A Net Load Scenario Generation Method Based on Probabilistic Distance Clustering

Abstract

This invention discloses a method for generating net load scenarios based on probabilistic distance clustering, comprising: acquiring historical load data and historical wind power and photovoltaic output data; clustering the historical load curves, historical wind power and photovoltaic output curves respectively using the k-means clustering algorithm to obtain probabilistic load, wind power, and photovoltaic output scenarios; combining the probabilistic load, wind power, and photovoltaic scenarios, calculating the net load curve and its probability for each combined scenario; setting inverse weighting values ​​according to the net load scenario probabilities, and clustering the net load scenarios based on probabilistic weighted distance to generate probabilistic typical net load scenarios. This method fully expands the possibilities of net load scenarios, and obtains probabilistic net load scenarios while retaining the characterization of extreme scenarios, providing a data foundation for the system to perform probabilistic power balance analysis.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method for generating net load scenarios based on probabilistic distance clustering. Background Technology

[0002] The output scenarios of new energy sources such as wind power and photovoltaics are inherently uncertain, and correspondingly, the net load scenarios also possess uncertainty. Characterizing this uncertainty through probabilistic scenario generation is an effective approach. However, when generating probabilistic net load scenarios, clustering is typically performed directly based on historical net load scenarios. This method only generates probabilistic scenarios based on existing matching scenarios for load, wind power, and photovoltaics. In reality, the emergence of existing matching scenarios is largely accidental, failing to encompass all possibilities of net load scenarios and not fully considering the individual uncertainties of load, wind power, and photovoltaics. Therefore, probabilistic net load scenarios generated in this way cannot comprehensively and completely reflect the uncertainty of net load scenarios, nor can they effectively characterize some extreme net load scenarios, making them unsuitable for probabilistic power balance analysis. Summary of the Invention

[0003] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a net load scenario generation method based on probabilistic distance clustering, which can solve the problems that traditional probabilistic net load scenarios cannot fully and completely reflect the uncertainty of net load scenarios, and are difficult to characterize some extreme net load scenarios, making it inconvenient to perform probabilistic power balance analysis.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for generating a net load scenario based on probabilistic distance clustering, comprising:

[0007] Acquire historical load data, historical wind power and solar power output data;

[0008] Based on the k-means clustering algorithm, the historical load curves, historical wind power and photovoltaic output curves are clustered to obtain probabilistic load, wind power and photovoltaic output scenarios.

[0009] The probabilistic load, wind power, and photovoltaic scenarios are combined to calculate the net load curve and scenario probability for each combined scenario.

[0010] Based on the probability of the net load scenario, an inverse weighting value is set, and the net load scenarios are clustered based on the probability weighted distance to generate probabilistic typical net load scenarios.

[0011] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the historical load data obtained consists of several 24-point daily load curves, the historical wind power output data consists of several 24-point daily wind power output curves, and the historical photovoltaic output data consists of several 24-point daily photovoltaic output curves.

[0012] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the acquisition of the probabilistic load, wind power, and photovoltaic output scenarios includes,

[0013] Using several historical daily load curves as cluster samples, and a single historical daily load curve L i ={l i,1 ,l i,2 ,…,l i,24 Perform k-means clustering on the basic clustering units, and set the number of clusters to N. L N is obtained through clustering L A probabilistic load scenario curve;

[0014] Using several historical daily wind power output curves as clustering samples, and using a single historical daily wind power output curve W j ={w j,1 ,w j,2 ,…,w j,24 Perform k-means clustering on the basic clustering units, and set the number of clusters to N. W N is obtained through clustering W Probabilistic wind power output scenario curves;

[0015] Using several historical daily photovoltaic power output curves as clustering samples, and using a single historical daily photovoltaic power output curve v k ={v k,1 ,v k,2 ,…,v k,24 Perform k-means clustering on the basic clustering units, and set the number of clusters to N. V N is obtained through clustering V A probabilistic photovoltaic power output scenario curve.

[0016] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the scenario combination includes N L A probabilistic load scenario, N W A probabilistic wind power output scenario and N VBy permuting and combining the probabilistic photovoltaic output scenarios, we obtain N. L *N W *N V Each combined scenario includes a load scenario, a wind power output scenario, and a photovoltaic power output scenario bundle.

[0017] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the net load curve of the combined scenario is calculated as follows:

[0018] ΔL ijk =L i -W j -V k

[0019] Where, ΔL ijk This represents the net load curve for a combined scenario under load scenario i, wind power scenario j, and photovoltaic scenario k.

[0020] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the scenario probability of the combined scenario is calculated as follows:

[0021]

[0022] in, Let i represent the probability of scenario i under load scenario. Let represent the scenario probability of wind power scenario j. Let k represent the probability of a photovoltaic scenario. This represents the probability of a combined scenario under load scenario i, wind power scenario j, and photovoltaic scenario k, which is the probability of a net load scenario under load scenario i, wind power scenario j, and photovoltaic scenario k.

[0023] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the setting of the inverse weighting value includes, for N L *N W *N V For each net load scenario, calculate the reciprocal of its scenario probability, as follows:

[0024]

[0025] In the formula, It represents the reciprocal of the scenario probability of the net load scenario ijk.

[0026] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, the reciprocal of the scenario probability is normalized to obtain the inverse weighted value of the net load scenario probability, calculated as follows:

[0027]

[0028]

[0029] Among them, p' max N represents L *N W *N V The maximum value of the reciprocal of the scenario probability for each net load scenario; This represents the normalized value of the reciprocal of the scenario probability of the net load scenario, which is the inverse weighted value of the net load scenario probability.

[0030] As a preferred embodiment of the net load scenario generation method based on probabilistic distance clustering described in this invention, it further includes: when performing clustering based on the k-means algorithm, calculating the distance between the cluster center and the net load scenario ijk, the calculation method is as follows:

[0031]

[0032] Among them, D ijk D' represents the raw distance between the cluster centers and the net load scenario ijk; ijk This represents the probability-weighted distance between the cluster center and the net load scenario ijk.

[0033] The beneficial effects of this invention are as follows: The method of this invention uses the k-means clustering algorithm to cluster historical load curves, historical wind power, and photovoltaic output curves respectively, obtaining probabilistic load, wind power, and photovoltaic output scenarios, fully characterizing the inherent uncertainties of these scenarios. Then, the probabilistic load, wind power, and photovoltaic scenarios are combined to obtain the net load curve and its probability for the combined scenarios. This considers the occurrence of extreme scenarios based on historical net load scenarios, expanding the possibilities of net load scenarios through scenario combination. Finally, a reverse weighting value is set according to the net load scenario probability, and probabilistic clustering of the net load scenarios is performed based on weighted clustering distance to generate typical net load scenarios. By converting scenario probabilities into weighted clustering distance, probabilistic clustering of net load scenarios is achieved, reducing scenarios while retaining their probabilities. The resulting typical net load scenarios provide a data foundation for probabilistic power balance analysis of the system. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0035] Figure 1 This is a schematic diagram of a net load scenario generation method based on probabilistic distance clustering, provided as an embodiment of the present invention.

[0036] Figure 2 This is a schematic diagram of a probabilistic wind power scenario curve in a southern province in July, provided as an embodiment of the present invention, for a net load scenario generation method based on probabilistic distance clustering.

[0037] Figure 3 A schematic diagram of a probabilistic photovoltaic scene curve in a southern province in July, provided as an embodiment of the present invention, for a net load scene generation method based on probabilistic distance clustering.

[0038] Figure 4 A schematic diagram of the probabilistic load scenario curve for a southern province in July, provided as an embodiment of the present invention, for a net load scenario generation method based on probabilistic distance clustering.

[0039] Figure 5 A schematic diagram of a probabilistic typical net load scenario curve provided by a net load scenario generation method based on probabilistic distance clustering in an embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of a typical net load scenario curve based on a traditional method, which is provided as an embodiment of the present invention for a net load scenario generation method based on probabilistic distance clustering. Detailed Implementation

[0041] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0042] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0043] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0044] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0045] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0046] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0047] Example 1

[0048] Reference Figure 1 This is the first embodiment of the present invention, which provides a method for generating a net load scenario based on probabilistic distance clustering, including:

[0049] S1: Obtain historical load data, historical wind power and photovoltaic power output data;

[0050] Furthermore, historical load data is obtained as several 24-point daily load curves, historical wind power output data is obtained as several 24-point daily wind power output curves, and historical photovoltaic output data is obtained as several 24-point daily photovoltaic output curves.

[0051] S2: Based on the k-means clustering algorithm, the historical load curve, historical wind power and photovoltaic output curves are clustered to obtain probabilistic load, wind power and photovoltaic output scenarios;

[0052] Furthermore, using several historical daily load curves as clustering samples, and taking a single historical daily load curve L... i ={l i,1 ,l i,2 ,…,l i,24 Perform k-means clustering on the basic clustering units, and set the number of clusters to N. L N is obtained through clustering L A probabilistic load scenario curve;

[0053] Using several historical daily wind power output curves as clustering samples, and using a single historical daily wind power output curve W j ={w j,1 ,w j,2 ,…,w j,24 Perform k-means clustering on the basic clustering units, and set the number of clusters to N. W N is obtained through clustering W Probabilistic wind power output scenario curves;

[0054] Using several historical daily photovoltaic power output curves as clustering samples, and using a single historical daily photovoltaic power output curve v k ={v k,1 ,v k,2 ,…,v k,24 Perform k-means clustering on the basic clustering units, and set the number of clusters to N. V N is obtained through clustering V A probabilistic photovoltaic power output scenario curve.

[0055] S3: Combine probabilistic load, wind power, and photovoltaic scenarios, and calculate the net load curve and scenario probability for each combined scenario;

[0056] Furthermore, N L A probabilistic load scenario, N W A probabilistic wind power output scenario and N V By permuting and combining the probabilistic photovoltaic output scenarios, we obtain N. L *N W *N V Each combined scenario includes a load scenario, a wind power output scenario, and a photovoltaic power output scenario.

[0057] It should be noted that the net load curve for the combined scenario is calculated as follows:

[0058] ΔL ijk =L i -W j -V k

[0059] Where, ΔL ijkThis represents the net load curve for a combined scenario under load scenario i, wind power scenario j, and photovoltaic scenario k.

[0060] Furthermore, the probability of a combined scenario is calculated as follows:

[0061]

[0062] in, Let i represent the probability of scenario i under load scenario. Let represent the scenario probability of wind power scenario j. Let k represent the probability of a photovoltaic scenario. This represents the probability of a combined scenario under load scenario i, wind power scenario j, and photovoltaic scenario k, which is the probability of a net load scenario under load scenario i, wind power scenario j, and photovoltaic scenario k.

[0063] S4: Set inverse weighting values ​​based on the probability of net load scenarios, cluster net load scenarios based on probability weighted distance, and generate probabilistic typical net load scenarios.

[0064] Furthermore, regarding N L *N W *N V For each net load scenario, calculate the reciprocal of its scenario probability, as follows:

[0065]

[0066] In the formula, It represents the reciprocal of the scenario probability of the net load scenario ijk.

[0067] Furthermore, the reciprocal of the scenario probability is normalized to obtain the inverse weighted value of the net load scenario probability, calculated as follows:

[0068]

[0069]

[0070] Among them, p' max N represents L *N W *N V The maximum value of the reciprocal of the scenario probability for each net load scenario; This represents the normalized value of the reciprocal of the scenario probability of the net load scenario, which is the inverse weighted value of the net load scenario probability.

[0071] It should be noted that, according to the above formula, the desired first-order reflection Green's function can be obtained by deconvolving the separated uplink and downlink waves in the frequency plane wave domain. During processing, the obtained uplink and downlink wave data need to be subjected to a Fast Fourier Transform for each channel, and after conversion to the frequency plane wave domain, the above deconvolution process is performed for each plane wave direction and each frequency.

[0072] Furthermore, when performing clustering based on the k-means algorithm and calculating the distance between the cluster centers and the net load scenario ijk, the distance is multiplied by the inverse weighted value of the net load scenario probability. The calculation method is as follows:

[0073]

[0074] Among them, D ijk D' represents the raw distance between the cluster centers and the net load scenario ijk; ijk This represents the probability-weighted distance between the cluster center and the net load scenario ijk.

[0075] Example 2

[0076] Reference Figure 2-6 As an embodiment of the present invention, a net load scenario generation method based on probabilistic distance clustering is provided. To verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0077] This example analyzes wind power, photovoltaic, and load data from a province in southern China in July 2020. The data type is 31×24 point curve data.

[0078] (1) Cluster the historical wind power, photovoltaic, and load curve data respectively to obtain probabilistic wind power, photovoltaic, and load scenario curves and their scenario probabilities. The probabilistic scenario curves are as follows: Figure 2-4 As shown in Table 1-3, the scenario probabilities are as follows.

[0079] Table 1 Probability Table of Wind Power Scenarios

[0080]

[0081] Table 2 Probability Table of Photovoltaic Scenarios

[0082]

[0083] Table 3. Probability Table of Load Scenarios

[0084]

[0085] (2) The probabilistic load, wind power and photovoltaic scenarios were combined to obtain a total of 125 combined scenarios. The net load curve and scenario probability of each combined scenario were calculated.

[0086] (3) Set inverse weighting values ​​based on the probability of net load scenarios, and perform probabilistic clustering of net load scenarios based on weighted clustering distance to generate typical net load scenarios, such as... Figure 5 As shown.

[0087] Table 4 Net Load Scenario Probability Table

[0088]

[0089] (4) Comparative Analysis

[0090] Clustering based on historical 31-day net load curves using traditional methods yields probabilistic typical net load scenarios, such as... Figure 6 As shown.

[0091] Table 5. Probability Table of Net Load Scenarios

[0092]

[0093] Table 6 compares typical net load scenarios generated using traditional methods with those generated using the probabilistic distance clustering method proposed in this patent. The table shows that the maximum value of the typical net load scenario curve generated by this patent method is 24066 MW, higher than the maximum value of 23943 MW generated by the traditional method; the minimum value is 12839 MW, lower than the minimum value of 12851 MW generated by the traditional method, indicating that the typical net load scenario generated by this patent method has a wider range of variation and fully considers extreme scenarios. Furthermore, the maximum peak-to-valley difference of the typical net load scenario curve generated by this patent method is 8470 MW, higher than the maximum peak-to-valley difference of 8451 MW generated by the traditional method; the minimum peak-to-valley difference is 5860 MW, lower than the minimum peak-to-valley difference of 6950 MW generated by the traditional method, indicating that for a single net load scenario, this patent method can construct more types of net load scenarios, effectively expanding the possibilities of net load scenarios.

[0094] Table 6 Comparison of results between traditional methods and the method of this patent.

[0095]

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0097] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0098] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0099] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0100] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0101] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0102] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for generating net load scenarios based on probabilistic distance clustering, characterized in that: include, Acquire historical load data, historical wind power and solar power output data; Based on the k-means clustering algorithm, the historical load data, historical wind power and photovoltaic output data are clustered respectively to obtain probabilistic load, wind power and photovoltaic output scenarios; The probabilistic load, wind power, and photovoltaic scenarios are combined to calculate the net load curve and scenario probability for each combined scenario. Based on the probability of the net load scenario, an inverse weighting value is set, and the net load scenarios are clustered based on the probability weighted distance to generate probabilistic typical net load scenarios. The scenario combination includes, A probabilistic load scenario, A probabilistic wind power output scenario and By permuting and combining several probabilistic photovoltaic power output scenarios, we obtain... Each combined scenario includes a load scenario, a wind power output scenario, and a photovoltaic output scenario. The net load curve for the combined scenario is calculated as follows: ， in, This represents the net load curve for a combined scenario under load scenario i, wind power scenario j, and photovoltaic scenario k. These are, respectively, a single historical daily load curve, a single historical daily wind power output curve, and a single historical daily photovoltaic power output curve; The probability of the combined scenarios is calculated as follows: ， in, Let i represent the probability of scenario i under load scenario. Let represent the scenario probability of wind power scenario j. Let k represent the probability of a photovoltaic scenario. This represents the probability of a combined scenario under load scenario i, wind power scenario j, and photovoltaic scenario k, which is the probability of a net load scenario under load scenario i, wind power scenario j, and photovoltaic scenario k. The setting of the reverse weighting value includes, for For each net load scenario, calculate the reciprocal of its scenario probability, as follows: in, This represents the reciprocal of the scenario probability of the net load scenario ijk; The reciprocal of the scenario probability is normalized to obtain the inverse weighted value of the net load scenario probability, calculated as follows: in, express The maximum value of the reciprocal of the scenario probability for each net load scenario; This represents the normalized value of the reciprocal of the scenario probability of the net load scenario, i.e., the inverse weighted value of the net load scenario probability. When performing clustering based on the k-means algorithm, the distance between the cluster centers and the net load scenario ijk is calculated as follows: in, This represents the original distance between the cluster centers and the net load scenario ijk; This represents the probability-weighted distance between the cluster center and the net load scenario ijk.

2. The net load scenario generation method based on probabilistic distance clustering as described in claim 1, characterized in that: The historical load data obtained consists of several 24-point daily load curves, the historical wind power output data consists of several 24-point daily wind power output curves, and the historical photovoltaic output data consists of several 24-point daily photovoltaic output curves.

3. The net load scenario generation method based on probabilistic distance clustering as described in claim 2, characterized in that: The acquisition of the probabilistic load, wind power, and photovoltaic output scenarios includes, Using several historical daily load curves as cluster samples, and using a single historical daily load curve as a cluster sample... Perform k-means clustering on the basic clustering units, and set the number of clusters to 1. Obtained through clustering A probabilistic load scenario curve; Using several historical daily wind power output curves as clustering samples, and using a single historical daily wind power output curve... Perform k-means clustering on the basic clustering units, and set the number of clusters to 1. Obtained through clustering Probabilistic wind power output scenario curves; Using several historical daily photovoltaic power output curves as cluster samples, and using a single historical daily photovoltaic power output curve as a cluster sample... Perform k-means clustering on the basic clustering units, and set the number of clusters to 1. Obtained through clustering A probabilistic photovoltaic power output scenario curve.

Images