Energy control method and system for wind energy storage power station

Abstract

The present application relates to the field of data processing, and in particular to an energy control method and system for a wind energy storage power station, the method comprising the steps of: calculating a correction value of an initial smoothing parameter; taking the product of the initial smoothing parameter and the correction value as a correction smoothing parameter for a future moment; obtaining a prediction result based on the correction smoothing parameter and an exponential smoothing algorithm, and performing power control based on the prediction result; the correction value is calculated by: taking wind force and time as a two-dimensional data point; calculating the local variation degree of wind force data and the noise degree of wind force data within a preset time range of the data point; taking the normalized result of the product of the local variation degree and the noise degree as the attention degree of the data point; setting a local window, and taking the ratio of the mean value of the data points in the local window whose attention degree is greater than a preset threshold to the maximum value of the data points in the local window as the correction value. The present application improves the accuracy of prediction by adaptively correcting the smoothing parameter.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to an energy control method and system for a wind energy storage power station. Background Art

[0002] In the power system, wind power generation has become an important and indispensable form. Wind energy resources are abundant and have strong environmental adaptability, so it is used in the power system. Due to the instability of wind power, there may even be large fluctuations in a short period of time. If wind energy is directly converted into electrical energy for output, it will cause instability of the power grid and may also cause harm to electrical equipment. Therefore, it is necessary to store and release the electrical energy converted from wind energy in a timely manner.

[0003] For example, when the wind speed is high, the energy storage system can store excess electrical energy, and when the wind speed decreases, the energy storage system can release electrical energy to ensure the stability of the output power. In stages where wind power fluctuates greatly, the instability of wind power may lead to an oversupply or undersupply of electricity, so it is necessary to predict wind power or power generation so that the storage station can adjust the energy storage output according to the predicted data, balance the supply and demand relationship in the power grid, and thus maintain the stability of the power grid. Parameters such as wind power or power generation can be predicted by setting a prediction model. For example, the patent document with publication number CN118114809A discloses a method and device for predicting wind power generation. A neural network model is trained with a historical data set to obtain a wind power generation prediction model to predict wind power generation for a wind farm.

[0004] In the prediction process, the closer the historical data is to the current moment, the more significant it is for the prediction. We can give the historical data a gradually decreasing weight by presetting a fixed smoothing parameter, and use the weighted calculation result as the prediction result for future data. This can be achieved through exponential smoothing, where the smoothing parameter determines the degree of reduction in the weight of the historical data. The larger the smoothing parameter, the greater the weight of the historical data closer to the current moment, and the smaller the weight of the historical data farther from the current moment. However, due to the instability of wind force, the fixed smoothing parameter cannot adapt well to the unstable wind factor, and there is a technical problem of poor prediction effect. Summary of the invention

[0005] In order to solve the technical problem of poor prediction effect caused by unstable wind factors, the present application provides an energy control method and system for a wind energy storage power station.

[0006] In a first aspect, the present application provides an energy control method for a wind energy storage power station, which adopts the following technical solution:

[0007] A method for regulating and controlling energy of a wind energy storage power station comprises the following steps: calculating a revised value of an initial smoothing parameter; taking the product of the initial smoothing parameter and the revised value as a revised smoothing parameter at a future moment; obtaining a prediction result according to the revised smoothing parameter and an exponential smoothing algorithm, and regulating and controlling electric energy according to the prediction result; the method for calculating the revised value is: taking wind force and time as a two-dimensional data point; calculating the local variation degree of wind force data and the noise degree of wind force data within a preset time range of the data point; taking the normalized result of the product of the local variation degree and the noise degree as the attention degree of the data point; setting a local window, and taking the ratio of the mean value of the data points in the local window whose attention degree is greater than a preset threshold value to the maximum value of the data points in the local window as the revised value.

[0008] The beneficial effects are as follows: by analyzing the local variation degree of any data point, the data is clustered, the noise degree of the data point is obtained, and the smoothing parameters of future data are obtained according to the local variation degree and the noise degree of the data point. The degree of attention is comprehensively quantified using the degree of local variation and the degree of noise. The degree of attention represents the prediction weight. The greater the degree of attention, the greater the weight of the data. The product of the degree of local variation and the degree of noise is normalized as the degree of attention, and the smoothing parameters are dynamically corrected to better adapt to the wind conditions of the real scene and improve the prediction effect.

[0009] Optionally, the local variation degree is calculated as:

[0010] , where Indicates The degree of local change of the data point at the moment; Represents the total number of data points within the time range; Indicates the time sequence; Indicates the ordinal number of the data point within the time range; Indicates Wind data value at the moment; express The corresponding time range Wind data values; Represents the standard normalization function.

[0011] The beneficial effect is: for wind data, due to its poor stability, for wind data collected at different times, it is necessary to obtain the difference between the data point and the surrounding data, that is, the degree of local change. The greater the degree of local change, the more drastic the change in wind data at that moment, which may mean that the instability of wind power output increases. Indicates The larger the mean value of the difference between the wind data at a certain moment and all the data points within its local range, the greater the degree of local change of the data point; conversely, the smaller the degree of local change of the data point.

[0012] Optionally, the local variation degree is calculated as:

[0013] , where Indicates The degree of local change of the data point at the moment; Represents the total number of data points within the time range; Indicates the time sequence; Indicates the ordinal number of the data point within the time range; Indicates Wind data value at the moment; express The corresponding time range Wind data values; Represents the standard normalization function.

[0014] The beneficial effects are: express and The larger the standard deviation, the greater the local variation of the data point; conversely, the smaller the local variation of the data point. The method of using squared differences to calculate the degree of local variation is more sensitive to outliers because squared differences will amplify the impact of larger differences.

[0015] Optionally, the method for calculating the degree of noise is: clustering the data points to obtain multiple clusters, and taking other clusters with the smallest distance values ​​from the cluster center of any cluster as similar clusters; calculating the degree of noise based on the Euclidean distance between the data point and the cluster center of the cluster where it is located and the Euclidean distance between the data point and the cluster center of the adjacent cluster.

[0016] The beneficial effect is: the purpose of clustering is to divide the samples in the data set into several groups, so that the similarity between data points in the same group is high, while the similarity between data points in different groups is low. The Euclidean distance between a data point and the center of its cluster reflects the similarity between the data point and other points in the cluster. A smaller distance means that the consistency between the data point and the center of the cluster is high, that is, it is more similar to other points in the cluster. The Euclidean distance between a data point and the center of a nearby cluster reflects the distinction between the data point and the adjacent cluster. A larger distance means that the data point is more different from the center of the adjacent cluster, that is, it is more different from the points in the adjacent cluster.

[0017] Optionally, the noise level is calculated as: , where Indicates The noise level of wind data at a given moment; Indicates The Euclidean distance between the wind data at a certain moment and the cluster center of the cluster to which it belongs; Indicates The Euclidean distance between the wind speed data at a certain moment and the cluster center of the nearest cluster.

[0018] The beneficial effect is that the degree of noise is quantified by the ratio of the distances. If the numerator and denominator are close, this may indicate that the data point does not belong to any obvious cluster and therefore has a high degree of noise.

[0019] Optionally, the noise level is calculated as: , where Indicates The noise level of wind data at a given moment; Indicates The Euclidean distance between the wind data at a certain moment and the cluster center of the cluster to which it belongs; Indicates The Euclidean distance between the wind speed data at the moment and the center of the clusters close to it; Indicates An exponential function with base .

[0020] The beneficial effect is that by calculating the absolute value of the difference between the two distances and applying exponential decay, a quantitative index to measure the impact of noise can be obtained. and The greater the difference between The smaller the value of , the more likely the data point is to be noise.

[0021] Optionally, the clustering algorithm adopts an iterative self-organizing clustering algorithm.

[0022] In a second aspect, the present application provides an energy control system for a wind energy storage power station, which adopts the following technical solution:

[0023] An energy control system for a wind energy storage power station comprises a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the energy control method for the wind energy storage power station described above is implemented.

[0024] The beneficial effect is that the energy regulation method of the above-mentioned wind energy storage power station is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0025] This application has the following technical effects:

[0026] 1. By analyzing the local variation degree of any data point, cluster the data, obtain the noise degree of the data point, and obtain the smoothing parameters of future data based on the local variation degree and noise degree of the data point. The degree of attention is comprehensively quantified using the local variation degree and the noise degree. The degree of attention represents the prediction weight. The greater the degree of attention, the greater the weight of the data. The product of the local variation degree and the noise degree is normalized as the degree of attention, and the smoothing parameters are dynamically corrected to better adapt to the wind conditions in real scenes and improve the prediction effect.

[0027] 2. For wind data, due to its poor stability, for wind data collected at different times, it is necessary to obtain the difference between the data point and the surrounding data, that is, the degree of local change. The greater the degree of local change, the more drastic the change in wind data at that moment, which may mean that the instability of wind power output increases.

[0028] 3. In the process of collecting wind force, due to the occurrence of instantaneous strong winds or gusts, the collected data is noise data. These data are greatly different from other data points and have little reference significance. Therefore, it is necessary to calculate the noise level of any data point. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-restrictive manner, and the same or corresponding numbers represent the same or corresponding parts.

[0030] Figure 1 It is a method flow chart of an energy control method of a wind energy storage power station according to an embodiment of the present application.

[0031] Figure 2 It is a method flow chart of a correction value calculation method in an energy control method of a wind energy storage power station in an embodiment of the present application. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0033] It should be understood that when the terms "first", "second", etc. are used in the claims, specification and drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0034] The present application embodiment discloses an energy control method for a wind energy storage power station, referring to Figure 1 , including steps S1 to S3, which are specifically as follows:

[0035] S1: Calculate the revised value of the initial smoothing parameter.

[0036] In one embodiment, wind data is acquired by building a wind tower near the wind energy storage station. The wind tower is built in the target wind farm to analyze the actual situation of wind energy resources in the wind farm. An anemometer is installed on the tower body to collect wind speed data, and the collected data is stored in a data recorder installed on the tower body. The preset collection frequency is 1 time / second.

[0037] Reference Figure 2 , the calculation method of the correction value includes steps S10 to S12:

[0038] S10: Taking the wind force and time as a two-dimensional data point; calculating the local change degree of the wind force data and the noise degree of the wind force data within a preset time range of the data point.

[0039] For wind data, due to its poor stability, for wind data collected at different times, it is necessary to obtain the difference between the data point and its surrounding data, that is, the degree of local change, so as to facilitate the subsequent acquisition of smoothing parameters based on the difference between the data point and its surrounding data, thereby making the prediction results more accurate.

[0040] In one embodiment, the preset experience range size ,Will As the The time range of the data point at the moment (excluding The local change degree of the data point is calculated based on the difference between the data point and the data points in the local range. The time range can be adjusted according to the actual application scenario, which will not be elaborated here.

[0041] In one embodiment, the calculation formula of the local change degree is:

[0042] , where Indicates The degree of local change of the data point at the moment; Represents the total number of data points within the time range; Indicates the time sequence; Indicates the ordinal number of the data point within the time range; Indicates Wind data value at the moment; express The corresponding time range Wind data values; Represents the standard normalization function.

[0043] in, Indicates The larger the mean value of the difference between the wind data at a certain moment and all the data points within its local range, the greater the degree of local change of the data point; conversely, the smaller the degree of local change of the data point.

[0044] In one embodiment, the calculation formula of the local change degree is:

[0045] , where Indicates The degree of local change of the data point at the moment; Represents the total number of data points within the time range; Indicates the time sequence; Indicates the ordinal number of the data point within the time range; Indicates Wind data value at the moment; express The corresponding time range Wind data values; Represents the standard normalization function.

[0046] express and The larger the standard deviation, the greater the local variation of the data point; conversely, the smaller the local variation of the data point. Compared with the previous embodiment, this embodiment is more sensitive to outliers because the squared difference will amplify the impact of larger differences.

[0047] In the process of collecting wind force, due to the occurrence of instantaneous strong winds or gusts, the collected data is noise data. These data are greatly different from other data points and have no reference significance. Therefore, it is necessary to calculate the noise level of any data point.

[0048] The noise level of wind data is calculated as:

[0049] Cluster the data points to obtain multiple clusters, and take the other clusters with the smallest distance value from the cluster center of any cluster as similar clusters; the clustering algorithm adopts the iterative self-organizing clustering algorithm. The iterative self-organizing clustering algorithm is a clustering algorithm based on self-organizing characteristics and iterative mechanisms. The algorithm gradually forms a stable clustering result by continuously adjusting the cluster center and the attribution of samples. The iterative self-organizing clustering algorithm is an existing technology and will not be described in detail here.

[0050] Cluster the data points to obtain multiple clusters, and take the other clusters with the smallest distance value from the cluster center of any cluster as similar clusters. Specifically, obtain the cluster centers of all clusters, and for any data point, in addition to the cluster where it is located, record the cluster where the cluster center with the closest Euclidean distance to the data point is located as the cluster closest to the data point (similar cluster).

[0051] For any data point, the degree of noise is calculated based on the Euclidean distance between the data point and the cluster center of the cluster in which it is located and the Euclidean distance between the data point and the cluster center of the adjacent cluster.

[0052] In one embodiment, the noise level is calculated as follows: , where Indicates The noise level of wind data at a given moment; Indicates The Euclidean distance between the wind data at a certain moment and the cluster center of the cluster to which it belongs; Indicates The Euclidean distance between the wind speed data at a certain moment and the cluster center of the nearest cluster.

[0053] Clustering will divide the data point into the cluster closest to the data point. If the numerator and denominator are close, it may indicate that the data point does not belong to any obvious cluster, so its noise level is high. On the contrary, the data point has a low noise level.

[0054] The smaller it is, the more similar the data point is to the data points in the cluster, that is, the lower the noise level of the data point is, otherwise the greater the noise level of the data point is; The larger the value is, the farther the data point is from the data points in other clusters, that is, the greater the difference is. It also means that the clustering effect is better. The higher the clustering effect is, the more it characterizes the molecular The higher the confidence level.

[0055] In one embodiment, the noise level is calculated as follows: , where Indicates The noise level of wind data at a given moment; Indicates The Euclidean distance between the wind data at a certain moment and the cluster center of the cluster to which it belongs; Indicates The Euclidean distance between the wind speed data at the moment and the center of the clusters close to it; Indicates An exponential function with base .

[0056] S11: The normalized result of the product of the local change degree and the noise degree is used as the attention degree of the data point.

[0057] The higher the local variation degree of a data point, the higher the attention degree of the data point, that is, the greater the weight of the data point. At the same time, in order to avoid the data point being a noise data point, the normalized result of the product of the local variation degree of the data point and the noise degree is taken as the attention degree of the data point.

[0058] S12: Setting a local window, and taking the ratio of the mean of the data points in the local window whose attention level is greater than a preset threshold to the maximum value of the data points in the local window as a correction value.

[0059] In one embodiment, for any moment that needs to be predicted, the size is set to local window, get the continuous The attention level of the data point at the moment, and obtain the attention level greater than the preset experience threshold The local window and the preset threshold value can be set by the implementer according to the specific implementation situation, for example , .

[0060] S2: The product of the initial smoothing parameter and the revised value is used as the revised smoothing parameter at the future moment.

[0061] According to the change of attention level, the corrected smoothing parameter corresponding to the data point is calculated. The calculation formula of the corrected smoothing parameter is: , where Indicates smoothing parameter at time instant; Indicates the ordinal number of the time that needs to be predicted; represents the preset initial smoothing parameter, for example ; Indicates The level of attention in the local window at the moment is greater than the preset threshold The mean of all data points at the time corresponding to ; Indicates The maximum value of the moment in the local window of the moment (the moment before the moment is the maximum value of the moment in the local window of the moment).

[0062] in, Indicates the position of the data with the highest degree of attention in the local window at that moment. The larger the value, the further back it is located, and the smaller the value, the more forward it is located. The degree of attention represents the prediction weight. The greater the degree of attention, the greater the weight of the data. For example, if The smaller it is, the more weighted the data is, and the earlier the corresponding time is, which means that a smaller smoothing parameter should be used to make the weight of the earlier data larger; on the contrary, if The larger it is, the later the data with greater weight is, which means that a larger smoothing parameter is needed to make the weight of the later data greater.

[0063] The smoothing parameter determines the extent to which the weight of historical data is reduced. The larger the smoothing parameter, the greater the weight of historical data closer to the current moment, and the smaller the weight of historical data farther away from the current moment.

[0064] S3: Obtain prediction results based on the modified smoothing parameters and the exponential smoothing algorithm, and perform power regulation based on the prediction results.

[0065] For any moment that needs to be predicted, the exponential smoothing method is used to predict according to the smoothing parameters corresponding to the moment, and the prediction result can be obtained. Exponential smoothing method prediction is an existing technology and will not be described here. The staff is asked to adjust according to the prediction results. For example, when the predicted wind speed is high, the wind energy storage station is asked to store electricity through wind charging to maximize cost-effectiveness. When the wind speed is low, the wind energy storage station is asked to inject the stored electricity into the power grid to maintain the stability of the power grid.

[0066] An embodiment of the present application also discloses an energy control system for a wind energy storage power station, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an energy control method for a wind energy storage power station according to the present application is implemented.

[0067] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0068] In the present application, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high bandwidth memory HBM (High Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0069] Although this specification has shown and described a plurality of embodiments of the present application, it is obvious to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, modifications and alternatives without departing from the thought and spirit of the present application. It should be understood that in the process of practicing the present application, various alternatives to the embodiments of the present application described herein may be adopted.

[0070] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for regulating energy in a wind energy storage power station, characterized in that: Includes steps: Calculate the correction value of the initial smoothing parameter; take the product of the initial smoothing parameter and the correction value as the corrected smoothing parameter at the future moment; obtain the prediction result based on the corrected smoothing parameter and the exponential smoothing algorithm, and perform power regulation based on the prediction result; the calculation method of the correction value is: Take wind force and time as a two-dimensional data point; calculate the local change degree of wind force data and the noise degree of wind force data within a preset time range; The normalized result of the product of the local change degree and the noise degree is used as the attention degree of the data point; A local window is set, and the ratio of the time mean of all data points in the local window at a certain moment whose attention level is greater than a preset threshold to the time maximum value of the data points in the local window at that moment is taken as the correction value.

2. The energy control method of the wind energy storage power station according to claim 1 is characterized in that: The calculation formula for the degree of local change is: , where Indicates The degree of local change of the data point at the moment; Represents the total number of data points within the time range; Indicates the time sequence; Indicates the ordinal number of the data point within the time range; Indicates Wind data value at the moment; express The corresponding time range Wind data values; Represents the standard normalization function.

3. The energy control method of the wind energy storage power station according to claim 1 is characterized in that: The calculation formula for the degree of local change is: , where Indicates The degree of local change of the data point at the moment; Represents the total number of data points within the time range; Indicates the time sequence; Indicates the ordinal number of the data point within the time range; Indicates Wind data value at the moment; express The corresponding time range Wind data values; Represents the standard normalization function.

4. The energy control method of the wind energy storage power station according to claim 1 is characterized in that: The noise level is calculated as: Cluster the data points to obtain multiple clusters, and take the other clusters with the smallest distance value from the cluster center of any cluster as similar clusters; The degree of noise is calculated based on the Euclidean distance between the data point and the cluster center of the cluster in which it belongs and the Euclidean distance between the data point and the cluster center of the adjacent cluster.

5. The energy control method of the wind energy storage power station according to claim 4 is characterized in that: The noise level is calculated as: , where Indicates The noise level of wind data at a given moment; Indicates The Euclidean distance between the wind data at a certain moment and the cluster center of the cluster to which it belongs; Indicates The Euclidean distance between the wind speed data at a certain moment and the cluster center of the nearest cluster.

6. The energy control method of the wind energy storage power station according to claim 4 is characterized in that: The noise level is calculated as: , where Indicates The noise level of wind data at a given moment; Indicates The Euclidean distance between the wind data at a certain moment and the cluster center of the cluster to which it belongs; Indicates The Euclidean distance between the wind speed data at the moment and the center of the clusters close to it; Indicates An exponential function with base .

7. The energy control method of the wind energy storage power station according to claim 4 is characterized in that: The clustering algorithm adopts iterative self-organizing clustering algorithm.

8. An energy control system for a wind energy storage power station, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the energy control method of the wind energy storage power station according to any one of claims 1-7 is implemented.

Images