power vector of each wind turbine generator set at all frequencies in the frequency domain and the corresponding energy amplitude, the power impact intensity of each wind turbine generator set is obtained;

[0008] S3, according to the comparison of the power influence intensity of each wind turbine group before and after the electric power vector combination with other wind turbine groups, the fluctuation significance of each wind turbine group relative to other wind turbine groups is obtained; according to the fluctuation significance of each wind turbine group and all other wind turbine groups, the fluctuation significance weight of each wind turbine group is obtained; according to the fluctuation significance weight and power influence intensity of all wind turbine groups, the array power fluctuation degree is obtained;

[0009] S4, obtaining the output power of the wind power array at each moment; training a neural network based on the output power of the wind power array at each moment and at a moment a period of time before and the array power fluctuation to obtain a neural prediction model, using the neural prediction model to predict the output power at a moment a period of time after the current moment to obtain an output power prediction value; dispatching power equipment based on the comparison result between the output power prediction value and the preset calibrated power.

[0010] Furthermore, the method for obtaining the power impact intensity of each wind turbine generator set includes:

[0011] Perform frequency domain analysis on the electric power vector of each wind turbine generator set to obtain the electric power spectrum of each wind turbine generator set; wherein each element in the electric power spectrum is the energy amplitude of the electric power vector at each frequency;

[0012] The ratio of each element except the first element in the electric power spectrum to the first element is recorded as the energy proportion at the frequency corresponding to each element; the mean value of all elements in the electric power vector of each wind turbine is calculated and recorded as the power mean value of each wind turbine;

[0013] Calculate the product of the frequency corresponding to each element and the energy proportion at the corresponding frequency; obtain the sum of the products of all elements; and take the product of the sum and the power average of each wind turbine as the power impact intensity of each wind turbine.

[0014] Further, the fluctuation significance of each wind turbine set relative to other wind turbine sets is obtained based on the comparison of the power impact intensity before and after the electric power vector combination of each wind turbine set with other wind turbine sets, including:

[0015] Calculate the sum vector of the electric power vectors of each wind turbine group and other wind turbine groups, and record it as the power sum vector of each wind turbine group and other wind turbine groups;

[0016] For the power and vector of each wind turbine group and other wind turbine groups, the combined impact intensity of each wind turbine group and other wind turbine groups is obtained according to the calculation method of the power impact intensity of each wind turbine group;

[0017] According to the comparison result of the sum of the power influence intensity of the electric power vector of each wind turbine set and other wind turbine sets and the combined influence intensity, the fluctuation significance degree of each wind turbine set relative to other wind turbine sets is obtained.

[0018] Further, the comparison result of the power influence intensity of each wind turbine group and the power vector of other wind turbine groups with the combined influence intensity is used to obtain the fluctuation significance of each wind turbine group relative to other wind turbine groups, including: recording the fluctuation significance of the nth wind turbine group relative to the hth wind turbine group as , ;in, is the power impact intensity of the nth wind turbine; is the power impact intensity of the h-th wind turbine; is the combined impact intensity of the nth and hth wind turbines.

[0019] Furthermore, the fluctuation significance weight of each wind turbine is obtained according to the fluctuation significance degree of each wind turbine and all other wind turbines, including:

[0020] The sum of the fluctuation significance of each wind turbine and all other wind turbines is calculated and recorded as the fluctuation significance feature of each wind turbine; the normalized value of the fluctuation significance feature of each wind turbine is used as the fluctuation significance weight of each wind turbine.

[0021] Furthermore, the array power fluctuation degree is obtained according to the fluctuation significance weights and power impact strengths of all wind turbines, including:

[0022] The weighted average of the power impact intensity of all wind turbines is calculated using the significant fluctuation weight of each wind turbine as the weight to obtain the array power fluctuation degree.

[0023] Furthermore, the neural network is trained according to the output power of the wind power array and the array power fluctuation at each moment and a period of time before, to obtain a neural prediction model, and the output power at a moment a period of time after the current moment is predicted using the neural prediction model to obtain an output power prediction value, including:

[0024] Record a preset number of moments before the current moment as training moments, arrange the output powers of the wind power arrays at all training moments in chronological order, and obtain a power label vector;

[0025] The output power, array power fluctuation and power label vector of the wind power array at all training moments before the preset time period are used to train the neural network to obtain a trained neural prediction model;

[0026] The output power and array power fluctuation of the wind power array at the current moment are predicted using a neural prediction model to obtain the output power prediction value after the preset time period.

[0027] Furthermore, the dispatching of electric power equipment according to the comparison result between the output power prediction value and the preset calibrated power includes: when the output power prediction value is lower than the preset value, starting the thermal power generating set.

[0028] In a second aspect, an embodiment of the present application provides a power optimization dispatching device in a new energy grid-connected scenario, and the power optimization dispatching device includes:

[0029] The power data acquisition module is used to collect the electric power vector of each wind turbine in the wind power array;

[0030] The power data processing module is used to obtain the power impact intensity of each wind turbine according to the numerical distribution and synchronous change of the electric power vector of each wind turbine in the frequency domain and the energy amplitude corresponding to it; obtain the fluctuation significance of each wind turbine relative to other wind turbines according to the comparison of the power impact intensity before and after the electric power vector of each wind turbine is combined with other wind turbines; obtain the fluctuation significance weight of each wind turbine according to the fluctuation significance of each wind turbine and all other wind turbines; obtain the array power fluctuation degree according to the fluctuation significance weight and power impact intensity of all wind turbines;

[0031] The power equipment dispatching module is used to obtain the output power of the wind power array at each moment; predict the output power at a moment after a preset time period at the current moment based on the array power fluctuation and output power at the current moment to obtain an output power prediction value; and dispatch the power equipment based on the output power prediction value.

[0032] In a third aspect, an embodiment of the present application also provides an electric power optimization scheduling device in a new energy grid-connected scenario, the device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of any one of the above methods when executing the computer program.

[0033] It can be seen from the above embodiments that the power optimization scheduling method, device and equipment in the renewable energy grid connection scenario provided by the embodiments of the present application have at least the following beneficial effects:

[0034] This application aims to solve the problem of inaccurate wind power generation prediction leading to power equipment scheduling errors, analyzes the influence of the stability of power generation data and frequency domain characteristics on the accuracy of power generation prediction results, first obtains the electric power vector of each wind turbine in the wind power array, and obtains the frequency spectrum of the electric power vector, and uses the energy amplitude of the non-DC component to characterize the output power fluctuation characteristics of a single wind turbine; further, through the distribution of power fluctuation sequences at different frequencies, the output power of a single wind turbine is collected at the same time, the power impact intensity is calculated, and the influence of a single wind turbine on the output power of the entire wind turbine array is characterized; further, through the power impact of different wind turbines Intensity, construct the degree of fluctuation significance, characterize the degree of complementarity of the output power characteristics of any two wind turbines, and use the wind turbine whose output characteristics are less complementary to the output characteristics of other wind turbines as the wind turbine that can better characterize the wind force change characteristics of the wind farm. Finally, combined with the power impact intensity, the array power fluctuation degree is obtained to characterize the power fluctuation of the wind power array at the current moment. This embodiment extracts features from the electric power data of the wind turbines in the wind power array, obtains the features of the output power fluctuation of the wind power array, and uses this to predict the electric power of the wind power array, so as to obtain more accurate wind power prediction results, optimize the scheduling of power equipment according to the prediction results, and improve the scheduling accuracy of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of the steps of a method for optimizing power dispatch in a renewable energy grid-connected scenario provided by an embodiment of the present application;

[0037] Figure 2 A schematic diagram of electric power vectors provided for one embodiment of the present application;

[0038] Figure 3 A structural block diagram of a power optimization and dispatching device in a renewable energy grid-connected scenario provided in one embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following is a detailed description of the power optimization dispatching method, device and equipment in the new energy grid connection scenario proposed in this application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless otherwise specified and limited, terms such as "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such articles or devices. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the article or device including the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items. All technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of this application.

[0041] The following is a detailed description of the specific scheme of the power optimization scheduling method, device and equipment in the renewable energy grid-connected scenario provided by the present application with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flowchart of the steps of a power optimization dispatching method in a new energy grid-connected scenario provided by an embodiment of the present application, the method comprising the following steps:

[0043] S1, collecting the electric power vector of each wind turbine in the wind power array.

[0044] Wind power generation in renewable energy grid connection usually installs multiple wind turbines in the same wind farm to form a wind power array, which supplies power to the grid as a unit. In the process of wind farm grid connection, in order to reduce the impact of wind power on grid stability, it is necessary to optimize the power dispatch of the entire wind power array. Traditional methods usually ignore the power characteristics of each unit in the wind power array and only predict the power based on the output characteristics of the entire wind power array.

[0045] Therefore, this embodiment extracts the internal features of the wind turbine array to obtain a more accurate wind power prediction result. There are a total of N = 100 wind turbines in the wind turbine array. At the current time r, the power generation data of the past 15 minutes is obtained for each wind turbine. The power generation data represents the output power of the wind turbine at different times. The sampling period of the power generation data is Ts = 1 second, and the data length of the power generation data is T = 900. Finally, for the nth wind turbine, a vector of length T is obtained at time r, which is recorded as the electric power vector, such as Figure 2 As shown, it is a visualization graph of the electric power vector, where the horizontal axis is time and the vertical axis is the wind power output by the nth wind turbine unit collected at each time point.

[0046] S2, according to the numerical distribution and synchronous change of the electric power vector of each wind turbine generator set at all frequencies in the frequency domain and the corresponding energy amplitude, the power impact intensity of each wind turbine generator set is obtained.

[0047] The electric power vector of a wind turbine contains information about the fluctuation of its generated electricity; the stronger the electric power fluctuation of a wind turbine, the stronger the fluctuation of the entire wind turbine array. The more the frequency component of the electric power vector of a wind turbine is biased towards high-frequency components, the stronger the fluctuation of its output power; at the same time, when judging the output power fluctuation intensity of a wind turbine through frequency domain information, the DC component of the electric power vector usually represents the size of the stable electric power provided by the wind turbine, so this embodiment compares the energy amplitude of other frequency components with the energy amplitude of the DC component to judge the power fluctuation degree of the wind turbine.

[0048] Specifically, the electric power vector of each wind turbine is analyzed in the frequency domain to obtain the electric power spectrum of each wind turbine. Taking the nth wind turbine as an example, the electric power vector of the nth wind turbine is used as input. This embodiment adopts the FFT fast Fourier transform algorithm for calculation. The algorithm parameters include: the frequency range is 0 to 100 kHz, and the frequency sampling interval is 100 Hz. The output is the electric power spectrum of the nth wind turbine.

[0049] Among them, the FFT algorithm is a common technology in the field of electrical signal processing, and the specific implementation process will not be repeated; the implementer can use other frequency domain analysis algorithms according to the actual situation, and this application does not limit it. According to the necessary parameters set, it can be seen that the output electric power spectrum is a vector with a length of M=1001, where the mth element of the electric power spectrum represents the electric power vector in The energy amplitude at a frequency.

[0050] The ratio of each element except the first element in the electric power spectrum to the first element is recorded as the energy proportion at the corresponding frequency of each element; the sequence composed of the energy proportions at the corresponding frequencies of all elements is recorded as the power fluctuation sequence. Specifically, for the electric power spectrum, the value of the first element represents the energy amplitude of the electric power vector on the DC component. The larger the amplitude of this part of energy, the more stable electric energy the nth wind turbine has provided in the past 15 minutes; therefore, this embodiment further divides the values ​​of the second to Mth elements of the electric power spectrum by the value of the first element of the electric power spectrum to obtain a sequence with a length of K=1000, which is recorded as the power fluctuation sequence. , and the kth element in the sequence The corresponding frequency position is .

[0051] Among them, the ratio of the non-DC component energy amplitude to the DC component energy amplitude in the electric power vector represents the ratio of the unstable electric energy provided by the nth wind turbine to the stable electric energy provided. The larger the ratio, the more unstable the electric energy provided by the nth wind turbine.

[0052] Finally, the larger the sequence value in the obtained power fluctuation sequence and the larger the sequence value at the end of the power fluctuation sequence, the more unstable the electric energy provided by the nth wind turbine is, and the more significant the fluctuation characteristics of the electric energy it provides are.

[0053] The power fluctuation sequence characterizes the ratio of stable power and unstable power provided by the wind turbine, and reflects the fluctuation of power provided by the wind turbine. To further estimate the fluctuation impact of the wind turbine on the entire wind power array, this embodiment combines the power fluctuation sequence with the power generation of the wind turbine to calculate the power impact intensity:

[0054] Specifically, the mean value of all elements in the electric power vector of each wind turbine is calculated and recorded as the mean power value of each wind turbine , the wind turbine with a larger power average value generates more electric energy, and its power impact intensity on the wind power array is greater. Further, the power impact intensity is calculated;

[0055]

[0056] in, is the power impact intensity of the nth wind turbine, is the mean power of the nth wind turbine, is the power fluctuation sequence length, is the kth element value of the power fluctuation sequence of the nth wind turbine, is the frequency corresponding to the kth element value of the power fluctuation sequence.

[0057] Among them, the larger the kth sequence value of the power fluctuation sequence, and the higher the frequency position frequency corresponding to the kth sequence value at this time, the more significant the power fluctuation characteristics of the nth wind turbine group, and the more unstable the power it provides; the larger the power mean value, the more total power provided by the nth wind turbine group, and the more severe the impact of its power fluctuation on the wind power array. Finally, the power impact intensity is obtained, which represents the impact intensity of the mth wind turbine group on the output power fluctuation characteristics of the wind power array. The greater the power impact intensity of the nth wind turbine group, the greater the power fluctuation of the wind power array output caused by the power provided by the nth wind turbine group.

[0058] S3, based on the comparison of the power impact intensity of each wind turbine before and after the electric power vector combination with other wind turbines, the fluctuation significance of each wind turbine relative to other wind turbines is obtained; based on the fluctuation significance of each wind turbine with all other wind turbines, the fluctuation significance weight of each wind turbine is obtained; based on the fluctuation significance weights and power impact intensity of all wind turbines, the array power fluctuation degree is obtained.

[0059] The above-mentioned power impact intensity characterizes the fluctuation effect of a single wind turbine on the array power. When the power fluctuation of the entire wind power array is characterized by the wind turbine, it is necessary to consider the complementary characteristics of the output power of different wind turbines in the same array. Usually, in a wind power array, when the output power of all wind turbines increases or decreases at the same time, it usually represents the change of wind force in the entire area of ​​the wind farm. At this time, the output power of the wind power array fluctuates greatly; when power complementarity occurs between different wind turbines in the wind power array, that is, the power output of one wind turbine increases, while the power output of another wind turbine decreases synchronously, it means that the wind force change in the wind farm is only the wind force change in the local area, and this wind force change has a small impact on the power output of the entire wind power array.

[0060] Therefore, this embodiment determines the wind force variation of the entire wind farm through the complementary characteristics of the output power of different wind turbines in the wind turbine array. If the output power of a wind turbine does not reflect the complementary characteristics compared with other wind turbines, its output characteristics can better represent the wind force variation in the wind farm, and can more significantly reflect the output power fluctuation characteristics of the wind turbine array.

[0061] Specifically, the power impact intensity comparison between each wind turbine and the other wind turbines before and after the electric power vector combination is compared to obtain the fluctuation significance of each wind turbine relative to the other wind turbines. Taking the nth wind turbine as an example, the non-complementarity of the output power of the nth wind turbine and the hth wind turbine is calculated to characterize the fluctuation significance weight of the nth wind turbine relative to the hth wind turbine. The calculation method is:

[0062] Obtain the electric power vectors of the nth and hth wind turbines respectively, and calculate the power impact intensity of the two respectively. , ; Further, the electric power vector sum vector of the nth and hth wind turbines is recorded as the power sum vector, which is equivalent to treating the nth and hth wind turbines as a virtual wind turbine, and using the power sum vector to characterize the output power characteristics of the virtual wind turbine. The power impact intensity calculation method of this embodiment is further used to obtain the power impact intensity of the power sum vector, which is recorded as the combined impact intensity According to the comparison result of the sum of the power influence intensity of the electric power vector of each wind turbine generator set and other wind turbine generator sets and the combined influence intensity, the fluctuation significance degree of each wind turbine generator set relative to other wind turbine generator sets is obtained.

[0063] In one embodiment of the present application, the calculation formula for the fluctuation significance is:

[0064]

[0065] in, is the fluctuation significance of the nth wind turbine relative to the hth wind turbine; is the power impact intensity of the nth wind turbine; is the power impact intensity of the h-th wind turbine; is the combined impact intensity of the nth and hth wind turbines.

[0066] The sum of the power impact intensities of the two wind turbines represents the degree of fluctuation of the output power when the two wind turbines are used as power output elements separately. Characterizes the degree of fluctuation of power output when two wind turbines are regarded as an integral power output element; if the output power of the two turbines reflects complementary characteristics, the combined impact strength will be lower than the sum of the power impact strengths of the two wind turbines, and the fluctuation significance will be small; if they are completely complementary, the combined impact strength will be equal to 0, and the fluctuation significance will be equal to 0. Therefore, the greater the fluctuation significance, the less complementary the power fluctuation of the nth wind turbine is to the power fluctuation of the hth wind turbine, and the more the nth wind turbine can reflect the wind power fluctuation in the entire wind farm.

[0067] Furthermore, the significant fluctuation weight is calculated for the nth wind turbine group, and the calculation method is:

[0068] For the nth wind turbine, obtain the fluctuation significance degree with all other wind turbines and sum them up, which is recorded as the fluctuation significance feature of the nth wind turbine. The larger the fluctuation significance feature, the more the nth wind turbine can reflect the wind fluctuation in the whole wind farm. The normalized value of the fluctuation significance feature of each wind turbine is used as the fluctuation significance weight of each wind turbine.

[0069] As an embodiment of the present application, the method for normalizing the significant fluctuation characteristics of each wind turbine is: taking the ratio of the significant fluctuation characteristics of each wind turbine to the sum of the significant fluctuation characteristics of all wind turbines as the significant fluctuation weight of each wind turbine.

[0070] As another embodiment of the present application, a linear normalization function may be used to calculate the normalized value, which is not limited in the present application.

[0071] Among them, the significant fluctuation weight is the normalized value of the significant fluctuation feature. The larger the value, the more the wind turbine can reflect the wind fluctuation in the entire wind farm.

[0072] Finally, taking the significant fluctuation weight of each wind turbine as the weight, the weighted average of the power impact intensity of all wind turbines is calculated to obtain the array power fluctuation degree. The larger the value, the more significant the overall trend of wind force change in the wind field where the wind turbine array is located at the current moment, and the more likely the wind turbine array is to have large output power fluctuations in the next period of time. In this case, other power equipment should be started in advance to generate electricity and supply energy to the power grid to ensure the stability of new energy grid connection.

[0073] S4, obtaining the output power of the wind power array at each moment; predicting the output power at a moment after a preset time period at the current moment based on the array power fluctuation and output power at the current moment, to obtain an output power prediction value; and dispatching power equipment based on the output power prediction value.

[0074] In traditional wind power prediction models, usually only the total output power of the entire wind power array is considered, without in-depth mining of the power data of each wind turbine in the array to more carefully evaluate the power generation status of the wind farm. This embodiment analyzes the differences in the output power of each wind turbine in the wind power array, extracts the characteristics reflecting the volatility of the array output power, obtains the electric energy fluctuation vector, and judges whether the power generation state of the wind power array is stable based on this. This method improves the accuracy of judging the power generation state of the wind power array, provides a basis for building a more accurate training data set, and makes the final wind power array output power prediction results more accurate and reliable. In turn, the output power of the wind power array can be further predicted based on this, and power optimization scheduling can be performed. The specific method is:

[0075] The prediction results are used to control the power generation of other power equipment to supplement the shortage of wind power generation and ensure the stability of power grid. In this embodiment, power optimization scheduling is completed by controlling the power generation of thermal power generators fueled by coal, oil or combustible gas. Since thermal power generators have slow response and long ramp-up time, it is necessary to predict the power grid fluctuations caused by wind power arrays in advance. In this embodiment, the wind power array output power in 15 minutes is predicted by the wind power array data at the current moment to complete the power prediction.

[0076] This embodiment uses LSTM network to predict electric energy, so a training data set is first obtained. The method for obtaining the training data set is as follows:

[0077] The G=10000 moments before the current moment are recorded as training moments, and the output power of the wind power array at all training moments is arranged in chronological order to obtain a power label vector; the output power and array power fluctuation of the wind power array at the moment 15 minutes before all training moments are used as the data input of the training data set, and the power label vector is used as the label part of the training data set to train the neural network to obtain a trained neural prediction model.

[0078] Specifically, the input part of the training data set is used as input to the LSTM neural network model, and the output data of the neural network model is recorded as the predicted power. Optional: The activation function uses the ReLU function, and the optimizer uses the Adam optimizer; the loss function is obtained by taking the mean square error between the predicted power and the element value of the power label vector; and finally a trained LSTM neural network model is obtained.

[0079] The output power and array power fluctuation of the wind power array at the current moment are predicted using the neural prediction model to obtain the output power prediction value 15 minutes later. The training process of the LSTM neural network model is a common technology in the field of power control, and the details will not be repeated here.

[0080] Finally, in the actual power optimization scenario, the array power fluctuation at time r is calculated according to the method described in this embodiment, and the output power of the wind power array at time r is obtained. Both are input into the trained LSTM neural network model to obtain the output power prediction value of the wind power array after 15 minutes.

[0081] Further, the power forecast is used to control the generation of other power equipment to supplement the shortage of wind power generation and ensure the stability of the power grid. Optionally, the optimization scheduling method is: set the rated power to 50 megawatts, and start the thermal power generator when the output power forecast value after 15 minutes is lower than 50 megawatts to provide additional power support for the power grid and ensure the stability of the power grid.

[0082] This embodiment further obtains the characteristics of wind power array output power fluctuation by extracting the electric power data of wind turbines in the wind power array, and uses it to predict the electric power of the wind power array to complete power optimization scheduling. Compared with the traditional power prediction based on the LSTM model, this embodiment extracts the output power fluctuation information inside the wind power array, obtains more detailed electric power fluctuation characteristics, and can obtain more accurate wind power prediction results to avoid errors in power optimization scheduling.

[0083] See also Figure 3 , Figure 3Schematic diagram of the structure of the power optimization dispatching device in the renewable energy grid-connected scenario provided by the embodiment of the present application. In this embodiment, each unit included in the terminal is used to execute each step in the embodiment corresponding to the power optimization dispatching method in the renewable energy grid-connected scenario. Figure 3 The power optimization dispatching equipment includes: power data acquisition module, power data processing module, and power equipment dispatching module.

[0084] The power data acquisition module is used to collect the electric power vector of each wind turbine in the wind power array;

[0085] The power data processing module is used to obtain the power impact intensity of each wind turbine according to the numerical distribution and synchronous change of the electric power vector of each wind turbine in the frequency domain and the energy amplitude corresponding to it; obtain the fluctuation significance of each wind turbine relative to other wind turbines according to the comparison of the power impact intensity before and after the electric power vector of each wind turbine is combined with other wind turbines; obtain the fluctuation significance weight of each wind turbine according to the fluctuation significance of each wind turbine and all other wind turbines; obtain the array power fluctuation degree according to the fluctuation significance weight and power impact intensity of all wind turbines;

[0086] The power equipment dispatching module is used to obtain the output power of the wind power array at each moment; predict the output power at a moment after a preset time period at the current moment based on the array power fluctuation and output power at the current moment to obtain an output power prediction value; and dispatch the power equipment based on the output power prediction value.

[0087] Based on the same inventive concept as the above method, an embodiment of the present application also provides an electric power optimization scheduling device in a new energy grid-connected scenario, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the electric power optimization scheduling methods in the above-mentioned new energy grid-connected scenario are implemented.

[0088] It is to be understood that the sequence of the embodiments of the present application described above is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of the present specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0090] The above content is only an implementation method of the present application and is not intended to limit the scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the protection scope of the present application.

Claims

1. The power optimization dispatching method in the scenario of new energy grid connection is characterized by: The method comprises the following steps: S1, collecting the electric power vector of each wind turbine in the wind power array; S2, according to the numerical distribution and synchronous change of the electric power vector of each wind turbine generator set at all frequencies in the frequency domain and the corresponding energy amplitude, the power impact intensity of each wind turbine generator set is obtained; S3, according to the comparison of the power influence intensity of each wind turbine group before and after the electric power vector combination with other wind turbine groups, the fluctuation significance of each wind turbine group relative to other wind turbine groups is obtained; according to the fluctuation significance of each wind turbine group and all other wind turbine groups, the fluctuation significance weight of each wind turbine group is obtained; according to the fluctuation significance weight and power influence intensity of all wind turbine groups, the array power fluctuation degree is obtained; S4, obtaining the output power of the wind power array at each moment; training a neural network based on the output power of the wind power array at each moment and the moment before a period of time and the array power fluctuation to obtain a neural prediction model, using the neural prediction model to predict the output power at a moment after a period of time from the current moment to obtain an output power prediction value; dispatching power equipment based on the comparison result between the output power prediction value and the preset calibrated power; The neural network is trained according to the output power of the wind power array and the array power fluctuation at each moment and the moment before a period of time to obtain a neural prediction model, and the output power at a moment after a period of time at the current moment is predicted using the neural prediction model to obtain an output power prediction value, including: Record a preset number of moments before the current moment as training moments, arrange the output powers of the wind power arrays at all training moments in chronological order, and obtain a power label vector; The output power, array power fluctuation and power label vector of the wind power array at all training moments before the preset time period are used to train the neural network to obtain a trained neural prediction model; The output power and array power fluctuation of the wind power array at the current moment are predicted using a neural prediction model to obtain the output power prediction value after the preset time period; The dispatching of electric power equipment according to the comparison result between the output power prediction value and the preset rated power includes: when the output power prediction value is lower than the preset rated power, starting the thermal power generating set; The method for obtaining the power impact intensity of each wind turbine generator set includes: Perform frequency domain analysis on the electric power vector of each wind turbine generator set to obtain the electric power spectrum of each wind turbine generator set; wherein each element in the electric power spectrum is the energy amplitude of the electric power vector at each frequency; The ratio of each element except the first element in the electric power spectrum to the first element is recorded as the energy proportion at the frequency corresponding to each element; the mean value of all elements in the electric power vector of each wind turbine is calculated and recorded as the power mean value of each wind turbine; Calculate the product of the frequency corresponding to each element and the energy proportion at the corresponding frequency; obtain the sum of the products of all elements; and take the product of the sum and the power average of each wind turbine as the power impact intensity of each wind turbine.

2. The power optimization dispatching method in the scenario of new energy grid connection according to claim 1, characterized in that: The comparison of the power impact intensity of each wind turbine generator set before and after the electric power vector combination with other wind turbine generator sets is used to obtain the fluctuation significance of each wind turbine generator set relative to other wind turbine generator sets, including: Calculate the sum vector of the electric power vectors of each wind turbine group and other wind turbine groups, and record it as the power sum vector of each wind turbine group and other wind turbine groups; For the power and vector of each wind turbine group and other wind turbine groups, the combined impact intensity of each wind turbine group and other wind turbine groups is obtained according to the calculation method of the power impact intensity of each wind turbine group; According to the comparison result of the sum of the power influence intensity of the electric power vector of each wind turbine set and other wind turbine sets and the combined influence intensity, the fluctuation significance degree of each wind turbine set relative to other wind turbine sets is obtained.

3. The power optimization dispatching method in the scenario of new energy grid connection as claimed in claim 2 is characterized in that: The method of obtaining the fluctuation significance of each wind turbine relative to other wind turbines according to the comparison result of the sum of the power influence strength of the electric power vector of each wind turbine and other wind turbines and the combined influence strength includes: recording the fluctuation significance of the nth wind turbine relative to the hth wind turbine as , ;in, is the power impact intensity of the nth wind turbine; is the power impact intensity of the h-th wind turbine; is the combined impact intensity of the nth and hth wind turbines.

4. The power optimization dispatching method in the renewable energy grid connection scenario according to claim 1, characterized in that: The fluctuation significance weight of each wind turbine is obtained according to the fluctuation significance degree of each wind turbine and all other wind turbines, including: The sum of the fluctuation significance of each wind turbine and all other wind turbines is calculated and recorded as the fluctuation significance feature of each wind turbine; the normalized value of the fluctuation significance feature of each wind turbine is used as the fluctuation significance weight of each wind turbine.

5. The power optimization dispatching method in the scenario of new energy grid connection according to claim 1, characterized in that: The array power fluctuation degree is obtained according to the fluctuation significance weights and power impact intensity of all wind turbines, including: The weighted average of the power impact intensity of all wind turbines is calculated using the significant fluctuation weight of each wind turbine as the weight to obtain the array power fluctuation degree.

6. A power optimization dispatching device in a renewable energy grid-connected scenario, implementing the steps of a power optimization dispatching method in a renewable energy grid-connected scenario as described in any one of claims 1 to 5, characterized in that: The power optimization dispatching equipment comprises: The power data acquisition module is used to collect the electric power vector of each wind turbine in the wind power array; The power data processing module is used to obtain the power impact intensity of each wind turbine according to the numerical distribution and synchronous change of the electric power vector of each wind turbine in the frequency domain and the energy amplitude corresponding to it; obtain the fluctuation significance of each wind turbine relative to other wind turbines according to the comparison of the power impact intensity before and after the electric power vector of each wind turbine is combined with other wind turbines; obtain the fluctuation significance weight of each wind turbine according to the fluctuation significance of each wind turbine and all other wind turbines; obtain the array power fluctuation degree according to the fluctuation significance weight and power impact intensity of all wind turbines; The power equipment dispatching module is used to obtain the output power of the wind power array at each moment; predict the output power at a moment after a preset time period at the current moment based on the array power fluctuation and output power at the current moment to obtain an output power prediction value; and dispatch the power equipment based on the output power prediction value.

7. A power optimization dispatching device in a new energy grid-connected scenario, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the power optimization scheduling method in the new energy grid-connected scenario as described in any one of claims 1 to 5 are implemented.

Images