Method and device for determining typical scene of wind and light output, electronic equipment and storage medium

By dividing and clustering of wind and light output data on long and short time scales, the problem of low accuracy in typical wind and light output scenarios is solved, and the energy planning and risk assessment capabilities of the distribution network system are improved.

CN118941409APending Publication Date: 2024-11-12GUANGDONG POWER GRID CO LTD +1
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410970637.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The timing and seasonal characteristics of wind and light output in the distribution network system have not been effectively considered, resulting in low accuracy of typical wind and light output scenarios, which poses challenges to the optimization and scheduling of the distribution network.

Method used

By dividing the load data into multiple time-continuous target scene sets on a long time scale, and clustering the scenery output data in each target scene set on a short time scale, the target is generated using K-Means clustering and other algorithms. A typical scenario of landscape output, eliminate abnormal data and fill in missing data.

Benefits of technology

It improves the accuracy of generation of typical scenes of wind and light output, retains the seasonal and sequential characteristics of load and wind and light output, helps to guide the energy planning and optimal allocation of resources of the distribution network system, and realizes economic analysis and risk assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941409A_ABST
    Figure CN118941409A_ABST
Patent Text Reader

Abstract

The invention discloses a wind and light output typical scene determination method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring load data and wind-light output data of a wind-light power station in a preset period; performing time division on the load data to obtain target scene sets in different time periods; and clustering the wind and light output data in each target scene set to obtain a target wind and light output typical scene corresponding to each target scene set. Load data are divided into a plurality of target scene sets with continuous time on a long time scale, and then the wind and light output data in each target scene set are clustered on a short time scale, so that a required target wind and light output typical scene is generated, the seasonal and sequential characteristics of load and wind and light output are effectively reserved, and the target wind and light output efficiency is improved. The generation accuracy of the wind and light output typical scene is improved, energy planning construction and resource optimization configuration of the power distribution network system are guided, and economic analysis and risk evaluation of the power distribution network system are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, device, electronic device and storage medium for determining typical wind and solar power output scenarios. Background Art

[0002] With the large-scale decentralized access of new energy to the distribution network system, the traditional source-follow-load distribution network system has been transformed into a smart distribution network system with mutual regulation between the source and the load. However, the access of a high proportion of new energy and other equipment has resulted in a huge amount of data generated by the distribution network terminals, which is difficult to calculate and analyze. In addition, the output of wind and solar energy is limited by time and space, and has the characteristics of high uncertainty and randomness, which has brought huge challenges to the optimization and scheduling of the distribution network system.

[0003] In the existing technology, scenario analysis is usually used to achieve optimal dispatching of distribution networks. However, the geographical location of distributed wind and solar power generation in distribution networks is complex, with typical characteristics of small capacity and large spatial dimensions. In addition, wind and solar output has obvious seasonal characteristics, and the output data trends vary greatly in different seasons, resulting in low accuracy of the generated typical wind and solar output scenarios. Summary of the invention

[0004] The present invention provides a method, device, electronic device and storage medium for determining a typical scene of wind and solar power output, so as to improve the generation accuracy of the typical scene of wind and solar power output.

[0005] According to one aspect of the present invention, a method for determining a typical scene of wind and solar power output is provided, the method comprising:

[0006] Obtain load data and wind and solar power output data within a preset period of the wind and solar power station;

[0007] Perform time segmentation on the load data to obtain target scenario sets in different time periods;

[0008] The wind and solar power output data in each target scene set are clustered to obtain the target wind and solar power output typical scenes corresponding to each target scene set.

[0009] According to another aspect of the present invention, a device for determining a typical scene of wind and solar power output is provided, the device comprising:

[0010] A data acquisition module is used to obtain load data and wind and solar power output data within a preset period of the wind and solar power station;

[0011] The scenario set division module is used to divide the load data into time segments to obtain target scenario sets in different time periods;

[0012] The typical scenario determination module is used to cluster the wind and solar power output data in each target scenario set to obtain the target wind and solar power output typical scenario corresponding to each target scenario set.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining a typical scene of wind and solar power output described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining a typical wind and solar power output scenario described in any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention divides the load data into multiple time-continuous target scenario sets on a long time scale, and then clusters the wind and solar power output data in each target scenario set on a short time scale, thereby generating the required target wind and solar power output typical scenarios. It solves the problem of low accuracy of the generated wind and solar power output typical scenarios due to failure to take into account the temporal characteristics of wind and solar power output, effectively retains the seasonal and temporal characteristics of the load and wind and solar power output, improves the generation accuracy of the wind and solar power output typical scenarios, and is conducive to guiding the energy planning and construction of the distribution network system and the optimal allocation of resources, and realizing the economic analysis and risk assessment of the distribution network system.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 is a flow chart of a method for determining a typical scene of wind and solar power output provided according to Embodiment 1 of the present invention;

[0022] Figure 2is a flow chart of a method for determining a typical scene of wind and solar power output provided according to Embodiment 2 of the present invention;

[0023] Figure 3 is a flowchart of another method for determining a typical scene of wind and solar power output provided according to Embodiment 2 of the present invention;

[0024] Figure 4 is a full-year photovoltaic output scenario diagram of a photovoltaic power station provided according to the second embodiment of the present invention;

[0025] Figure 5 is a diagram of wind power output scenarios of a wind power plant throughout the year provided according to the second embodiment of the present invention;

[0026] FIG6 is a clustering result diagram of photovoltaic output scenarios throughout the year provided according to the second embodiment of the present invention;

[0027] Figure 7 This is a result diagram of a typical photovoltaic output scenario provided according to the second embodiment of the present invention;

[0028] Figure 8 is a probability result diagram of a typical photovoltaic output scenario provided according to the second embodiment of the present invention;

[0029] FIG9 is a clustering result diagram of wind power output scenarios throughout the year provided according to the second embodiment of the present invention;

[0030] Fig.10 This is a result diagram of a typical wind power output scenario provided according to the second embodiment of the present invention;

[0031] Fig.11 is a probability result diagram of a typical wind power output scenario provided according to the second embodiment of the present invention;

[0032] Fig.12 2 is a schematic diagram of the structure of a typical scene determination device for wind and solar power output provided according to Embodiment 3 of the present invention;

[0033] Fig.13 It is a structural schematic diagram of an electronic device for implementing the method for determining a typical scene of wind and solar power output according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] Embodiment 1

[0037] Figure 1 A flowchart of a method for determining a typical wind-solar output scenario is provided for the first embodiment of the present invention. This embodiment is applicable to the case of determining a typical wind-solar output scenario in a large-scale distributed wind-solar output scenario of a distribution network. The method can be executed by a wind-solar output typical scenario determination device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device. Figure 1 As shown, the method for determining a typical wind and solar power output scene provided in the first embodiment specifically includes the following steps:

[0038] S110, obtaining load data and wind / solar power output data of a wind / solar power station within a preset period.

[0039] Among them, the wind and solar power station may include a wind power plant and a photovoltaic power station. The preset period may refer to a time period for determining a typical scene of wind and solar output, and the preset period may include at least 1 year (i.e. 365 days), etc. The load data may refer to the sampled data of the load power consumption. The wind and solar output data may include the sampled data of wind power generation power (i.e. wind power output value) and photovoltaic power generation power (i.e. photovoltaic output value). There may be multiple load data and wind and solar output data, and each load data or wind and solar output data corresponds to the sampled data of the preset collection time period. Exemplarily, the load data and wind and solar output data may be time series data. The preset collection time period may be set and adjusted according to the experience of the technicians, and the number of sampling points of the load data and wind and solar output data may also be set and adjusted according to the experience of the technicians. Exemplarily, the preset time period may be 1 day, each load data may be the sampled data of the load power consumption of 1 day, and each wind and solar output data may also include the sampled data of the wind power output value and the photovoltaic output value of 1 day. The number of sampling points may be 96, and the time interval between each sampling point may be 15 minutes, that is, load data and wind and solar power output data are collected every 15 minutes within a day.

[0040] In an embodiment of the present invention, load data and wind-solar output data for a wind-solar power station within a preset period can be obtained from a data storage location such as a local or cloud server. For example, historical data of load data and wind-solar output data of a wind-solar power station for a whole year can be obtained to serve as reference data for determining subsequent typical wind-solar output scenarios.

[0041] S120 , time segmenting the load data to obtain target scenario sets in different time periods.

[0042] Among them, the target scenario set can be understood as using a certain preset data classification algorithm to divide the load data within a preset period into multiple time-continuous scenario sets. The preset data classification algorithm can refer to a data classification algorithm that does not change the temporal sequence of the load data, for example, it can at least include: an ordered clustering algorithm, etc.

[0043] In an embodiment of the present invention, a data classification algorithm that does not change the temporal nature of the data, such as ordered clustering, may be used to divide the load data within a preset period, thereby obtaining a plurality of time-continuous target scene sets in different time periods. This embodiment effectively retains the seasonal and temporal characteristics of the load and wind and solar output by performing temporal division of the target scene set according to the load data on a long time scale, and then subsequently clustering the wind and solar output data in each target scene set to generate a corresponding target wind and solar output typical scene, which is beneficial to improving the accuracy of determining the wind and solar output typical scene.

[0044] S130 , clustering the wind and solar power output data in each target scene set to obtain a typical target wind and solar power output scene corresponding to each target scene set.

[0045] Among them, the target wind and solar output typical scenario can be understood as a set of data that can represent the wind and solar output data within the target scenario set. The target wind and solar output typical scenario can include a piece of wind power output value data and a piece of photovoltaic output value data under the corresponding target scenario set.

[0046] In an embodiment of the present invention, clustering algorithms such as K-Means clustering and Mean Shift clustering can be used to cluster the wind and solar output data in each divided target scene set, thereby generating a target wind and solar output typical scene corresponding to each target scene set. Based on the division of multiple time-continuous target scene sets on a long time scale, this embodiment clusters each target scene set on a short time scale, thereby generating the required target wind and solar output typical scene. The generation accuracy of the wind and solar output typical scene can be improved through the multi-time scale (long time scale + short time scale) clustering method, which helps to guide the optimization of the distribution network system scheduling and risk assessment, thereby promoting the consumption of new energy, reducing the operating cost of the power system, and having the advantages of both economic and clean power generation, and ultimately opening up new ideas for the power system to effectively consume new energy and realize economic, clean and low-carbon power supply.

[0047] Further, based on the above-mentioned embodiments of the invention, this embodiment provides a method for determining a typical scene of wind and solar power output, further comprising:

[0048] Call the density-based clustering algorithm to remove abnormal data and low-probability data in load data and wind and solar power output data;

[0049] The K nearest neighbor algorithm is called to fill in the missing data in the removed load data and wind and solar power output data.

[0050] In the embodiment of the present invention, since there may be some abnormal data and low-probability data in the acquired load data and wind and solar power output data, a density-based clustering algorithm, namely, a DBSCAN clustering algorithm, can be used to pre-process the acquired data, remove the abnormal data and low-probability data therein, and then call the K-Nearest Neighbor (KNN) algorithm to fill in the missing data. This embodiment can improve the accuracy and reliability of the data by removing the abnormal data and low-probability data in the load data and wind and solar power output data, and filling in the missing values, thereby helping to improve the accuracy of the generation of subsequent typical wind and solar power output scenarios.

[0051] The technical solution of the embodiment of the present invention divides the load data into multiple time-continuous target scenario sets on a long time scale, and then clusters the wind and solar power output data in each target scenario set on a short time scale, thereby generating the required target wind and solar power output typical scenarios. It solves the problem of low accuracy of the generated wind and solar power output typical scenarios due to failure to take into account the temporal characteristics of wind and solar power output, effectively retains the seasonal and temporal characteristics of the load and wind and solar power output, improves the generation accuracy of the wind and solar power output typical scenarios, and is conducive to guiding the energy planning and construction of the distribution network system and the optimal allocation of resources, and realizing the economic analysis and risk assessment of the distribution network system.

[0052] Embodiment 2

[0053] Figure 2 The flowchart of a method for determining a typical wind and solar power output scenario provided in the second embodiment of the present invention is further optimized and expanded based on the above implementation, and can be combined with various optional technical solutions in the above implementation. Figure 2 As shown, a method for determining a typical wind and solar power output scene provided in the second embodiment specifically includes the following steps:

[0054] S210: Obtain load data and wind / solar power output data of a wind / solar power station within a preset period.

[0055] S220: Obtain a first cluster number candidate value set corresponding to a preset ordered clustering algorithm; wherein the first cluster number candidate value set includes at least one first cluster number candidate value.

[0056] Among them, the preset ordered clustering algorithm may refer to a clustering algorithm that can ensure the temporal continuity of clustering results and the differences in different time periods. Different from the clustering algorithm that disrupts the time sequence, this embodiment uses the preset ordered clustering algorithm to effectively retain the seasonal and sequential characteristics of load and wind and solar output. The first cluster number candidate value set may refer to a set of several first cluster number candidate values ​​corresponding to the preset ordered clustering algorithm. The number and value of the first cluster number candidate values ​​can be set according to actual needs. This embodiment does not limit this. Exemplarily, the first cluster number candidate value set may include the following first cluster number candidate values: {1, 2, 3, 4, 5, 6, 7, 8, 9}.

[0057] In an embodiment of the present invention, a first set of candidate values ​​of the number of clusters corresponding to a preset ordered clustering algorithm can be obtained first, so that the silhouette coefficient can be used as an evaluation index of the quality of the clustering results, thereby determining the best clustering result for long-term scale division of load data based on the preset ordered clustering algorithm.

[0058] S230, selecting each first cluster number candidate value in turn, calling a preset ordered clustering algorithm to cluster the load data, and obtaining a first clustering result corresponding to the first cluster number candidate value when a preset time segmentation objective function reaches a minimum.

[0059] The preset time segmentation objective function may refer to a pre-configured time segmentation objective for a preset ordered clustering algorithm. In this embodiment, the preset time segmentation objective may be: ensuring that the sum of squares between the classified sample data and the average value of the sample data is minimized.

[0060] In an embodiment of the present invention, the preset ordered clustering algorithm can be called respectively to cluster the load data within a preset period under different values ​​of the first clustering number candidate value, and when the preset time division objective function is minimized, the iterative process of the preset ordered clustering algorithm is stopped to obtain the first clustering result under different first clustering number candidate values, wherein the first clustering result is a series of time-continuous scene sets obtained after time division.

[0061] Further, based on the above-mentioned embodiments of the invention, the preset time segmentation objective function can be expressed as:

[0062]

[0063] Where G represents the number of sample data; i represents the sample data sequence number; x i Represents the i-th sample data; Represents the mean of the sample data.

[0064] S240: For each first cluster number candidate value, determine a silhouette coefficient corresponding to the first clustering result.

[0065] Among them, the silhouette coefficient is an evaluation indicator used to measure the quality of clustering results. The larger the silhouette coefficient, the smaller the difference within the scene set and the larger the difference outside the scene set, that is, the better the clustering effect.

[0066] In the embodiment of the present invention, in order to evaluate the clustering effect of the first clustering result under different values ​​of the first cluster number candidate value, the silhouette coefficient can be used as an evaluation index to determine the silhouette coefficients of the first clustering result under different first cluster number candidate values. The silhouette coefficient can be determined by the following formula:

[0067]

[0068] Among them, s i Represents the silhouette coefficient corresponding to the i-th sample data; a i represents the average Euclidean distance between the i-th sample data and all other sample data in the same cluster; b iIt represents the average Euclidean distance between the i-th sample data and all sample data in the nearest cluster; max() represents the maximum value function.

[0069] The value range of the silhouette coefficient is (-1,1), where a positive silhouette coefficient indicates that the sample is more similar to the samples in the scene set in which it is located; a negative silhouette coefficient indicates that the sample is more similar to the samples in other scene sets; and a silhouette coefficient of 0 indicates that the samples in the two scene sets have the same similarity and should belong to the same scene set.

[0070] In this embodiment, the above formula may be used to sequentially calculate the silhouette coefficient corresponding to each sample data in the first clustering result, and then the average of the silhouette coefficients of all the sample data is used as the silhouette coefficient corresponding to the first clustering result.

[0071] S250: Determine the first cluster number candidate value corresponding to the maximum silhouette coefficient as the first optimal cluster number, and use the first clustering result corresponding to the first optimal cluster number as the target scene set for different time periods.

[0072] In an embodiment of the present invention, since a larger silhouette coefficient indicates a better clustering effect, the first clustering number candidate value corresponding to the largest silhouette coefficient can be determined as the first optimal clustering number among all first clustering results, and then the first clustering result corresponding to the first optimal clustering number is used as a series of time-continuous target scene sets after the final long time scale division.

[0073] S260, obtaining a second cluster number candidate value set corresponding to a preset K-Means clustering algorithm and a preset iteration termination condition; wherein the second cluster number candidate value set includes at least one second cluster number candidate value.

[0074] Among them, the preset K-Means clustering algorithm is a clustering analysis algorithm based on iterative solution, and its core idea is to divide the samples in the data set into K different cluster clusters, where each cluster cluster is represented by the average value of all samples in the cluster cluster (i.e., the cluster center). Similarly, the second cluster number candidate value set may refer to a set containing several second cluster number candidate values ​​corresponding to the preset K-Means clustering algorithm. The preset iteration termination condition may refer to a judgment condition configured for the preset K-Means clustering algorithm for determining whether to terminate the clustering algorithm iteration process, for example, it may include but is not limited to: whether the preset maximum number of iterations has been reached, whether all cluster centers no longer change, whether the offset values ​​of all cluster centers after the last iteration are less than the preset value, etc.

[0075] In an embodiment of the present invention, a second cluster number candidate value set corresponding to a preset K-Means clustering algorithm and a preset iteration termination condition can be obtained in advance, so that the preset K-Means clustering algorithm can be called subsequently to cluster the wind and solar power output data in each target scene set, thereby obtaining the corresponding typical wind and solar output scene.

[0076] S270, selecting each second cluster number candidate value in turn, calling the preset K-Means clustering algorithm to cluster the wind and solar power output data in each target scene set, and when the preset iteration termination condition is reached, obtaining the second clustering result under the corresponding second cluster number candidate value.

[0077] In an embodiment of the present invention, the preset K-Means clustering algorithm can be called respectively under different values ​​of the second cluster number candidate value to cluster the wind and solar power output data in each target scene set divided into short time scales, and when the preset iteration termination condition is met, the iteration process of the preset K-Means clustering algorithm is stopped to obtain the second clustering results under different second cluster number candidate values, wherein the second clustering result is the typical wind and solar power output scene under the corresponding scene set.

[0078] Further, based on the above-mentioned embodiment of the invention, S270 specifically includes the following steps:

[0079] S2701, based on the selected second cluster number candidate values, initializing a corresponding number of cluster centers in each target scene set;

[0080] S2702, determining the Euclidean distance between each wind and solar power output data in the target scene set and each cluster center, and assigning each wind and solar power output data to the cluster with the closest distance according to the Euclidean distance;

[0081] S2703, determining the average value of all wind and solar power output data in each cluster, and updating the cluster center of the corresponding cluster according to the average value;

[0082] S2704, determine whether the preset iteration termination condition is currently reached. If so, generate the second clustering result under the second cluster number candidate value corresponding to the target scene set; if not, return to execute the step of determining the Euclidean distance between each wind and solar power output data in the target scene set and each cluster center, and distribute each wind and solar power output data to the nearest cluster cluster according to the Euclidean distance.

[0083] Specifically, under different values ​​of the second clustering number candidate value, a corresponding number of cluster centers can be initialized in each target scene set. For example, a corresponding number of sample data can be randomly selected in the target scene set as the initial cluster center; then the Euclidean distance between each wind and solar output data in the target scene set and each cluster center is determined, and each wind and solar output data is assigned to the cluster cluster with the nearest distance according to the Euclidean distance; then the cluster center of the corresponding cluster cluster is updated according to the average value of all wind and solar output data in each cluster cluster; S2702 and S2703 are continuously iterated and executed until the preset iteration termination condition is reached, and the iterative process is terminated, thereby obtaining the second clustering result of each target scene set under different second clustering number candidate values.

[0084] S280. Determine a silhouette coefficient corresponding to a second clustering result for each second cluster number candidate value of the target scene set.

[0085] In the embodiment of the present invention, in order to evaluate the clustering effect of the second clustering result under different values ​​of the second cluster number candidate value, the silhouette coefficient can be used as an evaluation index to determine the silhouette coefficients corresponding to the second clustering result under different second cluster number candidate values. The process of determining the silhouette coefficient is the same as that in S240 above, and will not be repeated here.

[0086] S290. Determine the second cluster number candidate value corresponding to the maximum silhouette coefficient as the second optimal cluster number, and use the cluster center of the second cluster result corresponding to the second optimal cluster number as the target wind and solar output typical scene of the corresponding target scene set, and at the same time determine the occurrence probability corresponding to the target wind and solar output typical scene.

[0087] In an embodiment of the present invention, since a larger silhouette coefficient indicates a better clustering effect, the second cluster number candidate value corresponding to the largest silhouette coefficient can be determined as the second optimal cluster number among all second clustering results, and the cluster center of the second clustering result corresponding to the second optimal clustering number is used as the target wind and solar output typical scene of the corresponding target scene set, and the occurrence probability of each target wind and solar output typical scene is determined, that is, the sum of all scene probabilities in each cluster cluster in the second clustering result is used as the occurrence probability of the corresponding target wind and solar output typical scene.

[0088] The technical solution of the embodiment of the present invention divides the load data into multiple optimal and time-continuous target scenario sets by using a preset ordered clustering algorithm and a contour coefficient evaluation index on a long time scale; then clusters the wind and solar power output data in each target scenario set on a short time scale by using a preset K-Means clustering algorithm and a contour coefficient evaluation index, thereby generating the optimal target wind and solar power output typical scenario, thereby solving the problem of low accuracy of the generated wind and solar power output typical scenario due to failure to consider the temporal characteristics of the wind and solar power output, effectively retaining the seasonal and temporal characteristics of the load and wind and solar power output, and at the same time using the contour coefficient evaluation index to effectively improve the generation accuracy of the wind and solar power output typical scenario, which is beneficial to guiding the energy planning and construction of the distribution network system and the optimal allocation of resources, and realizing the economic analysis and risk assessment of the distribution network system.

[0089] Figure 3 This is a flow chart of another method for determining a typical wind-solar output scene provided in Embodiment 2 of the present invention. As shown in 3, the method for determining a typical wind-solar output scene is similar to the above embodiment, and the details can be referred to the above embodiment, which will not be repeated here.

[0090] In order to enable those skilled in the art to better understand the embodiments of the present invention, the technical solutions of the present invention are described below in conjunction with specific embodiments.

[0091] This embodiment uses a certain year's historical data of a wind power plant and a photovoltaic power plant with an installed capacity of 100kW in a certain area for simulation analysis, and uses MATLAB tools for calculation and analysis. The annual historical output scenarios of the photovoltaic power plant and the wind power plant are as follows: Figure 4 and Figure 5 shown.

[0092] In order to verify the method for determining typical wind and solar power output scenarios proposed in this embodiment, this embodiment pre-processes the data based on the annual historical output data of wind power plants and photovoltaic power stations in a certain area, excludes low-probability and abnormal scenarios, and fills in missing data. On this basis, based on the ordered clustering algorithm, the annual wind power and photovoltaic output data are segmented by time nodes. Finally, the K-Means clustering algorithm is used to aggregate the scene sets after time division, thereby obtaining typical wind and solar power output scenarios.

[0093] Figure 6 shows the clustering results after the photovoltaic output scenes throughout the year are aggregated into 6 categories. As can be seen from Figure 6, in the same cluster, multiple photovoltaic output scenes are not only similar in amplitude, but also similar in change trend. Among them, the photovoltaic output amplitudes of clusters 1, 3 and 6 are large and the fluctuation is small, while the amplitudes of clusters 2, 4 and 5 are small and the volatility is large. After K-Means clustering, the photovoltaic output scenes throughout the year can be effectively divided into several typical photovoltaic output scenes, which can well characterize the uncertainty and volatility characteristics of photovoltaic output.

[0094] Figure 7 The cluster center of each cluster after the photovoltaic output scene is clustered is shown, that is, the final typical output scene obtained after each cluster is reduced. Figure 7 It can be seen that the final typical output scenario is similar to the output scenario cluster in the corresponding cluster cluster in Figure 6 in terms of amplitude and change trend, which illustrates the effectiveness of this scheme. Aggregating and reducing a large number of photovoltaic output scenarios helps guide the optimal scheduling and risk assessment of the distribution network system. Figure 8 The probability of each typical photovoltaic output scenario after the aggregation of photovoltaic output scenarios is shown. Among them, the probability values ​​of scenario 3 and scenario 6 are relatively large, indicating that the probability of these two typical photovoltaic output scenarios occurring is the highest, while the probability of other scenarios occurring is relatively small.

[0095] Figure 9 shows the clustering results after the wind power output scenes throughout the year are aggregated into 4 categories. As can be seen from Figure 9, the wind power output scenes throughout the year are highly volatile and random, making it difficult to make accurate predictions. After clustering the wind power output scenes using the K-Means clustering algorithm, typical wind power output scenes can be effectively extracted, which is similar to the photovoltaic clustering effect. The wind power output scene characteristics in the same cluster are similar, but the differences between different clusters are small, indicating that the wind power output in the region is relatively stable that year.

[0096] Fig.10 The cluster center of each category after the wind power output scene clustering is shown, that is, the final typical output scene obtained after each category is reduced. Fig.10 It can be seen that the final typical output scenario is similar to the wind power output scenario in the corresponding cluster in Figure 9 in terms of amplitude and change trend, which further illustrates the effectiveness of this scheme. Fig.11 The probability of each typical output scenario after the aggregation of wind power scenarios is shown, among which the probability values ​​of each scenario are not much different, which also shows that the wind power output in the region is relatively stable that year.

[0097] Embodiment 3

[0098] Fig.12 This is a schematic diagram of the structure of a typical scene determination device for wind and solar output provided by Embodiment 3 of the present invention. Fig.12 As shown, the device comprises:

[0099] The data acquisition module 31 is used to acquire the load data and wind and solar power output data of the wind and solar power station within a preset period;

[0100] The scenario set division module 32 is used to divide the load data into time segments to obtain target scenario sets in different time periods;

[0101] The typical scene determination module 33 is used to cluster the wind and solar power output data in each target scene set to obtain the target wind and solar power output typical scene corresponding to each target scene set.

[0102] The technical solution of the embodiment of the present invention divides the load data into multiple time-continuous target scenario sets on a long time scale, and then clusters the wind and solar power output data in each target scenario set on a short time scale, thereby generating the required target wind and solar power output typical scenarios. It solves the problem of low accuracy of the generated wind and solar power output typical scenarios due to failure to take into account the temporal characteristics of wind and solar power output, effectively retains the seasonal and temporal characteristics of the load and wind and solar power output, improves the generation accuracy of the wind and solar power output typical scenarios, and is conducive to guiding the energy planning and construction of the distribution network system and the optimal allocation of resources, and realizing the economic analysis and risk assessment of the distribution network system.

[0103] Further, based on the above-mentioned embodiment of the invention, the scene set division module 32 includes:

[0104] A first candidate value set acquisition unit, used to acquire a first cluster number candidate value set corresponding to a preset ordered clustering algorithm; wherein the first cluster number candidate value set includes at least one first cluster number candidate value;

[0105] A first clustering unit is used to select each first clustering number candidate value in turn, call a preset ordered clustering algorithm to cluster the load data, and obtain a first clustering result corresponding to the first clustering number candidate value when the preset time segmentation objective function reaches a minimum;

[0106] A first silhouette coefficient determining unit, configured to determine a silhouette coefficient corresponding to a first clustering result for each first cluster number candidate value;

[0107] The target scene set determination unit is used to determine the first cluster number candidate value corresponding to the maximum silhouette coefficient as the first optimal cluster number, and use the first clustering result corresponding to the first optimal cluster number as the target scene set for different time periods.

[0108] Further, based on the above-mentioned embodiment of the invention, the preset time segmentation objective function is expressed as:

[0109]

[0110] Where G represents the number of sample data; i represents the sample data sequence number; x i Represents the i-th sample data; Represents the mean of the sample data.

[0111] Further, based on the above-mentioned embodiment of the invention, the typical scene determination module 33 includes:

[0112] A second candidate value set acquisition unit, used to acquire a second cluster number candidate value set corresponding to a preset K-Means clustering algorithm and a preset iteration termination condition; wherein the second cluster number candidate value set includes at least one second cluster number candidate value;

[0113] The second clustering unit is used to select each second clustering number candidate value in turn, call a preset K-Means clustering algorithm to cluster the wind and solar power output data in each target scene set, and obtain a second clustering result corresponding to the second clustering number candidate value when a preset iteration termination condition is reached;

[0114] A second silhouette coefficient determining unit, configured to determine a silhouette coefficient corresponding to a second clustering result for each second cluster number candidate value of the target scene set;

[0115] The target wind and solar power output typical scenario and probability generation unit is used to determine the second cluster number candidate value corresponding to the maximum silhouette coefficient as the second optimal cluster number, and the cluster center of the second clustering result corresponding to the second optimal cluster number as the target wind and solar power output typical scenario of the corresponding target scenario set, and at the same time determine the occurrence probability corresponding to the target wind and solar power output typical scenario.

[0116] Further, based on the above-mentioned embodiment of the invention, the second clustering unit is specifically used for:

[0117] Initializing a corresponding number of cluster centers in each target scene set based on the selected second cluster number candidate values;

[0118] Determine the Euclidean distance between each wind and solar power output data in the target scene set and each cluster center, and assign each wind and solar power output data to the cluster with the closest distance according to the Euclidean distance;

[0119] Determine the average value of all wind and solar power output data in each cluster, and update the cluster center of the corresponding cluster according to the average value;

[0120] Determine whether the preset iteration termination condition is currently reached. If so, generate the second clustering result under the second clustering number candidate value corresponding to the target scene set; if not, return to execute the step of determining the Euclidean distance between each wind and solar output data in the target scene set and each cluster center, and assign each wind and solar output data to the nearest cluster cluster according to the Euclidean distance.

[0121] Further, based on the above-mentioned embodiment of the invention, the silhouette coefficient is determined by the following formula:

[0122]

[0123] Among them, s i Represents the silhouette coefficient corresponding to the i-th sample data; a i represents the average Euclidean distance between the i-th sample data and all other sample data in the same cluster; b i It represents the average Euclidean distance between the ith sample data and all sample data in the nearest cluster; max(·) represents the maximum value function.

[0124] Further, based on the above-mentioned embodiment of the invention, the wind and solar power output typical scene determination device further includes:

[0125] Data elimination module, used to call density-based clustering algorithm to eliminate abnormal data and low-probability data in load data and wind and solar output data;

[0126] The data filling module is used to call the K nearest neighbor algorithm to fill in the missing data in the removed load data and wind and solar power output data.

[0127] The wind-solar output typical scene determination device provided in the embodiment of the present invention can execute the wind-solar output typical scene determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0128] Embodiment 4

[0129] Fig.13 A schematic diagram of the structure of an electronic device 40 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0130] like Fig.13As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0131] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0132] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as a typical scene determination method for wind and solar output.

[0133] In some embodiments, the method for determining a typical scene of wind and solar power output may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the method for determining a typical scene of wind and solar power output described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to execute the method for determining a typical scene of wind and solar power output in any other appropriate manner (e.g., by means of firmware).

[0134] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0136] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0138] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0139] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0140] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0141] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining typical wind and solar power output scenarios, characterized in that: The method comprises: Obtain load data and wind and solar power output data within a preset period of the wind and solar power station; Performing time segmentation on the load data to obtain target scene sets in different time periods; The wind and solar power output data in each of the target scene sets are clustered to obtain a typical target wind and solar power output scene corresponding to each of the target scene sets.

2. The method according to claim 1, characterized in that: The time segmentation of the load data to obtain target scene sets in different time periods includes: Obtaining a first cluster number candidate value set corresponding to a preset ordered clustering algorithm; wherein the first cluster number candidate value set includes at least one first cluster number candidate value; Selecting each of the first candidate values ​​of the number of clusters in turn, calling the preset ordered clustering algorithm to cluster the load data, and obtaining a first clustering result corresponding to the first candidate value of the number of clusters when the preset time segmentation objective function reaches a minimum; For each of the first cluster number candidate values, determining a silhouette coefficient corresponding to the first clustering result; The first clustering number candidate value corresponding to the maximum silhouette coefficient is determined as the first optimal clustering number, and the first clustering result corresponding to the first optimal clustering number is used as the target scene set for the different time periods.

3. The method according to claim 2, characterized in that The preset time segmentation objective function is expressed as: Among them, G represents the number of sample data; i represents the sample data sequence number; x i Represents the i-th sample data; Represents the mean of the sample data.

4. The method according to claim 1, characterized in that: The clustering of the wind and solar power output data in each target scene set to obtain a target wind and solar power output typical scene corresponding to each target scene set includes: Obtaining a second cluster number candidate value set corresponding to a preset K-Means clustering algorithm and a preset iteration termination condition; wherein the second cluster number candidate value set includes at least one second cluster number candidate value; Selecting each of the second cluster number candidate values ​​in turn, calling the preset K-Means clustering algorithm to cluster the wind and solar power output data in each of the target scene sets, and obtaining a second clustering result corresponding to the second cluster number candidate value when the preset iteration termination condition is reached; For each of the second clustering number candidate values ​​of the target scene set, determining a silhouette coefficient corresponding to the second clustering result; The second cluster number candidate value corresponding to the maximum silhouette coefficient is determined as the second optimal cluster number, and the cluster center of the second clustering result corresponding to the second optimal cluster number is used as the target wind and solar output typical scene of the corresponding target scene set, and the occurrence probability corresponding to the target wind and solar output typical scene is determined at the same time.

5. The method according to claim 4, characterized in that The step of sequentially selecting each of the second cluster number candidate values, calling the preset K-Means clustering algorithm to cluster the wind and solar power output data in each of the target scene sets, and obtaining a second clustering result corresponding to the second cluster number candidate value when the preset iteration termination condition is reached includes: Initializing a corresponding number of cluster centers in each of the target scene sets based on the selected second cluster number candidate values; Determine the Euclidean distance between each wind / solar power output data in the target scene set and each cluster center, and distribute each wind / solar power output data to the cluster with the closest distance according to the Euclidean distance; Determine an average value of all wind and solar power output data in each of the clusters, and update the cluster center of the corresponding cluster according to the average value; Determine whether the preset iteration termination condition is currently reached. If so, generate the second clustering result under the second clustering number candidate value corresponding to the target scene set; if not, return to execute the step of determining the Euclidean distance between each wind and solar power output data in the target scene set and each cluster center, and distribute each wind and solar power output data to the nearest cluster cluster according to the Euclidean distance.

6. The method according to claim 2 or 4, characterized in that: The silhouette coefficient is determined by the following formula: Among them, s i Represents the silhouette coefficient corresponding to the i-th sample data; a i represents the average Euclidean distance between the i-th sample data and all other sample data in the same cluster; b i It represents the average Euclidean distance between the ith sample data and all sample data in the nearest cluster; max(·) represents the maximum value function.

7. The method according to claim 1, characterized in that Also includes: Calling a density-based clustering algorithm to remove abnormal data and low-probability data from the load data and the wind and solar power output data; The K nearest neighbor algorithm is called to fill in the missing data in the removed load data and wind and solar power output data.

8. A device for determining typical wind and solar power output scenes, characterized in that: The device comprises: A data acquisition module is used to obtain load data and wind and solar power output data within a preset period of the wind and solar power station; A scene set division module, used to divide the load data into time segments to obtain target scene sets in different time periods; The typical scene determination module is used to cluster the wind and solar power output data in each target scene set to obtain the target wind and solar power output typical scene corresponding to each target scene set.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining a typical scene of wind and solar power output according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining a typical wind and solar power output scenario according to any one of claims 1-7 when executed.

Citation Information

Cited By

  • Charging strategy determination method and device, electronic equipment and nonvolatile storage medium

    CN119231604A

  • Charging strategy determination method, device, electronic device and non-volatile storage medium

    CN119231604B