Distributed photovoltaic power station grouping clustering method and system considering objective weighting

Abstract

The invention relates to a distributed photovoltaic power station grouping clustering method and system considering objective weighting, and belongs to the technical field of photovoltaic grouping, and the method comprises the following steps: constructing a resource clustering control index system through obtaining the operation data of a distributed photovoltaic power station, the resource clustering control index system comprises a technical index and an economic index. And after normalizing the indexes, calculating the proportion of each index of the distributed photovoltaic power station in the distributed photovoltaic power station group, and calculating an anti-entropy value and an entropy weight based on the proportion. And determining an index numerical value standard deviation of the photovoltaic group, and calculating a threshold value by using the standard deviation. And inputting the photovoltaic group into a grouping clustering algorithm, and outputting a grouping clustering result in combination with the threshold value, the anti-entropy value and the entropy weight. According to the method, different indexes are objectively weighted, so that the weight distribution of each photovoltaic in the clustering process is reasonable and accurate, the accuracy of the clustering result is improved, and the actual operation characteristics and the regulation and control demand difference of the photovoltaic are better reflected.

Description

Technical Field

[0001] The present invention relates to a method and system for clustering distributed photovoltaic power stations considering objective weighting, belonging to the technical field of photovoltaic clustering. Background Art

[0002] With the acceleration of the global energy transformation, distributed photovoltaic power generation, as a clean and renewable energy utilization form, has been widely applied and developed rapidly. Numerous distributed photovoltaic power station systems are spread across different regions and application scenarios, such as urban buildings, industrial parks, rural residences, etc., providing a strong supplement to energy supply and also having an important impact on the operation and management of the power grid. During the large-scale access and operation of distributed photovoltaic power stations, in order to effectively manage and control a large number of distributed photovoltaic power station resources, improve energy utilization efficiency, optimize power grid operation performance, and reduce operation costs, clustering them becomes a key issue. Clustering aims to classify distributed photovoltaic power stations with similar characteristics and operation rules, so as to formulate targeted regulation strategies and operation management plans.

[0003] When constructing a resource clustering control index system for traditional distributed photovoltaic power station clustering methods, the considered indicators are often not comprehensive enough, and may only focus on one or two aspects, such as only paying attention to single factors like the power generation of photovoltaic or the installation location, ignoring the multi-dimensional characteristics during the operation of distributed photovoltaic power stations, including technical indicators (such as downward regulation capacity, regulation time) and economic indicators (such as regulation cost), etc., resulting in the clustering results not being able to accurately reflect their comprehensive characteristics and being difficult to meet the requirements of refined management and optimized regulation.

[0004] The prior art, such as the Chinese patent application with publication number CN118313496A, discloses a method for predicting the power of a distributed photovoltaic cluster based on deep similarity clustering, including the following steps: obtaining the power data of a distributed photovoltaic power station group, preliminarily clustering the distributed photovoltaic power station group through a clustering algorithm to obtain multiple cluster centers as typical power stations; based on the typical power stations and their power data, determining the deep power characteristics of multiple typical power stations by training multiple preset autoencoders; using the typical power stations as the cluster centers for secondary clustering based on the power characteristics of the multiple typical power stations, and obtaining the optimal clustering result according to the deep similarity between distributed photovoltaic power station groups; based on the optimal clustering result, predicting the regional distributed photovoltaic power through a pre-constructed graph convolutional neural network prediction model. However, the above patent highly depends on the historical power data of distributed photovoltaic power stations to train the autoencoder and construct the graph convolutional neural network prediction model. If there are situations such as missing, noisy, or large error in the historical data, it will affect the training effect and prediction accuracy of the model. Summary of the Invention

[0005] To solve the problems existing in the above-mentioned prior art, the present invention proposes a method and system for clustering distributed photovoltaic power stations considering objective weighting.

[0006] The technical solution of the present invention is as follows: On the one hand, the present invention provides a method for clustering distributed photovoltaic power stations considering objective weighting, including the following steps: Obtain the operation data of the distributed photovoltaic power station, and construct a resource clustering control index system based on the operation data; Normalize each index in the resource clustering control index system; Calculate the proportion of each normalized index of the distributed photovoltaic power station in the sum of the corresponding indexes of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion; Determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values; Take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the clustering result of the distributed photovoltaic power station based on the threshold, the anti-entropy value and the entropy weight.

[0007] As a preferred embodiment, the specific steps for constructing a resource clustering control index system based on the operation data are as follows: Determine technical indexes, including the downward regulation capacity of the distributed photovoltaic power station and the regulation time of the distributed photovoltaic power station; Calculate the downward regulation capacity of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ; ; ; In the formula, represents the total number of distributed photovoltaic power stations in the distributed photovoltaic power station group, represents the th operating power of the th distributed photovoltaic power station at time , represents the time index, represents the index of the number of distributed photovoltaic power stations, represents the th regulation time of the Calculate the regulation cost of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ; In the formula, represents the curtailment cost of the th distributed photovoltaic power station, represents the curtailment amount of the th distributed photovoltaic power station, represents the regulation cost of the th distributed photovoltaic power station; Construct a resource clustering control index system based on the technical indicators and economic indicators.

[0008] As a preferred implementation, normalize each index in the resource clustering control index system. The specific steps are as follows: Perform positive normalization on the downward adjustment capacity, which is expressed by the formula: ; In the formula, represents the downward adjustment capacity of the th distributed photovoltaic power station after positive normalization, represents the maximum value of the downward adjustment capacity of the distributed photovoltaic power station group, represents the minimum value of the downward adjustment capacity of the distributed photovoltaic power station group; Perform negative normalization on the regulation time and regulation cost, which is expressed by the formula: ; ; In the formula, represents the regulation cost or regulation time of the th distributed photovoltaic power station after negative normalization, or the regulation cost or regulation time of the th distributed photovoltaic power station, represents the regulation cost or regulation time of the th distributed photovoltaic power station, or the regulation cost or regulation time of the th distributed photovoltaic power station, represents the minimum value of the regulation cost or regulation time of the distributed photovoltaic power station group, represents the maximum value of the regulation cost or regulation time of the distributed photovoltaic power station group.

[0009] As a preferred implementation, calculate the proportion of each index of the distributed photovoltaic power station after normalization in the index sum of the corresponding index of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion. The specific steps are as follows: Calculate the proportion of each normalized index of a distributed photovoltaic power station in the sum of corresponding indices of a distributed photovoltaic power station group, which is expressed by the formula: ; ; ; ; In the formula, represents the proportion of the th normalized index of the th distributed photovoltaic power station in the sum of corresponding indices of the distributed photovoltaic power station group, represents the th normalized index of the th distributed photovoltaic power station, represents the index index, represents the sum of the th index of the distributed photovoltaic power station group after normalization; Calculate the anti-entropy value of the corresponding index of the distributed photovoltaic power station group based on the said proportion, which is expressed by the formula: ; In the formula, represents the anti-entropy value of the th index of the distributed photovoltaic power station group, represents the logarithmic function; Calculate the entropy weight of the corresponding index of the distributed photovoltaic power station group based on the said anti-entropy value, which is expressed by the formula: ; In the formula, represents the entropy weight of the th index of the distributed photovoltaic power station group.

[0010] As a preferred embodiment, determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values. The specific steps are as follows: Determine the standard deviation of the index values of the distributed photovoltaic power station group, which is expressed by the formula: ; ; ; In the formula, represents the standard deviation of the index values of the distributed photovoltaic power station group, represents the standard deviation of the index values of the th index of the distributed photovoltaic power station group, The mean value of the th index representing the distributed photovoltaic power station group; Calculate the bandwidth of the distributed photovoltaic power station group based on the standard deviation of the index values, which is expressed by the formula: ; In the formula, represents the index dimension, represents the bandwidth of the distributed photovoltaic power station group; Set the threshold of the distributed photovoltaic power station group to be times of the bandwidth .

[0011] As a preferred implementation manner, take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the clustering result of the distributed photovoltaic power stations based on the threshold, the anti-entropy value, and the entropy weight. The specific steps are as follows: S1. Mark each distributed photovoltaic power station in the distributed photovoltaic power station group as unclassified; S2. Create an empty cluster set , and randomly select a distributed photovoltaic power station as the initial center point ; S3. Search for the distributed photovoltaic power stations within the range with the initial center point as the center and the bandwidth of the distributed photovoltaic power station group as the radius to construct a set ; S4. Calculate the vector from the distributed photovoltaic power stations in the set to the initial center point , which is expressed by the formula: ; In the formula, represents the exponential function, represents the vector from the th distributed photovoltaic power station in the set to the initial center point , represents the coordinate point of the th distributed photovoltaic power station in the set , represents the norm operator; S5. Calculate the weighted vector sum of the distributed photovoltaic power station group in the set based on the vector and the entropy weight, which is expressed by the formula: ; In the formula, represents the weighted vector sum of the distributed photovoltaic power station group in the set , Represents a set The th downward adjustment capacity of the distributed photovoltaic power station, Represents a set The th regulation time of the distributed photovoltaic power station, Represents a set The th regulation cost of the distributed photovoltaic power station, Represents the entropy weight of the downward adjustment capacity of the distributed photovoltaic power station group, Represents the entropy weight of the regulation time of the distributed photovoltaic power station group, Represents the entropy weight of the regulation cost of the distributed photovoltaic power station group, Represents the downward adjustment capacity of the initial center point , Represents the regulation time of the initial center point , Represents the regulation cost of the initial center point ; S6. Calculate the offset vector of the initial center point based on the weighted vector, which is expressed by the formula: ; In the formula, represents the offset vector of the initial center point ; S7. Update the initial center point based on the offset vector of the initial center point , which is expressed by the formula: ; In the formula, represents the updated center point; If the modulus of the offset vector of the initial center point is less than the threshold , then execute step S8, otherwise execute step S3 with the updated center point; S8. If there is an ideal cluster in the empty cluster set that satisfies , then incorporate the set into the ideal cluster , otherwise add the set as a new cluster to the empty cluster set ; Among them, represents the calculation of the Euclidean distance; S9. The step is completed, and the clustering result of the distributed photovoltaic power station grouping is output.

[0012] On the other hand, the present invention also provides a distributed photovoltaic power station clustering system considering objective weighting, including: System construction module: obtaining the operation data of the distributed photovoltaic power station, and constructing a resource clustering control index system based on the operation data; Preprocessing module: normalizing each index in the resource clustering control index system; Objective weighting module: calculating the proportion of each normalized index of the distributed photovoltaic power station in the sum of corresponding indexes of the distributed photovoltaic power station group, and calculating the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion; Threshold setting module: determining the standard deviation of the index values of the distributed photovoltaic power station group, and calculating the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values; Clustering module: taking the distributed photovoltaic power station group as the input of the clustering algorithm, and outputting the clustering result of the distributed photovoltaic power station based on the threshold, the anti-entropy value and the entropy weight.

[0013] As a preferred embodiment, the specific steps of constructing a resource clustering control index system based on the operation data are as follows: Determine technical indexes, including the downward regulation capacity of the distributed photovoltaic power station and the regulation time of the distributed photovoltaic power station; Calculate the downward regulation capacity of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ; ; ; In the formula, represents the total number of distributed photovoltaic power stations in the distributed photovoltaic power station group, represents the operating power of the th distributed photovoltaic power station at the th moment, represents the downward regulation capacity of the th distributed photovoltaic power station, represents the minimum operating boundary of the th distributed photovoltaic power station, represents the time index, represents the index of the number of distributed photovoltaic power stations, represents the th distributed photovoltaic power station's regulation time; Determine economic indexes, including the regulation cost of the distributed photovoltaic power station; Calculate the regulation cost of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ; In the formula, represents the curtailment cost of the th distributed PV power station, represents the curtailment amount of the th distributed PV power station, represents the regulation cost of the th distributed PV power station; Construct a resource clustering control index system based on the technical indicators and economic indicators.

[0014] As a preferred implementation, normalize each index in the resource clustering control index system. The specific steps are as follows: Perform positive normalization on the downward adjustment capacity, which is expressed by the formula: ; In the formula, represents the downward adjustment capacity of the th distributed PV power station after positive normalization, represents the maximum value of the downward adjustment capacity of the distributed PV power station group, represents the minimum value of the downward adjustment capacity of the distributed PV power station group; Perform negative normalization on the regulation time and regulation cost, which is expressed by the formula: ; ; In the formula, represents the regulation cost or regulation time of the th distributed PV power station after negative normalization, or the regulation cost or regulation time of the th distributed PV power station, represents the th distributed PV power station's regulation cost or the th distributed PV power station's regulation time, represents the minimum value of the regulation cost or regulation time of the distributed PV power station group, represents the maximum value of the regulation cost or regulation time of the distributed PV power station group.

[0015] As a preferred implementation, calculate the proportion of each index of the distributed PV power station after normalization in the index sum of the corresponding index of the distributed PV power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed PV power station group based on the proportion. The specific steps are as follows: Calculate the proportion of each index of the distributed PV power station after normalization in the index sum of the corresponding index of the distributed PV power station group, which is expressed by the formula: ; ; ; ; In the formula, represents the proportion of the th index after normalization of the th distributed photovoltaic power station in the sum of the corresponding indexes of the distributed photovoltaic power station group, represents the th index after normalization of the th distributed photovoltaic power station, represents the index index, represents the sum of the th index after normalization of the distributed photovoltaic power station group; Calculate the anti-entropy value of the corresponding index of the distributed photovoltaic power station group based on the said proportion, which is expressed by the formula: ; In the formula, represents the anti-entropy value of the th index of the distributed photovoltaic power station group, represents the logarithmic function; Calculate the entropy weight of the corresponding index of the distributed photovoltaic power station group based on the said anti-entropy value, which is expressed by the formula: ; In the formula, represents the entropy weight of the th index of the distributed photovoltaic power station group.

[0016] The present invention has the following beneficial effects: 1. By calculating the anti-entropy value and entropy weight of multiple key indexes (technical indexes such as regulation capacity and regulation time, economic indexes such as regulation cost) of the distributed photovoltaic power station respectively, the present invention realizes the objective weighting of different indexes. The present invention fully considers the importance degree of each index in measuring the characteristics and adaptability of the distributed photovoltaic power station, makes the weight distribution of each distributed photovoltaic power station in the clustering process more reasonable and accurate, thereby improving the accuracy of the clustering result of the grouping, and can more truly reflect the actual operation characteristics and regulation demand differences of the distributed photovoltaic power station.

[0017] 2. By normalizing each index in the resource clustering control index system, the present invention eliminates the influence brought by the differences in dimension and order of magnitude between different indexes, enables each index to be compared and analyzed on a unified scale, further improves the accuracy of the clustering of the grouping, and avoids the deviation of the clustering result caused by the too large difference in index values.

[0018] 3. Based on the actual operation data of distributed photovoltaic power stations, the present invention constructs a resource clustering control index system, which can truly reflect the performance and characteristics of distributed photovoltaic power stations during actual operation, making the basis for clustering more objective and scientific. The obtained clustering results are more in line with the actual operation of distributed photovoltaic power stations, providing a more targeted and effective basis for subsequent management and control of distributed photovoltaic power station groups.

[0019] 4. The present invention calculates the entropy weights of various indicators of distributed photovoltaic power station groups by using anti-entropy values. The entropy weights reflect the amount of information and importance of each indicator in distributed photovoltaic power station groups. The weights determined in this way can objectively measure the role of each indicator in the clustering process, avoiding the influence of human factors on weight allocation and making the clustering process more scientific and reasonable.

[0020] 5. The clustering algorithm adopted by the present invention can quickly divide distributed photovoltaic power station groups into different clusters through steps such as setting initial center points, calculating vectors, and updating center points by offset vectors. In the calculation process, this algorithm fully considers various factors such as the weights and thresholds of indicators of distributed photovoltaic power stations, can obtain relatively ideal clustering results within fewer iterations, improves the efficiency of clustering, saves calculation time and resources, and is suitable for the clustering needs of large-scale distributed photovoltaic power station groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flowchart of the method implementation of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] It should be understood that the step numbers used in the text are only for convenient description and do not limit the execution order of the steps.

[0024] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0025] The terms "comprising" and "including" 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 combinations.

[0026] The term "and / or" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0027] Example 1: See Figure 1 , this embodiment provides a distributed photovoltaic power station clustering method considering objective weighting, including the following steps: Obtain the operation data of the distributed photovoltaic power station, and construct a resource clustering control index system based on the operation data; wherein, the operation data includes operation power, curtailment cost and curtailment amount; Normalize each index in the resource clustering control index system; Calculate the proportion of each normalized index of the distributed photovoltaic power station in the sum of the corresponding indexes of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion; Determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values; Take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the distributed photovoltaic power station clustering result based on the threshold, the anti-entropy value and the entropy weight.

[0028] Selecting the capacity reduction, regulation time and regulation cost as the resource clustering control index system is because they can comprehensively and efficiently measure the efficiency, economy and feasibility of photovoltaic participating in power grid regulation.

[0029] As a preferred implementation manner, the specific steps for constructing a resource clustering control index system based on the operation data are as follows: Determine technical indexes, including the capacity reduction of the distributed photovoltaic power station and the regulation time of the distributed photovoltaic power station; Calculate the capacity reduction of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ; ; ; In the formula, represents the total number of distributed photovoltaic power stations in the distributed photovoltaic power station group, represents the th distributed photovoltaic power station at The operating power at a moment, represents the downward regulation capacity of the th distributed photovoltaic power station, represents the minimum operating boundary of the th distributed photovoltaic power station, represents the time index, represents the index of the number of distributed photovoltaic power stations, represents the th regulation time of the distributed photovoltaic power station; Determine the economic indicators, including the regulation cost of the distributed photovoltaic power station; Calculate the regulation cost of the distributed photovoltaic power station based on the operating data, which is expressed by the formula: ; In the formula, represents the curtailment cost of the th distributed photovoltaic power station, represents the curtailment amount of the th distributed photovoltaic power station, represents the regulation cost of the th distributed photovoltaic power station; Construct a resource clustering control index system based on the technical indicators and economic indicators.

[0030] The larger the adjustable capacity and the smaller the regulation time, the stronger the ability of the photovoltaic to participate in grid regulation, and the lower the regulation cost, the greater the benefit of the photovoltaic to participate in grid regulation.

[0031] As a preferred implementation method, normalize each index in the resource clustering control index system. The specific steps are as follows: Perform positive normalization on the downward regulation capacity, which is expressed by the formula: ; In the formula, represents the downward regulation capacity of the th distributed photovoltaic power station after positive normalization, represents the maximum value of the downward regulation capacity of the distributed photovoltaic power station group, represents the minimum value of the downward regulation capacity of the distributed photovoltaic power station group; Perform negative normalization on the regulation time and regulation cost, which is expressed by the formula: ; ; In the formula, represents the regulation cost of the th distributed photovoltaic power station after negative normalization or the The regulation time of a distributed photovoltaic power station represents the th regulation cost of a distributed photovoltaic power station or the th regulation time of a distributed photovoltaic power station, represents the minimum value of the regulation cost or regulation time of a distributed photovoltaic power station group, represents the maximum value of the regulation cost or regulation time of a distributed photovoltaic power station group.

[0032] As a preferred implementation mode, calculate the proportion of each index after normalization of a distributed photovoltaic power station in the sum of corresponding indexes of a distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion. The specific steps are as follows: Calculate the proportion of each index after normalization of a distributed photovoltaic power station in the sum of corresponding indexes of a distributed photovoltaic power station group, which is expressed by the formula: ; ; ; ; In the formula, represents the th proportion of the th index after normalization of the th distributed photovoltaic power station in the sum of corresponding indexes of the distributed photovoltaic power station group, represents the th index after normalization of the th distributed photovoltaic power station, When it is 1, it represents the downward capacity index, represents the th downward capacity after normalization of the th distributed photovoltaic power station, When it is 2, it represents the regulation time index, represents the th regulation time after normalization of the th distributed photovoltaic power station, When it is 3, it represents the regulation cost index, represents the th sum of indexes of the th index after normalization of the distributed photovoltaic power station group; ; In the formula, The anti-entropy value of the th index of the distributed PV power station group, represents the logarithmic function; Based on the anti-entropy value, calculate the entropy weight of the corresponding index of the distributed PV power station group, which is expressed by the formula: ; In the formula, The th index entropy weight of the distributed PV power station group.

[0033] As a preferred implementation, determine the standard deviation of the index values of the distributed PV power station group, and calculate the threshold of the distributed PV power station group based on the standard deviation of the index values. The specific steps are as follows: Determine the standard deviation of the index values of the distributed PV power station group, which is expressed by the formula: ; ; ; In the formula, represents the standard deviation of the index values of the distributed PV power station group, The th index value standard deviation of the distributed PV power station group, The th index mean value of the distributed PV power station group; Based on the standard deviation of the index values, calculate the bandwidth of the distributed PV power station group, which is expressed by the formula: ; In the formula, represents the index dimension, represents the bandwidth of the distributed PV power station group; Set the threshold of the distributed PV power station group to times of the bandwidth .

[0034] As a preferred implementation, take the distributed PV power station group as the input of the clustering algorithm, and output the clustering result of the distributed PV power stations based on the threshold, the anti-entropy value and the entropy weight. The specific steps are as follows: S1. Mark each distributed PV power station in the distributed PV power station group as unclassified; S2. Create an empty cluster set , and randomly select a distributed PV power station as the initial center point ; S3. With the initial center point as the center, the bandwidth of the distributed PV power station group Construct a set of distributed photovoltaic power stations within the range searched by the radius ; S4. Calculate the vector from the distributed photovoltaic power stations in the set to the initial center point . It is expressed by the formula: ; In the formula, represents the exponential function, represents the set The vector from the th distributed photovoltaic power station in to the initial center point represents the set The th coordinate point of the distributed photovoltaic power station in represents the norm operator; S5. Based on the vector and entropy weight, calculate the weighted vector sum of the distributed photovoltaic power station group in the set . It is expressed by the formula: ; In the formula, represents the weighted vector sum of the distributed photovoltaic power station group in the set , represents the downward regulation capacity of the th distributed photovoltaic power station in the set , represents the regulation time of the th distributed photovoltaic power station in the set , represents the regulation cost of the th distributed photovoltaic power station in the set , represents the entropy weight of the downward regulation capacity of the distributed photovoltaic power station group, represents the entropy weight of the regulation time of the distributed photovoltaic power station group, represents the entropy weight of the regulation cost of the distributed photovoltaic power station group, represents the downward regulation capacity of the initial center point , represents the regulation time of the initial center point , represents the regulation cost of the initial center point ; S6. Based on the weighted vector, calculate the offset vector of the initial center point . It is expressed by the formula: ; In the formula, represents the initial center point Offset vector; S7. Based on the initial center point Update the initial center point with the offset vector , which is expressed by the formula: ; In the formula, represents the updated center point; If the modulus value of the offset vector of the initial center point is less than the threshold , then execute step S8, otherwise execute step S3 with the updated center point; S8. If there is an ideal cluster in the empty cluster set that satisfies , then incorporate the set into the ideal cluster , otherwise use the set as a new cluster to join the empty cluster set ; Among them, represents calculating the Euclidean distance, which is expressed by the formula: ; In the formula, represents the -th index of the ideal cluster , represents the -th index of the updated center point; S9. Record the number of times all distributed photovoltaic power stations are accessed by all updated center points , and allocate them to the cluster where the center point with the most access times belongs. If the number of times is the same, then select the cluster where the center point with the closest Euclidean distance belongs; S10. The step is completed, and the clustering result of the distributed photovoltaic power stations is output.

[0035] Embodiment 2: This embodiment provides a distributed photovoltaic power station clustering system considering objective weighting, including: System construction module: Obtain the operation data of the distributed photovoltaic power stations, and construct a resource clustering control index system based on the operation data; Preprocessing module: Normalize each index in the resource clustering control index system; Objective weighting module: Calculate the proportion of each normalized index of the distributed photovoltaic power stations in the sum of the corresponding indexes of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion; Threshold setting module: Determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values; Clustering module: Take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the clustering result of the distributed photovoltaic power stations based on the threshold, the anti-entropy value and the entropy weight.

[0036] As a preferred implementation, construct a resource clustering control index system based on the operation data. The specific steps are as follows: Determine technical indicators, including the downward regulation capacity of the distributed photovoltaic power station and the regulation time of the distributed photovoltaic power station; Calculate the downward regulation capacity of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ; ; ; In the formula, represents the total number of distributed photovoltaic power stations in the distributed photovoltaic power station group, represents the th operating power of the th distributed photovoltaic power station at the th time, represents the time index, represents the distributed photovoltaic power station number index, represents the th regulation time of the th ; In the formula, represents the light curtailment cost of the th distributed photovoltaic power station, represents the light curtailment amount of the th distributed photovoltaic power station, represents the regulation cost of the

[0037] As a preferred embodiment, normalize each index in the resource clustering control index system. The specific steps are as follows: Perform positive normalization on the downward regulation capacity, which is expressed by the formula: ; In the formula, represents the downward regulation capacity of the th distributed photovoltaic power station after positive normalization, represents the maximum value of the downward regulation capacity of the distributed photovoltaic power station group, represents the minimum value of the downward regulation capacity of the distributed photovoltaic power station group; Perform negative normalization on the regulation time and regulation cost, which is expressed by the formula: ; ; In the formula, represents the regulation cost of the th distributed photovoltaic power station or the regulation time of the th distributed photovoltaic power station after negative normalization, represents the regulation cost of the th distributed photovoltaic power station or the regulation time of the th distributed photovoltaic power station, represents the minimum value of the regulation cost or regulation time of the distributed photovoltaic power station group, represents the maximum value of the regulation cost or regulation time of the distributed photovoltaic power station group.

[0038] As a preferred embodiment, calculate the proportion of each index of the distributed photovoltaic power station after normalization in the index sum of the corresponding index of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion. The specific steps are as follows: Calculate the proportion of each index of the distributed photovoltaic power station after normalization in the index sum of the corresponding index of the distributed photovoltaic power station group, which is expressed by the formula: ; ; ; ; In the formula, represents the proportion of the th index of the th distributed photovoltaic power station after normalization in the index sum of the corresponding index of the distributed photovoltaic power station group, represents the th distributed photovoltaic power station after normalization of the An index indicating the index of the indicator indicating the sum of the th indicator after the normalization processing of the distributed photovoltaic power station group; Based on the said ratio, calculate the anti-entropy value of the corresponding indicator of the distributed photovoltaic power station group, which is expressed by the formula as: ; In the formula, indicating the anti-entropy value of the th indicator of the distributed photovoltaic power station group, indicating the logarithmic function; Based on the said anti-entropy value, calculate the entropy weight of the corresponding indicator of the distributed photovoltaic power station group, which is expressed by the formula as: ; In the formula, indicating the entropy weight of the th indicator of the distributed photovoltaic power station group.

[0039] In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent the situation where A exists alone, A and B exist simultaneously, or B exists alone. Wherein A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one of the following" and its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0040] Those of ordinary skill in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be implemented by a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0041] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0042] In several embodiments provided by the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (hereinafter referred to as ROM), random access memories (hereinafter referred to as RAM), magnetic disks, or optical discs.

[0043] The above are only the embodiments of the present invention, and thus do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A distributed photovoltaic power station clustering method considering objective weighting, characterized in that It includes the following steps: Obtain the operation data of the distributed photovoltaic power station, and construct a resource clustering control index system based on the operation data; Normalize each index in the resource clustering control index system; Calculate the proportion of each index after normalization of the distributed photovoltaic power station in the sum of the corresponding indexes of the distributed photovoltaic power station group. Calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion; Determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values; Take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the clustering result of the distributed photovoltaic power station based on the threshold, the anti-entropy value and the entropy weight.

2. The method for clustering and grouping distributed photovoltaic power stations considering objective weighting according to claim 1, wherein Construct a resource clustering control index system based on the operation data. The specific steps are as follows: Determine technical indexes, including the downward regulation capacity of the distributed photovoltaic power station and the regulation time of the distributed photovoltaic power station; Calculate the downward regulation capacity of the distributed photovoltaic power station based on the operation data. It is expressed by the formula: ； ； ； In the formula, represents the total number of distributed photovoltaic power stations in the distributed photovoltaic power station group, represents the th distributed photovoltaic power station's operating power at moment, represents the th distributed photovoltaic power station's downward regulation capacity, represents the th distributed photovoltaic power station's minimum operating boundary, represents the time index, represents the distributed photovoltaic power station number index, represents the th distributed photovoltaic power station's regulation time; Determine economic indexes, including the regulation cost of the distributed photovoltaic power station; Calculate the regulation cost of the distributed photovoltaic power station based on the operation data. It is expressed by the formula: ； In the formula, represents the curtailment cost of the th distributed photovoltaic power station, represents the curtailment amount of the th distributed photovoltaic power station, represents the regulation cost of the th distributed photovoltaic power station; Construct a resource clustering control index system based on the technical indexes and economic indexes.

3. The distributed photovoltaic power station clustering method considering objective weighting according to claim 2, characterized in that, Normalize each index in the resource clustering control index system. The specific steps are as follows: Perform positive normalization on the downward regulation capacity. It is expressed by the formula: ； In the formula, represents the downward regulation capacity of the th distributed photovoltaic power station after positive normalization, represents the maximum value of the downward regulation capacity of the distributed photovoltaic power station group, represents the minimum value of the downward regulation capacity of the distributed photovoltaic power station group; Perform negative normalization on the regulation time and regulation cost. It is expressed by the formula: ； ； In the formula, represents the regulation cost of the th distributed photovoltaic power station after negative normalization or the regulation time of the th distributed photovoltaic power station, represents the regulation cost of the th distributed photovoltaic power station or the regulation time of the th distributed photovoltaic power station, represents the minimum value of the regulation cost or regulation time of the distributed photovoltaic power station group, represents the maximum value of the regulation cost or regulation time of the distributed photovoltaic power station group.

4. The distributed photovoltaic power station clustering method considering objective weighting according to claim 3, characterized in that Calculate the proportion of each index after normalization of the distributed photovoltaic power station in the sum of the corresponding indexes of the distributed photovoltaic power station group. Calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion. The specific steps are as follows: Calculate the proportion of each index after normalization of the distributed photovoltaic power station in the sum of the corresponding indexes of the distributed photovoltaic power station group. It is expressed by the formula: ； ； ； In the formula, represents the proportion of the -th index after normalization of the -th distributed photovoltaic power station in the sum of the corresponding indices of the distributed photovoltaic power station group, represents the -th index after normalization of the -th distributed photovoltaic power station, represents the index index, represents the sum of the -th index after normalization of the distributed photovoltaic power station group; Calculate the anti-entropy value of the corresponding index of the distributed photovoltaic power station group based on the proportion. It is expressed by the formula: ； In the formula, represents the anti-entropy value of the th index of the distributed photovoltaic power station group, represents the logarithmic function; Calculate the entropy weight of the corresponding index of the distributed photovoltaic power station group based on the anti-entropy value. It is expressed by the formula: ； In the formula, represents the entropy weight of the th index of the distributed photovoltaic power station group.

5. The method for clustering and grouping distributed photovoltaic power stations considering objective weighting according to claim 4, characterized in that, Determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values. The specific steps are as follows: Determine the standard deviation of the index values of the distributed photovoltaic power station group. It is expressed by the formula: ； ； ； In the formula, represents the standard deviation of the index values of the distributed photovoltaic power station group, represents the standard deviation of the index values of the th index of the distributed photovoltaic power station group, and represents the average value of the Calculate the bandwidth of the distributed photovoltaic power station group based on the standard deviation of the index values. It is expressed by the formula: ； In the formula, represents the index dimension, represents the bandwidth of the distributed photovoltaic power station group; Set the threshold of the distributed photovoltaic power station group to the bandwidth of .

6. The method for clustering and grouping distributed photovoltaic power stations considering objective weighting according to claim 5, wherein, Take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the clustering result of the distributed photovoltaic power station based on the threshold, the anti-entropy value and the entropy weight. The specific steps are as follows: S1. Mark each distributed photovoltaic power station in the distributed photovoltaic power station group as unclassified; S2. Create an empty cluster set , randomly select a distributed photovoltaic power station as the initial center point ; S3. With the initial center point as the center, and the bandwidth of the distributed photovoltaic power station group as the radius, search for distributed photovoltaic power stations within the range to construct a set ; S4. Calculate the set of vectors from the distributed photovoltaic power stations to the initial center point which is expressed by the formula as follows: ； In the formula, represents the exponential function, represents the set the th vector from the distributed photovoltaic power station to the initial center point ; represents the set the th coordinate point of the distributed photovoltaic power station; represents the norm operator; S5. Calculate the weighted vector sum of the distributed PV power plant group in the set based on the vector and entropy weight, which is expressed by the formula as follows: ​ ； In the formula, represents the weighted vector sum of the distributed photovoltaic power station groups in the set represents the set in the th downward adjustment capacity of the distributed photovoltaic power station represents the set in the th regulation time of the distributed photovoltaic power station represents the set in the th regulation cost of the distributed photovoltaic power station represents the entropy weight of the downward adjustment capacity of the distributed photovoltaic power station group represents the entropy weight of the regulation time of the distributed photovoltaic power station group represents the entropy weight of the regulation cost of the distributed photovoltaic power station group represents the downward adjustment capacity of the initial center point represents the regulation time of the initial center point represents the regulation cost of the initial center point ;​​ S6. Calculate the offset vector of the initial center point based on the weighted vector, which is expressed by the formula: as follows: ； In the formula, represents the offset vector of the initial center point ; S7. Update the initial center point based on the offset vector of the initial center point, which is expressed by the formula: as follows: ； In the formula, represents the updated center point; If the initial center point The offset vector The modulus value is less than the threshold , then execute step S8, otherwise execute step S3 with the updated center point; S8. If there is an ideal cluster in the empty cluster set that satisfies , then incorporate the set into the ideal cluster ; otherwise, use the set as a new cluster and add it to the empty cluster set ;​ Among them, represents calculating the Euclidean distance; S9. The step is completed, and the clustering result of the distributed photovoltaic power station is output.

7. A distributed photovoltaic power station clustering system considering objective weighting, characterized in that, It includes: System construction module: Obtain the operation data of the distributed photovoltaic power station, and construct a resource clustering control index system based on the operation data; Preprocessing module: Normalize each index in the resource clustering control index system; Objective weighting module: Calculate the proportion of each normalized index of a distributed photovoltaic power station in the sum of corresponding indexes of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion. Threshold setting module: Determine the standard deviation of the index values of the distributed photovoltaic power station group, and calculate the threshold of the distributed photovoltaic power station group based on the standard deviation of the index values. Clustering module: Take the distributed photovoltaic power station group as the input of the clustering algorithm, and output the clustering result of the distributed photovoltaic power stations based on the threshold, the anti-entropy value, and the entropy weight.

8. The distributed photovoltaic power station clustering system considering objective weighting according to claim 7, characterized in that Construct a resource clustering control index system based on the operation data. The specific steps are as follows: Determine technical indexes, including the downward regulation capacity of the distributed photovoltaic power station and the regulation time of the distributed photovoltaic power station. Calculate the downward regulation capacity of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ； ； ； Wherein, represents the total number of distributed photovoltaic power stations in the distributed photovoltaic power station group, represents the th distributed photovoltaic power station at the operating power at the moment, represents the th distributed photovoltaic power station's downward regulation capacity, represents the th distributed photovoltaic power station's minimum operating boundary, represents the time index, represents the distributed photovoltaic power station number index, represents the th distributed photovoltaic power station's regulation time; Determine economic indexes, including the regulation cost of the distributed photovoltaic power station. Calculate the regulation cost of the distributed photovoltaic power station based on the operation data, which is expressed by the formula: ； In the formula, represents the curtailment cost of the th distributed photovoltaic power station, represents the curtailment amount of the th distributed photovoltaic power station, represents the regulation cost of the th distributed photovoltaic power station; Construct a resource clustering control index system based on the technical indexes and economic indexes.

9. The distributed photovoltaic power station clustering system considering objective weighting according to claim 8, wherein Normalize each index in the resource clustering control index system. The specific steps are as follows: Perform positive normalization on the downward regulation capacity, which is expressed by the formula: ； In the formula, represents the downward regulation capacity of the th distributed photovoltaic power station after positive normalization, represents the maximum value of the downward regulation capacity of the distributed photovoltaic power station group, represents the minimum value of the downward regulation capacity of the distributed photovoltaic power station group; Perform negative normalization on the regulation time and regulation cost, which is expressed by the formula: ； ； In the formula, represents the regulation cost of the th distributed photovoltaic power station after negative normalization processing or the regulation time of the th distributed photovoltaic power station, represents the regulation cost of the th distributed photovoltaic power station or the regulation time of the th distributed photovoltaic power station, represents the minimum value of the regulation cost or regulation time of the distributed photovoltaic power station group, represents the maximum value of the regulation cost or regulation time of the distributed photovoltaic power station group.

10. The distributed photovoltaic power station clustering system considering objective weighting according to claim 9, characterized in that Calculate the proportion of each normalized index of the distributed photovoltaic power station in the sum of corresponding indexes of the distributed photovoltaic power station group, and calculate the anti-entropy value and entropy weight of each index of the distributed photovoltaic power station group based on the proportion. The specific steps are as follows: Calculate the proportion of each normalized index of the distributed photovoltaic power station in the sum of corresponding indexes of the distributed photovoltaic power station group, which is expressed by the formula: ； ； ； In the formula, represents the proportion of the -th index after normalization of the -th distributed photovoltaic power station in the sum of the corresponding indices of the distributed photovoltaic power station group. represents the -th index after normalization of the -th distributed photovoltaic power station. represents the index index. represents the sum of the -th index after normalization of the distributed photovoltaic power station group. Calculate the anti-entropy value of the corresponding index of the distributed photovoltaic power station group based on the proportion, which is expressed by the formula: ； In the formula, represents the anti-entropy value of the th index of the distributed photovoltaic power station group, represents the logarithmic function; Calculate the entropy weight of the corresponding index of the distributed photovoltaic power station group based on the anti-entropy value, which is expressed by the formula: ； In the formula, represents the entropy weight of the th index of the distributed photovoltaic power station group.

Images