A Method for Risk Assessment of Distribution Network Operation in New Distributed Energy Scenarios

Abstract

This invention relates to the field of distribution network operation assessment technology, specifically a method for assessing distribution network operation risks in new distributed energy scenarios. The method includes the following steps: establishing a joint probability distribution function for wind and solar power output based on historical photovoltaic (PV) and wind power data; sampling the joint probability distribution function and performing an inverse transformation to obtain PV and solar power output data for each time period considering correlation; clustering the sampled PV and solar power output data for each time period to generate typical daily scenarios for wind and PV power; and performing improved Latin hypercube sampling on the generated typical daily scenarios for wind and PV power, calculating the voltage exceedance risk at distribution network nodes through power flow calculations. Compared with existing technologies, this invention fully considers the correlation characteristics between distributed energy sources and takes into account the correlation of random variables in the sampling method, making it more consistent with the current operational status of new distribution networks.

Description

Technical Field

[0001] This invention relates to the field of distribution network operation, and in particular to a method for assessing the operation risk of distribution networks in new scenarios of distributed energy. Background Technology

[0002] In power distribution networks, both distributed energy resources and loads exhibit randomness. The output of wind turbines and solar power, particularly distributed energy resources, is highly volatile. Current research rarely considers the correlation between wind and solar power in power distribution network operation assessments. While there is a weak correlation between distributed energy resources such as distributed solar and distributed wind power, considering individual distributed energy resource connections ignores the impact of this correlation. The integration of electric vehicles into the power distribution network significantly impacts the voltage at network nodes. Distributed energy resources can serve as voltage support for the connected electric vehicles, significantly improving the network voltage level. The connection of electric vehicles at different nodes has varying impacts on the operational risk of the power distribution network, leading to inaccurate risk assessment results. Traditional assessments of power distribution network operation risk have not considered the correlation between distributed solar and wind power, and rarely take into account the impact of electric vehicle connections at different nodes on the operational risk of the power distribution network. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a method for assessing the operation risk of distribution networks that takes into account new scenarios of distributed energy.

[0004] The objective of this invention can be achieved through the following technical solutions:

[0005] A method for assessing the operational risks of distribution networks in new scenarios of distributed energy resources, comprising the following steps:

[0006] Based on historical photovoltaic and wind power data, a joint probability distribution function for wind and solar power output is established.

[0007] The joint probability distribution function is sampled and inversely transformed to obtain the wind and solar power output data generated considering the correlation in each time period;

[0008] Clustering is performed on the wind and solar power output data generated from sampling at different time periods, taking into account correlation, to generate typical daily scenarios for wind power and photovoltaic power;

[0009] Improved Latin hypercube sampling is applied to the generated typical daily scenarios of wind power and photovoltaic power, and the operation risk of the distribution network is calculated through power flow calculation.

[0010] As a preferred technical solution, the specific steps for establishing the joint probability distribution function include:

[0011] The kernel density estimation method is used to generate the probability density function of wind and solar power output for each time period of 24 hours based on historical photovoltaic and wind power data;

[0012] For each time period, a joint probability distribution function of photovoltaic and wind power based on the Frank-Copula function is established.

[0013] As a preferred technical solution, the kernel density estimation method employs a Gaussian kernel function:

[0014]

[0015] In the formula, d represents the total number of days of historical data input, n represents any hourly segment within a 24-hour period, and x n X represents the expected probability density point. n These are points in the sampled data.

[0016] As a preferred technical solution, the sampling and inverse transformation of the joint probability distribution function to obtain the wind and solar power output for each time period specifically involves:

[0017] The joint probability distribution function of each time period is sampled using cubic spline interpolation to solve for the wind and solar power output of each time period corresponding to the cumulative probability.

[0018] Based on the sampling results and the inverse transformation of the joint probability distribution function, the output of distributed photovoltaic and wind power is calculated to obtain the wind and photovoltaic output generated considering the correlation in each time period.

[0019] As a preferred technical solution, the spline interpolation method is expressed as follows:

[0020] S i (x)=a i +b i (xx i )+c i (xx i ) 2 +d i (xx i ) 3

[0021] In the formula, S i (x) represents the sample multi-style, x i Represents a node.

[0022] As a preferred technical solution, the distributed photovoltaic power output modeling is as follows:

[0023] P = AG·η

[0024] In the formula, P represents the instantaneous power output of the photovoltaic system, A represents the effective area of ​​the photovoltaic module, G represents the solar radiation intensity, and η represents the conversion efficiency of the photovoltaic module.

[0025] As a preferred technical solution, the distributed wind power output modeling is as follows:

[0026]

[0027] In the formula, P w For active power output of wind power; k and c represent the shape and scale parameters of the Weibull distribution, respectively; μ w and σ w These represent the mean and standard deviation of wind power active power output, respectively.

[0028] As a preferred technical solution, the wind and solar power output data are clustered using the K-means clustering algorithm, and the specific steps are as follows:

[0029] Randomly select k data points as initial cluster centers; assign data points to each data point in the dataset to the nearest cluster center; update cluster centers by calculating the average value of all points in each cluster and updating the average value as the new cluster center; repeat the above steps until the cluster centers no longer change or the preset number of iterations is reached.

[0030] As a preferred technical solution, the improved Latin hypercube sampling comprises the following steps:

[0031] Based on the type of input variables and the linear correlation coefficient matrix C X The correlation coefficient matrix C of the standard normal variables after Nataf transformation is obtained. Z And the correlation coefficient matrix C W Perform Cholesky decomposition to obtain the lower triangular matrix B;

[0032] The first sample matrix W is obtained by sampling M standard normal variables. N×M And obtain the correlation coefficient matrix C between each column of the first sample matrix. W For the correlation coefficient matrix C W Cholesky decomposition yields the lower triangular matrix Q;

[0033] According to Z = W(Q) T ) -1 B T The correlation matrix is ​​C. Z The second sample matrix Z is used to obtain the order matrix L based on the order of its elements in each column. S ;

[0034] The first sample matrix W N×M According to the order matrix L S After sorting, Latin hypercube sampling is performed to obtain the final sampling matrix S.

[0035] As a preferred technical solution, the final sampling matrix S is used to calculate the node voltage over-limit risk, as follows:

[0036] The risk of node voltage exceeding limits considers the probability of voltage exceeding limits, which is the probability of exceeding the allowable voltage deviation range under the current voltage level of the node.

[0037]

[0038] In the formula, V i V represents the voltage magnitude at node i; i max and V i min The upper and lower limits of the node voltage are represented; F(V) represents the voltage accumulation function.

[0039] Severity index of node voltage over-limit risk Sev(V) i The following is represented:

[0040]

[0041]

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention provides a novel power distribution network operation evaluation method that takes into account new scenarios of distributed energy. It considers the correlation characteristics between distributed energy sources, establishes a joint probability distribution function based on the Frank-Copula function, and then obtains the wind and solar power output data generated considering the correlation in each time period through sampling and inverse transformation. Furthermore, it proposes an improved Latin hypercube sampling method that takes into account the correlation of random variables in the sampling method, which is more in line with the current operation status of new power distribution networks. Attached Figure Description

[0044] Figure 1 This is a flowchart of the distribution network operation risk assessment process for new distributed energy scenarios according to the present invention.

[0045] Figure 2 This is a topology diagram of an IEEE 34-node distribution network.

[0046] Figure 3 The photovoltaic time series distribution is shown based on the Frank-Copula correlation function.

[0047] Figure 4 The wind time series distribution map is based on the Frank-Copula correlation function;

[0048] Figure 5 To improve the flowchart of the Latin hypercube sampling method;

[0049] Figure 6A diagram showing the risk of voltage exceeding limits at each node of the distribution network. Detailed Implementation

[0050] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0051] Example 1

[0052] This invention provides a method for assessing the operational risks of distribution networks in new scenarios of distributed energy resources, such as... Figure 1 As shown, the specific implementation method is as follows:

[0053] S1: Generate the probability density function of wind and solar power output for each time period of 24 hours using the kernel density estimation method based on historical photovoltaic and wind power data;

[0054] S2: Establish a joint probability distribution function based on the Frank-Copula function for each time period;

[0055] S3: Sample the joint Copula function and joint probability distribution function for each time period. Based on the sampling results and the inverse transformation of the joint probability distribution function, finally obtain the wind and solar power output for each time period, i.e., wind and solar power output generated considering correlation.

[0056] S4: Cluster the sampled data to generate 20 typical daily scenarios for wind power and photovoltaic power;

[0057] S5: Improved Latin hypercube sampling is performed on various typical daily scenarios of wind power and photovoltaic power to calculate the operation risk of the distribution network through power flow calculation.

[0058] In step S1, the kernel density estimation method is used to analyze the historical photovoltaic and wind power data for each time period.

[0059] The kernel density estimation method selects a Gaussian kernel function:

[0060]

[0061] In the formula, d represents the total number of days of historical data input, n represents any hourly segment within a 24-hour period, and x n X represents the expected probability density point. n These are points in the sampled data.

[0062] In step S2, based on the Frank-Copula function joint probability distribution function, considering the weak correlation between photovoltaic and wind power, and according to the research literature, the correlation function between photovoltaic and wind power is close to the Frank-Copula empirical function. Therefore, a Frank-Copula function-based photovoltaic and wind power joint probability distribution function is established.

[0063] Frank-Copula distribution function:

[0064]

[0065] Frank-Copula density function:

[0066]

[0067] Wind-solar joint probability model for each period

[0068] F(x, y) = C(F X (x), F Y (y))

[0069] In step S3, the joint distribution function is sampled, and the wind-solar output for each period is obtained by inverse transformation. The 3rd-order spline interpolation method is used to solve the wind-solar output corresponding to the cumulative probability for each period of sampling.

[0070] Spline interpolation method:

[0071] S i (x) = a i + b i (x - x i ) + c i (x - x i ) 2 + d i (x - x i ) 3

[0072] In the formula, S i (x) represents the sample polynomial, and x i represents the node.

[0073] Secondly, the output of distributed photovoltaic and wind power is calculated. The modeling method of distributed photovoltaic output is as follows:

[0074] P = A·G·η

[0075] In the formula, P represents the instantaneous power output of the photovoltaic system, A represents the effective area of the photovoltaic module, G represents the solar radiation intensity, and η represents the conversion efficiency of the photovoltaic module, usually between 15% and 20%.

[0076] The modeling method of distributed wind power output is as follows:

[0077]

[0078] In the formula, P w For active power output of wind power; k and c represent the shape and scale parameters of the Weibull distribution, respectively; μ w and σ w These represent the mean and standard deviation of wind power active power output, respectively.

[0079] In step S4, a clustering algorithm generates typical daily scenes, and the K-means clustering algorithm is used to cluster the N groups of sampling results.

[0080] The specific steps of the K-means clustering algorithm are as follows: 1. Randomly select k data points as initial cluster centers (centroids); 2. Assign data points to the nearest cluster center for each data point in the dataset. Euclidean distance is typically used to measure this distance; 3. Update cluster centers by calculating the average value of all points within each cluster and updating this average value as the new cluster center; 4. Iterate, repeating the above steps until the cluster centers no longer change or the preset number of iterations is reached.

[0081] In step S5, improved Latin hypercube sampling is performed for various wind power and photovoltaic scenarios, and power flow calculation is used to determine the operational risks of the distribution network.

[0082] This invention considers improving the Latin Hypercube Sampling (LHS) sampling process, mainly based on the inverse function transformation method. It assumes a sampling size of N and the number of random variables M, namely X1, X2, ..., X... M random variable X K The cumulative probability distribution function is:

[0083] Y K =F K (X K K = 1, 2, ..., M

[0084] The range of the distribution function [0, 1] is divided into N non-overlapping, equally spaced subintervals [n / N, (n+1) / N], where n = 0, 1, ..., (N-1). Therefore, the length of each subinterval is 1 / N. Then, a T is selected in each subinterval. n That is, T n = (n+1-ξ) / N. ξ is selected primarily through the following methods: 0.5, Hammersley point set, interval boundary, and random number. Where ξ = 0.5 represents lattice sampling. This yields T. n The sampled values ​​are then obtained using an inverse transform:

[0085]

[0086] in, It is F k Inverse transformation of (·).

[0087] However, when dealing with multi-input random variables, the accuracy of the LHS simulation is affected not only by the sampled values ​​but also by the correlation between the sampled values ​​of different input random variables. Different correlation control methods use different correlation coefficients, including: CLMCS, single-bond optimization algorithm, Cholesky decomposition, genetic algorithm, simulated annealing algorithm, etc. After obtaining a sorted matrix with a certain correlation, arranging the sampled matrices according to the sorted matrix yields sampled matrices with the same correlation.

[0088] This invention employs a more accurate LHCS ranking method for handling the correlation of random variables. The specific steps are as follows:

[0089] 5.1) Calculate the cumulative probability distribution function of historical photovoltaic and wind power data and input it into the system parameters;

[0090] 5.2) Input system parameters, distribution functions of random variables, and linear correlation coefficient matrix C of variables. X Among them, the linear correlation coefficient matrix C of the variables X It consists of the Pearson correlation coefficient between the two variables;

[0091] 5.3) Based on the type of input variables and the linear correlation coefficient matrix C X The correlation coefficient matrix C of the standard normal variables after Nataf transformation is obtained. Z And the correlation coefficient matrix C Z Perform Cholesky decomposition to obtain the lower triangular matrix B;

[0092] 5.4) Sample matrix W is obtained by sampling from M standard normal variables. N×M The correlation coefficient matrix C between its columns is obtained by solving. W And the correlation coefficient matrix C Z Cholesky decomposition yields the lower triangular matrix Q;

[0093] 5.5) According to Z = W(Q) T ) -1 B T The correlation matrix is ​​C. Z The sample matrix Z is then used to obtain the order matrix L based on the order of its elements in each column. S ;

[0094] 5.6) The sample matrix obtained from the previous sampling is arranged according to the order matrix L. SAfter sorting, Latin hypercube sampling is performed to obtain the final sampling matrix S.

[0095] The final sampling matrix S is used for power flow calculation to obtain the node voltage over-limit risk value. In this method, the node voltage over-limit risk considers the voltage over-limit probability, which is the probability of exceeding the allowable voltage deviation range under the current voltage level of the node.

[0096]

[0097] In the formula, V i V represents the voltage magnitude at node i. i max and V i min The upper and lower limits of the node voltage are 1.05 and 0.95, respectively, and F(V) represents the voltage accumulation function.

[0098] Using Sev(V) i The following formula represents the severity index of node voltage over-limit risk:

[0099]

[0100] Traditional risk assessments of power distribution networks rarely consider the correlation between distributed photovoltaic (PV) and wind power. This invention provides a novel power distribution network operation assessment method that considers new scenarios involving distributed energy resources. It takes into account the correlation characteristics between distributed energy sources and incorporates the correlation of random variables in the sampling method, making it more consistent with the current operational status of modern power distribution networks.

[0101] Example 2

[0102] As one specific embodiment of the present invention, such as Figure 2 As shown, this example uses the IEEE 34-bus distribution system and is based on historical photovoltaic and wind power data from a region in eastern China. The total area of ​​the photovoltaic panels is 16,000 m². 2 The photoelectric conversion efficiency is 14%, the wind farm cut-in wind speed is 3 m / s, the cut-out wind speed is 15 m / s, the rated wind speed is 13.5 m / s, and the rated power of each wind farm is 1 MW.

[0103] Based on historical wind and solar power output data, kernel density estimation is used to generate probability density functions for wind and solar power output in each hour of the 24-hour period. The Frank-Copula function is selected to describe the correlation between wind and solar power output, and the joint probability distribution function for each hour is sampled. Finally, inverse transformation is performed to obtain the wind and solar power output scenarios and their probabilities. Typical daily scenarios are then generated using the K-means clustering method, such as... Figure 3 , 4 As shown.

[0104] like Figure 5This invention employs an LHCS sampling method that considers correlation, using the sampled data as input parameters for power flow calculation, and finally outputs the risk of node voltage exceeding limits.

[0105] like Figure 6 As shown in the results of the voltage limit exceedance risk calculation, the maximum voltage limit exceedance risk occurs at node 34, which is also the solar and wind power access node, with a voltage limit exceedance risk value of 0.035. The closer to node 34, the greater the voltage limit exceedance risk value, while the voltage amplitude is lower closer to the end nodes. Nodes 1-6 basically do not generate voltage limit exceedances.

[0106] In summary, this invention proposes a novel distribution network operation risk assessment method considering new scenarios of distributed energy resources. First, it models the correlation scenario of distributed energy resources. Second, based on correlation control, this invention proposes an LHCS sampling method that considers correlation, and performs power flow calculations to summarize the node voltage exceedance risk. The novel distribution network operation risk assessment method considering new scenarios of distributed energy resources provided by this invention proposes a sampling method for distributed energy resources that better aligns with considering correlation, and can provide a reference for current research on sampling in scenarios considering correlation.

[0107] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for assessing the operational risks of distribution networks in new scenarios of distributed energy, characterized by the following steps: include: Based on historical photovoltaic and wind power data, a joint probability distribution function for wind and solar power output is established. The joint probability distribution function is sampled and inversely transformed to obtain the wind and solar power output data generated considering the correlation in each time period; Cluster the wind and solar power output data obtained from sampling at different time periods to generate typical daily scenarios for wind power and photovoltaic power; Improved Latin hypercube sampling is applied to the generated typical daily scenarios of wind power and photovoltaic power, and the risk of voltage over-limit at distribution network nodes is calculated through power flow calculation. The specific steps of the improved Latin hypercube sampling are as follows: Based on the type of input variables and the linear correlation coefficient matrix The correlation coefficient matrix of the standard normal variables after Nataf transformation is obtained. And the correlation coefficient matrix Perform Cholesky decomposition to obtain the lower triangular matrix. B ; right M The first sample matrix is ​​obtained by sampling from standard normal variables. And obtain the correlation coefficient matrix between each column of the first sample matrix. For the correlation coefficient matrix Cholesky decomposition yields the lower triangular matrix. Q ; according to The correlation matrix is ​​obtained as follows The second sample matrix Z Then, the order matrix is ​​obtained from the second sample matrix Z according to the order of its elements in each column. ; The first sample matrix According to the order matrix After sorting, Latin hypercube sampling is performed to obtain the final sampling matrix. S ; For the final sampling matrix S The risk of voltage exceeding limits at distribution network nodes is obtained through power flow calculations, as detailed below: The risk of node voltage exceeding limits considers the probability of voltage exceeding limits, which is the probability of exceeding the allowable voltage deviation range at the current voltage level of the node. In the formula, Represents a node i The voltage amplitude; and Indicates the upper and lower limits of the node voltage; Represents the voltage accumulation function; Severity index of node voltage over-limit risk It is expressed as follows: 。 2. The method for assessing the operational risks of distribution networks in new scenarios of distributed energy, as described in claim 1, is characterized in that... The specific steps for establishing the joint probability distribution function include: The kernel density estimation method is used to generate the probability density function of wind and solar power output for each time period of 24 hours based on historical photovoltaic and wind power data; For each time period, a joint probability distribution function of photovoltaic and wind power based on the Frank-Copula function is established.

3. The method for assessing the operational risks of distribution networks in new scenarios of distributed energy, as described in claim 2, is characterized in that... The kernel density estimation method described above uses a Gaussian kernel function: In the formula, d represents the total number of days of the input historical data, and n represents any hourly segment within a 24-hour period. This represents the expected probability density point. These are points in the sampled data.

4. The method for assessing the operational risks of distribution networks in new scenarios of distributed energy, as described in claim 2, is characterized in that... The sampling and inverse transformation of the joint probability distribution function to obtain the wind and solar power output for each time period are specifically as follows: The joint probability distribution function of each time period is sampled using cubic spline interpolation to solve for the wind and solar power output of each time period corresponding to the cumulative probability. Based on the sampling results and the inverse transformation of the joint probability distribution function, the output of distributed photovoltaic and wind power is calculated to obtain the wind and photovoltaic output generated considering the correlation in each time period.

5. The method for assessing the operational risks of distribution networks in new scenarios of distributed energy, as described in claim 4, is characterized in that... The spline interpolation method is expressed as follows: In the formula, This indicates that the sample has multiple styles. Represents a node.

6. The method for assessing the operation risk of distribution networks in new scenarios of distributed energy resources according to claim 4, characterized in that, The distributed photovoltaic power output modeling is as follows: In the formula, P Indicates the instantaneous power output of the photovoltaic system. A Indicates the effective area of ​​the photovoltaic module. G Indicates the intensity of solar radiation. This indicates the conversion efficiency of the photovoltaic module.

7. The method for assessing the operation risk of distribution networks in new scenarios of distributed energy, as described in claim 4, is characterized in that... The distributed wind power output modeling is as follows: In the formula, Contribute to wind power; k and c These represent the shape and scale parameters of the Weibull distribution, respectively. and These represent the mean and standard deviation of wind power active power output, respectively.

8. The method for assessing the operational risks of distribution networks in new scenarios of distributed energy resources according to claim 1, characterized in that, The wind and solar power output data were clustered using the K-means clustering algorithm. The specific steps are as follows: Random selection k 1. Select 1 data point as the initial cluster center; 2. Assign data points to the nearest cluster center for each data point in the dataset. Update the cluster center by calculating the average value of all points in each cluster and updating the average value as the new cluster center; repeat the above steps until the cluster center no longer changes or the preset number of iterations is reached.

Images