Power grid frequency adjusting method and device based on wind storage collaborative inertia response

Abstract

The invention discloses a power grid frequency adjustment method and device based on wind storage collaborative inertia response, relates to the technical field of power grid safety, and mainly aims to improve the timeliness and accuracy of power grid frequency adjustment. Comprising the following steps: acquiring operation state time sequence data and environment time sequence data of a local power grid under different frequency regulation nodes; based on the operation state time sequence data and the environment time sequence data, performing high-frequency prediction on the frequency of each local power grid in a first preset time window in the future and performing low-frequency prediction on the frequency of each local power grid in a second preset time window in the future to obtain a high-frequency fluctuation extreme value and a low-frequency trend mean value; and determining a wind power output power regulation value of the wind power supply equipment and an energy storage output power regulation value of the energy storage equipment based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, and cooperatively regulating the local power grid in a distributed manner based on the wind power output power regulation value and the energy storage output power regulation value. The method is suitable for a power grid frequency adjustment scene.

Description

Technical Field

[0001] The present invention relates to the field of power grid safety technology, and in particular to a power grid frequency regulation method and device based on wind-storage coordinated inertia response. Background Art

[0002] As the global demand for renewable energy increases, wind power, as an important renewable energy source, continues to expand in scale. However, the randomness and intermittent characteristics of wind power lead to unstable output power, which in turn exacerbates grid frequency fluctuations. Especially when large-scale wind power is connected, the frequency regulation problem becomes more complicated. This fluctuation not only affects the stability of the grid, but also brings huge challenges to the traditional power system.

[0003] At present, the grid frequency is usually adjusted based on the grid frequency fluctuations that have already occurred. However, this adjustment method has a certain hysteresis, which may cause the grid frequency to remain unstable during the adjustment process, causing adverse effects on power equipment and user experience. At the same time, new fluctuations may occur during the adjustment process, which may lead to inaccurate grid frequency adjustment. Summary of the invention

[0004] The present invention provides a method and device for regulating the frequency of a power grid based on the coordinated inertia response of wind and storage, which can mainly improve the timeliness and accuracy of the frequency regulation of the power grid and avoid adverse effects on power equipment and user electricity consumption.

[0005] According to a first aspect of the present invention, a method for regulating power grid frequency based on wind-storage coordinated inertia response is provided, comprising: In response to the frequency regulation signal of the target power grid, the operating state time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each of the local power grids is located are obtained, wherein the target power grid is composed of multiple local power grids, and different local power grids perform frequency regulation through different frequency regulation nodes; Based on the operation status time series data and the environment time series data, a high-frequency prediction is performed in real time on the frequency of each of the local power grids in a first preset time window in the future to obtain a high-frequency fluctuation extreme value of the local power grid frequency; and based on the operation status time series data and the environment time series data, a low-frequency prediction is performed in real time on the frequency of each of the local power grids in a second preset time window in the future to obtain a low-frequency trend mean value of the local power grid frequency; Based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids are determined, and based on the wind power output power adjustment value, the corresponding wind power supply equipment is coordinated and adjusted in real time to transmit power to the corresponding local power grid, and based on the energy storage output power adjustment value, the corresponding energy storage equipment is coordinated and controlled in real time to charge and discharge the corresponding local power grid.

[0006] Optionally, the operating status time series data includes grid frequency time series data and grid load time series data of the local power grid, and the environmental time series data includes wind speed time series data; The method of performing high-frequency prediction on the frequency of each local power grid in a first preset time window in the future in real time based on the operating state time series data and the environmental time series data to obtain a high-frequency fluctuation extreme value of the local power grid frequency includes: Based on the grid frequency time series data, determine the grid frequency change rate, based on the grid load time series data, determine the grid load change amount, based on the wind speed time series data, determine the wind speed change rate; Respectively determine a frequency characteristic vector corresponding to the grid frequency change rate, a load characteristic vector corresponding to the grid load change amount, and a wind speed characteristic vector corresponding to the wind speed change rate, and determine a frequency prediction characteristic vector based on the frequency characteristic vector, the load characteristic vector, and the wind speed characteristic vector; The frequency prediction feature vector is input into a preset high-frequency prediction model for frequency prediction to obtain the high-frequency fluctuation extreme value of each local power grid within a first preset time window in the future, wherein the preset high-frequency prediction model is pre-trained based on a high-frequency sample data set with high-frequency fluctuation extreme value labels.

[0007] Optionally, based on the operating state time series data and the environmental time series data, performing low-frequency prediction on the frequency of each local power grid in a second preset time window in the future in real time to obtain a low-frequency trend mean of the local power grid frequency includes: The grid frequency time series data in the operating status time series data and the wind speed time series data in the environmental time series data are input into a preset low-frequency prediction model for frequency prediction to obtain the low-frequency trend mean of each local power grid in a second preset time window in the future, wherein the second preset time window is larger than the first preset time window, and the preset low-frequency prediction model is pre-trained based on a low-frequency sample data set with a low-frequency trend mean label.

[0008] Optionally, determining the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids based on the high-frequency fluctuation extreme value and the low-frequency trend mean value includes: Determine the high-frequency prediction error of the high-frequency fluctuation extreme value in the prediction process and the low-frequency prediction error of the low-frequency trend mean in the prediction process, and determine the high-frequency fusion coefficient corresponding to the high-frequency fluctuation extreme value and the low-frequency fusion coefficient corresponding to the low-frequency trend mean based on the high-frequency prediction error and the low-frequency prediction error respectively; Based on the high frequency fusion coefficient and the low frequency fusion coefficient, the high frequency fluctuation extreme value and the low frequency trend mean value are fused to obtain a fused grid frequency value. ; Based on the fused grid frequency value , determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids.

[0009] Optionally, the frequency value based on the fusion grid , determining the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids, including: Determine the virtual inertia reference value of wind power supply equipment i and the virtual inertia response adjustment factor at the current time t ; Based on the virtual inertia reference value , the virtual inertia response adjustment factor , the fused grid frequency value , determine the wind power supply equipment under each local power grid The wind power output power adjustment value at the current time t ,in, ; Determine the rated power of the energy storage device and the energy storage adjustment factor at the current time t ; Based on the rated power , the energy storage adjustment factor and the fused grid frequency value , determine the energy storage output power adjustment value of each energy storage device under the local power grid at the current time t ,in, .

[0010] Optionally, determine the virtual inertia response adjustment factor at the current time t ,include: Determine a frequency relationship function between the grid frequency deviation of the local grid at the current time t and the virtual inertia response adjustment factor to be optimized, and determine the optimization constraint condition of the virtual inertia response adjustment factor to be optimized based on the output power constraint of the wind power supply equipment and the power relationship function between the virtual inertia response adjustment factor to be optimized and the output power of the wind power supply equipment; Based on the optimization constraint condition, the virtual inertia response adjustment factor to be optimized is optimized with the grid frequency deviation satisfying the preset condition as the optimization index, and the virtual inertia response adjustment factor of the wind power supply equipment at the current time t is determined based on the optimization result. , wherein the preset condition is that the grid frequency deviation is within a preset deviation range.

[0011] Optionally, the real-time distributed coordinated adjustment based on the wind power output power adjustment value to the corresponding wind power supply equipment to perform power transmission operations to the corresponding local power grid, and the real-time distributed coordinated control based on the energy storage output power adjustment value to the corresponding energy storage equipment to perform charging and discharging operations to the corresponding local power grid includes: Any frequency regulation node in each of the frequency regulation nodes is respectively used as a target frequency regulation node, the wind power output power regulation value and the energy storage output power regulation value under the target frequency regulation node are respectively determined as the reference wind power output power regulation value and the reference energy storage output power regulation value, and the reference wind power output power regulation value and the reference energy storage output power regulation value are respectively sent to other frequency regulation nodes, so that the other frequency regulation nodes perform distributed collaborative frequency regulation on the corresponding local power grid based on the corresponding wind power output power regulation value, the energy storage output power regulation value, the reference wind power output power regulation value, and the reference energy storage output power regulation value, wherein the other frequency regulation node refers to the node excluding the target frequency regulation node in each of the frequency regulation nodes.

[0012] According to a second aspect of the present invention, there is provided a power grid frequency regulation device based on wind-storage coordinated inertia response, comprising: An acquisition unit is used to acquire, in response to a frequency adjustment signal of a target power grid, time series data of operation status of local power grids under different frequency adjustment nodes and time series data of environment in which a wind power supply device corresponding to each local power grid is located, wherein the target power grid is composed of a plurality of local power grids, and different local power grids perform frequency adjustment through different frequency adjustment nodes; A frequency prediction unit, configured to perform a high-frequency prediction on the frequency of each of the local power grids in a first preset time window in the future in real time based on the operation status time series data and the environment time series data, so as to obtain a high-frequency fluctuation extreme value of the frequency of the local power grid, and to perform a low-frequency prediction on the frequency of each of the local power grids in a second preset time window in the future in real time based on the operation status time series data and the environment time series data, so as to obtain a low-frequency trend mean value of the frequency of the local power grid; The frequency regulation unit is used to determine the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment under each of the local power grids based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, and to coordinately regulate the corresponding wind power supply equipment to transmit power to the corresponding local power grid in real time and in a distributed manner based on the wind power output power regulation value, and to coordinately control the corresponding energy storage equipment to charge and discharge the corresponding local power grid in real time and in a distributed manner based on the energy storage output power regulation value.

[0013] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned grid frequency regulation method based on wind-storage coordinated inertia response.

[0014] According to a fourth aspect of the present invention, there is provided a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned grid frequency regulation method based on the coordinated inertia response of wind and storage when executing the program.

[0015] According to a grid frequency regulation method and device based on wind-storage collaborative inertia response provided by the present invention, compared with the current method of regulating the grid frequency based on the grid frequency fluctuation that has already occurred, the present invention uses the operating status time series data of the local grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each local grid is located to perform high-frequency prediction and low-frequency prediction of the grid frequency at future moments respectively, and determines the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment according to the predicted high-frequency fluctuation extreme value and low-frequency trend mean value, and finally performs distributed collaborative regulation of the frequency of the local power grid based on the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment. Therefore, by adjusting the grid frequency according to the predicted grid frequency value at a future moment, potential frequency fluctuation trends can be discovered in time, so that preventive measures can be taken in advance, thereby improving the timeliness of frequency regulation and ensuring the stable operation of the grid; at the same time, the grid frequency is adjusted in combination with the high-frequency prediction and low-frequency prediction of the grid frequency, which can capture both the short-term rapid changes in the grid frequency and the long-term change trends in the grid frequency, thereby achieving comprehensive monitoring and analysis of the grid frequency, thereby improving the regulation accuracy of the grid frequency; at the same time, the present invention can ensure that when a frequency regulation node fails, the control process of other frequency regulation nodes will not be affected by the distributed and coordinated regulation of the frequency of the corresponding local grid by multiple frequency regulation nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 A flow chart of a method for regulating power grid frequency based on wind-storage coordinated inertia response provided by an embodiment of the present invention is shown; Figure 2 A schematic diagram of frequency fluctuation for high-frequency prediction of a local power grid provided by an embodiment of the present invention is shown; Figure 3 A schematic diagram of a frequency fluctuation trend for low-frequency prediction of a local power grid provided by an embodiment of the present invention is shown; Figure 4A A schematic diagram of a local power grid frequency fluctuation situation provided by an embodiment of the present invention is shown; Figure 4B An embodiment of the present invention provides a method for Figure 4A A schematic diagram of the power output of wind power supply equipment when adjusting the local power grid frequency fluctuation is shown; Figure 4C An embodiment of the present invention provides a method for Figure 4A A schematic diagram showing the power output of the energy storage device when adjusting the local grid frequency fluctuation; Figure 5 Another flow chart of a method for adjusting the frequency of a power grid based on wind-storage coordinated inertia response provided by an embodiment of the present invention is shown; Figure 6 A schematic diagram of the structure of a power grid frequency regulation device based on wind-storage coordinated inertia response provided by an embodiment of the present invention is shown; Figure 7 A schematic diagram of the structure of another power grid frequency regulation device based on wind-storage coordinated inertia response provided by an embodiment of the present invention is shown; Figure 8 A schematic diagram of the physical structure of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0018] At present, the method of adjusting the grid frequency based on the grid frequency fluctuations that have already occurred has a certain lag, which may cause the grid frequency to remain in an unstable state during the adjustment process, causing adverse effects on power equipment and users' electricity consumption experience.

[0019] In order to solve the above problems, an embodiment of the present invention provides a grid frequency regulation method based on wind-storage coordinated inertia response, such as Figure 1 As shown, the method includes: 101. In response to a frequency regulation signal of a target power grid, obtain the operating status time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each local power grid is located, wherein the target power grid is composed of multiple local power grids, and different local power grids perform frequency regulation through different frequency regulation nodes.

[0020] Among them, the operating status time series data refers to the grid frequency time series data composed of the grid frequency from the historical moment to the current moment and the grid load time series data composed of the grid load; the environmental time series data refers to the time series data composed of the wind speed around the wind power supply equipment from the historical moment to the current moment.

[0021] For the embodiment of the present invention, the power output of the wind power supply equipment and the energy storage equipment is dynamically adjusted through the deep intelligent scheduling algorithm in the power grid frequency regulation system to realize the dynamic real-time frequency regulation of the target power grid. The power grid frequency regulation system includes a wide-area measurement module, a wind power monitoring module, and an energy storage management module. The wide-area measurement module is used to measure data such as power grid frequency, power grid load, and wind speed; the wind power monitoring module is used to control the output power of the wind power supply equipment and collect the wind turbine speed, etc., and the energy storage management module is used to control the charging and discharging operations of the energy storage equipment. In order to improve the accuracy of power grid frequency regulation, before the power grid frequency regulation is performed, the embodiment of the present invention can also construct a power grid frequency regulation simulation platform, which is configured with a wind power-energy storage joint simulation model, and uses a deep learning framework for model training and optimization. In terms of hardware deployment, edge intelligent gateways and industrial modules (such as end-to-end delay ≤3 ms) can be used to ensure the real-time and low latency of data processing. In addition, in terms of core parameters, the virtual inertia reference value can be set to an adjustable range of 1.8 seconds to 6.5 seconds, and the output upper limit of the energy storage device is 40% of the output of the wind power supply device. Afterwards, the grid frequency regulation is simulated based on the grid frequency regulation simulation platform to verify the feasibility and accuracy of the grid frequency regulation method in the embodiment of the present invention.

[0022] Specifically, when the grid frequency is adjusted using the grid frequency regulation system, the target grid is divided into multiple local grids according to factors such as region, load type, and grid structure, and each local grid is controlled by different frequency regulation nodes. Furthermore, when a frequency regulation signal for the target grid is received, the operating status time series data and environmental time series data of the local grid under each frequency regulation node can be collected based on modules such as wide area measurement, and distributed collaborative frequency regulation is performed on each local grid through the operating status time series data and environmental time series data. Therefore, by performing distributed collaborative regulation on multiple local grids, the impact of single point failures on the entire grid is avoided.

[0023] 102. Based on the operation status time series data and the environmental time series data, a high-frequency frequency prediction is performed in real time on each local power grid in a first preset time window in the future to obtain the high-frequency fluctuation extreme value of the local power grid frequency; and based on the operation status time series data and the environmental time series data, a low-frequency frequency prediction is performed in real time on each local power grid in a second preset time window in the future to obtain the low-frequency trend mean value of the local power grid frequency.

[0024] Among them, the first preset time window and the second preset time window are set according to actual needs. The size of the first preset time window is smaller than the size of the second preset time window. For example, the first preset time window can be set to a 3-second time window, and the second preset time window can be set to a 10-minute time window.

[0025] For the embodiment of the present invention, a high frequency prediction within a first preset time window and a low frequency prediction within a second preset time window are performed for each local power grid, such as Figure 2 The figure shows the frequency fluctuation situation obtained by high-frequency prediction of a certain local power grid. Based on the frequency fluctuation situation, the high-frequency fluctuation extreme value of the local power grid within the first preset time window can be determined. Figure 3 The frequency fluctuation trend obtained by low-frequency prediction of a certain local power grid is shown. Based on the frequency fluctuation trend, the low-frequency trend mean of the local power grid within the second preset time window can be determined. The embodiment of the present invention adjusts the power grid frequency by combining high-frequency prediction and low-frequency prediction of the power grid frequency, which can capture the rapid changes of the power grid frequency and the long-term change trend of the power grid frequency at the same time, and realizes comprehensive monitoring and analysis of the power grid frequency, thereby improving the accuracy of power grid frequency regulation. At the same time, the embodiment of the present invention adjusts the power grid frequency by predicting future power grid frequency fluctuations, which can achieve the timeliness of power grid frequency regulation.

[0026] 103. Based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each local power grid, and based on the wind power output power adjustment value, coordinately adjust the corresponding wind power supply equipment in a real-time distributed manner to transmit power to the corresponding local power grid, and based on the energy storage output power adjustment value, coordinately control the corresponding energy storage equipment in a real-time distributed manner to charge and discharge the corresponding local power grid.

[0027] For the embodiment of the present invention, after determining the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each local power grid, each wind power supply equipment and energy storage equipment can independently make adjustment decisions based on local information (wind power output power adjustment value and energy storage output power adjustment value). That is, each frequency adjustment node executes an autonomous grid frequency power adjustment strategy by real-time monitoring of the local grid frequency, grid load and wind speed changes. This decentralized control method ensures the high response speed and flexibility of the system. At the same time, each frequency adjustment node regularly sends local adjustment status information (such as the current grid frequency, wind power output power adjustment value, and energy storage output power adjustment value) to other frequency adjustment nodes. Through the information exchange mechanism, each frequency adjustment node can distribute and coordinately adjust the power output of the corresponding wind power supply equipment and energy storage equipment, ensure the adjustment consistency between the wind power supply equipment and energy storage equipment under different frequency adjustment nodes, and avoid over-adjustment or inconsistent response. At the same time, the adjustment of the wind power supply equipment and energy storage equipment in the embodiment of the present invention also adopts a collaborative adjustment method to achieve optimal frequency adjustment. Therefore, the embodiment of the present invention uses multiple frequency regulation nodes to coordinately regulate the frequency of the corresponding local power grid in a distributed manner, which can ensure that when a frequency regulation node fails, the control process of other frequency regulation nodes is not affected.

[0028] The embodiment of the present invention proposes an adaptive frequency response strategy, which adopts different adjustment methods according to different types of grid frequency fluctuations (such as low-frequency events and high-frequency events) and the magnitude of the fluctuation amplitude. For example, low-frequency event adjustment: when the grid frequency is lower than a preset threshold (the preset threshold is a value set according to actual needs), the system enters a low-frequency response mode, at which time the wind power supply equipment and the energy storage equipment should respond first. The virtual inertia output power of the wind power supply equipment and the power output of the energy storage equipment will increase to help the grid restore the frequency. The energy storage equipment can dynamically adjust the charging and discharging strategy according to the frequency deviation and the predicted low-frequency fluctuations to provide the necessary power support. High-frequency event adjustment: when the grid frequency is higher than a preset threshold, the system enters a high-frequency response mode, at which time the wind power supply equipment and the energy storage equipment will reduce the power output to prevent the grid frequency from being too high. The output power of the wind power supply equipment and the energy storage equipment will be dynamically adjusted according to the real-time frequency of the grid and the predicted high-frequency fluctuations to balance the excessive frequency fluctuations of the grid. Frequency fluctuation amplitude adjustment: The grid frequency adjustment system can adjust the response strategy according to the amplitude and duration of the frequency fluctuation. When the frequency fluctuation amplitude is large, the wind power supply equipment and energy storage equipment will increase the response strength and adjust the power output in time. When the frequency fluctuation amplitude is small, the system will reduce the response strength to avoid instability caused by over-adjustment. Through the above-mentioned dynamic adjustment mechanism, the grid frequency adjustment system can ensure the accuracy and timeliness of grid frequency adjustment and avoid excessive or delayed response. Furthermore, each adjustment of the wind power supply equipment and energy storage equipment will be transmitted back to the grid frequency adjustment system in real time through the feedback mechanism to ensure the consistency and stability of equipment adjustment at all frequency adjustment nodes. For example, Figure 4A A schematic diagram showing the fluctuation of power grid frequency over time is shown. Figure 4B Shows the Figure 4A Schematic diagram of wind power supply equipment power output when adjusting the grid frequency fluctuations. Figure 4C Shows the Figure 4A Schematic diagram of the power output of the energy storage device when adjusting to the grid frequency fluctuations.

[0029] According to a grid frequency regulation method based on wind-storage collaborative inertia response provided by the present invention, compared with the current way of regulating the grid frequency based on the grid frequency fluctuation that has already occurred, the present invention uses the operating status time series data of the local grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each local grid is located to perform high-frequency prediction and low-frequency prediction of the grid frequency at future moments respectively, and determines the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment according to the predicted high-frequency fluctuation extreme value and low-frequency trend mean value, and finally performs distributed collaborative regulation of the frequency of the local power grid based on the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment. By adjusting the grid frequency according to the predicted grid frequency value at a future moment, potential frequency fluctuation trends can be discovered in time, so that preventive measures can be taken in advance, thereby improving the timeliness of frequency regulation and ensuring the stable operation of the grid; at the same time, the grid frequency is adjusted in combination with the high-frequency prediction and low-frequency prediction of the grid frequency, so that the rapid changes in the grid frequency and the long-term change trends of the grid frequency can be captured simultaneously, and comprehensive monitoring and analysis of the grid frequency can be achieved, thereby improving the regulation accuracy of the grid frequency; at the same time, the present invention can distribute and coordinately adjust the frequency of the corresponding local grid through multiple frequency regulation nodes, so as to ensure that when a frequency regulation node fails, the control process of other frequency regulation nodes will not be affected.

[0030] Further, in order to better illustrate the above process of adjusting the grid frequency, as a refinement and extension of the above embodiment, the embodiment of the present invention provides another grid frequency adjustment method based on wind-storage coordinated inertia response, such as Figure 5 As shown, the method includes: 201. In response to a frequency regulation signal of a target power grid, obtain the operating status time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each local power grid is located, wherein the target power grid is composed of multiple local power grids, and different local power grids perform frequency regulation through different frequency regulation nodes.

[0031] 202. Based on the operating status time series data and the environmental time series data, a high-frequency frequency prediction is performed in real time on each local power grid in a first preset time window in the future to obtain the high-frequency fluctuation extreme value of the local power grid frequency; and based on the operating status time series data and the environmental time series data, a low-frequency frequency prediction is performed in real time on each local power grid in a second preset time window in the future to obtain the low-frequency trend mean value of the local power grid frequency.

[0032] For the embodiment of the present invention, in order to be able to adjust the grid frequency in time, it is first necessary to perform high-frequency prediction on the frequency of each local grid within the first preset time window in the future. Based on this, step 202 specifically includes: determining the grid frequency change rate based on the grid frequency time series data, determining the grid load change amount based on the grid load time series data, and determining the wind speed change rate based on the wind speed time series data; respectively determining the frequency characteristic vector corresponding to the grid frequency change rate, the load characteristic vector corresponding to the grid load change amount, and the wind speed characteristic vector corresponding to the wind speed change rate, and determining the frequency prediction characteristic vector based on the frequency characteristic vector, the load characteristic vector, and the wind speed characteristic vector; inputting the frequency prediction characteristic vector into a preset high-frequency prediction model for frequency prediction to obtain the high-frequency fluctuation extreme value of each local grid within the first preset time window in the future, wherein the preset high-frequency prediction model is pre-trained based on a high-frequency sample data set with high-frequency fluctuation extreme value labels.

[0033] Specifically, first determine the time interval between adjacent data points in the grid frequency time series data, calculate their frequency difference for two adjacent data points, divide the frequency difference by the corresponding time interval, and obtain the rate of change of the grid frequency within the time interval, and finally determine the mean of the rate of change of each time interval to obtain the rate of change of the grid frequency. Similarly, the wind speed change rate can be determined in the above manner. At the same time, determine the grid load deviation between adjacent data points in the grid load time series data, and determine the mean of each grid load deviation to obtain the grid load change. Furthermore, in order to ensure the real-time nature of high-frequency prediction, the grid frequency change rate can be determined directly based on the grid frequency at the current moment and the grid frequency at the previous moment corresponding to the current moment, the grid load change can be determined based on the grid load at the current moment and the grid load at the previous moment corresponding to the current moment, and the wind speed change rate can be determined based on the wind speed at the current moment and the wind speed at the previous moment corresponding to the current moment. Then determine the frequency prediction feature vector as follows :

[0034] in, is the grid frequency change rate, is the grid load variation, is the wind speed change rate.

[0035] In another embodiment of the present invention, the method for determining the frequency prediction feature vector further includes: performing feature-level cross-processing on the frequency feature vector, the load feature vector, and the wind speed feature vector to obtain a feature cross vector; performing element-level cross-processing on the frequency feature vector, the load feature vector, and the wind speed feature vector to obtain an element cross vector; performing low-order cross-processing on the frequency feature vector, the load feature vector, and the wind speed feature vector to obtain a low-order cross vector; and combining the feature cross vector, the element cross vector, and the low-order cross vector to obtain the frequency prediction feature vector. Specifically, the frequency feature vector corresponding to the rate of change of the grid frequency, the load feature vector corresponding to the change in grid load, and the wind speed feature vector corresponding to the rate of change of wind speed are respectively determined by means of word embedding and the like. Then, in order to make full use of the relationship between the data, extract more implicit features, take into account both high-order and low-order processing, make the data utilization more sufficient, and make the subsequent prediction results more accurate to meet the requirements of the actual application scenario, it is necessary to perform cross-processing on the frequency feature vector, the load feature vector, and the wind speed feature vector. The specific cross-processing methods are as follows: If the frequency feature vector is (a 1 ,a 2 ), the load feature vector is (b 1 ,b 2 ), and the wind speed feature vector is (c 1 ,c 2 ), the specific cross-processing methods include: performing cross-processing at the feature level between different feature vectors, that is, after performing the Hadamard product on all the elements between the vectors, performing a convolution transformation under a certain weight to obtain a feature cross vector; at the same time, performing element-level cross-processing on all the feature vector data, that is, performing the Hadamard product on each element between the vectors, assigning different weight values to each product result, and then performing a linear transformation to obtain an element cross vector; in addition, performing low-order cross-processing on all the feature vectors, assigning a weight coefficient w to the result of the cross-processing, and then performing a linear transformation. The obtained low-order cross vector is f(w(a 1 ,a 2 ,b 1 ,b 2 ,c 1 ,c 2)); finally, the above feature cross vectors, element cross vectors, and low-order cross vectors are combined together, such as horizontal splicing, to obtain a frequency prediction feature vector. It should be noted that the above examples are only illustrative and do not limit the embodiments of the present application. Therefore, by cross-processing the frequency feature vector, the load feature vector, and the wind speed feature vector, different features can be automatically or explicitly combined to generate new feature combinations. These combined features may contain complex nonlinear relationships between the original features, and can capture more detailed and rich information in the data, that is, they can make full use of the relationship between various data, extract more implicit features, and take into account both high-order and low-order processing, so that data utilization is more sufficient, and the subsequent grid frequency adjustment is more accurate to meet the stable operation of the grid.

[0036] Furthermore, after determining the frequency prediction feature vector, a preset high-frequency prediction model can be used to predict the high-frequency fluctuation extreme value of the local power grid within the first preset time window. In order to improve the prediction accuracy of the preset high-frequency prediction model, it is first necessary to train and construct the preset high-frequency prediction model. Based on this, the method includes: constructing a preset initial high-frequency prediction model, and obtaining a high-frequency sample data set, wherein the high-frequency sample data set includes sample power grid frequency time series data, sample power grid load time series data, and sample wind speed time series data with high-frequency fluctuation extreme value labels of the sample power grid; dividing the high-frequency sample data set into training data and test data, using the training data to train the preset initial high-frequency prediction model, and using the test data to test the trained preset initial high-frequency prediction model, and finally determining the trained preset initial high-frequency prediction model that meets the test conditions as the preset high-frequency prediction model. The test conditions may be the prediction accuracy, prediction error, number of iterative training, etc. of the trained preset initial high-frequency prediction model. For example, if the sample frequency prediction feature vector composed of sample power grid frequency time series data, sample power grid load time series data, and sample environment time series data in the test data is input into the trained preset initial high-frequency prediction model, and the absolute value of the difference between the prediction result (predicted high-frequency fluctuation extreme value) and the actual high-frequency fluctuation extreme value is less than or equal to the preset threshold value (wherein the preset threshold value is set according to actual needs), then it is determined that the trained preset initial high-frequency prediction model meets the test conditions. Finally, the frequency prediction feature vectors under each local power grid are respectively input into the constructed preset high-frequency prediction model, and the preset high-frequency prediction model can output the high-frequency fluctuation extreme value of each local power grid in the first preset time window in the future.

[0037] At the same time, in order to be able to adjust the grid frequency in time, it is also necessary to perform low-frequency prediction on the frequency of each local grid in the second preset time window in the future in real time. Based on this, step 202 specifically includes: inputting the grid frequency time series data in the operating status time series data and the wind speed time series data in the environmental time series data into the preset low-frequency prediction model for frequency prediction, and obtaining the low-frequency trend mean of each local grid in the second preset time window in the future, wherein the second preset time window is larger than the first preset time window, and the preset low-frequency prediction model is pre-trained based on a low-frequency sample data set with a low-frequency trend mean label.

[0038] Specifically, in order to improve the prediction accuracy of the preset low-frequency prediction model, it is first necessary to train and construct the preset low-frequency prediction model. Based on this, the method includes: constructing a preset initial low-frequency prediction model, and obtaining a low-frequency sample data set, wherein the low-frequency sample data set includes sample power grid frequency time series data and sample wind speed time series data with a low-frequency trend mean label of the sample power grid; dividing the low-frequency sample data set into training data and test data, using the training data to train the preset initial low-frequency prediction model, and using the test data to test the trained preset initial low-frequency prediction model, and finally determining the trained preset initial low-frequency prediction model that meets the test conditions as the preset low-frequency prediction model. Among them, the test conditions may be that the prediction accuracy, prediction error, and iterative training times of the trained preset initial low-frequency prediction model meet corresponding conditions. The preset low-frequency prediction model is used to capture the long-term trend of power grid frequency fluctuations and is suitable for low-frequency prediction. The network structure of the preset low-frequency prediction model can adopt a bidirectional long short-term memory network layer (such as 64 neurons per layer) and an attention mechanism to focus on the key time period in the time series data so as to output the frequency trend mean in the future second preset time window. Furthermore, after the preset low-frequency prediction model is constructed, the preset low-frequency prediction model The process of making a low-frequency prediction is as follows:

[0039] in, is the low-frequency trend mean, For the current moment, is the historical moment (i.e. the start time of data recording for the power grid frequency time series data and wind speed time series data), is the grid frequency time series data, It is the wind speed time series data.

[0040] 203. Determine the high-frequency prediction error of the high-frequency fluctuation extreme value in the prediction process and the low-frequency prediction error of the low-frequency trend mean in the prediction process, and based on the high-frequency prediction error and the low-frequency prediction error, respectively determine the high-frequency fusion coefficient corresponding to the high-frequency fluctuation extreme value and the low-frequency fusion coefficient corresponding to the low-frequency trend mean.

[0041] 204. Based on the high-frequency fusion coefficient and the low-frequency fusion coefficient, the high-frequency fluctuation extreme value and the low-frequency trend mean value are fused to obtain a fused power grid frequency value.

[0042] Specifically, first determine the variance of the high-frequency prediction error of the high-frequency fluctuation extreme value in the prediction process , and the variance of the low-frequency forecast error of the low-frequency trend mean during the forecast process Then, the high-frequency fusion coefficient at the current time t is determined according to the following formula: and low frequency fusion coefficient :

[0043]

[0044] Furthermore, the frequency value of the integrated power grid is determined according to the following formula: :

[0045] in, is the high-frequency fluctuation extreme value of the local power grid within the first preset time window, It is the low-frequency trend mean of the local power grid within the second preset time window.

[0046] 205. Based on the integrated grid frequency value, determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each local grid.

[0047] For the embodiment of the present invention, after determining the fusion grid frequency value, it is necessary to determine the wind power output power adjustment value and the energy storage output power adjustment value based on the fusion grid frequency value. Based on this, step 205 specifically includes: determining the virtual inertia reference value of the wind power supply device i and the virtual inertia response adjustment factor at the current time t ; Based on the virtual inertia reference value , the virtual inertia response adjustment factor , the fused grid frequency value , determine the wind power supply equipment under each local power grid The wind power output power adjustment value at the current time t ,in, ; Determine the rated power of the energy storage device and the energy storage adjustment factor at the current time t ; Based on the rated power , the energy storage adjustment factor and the fused grid frequency value , determine the energy storage output power adjustment value of each energy storage device under the local power grid at the current time t ,in, .in, is the rate of change of the integrated grid frequency, and 0.7 is the nonlinear relationship between the grid frequency fluctuation and the energy storage output power regulation value. It can be adjusted dynamically according to the fluctuation of grid frequency, or it can be set to a fixed value according to actual needs. The adjustment is related to the rate of change of the grid frequency. The system adjusts according to the real-time frequency change of the grid. The value of ensures that the wind power supply equipment adapts to the frequency change in a short time, thereby providing appropriate inertia support. This process is adjusted through continuous feedback to ensure that when the grid frequency fluctuates, the energy storage equipment can respond quickly and achieve accurate frequency regulation. The embodiment of the present invention determines the wind power output power adjustment value and the energy storage output power adjustment value by integrating the grid frequency value, which can improve the overall prediction results and ensure that accurate grid regulation decisions can be obtained in the dual environment of high-frequency fluctuations and low-frequency trends.

[0048] Furthermore, the virtual inertia response adjustment factor at the current time t is determined as follows: :Determine the frequency relationship function between the grid frequency deviation of the local grid at the current time t and the virtual inertia response adjustment factor to be optimized, and determine the optimization constraint condition of the virtual inertia response adjustment factor to be optimized based on the output power constraint of the wind power supply equipment, the power relationship function between the virtual inertia response adjustment factor to be optimized and the output power of the wind power supply equipment; based on the optimization constraint condition, optimize the virtual inertia response adjustment factor to be optimized with the grid frequency deviation satisfying the preset condition as the optimization index, and determine the virtual inertia response adjustment factor of the wind power supply equipment at the current time t based on the optimization result. , wherein the preset condition is that the grid frequency deviation is within a preset deviation range.

[0049] Among them, the grid frequency deviation at the current time t is and the virtual inertia response adjustment factor to be optimized for the i-th wind power supply equipment The frequency relationship function between As shown below:

[0050] in, , The grid frequency deviation is and the virtual inertia response adjustment factor to be optimized The corresponding weight coefficients are used to control the grid frequency deviation respectively and the virtual inertia response adjustment factor to be optimized The importance of is the virtual inertia response adjustment factor to be optimized The upper limit of optimization time. At the same time, the output power constraints of wind power supply equipment are as follows:

[0051] in, Power supply equipment for wind power The output power, is the minimum output power of wind power supply equipment, is the maximum output power of the wind power supply equipment. Further, the virtual inertia response adjustment factor to be optimized is The optimization constraints are as follows:

[0052] in, is the power coefficient, It is the ratio of the blade tip speed of the wind power supply equipment to the wind speed. is the air density, is the cross-sectional area of ​​the wind power supply equipment, is the wind speed. Finally, based on the above optimization constraints, the frequency relationship function The virtual inertia response adjustment factor to be optimized in Continuously iterate the optimization until the grid frequency deviation is within the preset deviation range (the preset deviation range is set according to actual needs), or the iterative optimization is stopped when the grid frequency deviation reaches the minimum value, and the optimized virtual inertia response adjustment factor at this time is used as the virtual inertia response adjustment factor that finally participates in the calculation of the wind power output power adjustment value.

[0053] 206. Based on the wind power output power adjustment value, the corresponding wind power supply equipment is coordinated and adjusted in real time to transmit power to the corresponding local power grid, and based on the energy storage output power adjustment value, the corresponding energy storage equipment is coordinated and controlled in real time to charge and discharge the corresponding local power grid.

[0054] For the embodiment of the present invention, after determining the wind power output power adjustment value and the energy storage output power adjustment value, it is necessary to implement frequency adjustment of the local power grid based on the above adjustment values. Based on this, step 206 specifically includes: taking any frequency adjustment node in each of the frequency adjustment nodes as a target frequency adjustment node, determining the wind power output power adjustment value and the energy storage output power adjustment value under the target frequency adjustment node as the reference wind power output power adjustment value and the reference energy storage output power adjustment value, respectively, and sending the reference wind power output power adjustment value and the reference energy storage output power adjustment value to other frequency adjustment nodes, so that the other frequency adjustment nodes perform distributed collaborative frequency adjustment on the corresponding local power grid based on the corresponding wind power output power adjustment value, the energy storage output power adjustment value, the reference wind power output power adjustment value, and the reference energy storage output power adjustment value, wherein the other frequency adjustment node refers to a node in each of the frequency adjustment nodes excluding the target frequency adjustment node.

[0055] Specifically, after calculating the wind power output power adjustment value and the energy storage output power adjustment value under each frequency adjustment node, each frequency adjustment node can control the wind power supply equipment under this node to output the corresponding power to the local power grid based on the wind power output power adjustment value corresponding to this node. At the same time, each frequency adjustment node can control the energy storage equipment under this node to output or absorb the corresponding power to the local power grid based on the energy storage output power adjustment value corresponding to this node, so as to realize the distributed adjustment of the local power grid corresponding to this node. At the same time, when each frequency adjustment node adjusts the power grid frequency for its own node, it can also send the wind power output power adjustment value and energy storage output power adjustment value of its own node to other frequency adjustment nodes. Therefore, each frequency adjustment node can perform coordinated frequency adjustment of the corresponding local power grid based on the adjustment information such as the wind power output power adjustment value and energy storage output power adjustment value of this node and other nodes.

[0056] According to another grid frequency regulation method based on wind-storage collaborative inertia response provided by the present invention, compared with the current way of regulating the grid frequency based on the grid frequency fluctuation that has already occurred, the present invention uses the operating status time series data of the local grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each local grid is located to perform high-frequency prediction and low-frequency prediction of the grid frequency at future moments respectively, and determines the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment according to the predicted high-frequency fluctuation extreme value and low-frequency trend mean value, and finally performs distributed collaborative regulation of the frequency of the local power grid based on the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment. By adjusting the grid frequency according to the predicted grid frequency value at a future moment, potential frequency fluctuation trends can be discovered in time, so that preventive measures can be taken in advance, thereby improving the timeliness of frequency regulation and ensuring the stable operation of the grid; at the same time, the grid frequency is adjusted in combination with the high-frequency prediction and low-frequency prediction of the grid frequency, so that the rapid changes in the grid frequency and the long-term change trends of the grid frequency can be captured simultaneously, and comprehensive monitoring and analysis of the grid frequency can be achieved, thereby improving the regulation accuracy of the grid frequency; at the same time, the present invention can distribute and coordinately adjust the frequency of the corresponding local grid through multiple frequency regulation nodes, so as to ensure that when a frequency regulation node fails, the control process of other frequency regulation nodes will not be affected.

[0057] Further, as Figure 1 In a specific implementation, an embodiment of the present invention provides a power grid frequency regulation device based on wind-storage coordinated inertia response, such as Figure 6 As shown, the device includes: an acquisition unit 31, a frequency prediction unit 32, and a frequency adjustment unit 33.

[0058] The acquisition unit 31 can be used to respond to the frequency regulation signal of the target power grid, obtain the operating status time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each of the local power grids is located, wherein the target power grid is composed of multiple local power grids, and different local power grids perform frequency regulation through different frequency regulation nodes.

[0059] The frequency prediction unit 32 can be used to perform high-frequency prediction on the frequency of each of the local power grids in a first preset time window in the future in real time based on the operating status time series data and the environmental time series data, so as to obtain the high-frequency fluctuation extreme value of the local power grid frequency, and to perform low-frequency prediction on the frequency of each of the local power grids in a second preset time window in the future in real time based on the operating status time series data and the environmental time series data, so as to obtain the low-frequency trend mean value of the local power grid frequency.

[0060] The frequency adjustment unit 33 can be used to determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, and to perform real-time distributed collaborative adjustment of the corresponding wind power supply equipment to transmit power to the corresponding local power grid based on the wind power output power adjustment value, and to perform real-time distributed collaborative control of the corresponding energy storage equipment to perform charging and discharging operations on the corresponding local power grid based on the energy storage output power adjustment value.

[0061] In a specific application scenario, the operating status time series data includes the grid frequency time series data and grid load time series data of the local grid, and the environmental time series data includes wind speed time series data; Figure 7 As shown, in order to perform high frequency prediction on the local power grid, the frequency prediction unit 32 includes a first determination module 321 and a prediction module 322 .

[0062] The first determination module 321 can be used to determine the grid frequency change rate based on the grid frequency time series data, determine the grid load change amount based on the grid load time series data, and determine the wind speed change rate based on the wind speed time series data.

[0063] The first determination module 321 can also be used to respectively determine the frequency characteristic vector corresponding to the grid frequency change rate, the load characteristic vector corresponding to the grid load change, and the wind speed characteristic vector corresponding to the wind speed change rate, and determine the frequency prediction characteristic vector based on the frequency characteristic vector, the load characteristic vector, and the wind speed characteristic vector.

[0064] The prediction module 322 can be used to input the frequency prediction feature vector into a preset high-frequency prediction model for frequency prediction, and obtain the high-frequency fluctuation extreme value of each local power grid within a first preset time window in the future, wherein the preset high-frequency prediction model is pre-trained based on a high-frequency sample data set with high-frequency fluctuation extreme value labels.

[0065] In a specific application scenario, in order to perform low-frequency prediction on a local power grid, the prediction module 322 can also be used to input the power grid frequency time series data in the operating status time series data and the wind speed time series data in the environmental time series data into a preset low-frequency prediction model for frequency prediction, and obtain the low-frequency trend mean of each local power grid in a second preset time window in the future, wherein the second preset time window is larger than the first preset time window, and the preset low-frequency prediction model is pre-trained based on a low-frequency sample data set with a low-frequency trend mean label.

[0066] In a specific application scenario, in order to determine the wind power output power adjustment value and the energy storage output power adjustment value, the frequency adjustment unit 33 includes a second determination module 331 and a fusion module 332 .

[0067] The second determination module 331 can be used to determine the high-frequency prediction error of the high-frequency fluctuation extreme value in the prediction process and the low-frequency prediction error of the low-frequency trend mean in the prediction process, and based on the high-frequency prediction error and the low-frequency prediction error, respectively determine the high-frequency fusion coefficient corresponding to the high-frequency fluctuation extreme value and the low-frequency fusion coefficient corresponding to the low-frequency trend mean.

[0068] The fusion module 332 can be used to fuse the high-frequency fluctuation extreme value and the low-frequency trend mean value based on the high-frequency fusion coefficient and the low-frequency fusion coefficient to obtain a fused power grid frequency value. .

[0069] The second determination module 331 may also be used to determine the frequency value of the fused power grid based on the , determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids.

[0070] In a specific application scenario, in order to determine the wind power output power adjustment value and the energy storage output power adjustment value, the second determination module 331 can be specifically used to determine the virtual inertia reference value of the wind power supply device i. and the virtual inertia response adjustment factor at the current time t ; Based on the virtual inertia reference value , the virtual inertia response adjustment factor , the fused grid frequency value , determine the wind power supply equipment under each local power grid The wind power output power adjustment value at the current time t ,in, ; Determine the rated power of the energy storage device and the energy storage adjustment factor at the current time t ; Based on the rated power , the energy storage adjustment factor and the fused grid frequency value , determine the energy storage output power adjustment value of each energy storage device under the local power grid at the current time t ,in, .

[0071] In a specific application scenario, in order to determine the virtual inertia response adjustment factor at the current time t, the second determination module 331 can be specifically used to determine the frequency relationship function between the grid frequency deviation of the local grid at the current time t and the virtual inertia response adjustment factor to be optimized, and determine the optimization constraint condition of the virtual inertia response adjustment factor to be optimized based on the output power constraint of the wind power supply equipment and the power relationship function between the virtual inertia response adjustment factor to be optimized and the output power of the wind power supply equipment; based on the optimization constraint condition, the virtual inertia response adjustment factor to be optimized is optimized with the grid frequency deviation satisfying the preset condition as the optimization index, and the virtual inertia response adjustment factor of the wind power supply equipment at the current time t is determined based on the optimization result. , wherein the preset condition is that the grid frequency deviation is within a preset deviation range.

[0072] In a specific application scenario, in order to perform distributed collaborative frequency regulation on a local power grid, the frequency regulation unit 33 can be specifically used to take any frequency regulation node in each of the frequency regulation nodes as a target frequency regulation node, determine the wind power output power regulation value and the energy storage output power regulation value under the target frequency regulation node as the reference wind power output power regulation value and the reference energy storage output power regulation value, respectively, and send the reference wind power output power regulation value and the reference energy storage output power regulation value to other frequency regulation nodes, so that the other frequency regulation nodes perform distributed collaborative frequency regulation on the corresponding local power grid based on the corresponding wind power output power regulation value, the energy storage output power regulation value, the reference wind power output power regulation value, and the reference energy storage output power regulation value, wherein the other frequency regulation node refers to a node in each of the frequency regulation nodes excluding the target frequency regulation node.

[0073] It should be noted that for other corresponding descriptions of the functional modules involved in the power grid frequency regulation device based on wind-storage coordinated inertia response provided in an embodiment of the present invention, reference can be made to Figure 1 The corresponding description of the method shown will not be repeated here.

[0074] Based on the above Figure 1The method shown, accordingly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the following steps are implemented: in response to a frequency adjustment signal of a target power grid, the operating state time series data of a local power grid under different frequency adjustment nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each of the local power grids is located are obtained, wherein the target power grid is composed of a plurality of local power grids, and different local power grids perform frequency adjustment through different frequency adjustment nodes; based on the operating state time series data and the environmental time series data, the frequency of each of the local power grids in a first preset time window in the future is predicted in real time at a high frequency to obtain the frequency of the local power grid. High-frequency fluctuation extreme values, and based on the operating status time series data and the environmental time series data, real-time low-frequency prediction of the frequency of each of the local power grids in the second preset time window in the future to obtain the low-frequency trend mean of the local power grid frequency; based on the high-frequency fluctuation extreme values ​​and the low-frequency trend mean, determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids, based on the wind power output power adjustment value, real-time distributed and coordinated adjustment of the corresponding wind power supply equipment to the corresponding local power grid for power transmission operations, and based on the energy storage output power adjustment value, real-time distributed and coordinated control of the corresponding energy storage equipment to perform charging and discharging operations on the corresponding local power grid.

[0075] Based on the above Figure 1 The method shown and Figure 6 The embodiment of the device shown in the figure, the embodiment of the present invention also provides a physical structure diagram of a computer device, such as Figure 8As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43, and the processor 41 implements the following steps when executing the program: in response to the frequency regulation signal of the target power grid, obtaining the operating state time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each of the local power grids is located, wherein the target power grid is composed of a plurality of local power grids, and different local power grids perform frequency regulation through different frequency regulation nodes; based on the operating state time series data and the environmental time series data, real-time frequency regulation of each of the local power grids within a first preset time window in the future The method comprises the following steps: performing high-frequency prediction on the frequency to obtain the high-frequency fluctuation extreme value of the local power grid frequency, and performing low-frequency prediction on the frequency of each local power grid in a second preset time window in the future in real time based on the operation status time series data and the environmental time series data to obtain the low-frequency trend average value of the local power grid frequency; determining the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each local power grid based on the high-frequency fluctuation extreme value and the low-frequency trend average value, and performing real-time distributed collaborative adjustment on the wind power supply equipment to perform power transmission operations to the corresponding local power grid based on the wind power output power adjustment value, and performing real-time distributed collaborative control on the corresponding energy storage equipment to perform charging and discharging operations to the corresponding local power grid based on the energy storage output power adjustment value.

[0076] Through the technical solution of the present invention, the present invention performs high-frequency prediction and low-frequency prediction on the grid frequency at future moments respectively through the operating status time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each local power grid is located, and determines the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment according to the predicted high-frequency fluctuation extreme value and low-frequency trend mean value, and finally performs distributed collaborative regulation of the frequency of the local power grid based on the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment. By adjusting the grid frequency according to the predicted grid frequency value at a future moment, potential frequency fluctuation trends can be discovered in time, so that preventive measures can be taken in advance, thereby improving the timeliness of frequency regulation and ensuring the stable operation of the grid; at the same time, the grid frequency is adjusted in combination with the high-frequency prediction and low-frequency prediction of the grid frequency, so that the rapid changes in the grid frequency and the long-term change trends of the grid frequency can be captured simultaneously, and comprehensive monitoring and analysis of the grid frequency can be achieved, thereby improving the regulation accuracy of the grid frequency; at the same time, the present invention can distribute and coordinately adjust the frequency of the corresponding local grid through multiple frequency regulation nodes, so as to ensure that when a frequency regulation node fails, the control process of other frequency regulation nodes will not be affected.

[0077] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for regulating power grid frequency based on wind-storage coordinated inertia response, characterized in that: include: In response to the frequency regulation signal of the target power grid, the operating state time series data of the local power grid under different frequency regulation nodes and the environmental time series data of the environment in which the wind power supply equipment corresponding to each of the local power grids is located are obtained, wherein the target power grid is composed of multiple local power grids, and different local power grids perform frequency regulation through different frequency regulation nodes; Based on the operation status time series data and the environment time series data, a high-frequency prediction is performed in real time on the frequency of each of the local power grids in a first preset time window in the future to obtain a high-frequency fluctuation extreme value of the local power grid frequency; and based on the operation status time series data and the environment time series data, a low-frequency prediction is performed in real time on the frequency of each of the local power grids in a second preset time window in the future to obtain a low-frequency trend mean value of the local power grid frequency; Based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids are determined, and based on the wind power output power adjustment value, the corresponding wind power supply equipment is coordinated and adjusted in real time to transmit power to the corresponding local power grid, and based on the energy storage output power adjustment value, the corresponding energy storage equipment is coordinated and controlled in real time to charge and discharge the corresponding local power grid.

2. The method according to claim 1, characterized in that The operating status time series data includes the grid frequency time series data and grid load time series data of the local grid, and the environmental time series data includes wind speed time series data; The method of performing high-frequency prediction on the frequency of each local power grid in a first preset time window in the future in real time based on the operating state time series data and the environmental time series data to obtain a high-frequency fluctuation extreme value of the local power grid frequency includes: Based on the grid frequency time series data, determine the grid frequency change rate, based on the grid load time series data, determine the grid load change amount, based on the wind speed time series data, determine the wind speed change rate; Respectively determine a frequency characteristic vector corresponding to the grid frequency change rate, a load characteristic vector corresponding to the grid load change amount, and a wind speed characteristic vector corresponding to the wind speed change rate, and determine a frequency prediction characteristic vector based on the frequency characteristic vector, the load characteristic vector, and the wind speed characteristic vector; The frequency prediction feature vector is input into a preset high-frequency prediction model for frequency prediction to obtain the high-frequency fluctuation extreme value of each local power grid within a first preset time window in the future, wherein the preset high-frequency prediction model is pre-trained based on a high-frequency sample data set with high-frequency fluctuation extreme value labels.

3. The method according to claim 1, characterized in that The method of performing low-frequency prediction on the frequency of each local power grid in a second preset time window in the future in real time based on the operating state time series data and the environmental time series data to obtain a low-frequency trend mean value of the local power grid frequency includes: The grid frequency time series data in the operating status time series data and the wind speed time series data in the environmental time series data are input into a preset low-frequency prediction model for frequency prediction to obtain the low-frequency trend mean of each local power grid in a second preset time window in the future, wherein the second preset time window is larger than the first preset time window, and the preset low-frequency prediction model is pre-trained based on a low-frequency sample data set with a low-frequency trend mean label.

4. The method according to claim 1, characterized in that: The determining of the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids based on the high-frequency fluctuation extreme value and the low-frequency trend mean value includes: Determine the high-frequency prediction error of the high-frequency fluctuation extreme value in the prediction process and the low-frequency prediction error of the low-frequency trend mean in the prediction process, and determine the high-frequency fusion coefficient corresponding to the high-frequency fluctuation extreme value and the low-frequency fusion coefficient corresponding to the low-frequency trend mean based on the high-frequency prediction error and the low-frequency prediction error respectively; Based on the high frequency fusion coefficient and the low frequency fusion coefficient, the high frequency fluctuation extreme value and the low frequency trend mean value are fused to obtain a fused grid frequency value. ; Based on the fused grid frequency value , determine the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids.

5. The method according to claim 4, characterized in that The fusion grid frequency value based on , determining the wind power output power adjustment value of the wind power supply equipment and the energy storage output power adjustment value of the energy storage equipment under each of the local power grids, including: Determine the virtual inertia reference value of wind power supply equipment i and the virtual inertia response adjustment factor at the current time t ; Based on the virtual inertia reference value , the virtual inertia response adjustment factor , the fused grid frequency value , determine the wind power supply equipment under each local power grid The wind power output power adjustment value at the current time t ,in, ; Determine the rated power of the energy storage device and the energy storage adjustment factor at the current time t ; Based on the rated power , the energy storage adjustment factor and the fused grid frequency value , determine the energy storage output power adjustment value of each energy storage device under the local power grid at the current time t ,in, .

6. The method according to claim 5, characterized in that Determine the virtual inertia response adjustment factor at the current time t ,include: Determine a frequency relationship function between the grid frequency deviation of the local grid at the current time t and the virtual inertia response adjustment factor to be optimized, and determine the optimization constraint condition of the virtual inertia response adjustment factor to be optimized based on the output power constraint of the wind power supply equipment and the power relationship function between the virtual inertia response adjustment factor to be optimized and the output power of the wind power supply equipment; Based on the optimization constraint condition, the virtual inertia response adjustment factor to be optimized is optimized with the grid frequency deviation satisfying the preset condition as the optimization index, and the virtual inertia response adjustment factor of the wind power supply equipment at the current time t is determined based on the optimization result. , wherein the preset condition is that the grid frequency deviation is within a preset deviation range.

7. The method according to claim 1, characterized in that The real-time distributed coordinated adjustment based on the wind power output power adjustment value for corresponding wind power supply equipment to perform power transmission operation to the corresponding local power grid, and the real-time distributed coordinated control based on the energy storage output power adjustment value for corresponding energy storage equipment to perform charging and discharging operation to the corresponding local power grid, includes: Any frequency regulation node in each of the frequency regulation nodes is respectively used as a target frequency regulation node, the wind power output power regulation value and the energy storage output power regulation value under the target frequency regulation node are respectively determined as the reference wind power output power regulation value and the reference energy storage output power regulation value, and the reference wind power output power regulation value and the reference energy storage output power regulation value are respectively sent to other frequency regulation nodes, so that the other frequency regulation nodes perform distributed collaborative frequency regulation on the corresponding local power grid based on the corresponding wind power output power regulation value, the energy storage output power regulation value, the reference wind power output power regulation value, and the reference energy storage output power regulation value, wherein the other frequency regulation node refers to the node excluding the target frequency regulation node in each of the frequency regulation nodes.

8. A power grid frequency regulation device based on wind-storage coordinated inertia response, characterized in that: include: An acquisition unit is used to acquire, in response to a frequency adjustment signal of a target power grid, time series data of operation status of local power grids under different frequency adjustment nodes and time series data of environment in which a wind power supply device corresponding to each local power grid is located, wherein the target power grid is composed of a plurality of local power grids, and different local power grids perform frequency adjustment through different frequency adjustment nodes; A frequency prediction unit, configured to perform a high-frequency prediction on the frequency of each of the local power grids in a first preset time window in the future in real time based on the operation status time series data and the environment time series data, so as to obtain a high-frequency fluctuation extreme value of the frequency of the local power grid, and to perform a low-frequency prediction on the frequency of each of the local power grids in a second preset time window in the future in real time based on the operation status time series data and the environment time series data, so as to obtain a low-frequency trend mean value of the frequency of the local power grid; The frequency regulation unit is used to determine the wind power output power regulation value of the wind power supply equipment and the energy storage output power regulation value of the energy storage equipment under each of the local power grids based on the high-frequency fluctuation extreme value and the low-frequency trend mean value, and to coordinately regulate the corresponding wind power supply equipment to transmit power to the corresponding local power grid in real time and in a distributed manner based on the wind power output power regulation value, and to coordinately control the corresponding energy storage equipment to charge and discharge the corresponding local power grid in real time and in a distributed manner based on the energy storage output power regulation value.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images