Method and system for calculating probabilistic load flow of power distribution network driven by multi-source scene

Abstract

The invention relates to a multi-source scene-driven power distribution network probabilistic load flow calculation method and system, and the method comprises the steps: obtaining the historical data of 24 hours per day in a complete year of load and photovoltaic, and carrying out the clustering division based on a comprehensive similarity index, thereby obtaining a plurality of optical load typical daily output scene sets and the probability distribution thereof; processing data of each scene set to obtain an expected value and a variance of optical charge power; combining the network topology of the power distribution network, taking an expected value as a node injection amount, and performing probabilistic load flow calculation by adopting a two-point estimation method; and carrying out statistical analysis on a calculation result to obtain related statistical data of branch power and node voltage in each typical scene. According to the method, the typical scene probability distribution set is generated by using historical data, the estimation point selection precision and the calculation result reliability are improved, the calculation times are reduced, the efficiency is improved, the actual power grid operation behavior can be better reflected, decision support is provided for a power grid operator, resource allocation is optimized, the cost is reduced, and renewable energy utilization is promoted.

Description

Technical Field

[0001] This invention relates to the field of distribution network probabilistic power flow analysis technology, and in particular to a multi-source scenario-driven distribution network probabilistic power flow calculation method and system. Background Technology

[0002] In recent years, with the large-scale integration of renewable energy sources such as photovoltaic power generation and the rapid development of smart grid technology, the operating characteristics of distribution networks have become increasingly complex and variable. The intermittency and uncertainty of photovoltaic power generation, as well as the dynamic changes in load demand, have brought new challenges to the planning, operation, and management of distribution networks. To effectively assess the impact of various uncertainties in the distribution network on node voltage, line losses, and harmonics, researching corresponding probabilistic power flow calculation methods is particularly important. The statistical results of distribution network state variables obtained by considering uncertainties in the distribution network will have significant practical value.

[0003] Existing probabilistic power flow calculation methods are mainly divided into simulation methods and analytical methods. Simulation methods, represented by Monte Carlo simulation, generate a large number of random samples and perform deterministic power flow calculations on each sample, ultimately obtaining the statistical results. A common analytical method is the point estimation method, which selects a small number of sample points, utilizes data characteristics to determine estimation points, and performs deterministic power flow calculations to approximate the statistical distribution characteristics of various variables in the distribution network. Furthermore, existing methods often employ static or simplified models to predict and analyze photovoltaic power generation and load output.

[0004] However, the Monte Carlo method suffers from low computational efficiency due to the need for a large number of samples, making it difficult to meet the needs of real-time analysis. While the point estimation method reduces computational load, it ignores higher-order statistical properties, limiting its accuracy. Existing models cannot accurately characterize the intermittency of photovoltaic output, dynamic load changes, and differences in scenarios over multiple days, causing analysis results to deviate from actual operating conditions. Furthermore, the lack of consideration for the synergistic effects of multiple source scenarios makes it difficult to comprehensively reflect the complex behavior of the distribution network, limiting the practicality and reliability of decision support. Summary of the Invention

[0005] In view of this, it is necessary to provide a method and system for calculating the probabilistic power flow of a distribution network driven by multiple sources, so as to solve the above-mentioned problems of the existing technology.

[0006] To address the aforementioned problems, in a first aspect, embodiments of the present invention provide a multi-source scenario-driven probabilistic power flow calculation method for distribution networks, comprising:

[0007] S1. Obtain historical data of load and photovoltaic power generation for 24 hours every day throughout the year. Based on the comprehensive similarity index, cluster the data to obtain multiple typical solar load power generation scenario sets and their probability distributions.

[0008] S2, perform data processing on typical solar power generation scenarios for each solar charge to obtain the expected value and variance of solar charge power;

[0009] S3, combined with the distribution network topology, uses the expected value of optical load power as the node injection quantity, and adopts the two-point estimation method to perform probabilistic power flow calculation.

[0010] S4 performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

[0011] Preferably, the comprehensive similarity index includes the photovoltaic Euclidean distance, the load Euclidean distance, and the photovoltaic load correlation distance between the two scenarios;

[0012] In S1, the clustering of data based on a comprehensive similarity index includes:

[0013] Calculate the photovoltaic Euclidean distance, load Euclidean distance, and photovoltaic load correlation distance between the two scenarios;

[0014] The comprehensive similarity index is obtained by weighting and summing the photovoltaic Euclidean distance, load Euclidean distance, and photovoltaic load correlation distance between the two scenarios;

[0015] When the comprehensive similarity index of two operating scenarios is lower than the preset similarity threshold, they are determined to have the same origin and are classified into the same typical scenario set, and the total number of typical output scenario sets is maintained within the set range.

[0016] Preferably, the calculation of the photovoltaic Euclidean distance and the load Euclidean distance between the two scenarios includes:

[0017] For any two scenarios S i With S j Calculate scenario S i With S j The difference matrix D between E Difference matrix D E The calculation formula is:

[0018] D E =S i -S j (1)

[0019] In the formula, S i With S j These are Scene S i With Scene S j The description matrix.

[0020] Scene S i With S j The photovoltaic Euclidean distance and the load Euclidean distance are respectively:

[0021]

[0022] In the formula, d E,pv (S i S j ) for scene S i With S j The photovoltaic Euclidean distance between them; d E,load (S i S j ) for scene S i With S j The load-bearing Euclidean distance between them;

[0023] Calculating the photovoltaic load correlation distance between two scenarios includes:

[0024] For any two scenarios S i With S j Calculate the light charge difference matrix between the two scenes. The light charge difference matrix has 24 rows and n columns. The calculation formula for each row is as follows:

[0025] C i(t) =P DG,i(t) -P load,i(t)

[0026] C j(t) =P DG,j(t) -P load,j(t) (4)

[0027] In the formula, P DG,i(t) P load,i(t) These are Scene S i At time t, the photovoltaic output and load power, P DG,i(t) P load,i(t) These are Scene S j Photovoltaic output and load power at time t;

[0028] The photovoltaic load difference matrix between the two scenarios is as follows:

[0029] D C =C i -C j (5)

[0030] Based on the photovoltaic load difference matrix, scenario S is obtained. i With Scene S j The expression for the correlation distance between photovoltaic loads is:

[0031]

[0032] Preferably, the expression for the comprehensive similarity index is:

[0033] θ=p1d E,pv(S i S j )+p2d E,load (S i S j )+p3d C (S i S j (7)

[0034] In the formula, p1, p2, and p3 are weighting coefficients.

[0035] Preferably, in S2, the data processing of each typical solar power generation scenario set to obtain the expected value and variance of the solar power includes:

[0036] For each typical solar load power generation scenario set, extract the photovoltaic power output and load power data contained therein;

[0037] Assuming the distribution network system has n nodes, and each scenario set corresponds to an n×24 dimension matrix, the expected net load value matrix for each time period of the k-th typical output scenario is as follows:

[0038]

[0039] In the formula, X k Let be the net load expectation matrix for the k-th typical scenario of the distribution network. Let be the expected net load of the i-th node in the distribution network at time j;

[0040] The net load variance matrix for each time period of the k-th typical output scenario is:

[0041]

[0042] In the formula, Y k Let be the net load variance matrix for the k-th typical scenario of the distribution network. Let be the net load variance of the i-th node in the distribution network at time j.

[0043] Preferably, in S3, the step of combining the distribution network topology, using the expected value of optical load power as the node injection quantity, and employing a two-point estimation method for probabilistic power flow calculation includes:

[0044] S31, Obtain the network topology of the distribution network node system and determine the number of nodes as n, of which the number of load nodes is n-1;

[0045] S32, take the expected value of the optical charge power as the initial value of the node injection amount, take the injection power of the load node as the uncertain variable, and determine two estimation points on both sides of the mean of the uncertain variable;

[0046] S33 decomposes the problems of solving the node injection quantity and branch power into 2(n-1) sub-problems. For each sub-problem, the estimated point in the two-point estimation method is used to replace the corresponding uncertain variable, while the other uncertain variables are taken as the expected value of the net load injection quantity. 2(n-1) deterministic power flow calculations are performed on the load node.

[0047] S34. For a typical solar power scene set with solar charge, repeat steps S32 to S33 to perform a 24-hour probabilistic power flow calculation.

[0048] Preferably, in S4, the statistical analysis of the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios includes:

[0049] By statistically analyzing node voltage and branch power flow data for each time period within 24 hours, we construct probability density functions and cumulative distribution functions for node voltage and branch power to reflect the probability distribution characteristics of power grid state variables under multiple typical scenarios.

[0050] Secondly, embodiments of the present invention provide a multi-source scenario-driven distribution network probabilistic power flow calculation system, comprising:

[0051] The solar load data clustering module acquires historical data of load and photovoltaic power generation for 24 hours a day throughout the entire year, and clusters the data based on a comprehensive similarity index to obtain multiple typical solar load power generation scenario sets and their probability distributions.

[0052] The scene data processing module processes data from typical solar power scene sets to obtain the expected value and variance of solar power.

[0053] The power flow calculation module, combined with the distribution network topology, uses the expected value of optical load power as the node injection quantity and adopts a two-point estimation method to perform probabilistic power flow calculation.

[0054] The power flow results statistics module performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

[0055] Thirdly, the present invention also provides an electronic device, including a memory and a processor, wherein,

[0056] The memory is used to store programs;

[0057] The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the multi-source scenario-driven distribution network probabilistic power flow calculation method as described in the first aspect embodiment of the present invention.

[0058] Fourthly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instructions, which, when executed by a processor, can implement the steps in the multi-source scenario-driven distribution network probabilistic power flow calculation method as described in the first aspect embodiment of the present invention.

[0059] The multi-source scenario-driven probabilistic power flow calculation method and system for distribution networks provided by this invention have the following advantages compared with the prior art:

[0060] This invention significantly improves the accuracy of estimation point selection and enhances the reliability of calculation results by generating a probability distribution set of typical daily power generation scenarios for source loads using real historical solar load data. While maintaining high accuracy, this method drastically reduces the number of calculations, improving computational efficiency and making it suitable for real-time analysis of large-scale distribution networks. By considering the operating states of multiple typical daily scenarios in distribution networks, this invention can better reflect the grid behavior under actual operating conditions, providing grid operators with data-driven decision support tools to help them make more rational scheduling and management decisions in the face of uncertainty. Furthermore, through precise scenario extraction and clustering, this invention improves grid operating efficiency, optimizes the allocation of power resources, reduces operating costs, and increases economic benefits, while also promoting the utilization of renewable energy and supporting sustainable energy development goals. Attached Figure Description

[0061] Figure 1 Flowchart of the multi-source scenario-driven distribution network probabilistic power flow calculation method provided by the present invention;

[0062] Figure 2 A schematic diagram of the multi-source scenario-driven probabilistic power flow calculation method for distribution networks provided by this invention;

[0063] Figure 3 The structural block diagram of the multi-source scenario-driven distribution network probabilistic power flow calculation system provided by the present invention is shown below.

[0064] Figure 4 This is a structural block diagram of the electronic device provided by the present invention. Detailed Implementation

[0065] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0066] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0067] Existing technologies for probabilistic power flow calculations in distribution networks primarily employ Monte Carlo simulation and point estimation methods. However, the Monte Carlo method suffers from low computational efficiency due to the need for a large number of samples, making it difficult to meet real-time analysis requirements. While the point estimation method reduces computational load, it ignores higher-order statistical properties, limiting its accuracy. Furthermore, existing methods often use static or simplified models to predict optical load output, failing to accurately capture the fluctuations and variability of optical load, thus limiting the accuracy and practicality of scenario analysis, and neglecting to consider the overall operating status of the distribution network at the scale of daily operational scenarios.

[0068] In view of this, the present invention aims to provide a probabilistic power flow calculation method for distribution networks driven by multiple source scenarios. This method improves the accuracy of point selection in the two-point estimation method by generating a probability distribution set of typical daily power generation scenarios based on historical optical load data, thereby enhancing the reliability of the calculation results. Furthermore, it considers the operating states of multiple typical daily scenarios of the distribution network, enabling the model to better reflect the grid behavior under actual operating conditions. The following will elaborate and describe this method through several embodiments.

[0069] Figure 1 The flowchart shows the multi-source scenario-driven probabilistic power flow calculation method for distribution networks provided by this invention. Figure 2 This is a schematic diagram of the multi-source scenario-driven distribution network probabilistic power flow calculation method provided by the present invention, with reference to... Figure 1 and Figure 2 The multi-source scenario-driven probabilistic power flow calculation method for distribution networks provided by this invention includes at least the following steps:

[0070] Step S1: Obtain historical data of load and photovoltaic power generation for 24 hours every day throughout the year, and cluster the data based on the comprehensive similarity index to obtain multiple typical solar load power generation scenario sets and their probability distributions.

[0071] Specifically, historical data on load and photovoltaic power generation for each day of the past full year is collected. Missing data needs to be supplemented. In this embodiment, methods such as Lagrange interpolation can be used for data supplementation to ensure data integrity and continuity.

[0072] In this embodiment, the comprehensive similarity index includes the photovoltaic Euclidean distance, the load Euclidean distance, and the photovoltaic load correlation distance between the two scenarios. In step S1, the data is clustered based on the comprehensive similarity index, including:

[0073] S11, calculate the photovoltaic Euclidean distance, load Euclidean distance, and photovoltaic load correlation distance between the two scenarios;

[0074] For any two scenarios S i With S j Calculate scenario S i With S j The difference matrix D between E Difference matrix D E The calculation formula is:

[0075] D E =S i -S j (1)

[0076] In the formula, S i With S j These are Scene S i With Scene S j The description matrix.

[0077] Scene S i With S j The photovoltaic Euclidean distance and the load Euclidean distance are respectively:

[0078]

[0079] In the formula, d E,pv (S i S j ) for scene S i With S j The photovoltaic Euclidean distance between them; d E,load (S i S j ) for scene S i With S j The load-bearing Euclidean distance between them;

[0080] Calculating the photovoltaic load correlation distance between two scenarios includes:

[0081] For any two scenarios S i With S j Calculate the light charge difference matrix between the two scenes. The light charge difference matrix has 24 rows and n columns. The calculation formula for each row is as follows:

[0082] C i(t) =P DG,i(t) -P load,i(t)

[0083] C j(t) =P DG,j(t) -P load,j(t) (4)

[0084] In the formula, P DG,i(t) P load,i(t) These are Scene S i At time t, the photovoltaic output and load power, P DG,i(t) P load,i(t) These are Scene S j Photovoltaic output and load power at time t;

[0085] The photovoltaic load difference matrix between the two scenarios is as follows:

[0086] D C =C i -C j (5)

[0087] Based on the photovoltaic load difference matrix, scenario S is obtained. i With Scene S j The expression for the correlation distance between photovoltaic loads is:

[0088]

[0089] S12, the photovoltaic Euclidean distance, load Euclidean distance and photovoltaic load correlation distance between the two scenarios are weighted and summed to obtain the comprehensive similarity index;

[0090] The expression for the comprehensive similarity index is:

[0091] θ=p1d E,pv (S i S j )+p2d E,load (S i S j )+p3d C (S i S j (7)

[0092] In the formula, p1, p2, and p3 are weighting coefficients.

[0093] S13, when the comprehensive similarity index value of two operating scenarios is lower than the preset similarity threshold, it is determined that they have the same source characteristics and are classified into the same typical scenario set, and the total number of typical output scenario sets is maintained within the set range.

[0094] In this embodiment, based on a multi-dimensional comprehensive similarity index, clustering algorithms such as K-means can be used to cluster historical data. Through clustering, similar solar power output scenarios are grouped into one category, forming m typical daily solar power output scenario sets. By setting a reasonable similarity threshold, the total number m of typical output scenario sets is maintained between 6 and 9, thus reducing model complexity while fully covering the main operating modes of power distribution network source-load fluctuations. For each typical scenario set, its probability of occurrence is calculated, i.e., the ratio of the frequency of occurrence of this scenario set in the annual data to the total number of days, resulting in multiple typical daily solar power output scenario sets and their corresponding probability distributions, providing a foundation for subsequent probabilistic power flow calculations.

[0095] Step S2: Perform data processing on each typical solar power generation scenario set to obtain the expected value and variance of solar power.

[0096] Specifically, for each typical solar load power generation scenario set, extract the photovoltaic power output and load power data contained therein;

[0097] Assuming the distribution network system has n nodes, for each typical daily power output scenario set, calculate its expected net load value for each time period within a 24-hour period. Construct an n×24 dimension matrix for each scenario set, obtaining the expected net load value matrix for each time period of the k-th typical power output scenario as follows:

[0098]

[0099] In the formula, X k Let be the net load expectation matrix for the k-th typical scenario of the distribution network. Let be the expected net load of the i-th node in the distribution network at time j;

[0100] The net load variance matrix for each time period of the k-th typical output scenario is:

[0101]

[0102] In the formula, Y k Let be the net load variance matrix for the k-th typical scenario of the distribution network. Let be the net load variance of the i-th node in the distribution network at time j.

[0103] Step S3: Combining the distribution network topology, the expected value of optical load power is used as the node injection quantity, and a two-point estimation method is used to calculate the probabilistic power flow.

[0104] In power distribution networks, the injected power at load nodes is uncertain due to the intermittency and uncertainty of photovoltaic power generation, as well as the dynamic changes in load demand. This uncertainty poses new challenges to the planning, operation, and management of power distribution networks, thus requiring probabilistic power flow calculations to assess its impact.

[0105] Step S3 of the present invention specifically includes the following steps:

[0106] S31, obtain the network topology of the distribution network node system, determine the number of nodes as n, of which the number of load nodes is n-1; the other node is used as a reference node or a balancing node.

[0107] S32, the expected value of the optical load power (the net load expected value matrix X calculated in step S2) is used. k Using the injected power of the load nodes as the initial value of the node injection, and taking the injected power of the load nodes as an uncertain variable, two estimation points are determined on both sides of the mean of the uncertain variable. and

[0108] Using two variables on either side of the mean and To match random quantities The first three moments are used to replace the probability density function of the injection amount. and Defined as:

[0109]

[0110] In the formula, sp ik For location measurement, the expression is as follows:

[0111]

[0112] In the formula, λ i,3 for skewness coefficient, for The third central moment, Ω a For the a-th typical scene set, Let be the net load of node i at time j in scenario o.

[0113] S33 decomposes the problems of solving the node injection quantity and branch power into 2(n-1) sub-problems. For each sub-problem, the estimated point in the two-point estimation method is used to replace the corresponding uncertain variable, while the other uncertain variables are taken as the expected value of the net load injection quantity. 2(n-1) deterministic power flow calculations are performed on the load node.

[0114] S34. For a typical solar power scene set with solar charge, repeat steps S32 to S33 to perform a 24-hour probabilistic power flow calculation.

[0115] In this embodiment, when performing probabilistic power flow calculations using the two-point method for various typical scenarios of the distribution network, two values ​​are determined on both sides of the mean of each uncertain variable (i.e., the injected power at the load node), dividing the problem of solving for the node injection quantity and the branch power into several sub-problems. The values ​​on both sides of the mean are used to replace each uncertain variable, while other uncertain variables are represented by the expected value of the net load injection quantity, thereby performing deterministic power flow calculations.

[0116] Step S4: Perform statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under each typical scenario.

[0117] Specifically, for each typical solar load power generation scenario set, node voltage and branch power flow data for each time period within 24 hours are statistically analyzed. Probability Density Functions (PDFs) and Cumulative Distribution Functions (CDFs) for node voltage and branch power are constructed to reflect the probability distribution characteristics of grid state variables under multiple typical scenarios. The probability density function describes the probability density of node voltage and branch power occurring near a specific value. This helps grid operators identify potential risks in grid operation, such as excessively high or low voltage, or branch overload, allowing for proactive preventative and adjustment measures. The cumulative distribution function describes the cumulative probability of node voltage and branch power being less than or equal to a specific value. This provides decision support for grid operators, such as determining voltage and power thresholds to ensure timely intervention during grid operation and prevent exceeding safe limits.

[0118] By constructing the probability density function and cumulative distribution function of node voltage and branch power, the probability distribution characteristics of node voltage and branch power, such as the probability of voltage exceeding the limit and the risk of branch overload, can be presented intuitively, and the impact of source-load uncertainty on the operating status of the distribution network can be quantified.

[0119] Figure 3 The structural block diagram of the multi-source scenario-driven distribution network probabilistic power flow calculation system provided by the present invention is shown below. Figure 3 The multi-source scenario-driven distribution network probabilistic power flow calculation system 300 includes:

[0120] The solar load data clustering module 301 acquires historical data of the load and photovoltaic data for 24 hours a day throughout the year, and clusters the data based on a comprehensive similarity index to obtain multiple typical solar load power generation scenario sets and their probability distributions.

[0121] The scene data processing module 302 processes the data of each typical solar power scene set to obtain the expected value and variance of the solar power.

[0122] The power flow calculation module 303, in conjunction with the distribution network topology, uses the expected value of optical load power as the node injection quantity and employs a two-point estimation method to perform probabilistic power flow calculation.

[0123] The power flow result statistics module 304 performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

[0124] The multi-source scenario-driven distribution network probabilistic power flow calculation system provided by the present invention is used to execute the multi-source scenario-driven distribution network probabilistic power flow calculation method provided in the foregoing embodiments. The multi-source scenario-driven distribution network probabilistic power flow calculation method has been described in detail in the foregoing embodiments, and will not be repeated here.

[0125] The multi-source scenario-driven probabilistic power flow calculation method and system for distribution networks provided by this invention have the following advantages compared with the prior art:

[0126] (1) This invention significantly improves the accuracy of the selection of estimation points and enhances the reliability of the calculation results by generating a probability distribution set of typical daily power generation scenarios of source load using real historical solar load data. While maintaining high accuracy, this method greatly reduces the number of calculations and improves computational efficiency, making it suitable for real-time analysis of large-scale distribution networks.

[0127] (2) By considering the operating status of multiple typical daily scenarios of distribution networks, the present invention can better reflect the behavior of the power grid under actual operating conditions, provide power grid operators with data-driven decision support tools, and help them make more reasonable scheduling and management decisions in the face of uncertainty.

[0128] (3) This invention improves the operating efficiency of the power grid, optimizes the allocation of power resources, reduces operating costs, and improves economic benefits through accurate scene extraction and clustering. At the same time, it promotes the utilization of renewable energy and supports the goal of sustainable energy development.

[0129] Figure 4 A structural block diagram of the electronic device provided by the present invention, such as Figure 4 As shown, the present invention also provides an electronic device 400, which can be a mobile terminal, desktop computer, laptop, handheld computer, server, or other computing device. The electronic device 400 includes a processor 401 and a memory 402, wherein the memory 402 stores a multi-source scenario-driven distribution network probabilistic power flow calculation program 403.

[0130] In some embodiments, memory 402 may be an internal storage unit of a computer device, such as a hard disk or memory. In other embodiments, memory 402 may be an external storage device of a computer device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, memory 402 may include both internal and external storage units of the computer device. Memory 402 is used to store application software and various types of data installed on the computer device, such as program code installed on the computer device. Memory 402 can also be used to temporarily store data that has been output or will be output. In one embodiment, when the multi-source scenario-driven distribution network probabilistic power flow calculation program 403 is executed by processor 401, the following steps are implemented:

[0131] S1. Obtain historical data of load and photovoltaic power generation for 24 hours every day throughout the year. Based on the comprehensive similarity index, cluster the data to obtain multiple typical solar load power generation scenario sets and their probability distributions.

[0132] S2, perform data processing on typical solar power generation scenarios for each solar charge to obtain the expected value and variance of solar charge power;

[0133] S3, combined with the distribution network topology, uses the expected value of optical load power as the node injection quantity, and adopts the two-point estimation method to perform probabilistic power flow calculation.

[0134] S4 performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

[0135] In some embodiments, processor 401 may be a central processing unit (CPU), microprocessor or other data processing chip, used to run program code stored in memory 402 or process data, such as executing a distribution network probabilistic power flow calculation program driven by multi-source scenarios.

[0136] This embodiment also provides a computer-readable storage medium storing a multi-source scenario-driven distribution network probabilistic power flow calculation program. When executed by a processor, the multi-source scenario-driven distribution network probabilistic power flow calculation program performs the following steps:

[0137] S1. Obtain historical data of load and photovoltaic power generation for 24 hours every day throughout the year. Based on the comprehensive similarity index, cluster the data to obtain multiple typical solar load power generation scenario sets and their probability distributions.

[0138] S2, perform data processing on typical solar power generation scenarios for each solar charge to obtain the expected value and variance of solar charge power;

[0139] S3, combined with the distribution network topology, uses the expected value of optical load power as the node injection quantity, and adopts the two-point estimation method to perform probabilistic power flow calculation.

[0140] S4 performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

[0141] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A probabilistic power flow calculation method for distribution networks driven by multiple sources, characterized in that, include: S1. Obtain historical data of load and photovoltaic power generation for 24 hours every day throughout the year. Based on the comprehensive similarity index, cluster the data to obtain multiple typical solar load power generation scenario sets and their probability distributions. S2, perform data processing on typical solar power generation scenarios for each solar charge to obtain the expected value and variance of solar charge power; S3, combined with the distribution network topology, uses the expected value of optical load power as the node injection quantity, and adopts the two-point estimation method to perform probabilistic power flow calculation. S4 performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

2. The multi-source scenario-driven probabilistic power flow calculation method for distribution networks according to claim 1, characterized in that, The comprehensive similarity index includes the photovoltaic Euclidean distance, the load Euclidean distance, and the photovoltaic load correlation distance between the two scenarios; In S1, the clustering of data based on a comprehensive similarity index includes: Calculate the photovoltaic Euclidean distance, load Euclidean distance, and photovoltaic load correlation distance between the two scenarios; The comprehensive similarity index is obtained by weighting and summing the photovoltaic Euclidean distance, load Euclidean distance, and photovoltaic load correlation distance between the two scenarios; When the comprehensive similarity index of two operating scenarios is lower than the preset similarity threshold, they are determined to have the same origin and are classified into the same typical scenario set, and the total number of typical output scenario sets is maintained within the set range.

3. The multi-source scenario-driven probabilistic power flow calculation method for distribution networks according to claim 2, characterized in that, The calculation of the photovoltaic Euclidean distance and the load Euclidean distance between the two scenarios includes: For any two scenarios S i With S j Calculation scenario S i With S j The difference matrix D between E Difference matrix D E The calculation formula is: D E =S i -S j (1) In the formula, S i With S j These are Scene S i With Scene S j The description matrix. Scene S i With S j The photovoltaic Euclidean distance and the load Euclidean distance are respectively: In the formula, d E,pv (S i S j ) for scene S i With S j The photovoltaic Euclidean distance between them; d E,load (S i S j ) for scene S i With S j The load-bearing Euclidean distance between them; Calculating the photovoltaic load correlation distance between two scenarios includes: For any two scenarios S i With S j Calculate the light charge difference matrix between the two scenes. The light charge difference matrix has 24 rows and n columns. The calculation formula for each row is as follows: In the formula, P DG,i(t) P load,i(t) These are Scene S i At time t, the photovoltaic output and load power, P DG,i(t) P load,i(t) These are Scene S j Photovoltaic output and load power at time t; The photovoltaic load difference matrix between the two scenarios is as follows: D C =C i -C j (5) Based on the photovoltaic load difference matrix, scenario S is obtained. i With Scene S j The expression for the correlation distance between photovoltaic loads is:

4. The multi-source scenario-driven probabilistic power flow calculation method for distribution networks according to claim 3, characterized in that, The expression for the comprehensive similarity index is: In the formula, p1, p2, and p3 are weighting coefficients.

5. The multi-source scenario-driven probabilistic power flow calculation method for distribution networks according to claim 1, characterized in that, In S2, the data processing of each typical solar power generation scenario set to obtain the expected value and variance of solar power includes: For each typical solar load power generation scenario set, extract the photovoltaic power output and load power data contained therein; Assuming the distribution network system has n nodes, and each scenario set corresponds to an n×24 dimension matrix, the expected net load value matrix for each time period of the k-th typical output scenario is as follows: In the formula, X k Let be the net load expectation matrix for the k-th typical scenario of the distribution network. Let be the expected net load of the i-th node in the distribution network at time j; The net load variance matrix for each time period of the k-th typical output scenario is: In the formula, Y k Let be the net load variance matrix for the k-th typical scenario of the distribution network. Let be the net load variance of the i-th node in the distribution network at time j.

6. The multi-source scenario-driven probabilistic power flow calculation method for distribution networks according to claim 1, characterized in that, In S3, the method of combining the distribution network topology, using the expected value of optical load power as the node injection quantity, and employing a two-point estimation method for probabilistic power flow calculation includes: S31, Obtain the network topology of the distribution network node system and determine the number of nodes as n, of which the number of load nodes is n-1; S32, take the expected value of the optical charge power as the initial value of the node injection amount, take the injection power of the load node as the uncertain variable, and determine two estimation points on both sides of the mean of the uncertain variable; S33 decomposes the problems of solving the node injection quantity and branch power into 2(n-1) sub-problems. For each sub-problem, the estimated point in the two-point estimation method is used to replace the corresponding uncertain variable, while the other uncertain variables are taken as the expected value of the net load injection quantity. 2(n-1) deterministic power flow calculations are performed on the load node. S34. For a typical solar power scene set with solar charge, repeat steps S32 to S33 to perform a 24-hour probabilistic power flow calculation.

7. The multi-source scenario-driven probabilistic power flow calculation method for distribution networks according to claim 6, characterized in that, In S4, the statistical analysis of the probabilistic power flow calculation results yields relevant statistical data on branch power and node voltage under various typical scenarios, including: By statistically analyzing node voltage and branch power flow data for each time period within 24 hours, we construct probability density functions and cumulative distribution functions for node voltage and branch power to reflect the probability distribution characteristics of power grid state variables under multiple typical scenarios.

8. A multi-source scenario-driven probabilistic power flow calculation system for distribution networks, characterized in that, include: The solar load data clustering module acquires historical data of load and photovoltaic power generation for 24 hours every day throughout the year, and clusters the data based on a comprehensive similarity index to obtain multiple typical solar load power generation scenario sets and their probability distributions. The scene data processing module processes data from typical solar power scene sets to obtain the expected value and variance of solar power. The power flow calculation module, combined with the distribution network topology, uses the expected value of optical load power as the node injection quantity and adopts a two-point estimation method to perform probabilistic power flow calculation. The power flow results statistics module performs statistical analysis on the probabilistic power flow calculation results to obtain relevant statistical data on branch power and node voltage under various typical scenarios.

9. An electronic device, Its features are, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the multi-source scenario-driven distribution network probabilistic power flow calculation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the multi-source scenario-driven distribution network probabilistic power flow calculation method according to any one of claims 1 to 7.

Images