A probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output

Abstract

A probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output belongs to the field of power grid dispatching. In the process of power flow calculation, an uncertainty model for distributed photovoltaic access is established; the access node of the distributed photovoltaic power station is regarded as a PV node, the load is set as a ZIP model, and the node injection power equation is used as the original power flow equation; the improved particle swarm algorithm is used to estimate the parameters of the Gaussian mixture model; the likelihood function is constructed; the iterative optimization is performed with the maximum likelihood function as the optimization goal; the node voltage and cross-section power flow probability distribution description are obtained; the node voltage joint cumulative distribution function and joint probability density function are obtained when the photovoltaic output power is modeled using the Gaussian mixture model. It considers the randomness and correlation of the photovoltaic power station output when performing power flow calculation, which can improve the accuracy of the power flow calculation of the relevant lines, achieve a balance between the global optimization and local optimization capabilities, and thus make it easier to obtain the global optimal solution.

Description

Technical Field

[0001] The present invention belongs to the field of power flow calculation in power operation management, and in particular relates to a probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output. Background Art

[0002] The access of distributed photovoltaic power generation to the distribution network causes node voltage and line power flow to become volatile and uncertain.

[0003] The uncertainty of photovoltaic power station output will cause changes in system power flow. Traditional deterministic models can no longer meet the distribution network requirements of renewable energy access. At the same time, due to the correlation between photovoltaic power stations in time and space scales, traditional deterministic power flow calculations can no longer be applied to steady-state operation evaluation and analysis that takes randomness into account. A large number of correlation factors will be introduced during operation, such as the strong correlation between the outputs of multiple wind farms. The correlation will have an important impact on the safe and stable operation of the power system.

[0004] However, some probabilistic power flow methods such as the traditional semi-invariant method cannot be directly applied to the probabilistic power flow calculation of renewable energy power systems taking into account the correlation of input variables.

[0005] Therefore, how to take into account the randomness and correlation of renewable energy output, establish a renewable energy output uncertainty model, and on this basis, establish a random power flow calculation method, has become an important part of the operation and management of the power system. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a probabilistic power flow calculation method that takes into account the randomness and correlation of photovoltaic power station output. It first establishes an uncertainty model for distributed photovoltaic access, solves the parameters of the Gaussian mixture model based on the improved particle swarm algorithm, obtains the node voltage and cross-sectional power flow probability distribution description, and then obtains the node voltage joint cumulative distribution function (CDF) and joint probability density function (PDF) of the photovoltaic output power under the GMM modeling. It adopts an analytical probabilistic power flow analysis method based on the linearized power flow equation, and uses a typical IEEE 33-node simulation system and a certain area's actual distribution network system for verification, proving that it has higher accuracy than the EM algorithm and can be better applied to power operation and power dispatching systems.

[0007] The technical solution of the present invention is to provide a probabilistic power flow calculation method considering the randomness and correlation of the output of a photovoltaic power station, including power flow calculation, which is characterized in that in the process of performing the power flow calculation, the following steps are performed:

[0008] 1) First, establish an uncertainty model for distributed photovoltaic access;

[0009] 2) The access nodes of distributed photovoltaic power stations are regarded as PV nodes, the load is set as ZIP model, and the node injection power equation is used as the original power flow equation;

[0010] 3) Using improved particle swarm algorithm to estimate Gaussian mixture model parameters;

[0011] 4) Based on the actual collected active power output of the photovoltaic power station, a sample X={X1,…,X N}, and based on this, construct the likelihood function, and iterate and optimize the result parameters with the maximum likelihood function as the optimization goal;

[0012] 5) Obtain the linear relationship between the node voltage amplitude V and line flow Pij taking into account the PV node and ZIP load model and the output active power X of the PV power station;

[0013] 6) The active power output X of the photovoltaic power station obtained based on the multivariate Gaussian distribution Nm(X) model;

[0014] 7) Obtain the probability distribution description of node voltage and section power flow;

[0015] 8) Obtaining the joint cumulative distribution function CDF and joint probability density function PDF of node voltage when the photovoltaic output power is modeled using the Gaussian mixture model GMM;

[0016] The node voltage joint cumulative distribution function CDF of its node voltage V is expressed as:

[0017]

[0018] The joint probability density function PDF of its node voltage V is expressed as:

[0019]

[0020] 9) On this basis, a random power flow calculation method is established which takes into account the randomness and correlation of the output of photovoltaic power stations.

[0021] Specifically, the probabilistic power flow calculation method considering the randomness and correlation of the output of the photovoltaic power station, when performing power flow calculation, considers the randomness and correlation of the output of the photovoltaic power station, and uses the improved particle swarm algorithm to estimate the parameters of the Gaussian mixture model GMM, and the specific steps are as follows:

[0022] 2.1. First, determine x = [ω m ,μ m ,Σ m ] Form a particle group to be determined and initialize the particle position and velocity;

[0023] 2.2. Then update the position and velocity of the particle in each iteration;

[0024] 2.3. Calculate the updated particles according to the set objective function, update and record the current optimal value and global optimal value of the particle;

[0025] 2.4. Check whether the upper limit of the number of cycles is reached or whether the difference between the generations meets the error constraint. If so, the calculation ends and the result is output; otherwise, go to step 2.2 to continue the calculation.

[0026] Furthermore, the specific steps of updating are as follows:

[0027] Each iteration updates the particle's position and velocity according to the following two formulas:

[0028] v i =λv i +c1random(0,1)(p i -x i )

[0029] +c2random(0,1)(g i -x i );

[0030] x i =x i +v i ;

[0031] In the formula, λ is the inertia factor, c1 and c2 are acceleration constants, vi and xi are the velocity and current position of the particle, pi and gi correspond to the optimal position and global optimal position searched by the current particle, and random(0,1) represents a random number in the interval [0,1]

[0032] Furthermore, the inertia factor adopts a linear decreasing weight strategy, namely:

[0033]

[0034] In the formula, λ start and λ end are the initial weight and the final weight respectively, k is the current iteration number, T max is the maximum number of iterations.

[0035] Furthermore, when the position and velocity of the particle are updated in each iteration, the inertia factor adopts a dynamic λ value to obtain a better optimization effect compared with a fixed value.

[0036] Specifically, the linear relationship between the node voltage amplitude V and the line flow Pij of the PV node and ZIP load model and the output active power X of the photovoltaic power station is expressed by the following formula:

[0037]

[0038] Where X represents the active power output of the photovoltaic power station; A and B represent the linearization coefficients respectively.

[0039] The probabilistic power flow calculation method considering the randomness and correlation of the output of photovoltaic power stations described in the technical solution of the present invention adopts an analytical probabilistic power flow analysis method based on a linearized power flow equation, considers the randomness and correlation of the output of photovoltaic power stations, and establishes an uncertainty model for distributed photovoltaic access. The parameters of the Gaussian mixture model based on the improved particle swarm algorithm are solved to obtain a description of the probability distribution of node voltage and cross-sectional power flow, and then the photovoltaic output power is obtained by using the joint cumulative distribution function and the joint probability density function of the node voltage under the GMM modeling condition, thereby improving the accuracy of the power flow calculation of related lines, achieving a balance between global optimization and local optimization capabilities, and making it easier to obtain the global optimal solution.

[0040] Furthermore, the probabilistic power flow calculation method considering the randomness and correlation of the output of the photovoltaic power station adopts the line power flow joint probability density as the probability of evaluating the simultaneous over-limit of two or more line power flows.

[0041] Compared with the prior art, the advantages of the present invention are:

[0042] 1. The technical solution of the present invention adopts an analytical probability power flow analysis method based on a linearized power flow equation in the process of power flow calculation, takes into account the randomness and correlation of the output of photovoltaic power stations, establishes an uncertainty model for distributed photovoltaic access, and obtains a description of the probability distribution of node voltage and cross-section power flow;

[0043] 2. The technical solution of the present invention improves the accuracy of related line power flow calculation by using the node voltage joint cumulative distribution function and joint probability density function under the GMM modeling of photovoltaic output power;

[0044] 3. The technical solution of the present invention takes into account the randomness and correlation of renewable energy output, establishes a renewable energy output uncertainty model, and on this basis, establishes a random power flow calculation method, and uses the joint probability density of line power flow as a measure to evaluate the probability of two or more line power flows exceeding the limit at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic flow chart of the method of the present invention;

[0046] Figure 2 is an IEEE 33-node system diagram of an embodiment of the present invention;

[0047] FIG3( a ) is a schematic diagram of a probability density function of the voltage amplitude at node 5 in an embodiment;

[0048] FIG3( b ) is a schematic diagram of the cumulative distribution function of the voltage amplitude at node 5 in the embodiment;

[0049] Figure 4(a) is a schematic diagram of the joint probability density of line power flows based on the EM algorithm;

[0050] FIG4( b ) is a schematic diagram of the joint probability density of line power flows based on the algorithm of the present invention;

[0051] Figure 4(c) is a schematic diagram of the joint probability density of line power flows based on the MCS algorithm. DETAILED DESCRIPTION

[0052] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0053] The access of distributed photovoltaic power generation to the distribution network causes node voltage and line power flow to fluctuate, and traditional deterministic models cannot fully describe the fluctuation characteristics of renewable energy.

[0054] Therefore, taking into account the randomness and correlation of the output of photovoltaic power stations, the technical solution of the present invention first establishes an uncertainty model for distributed photovoltaic access.

[0055] 1.1. Gaussian mixture model:

[0056] The Gaussian Mixture Model (GMM) can accurately model non-Gaussian random variables, and can approximate any probability density distribution by using a linear combination of a certain amount of Gaussian density functions. Its expression is as follows:

[0057]

[0058] Where X is the input variable of the expression, representing the output power vector of the photovoltaic power station; ω m Represents the weight coefficient of the mth Gaussian distribution subcomponent; μ m and Σ m Represent the expected vector and covariance matrix of the mth Gaussian subcomponent, respectively. The off-diagonal elements of the covariance matrix are used to describe the correlation between different photovoltaic power plants. The parameters that need to be further estimated for the Gaussian mixture model GMM are ω and m , μ m and Σ m .

[0059] When performing parameter estimation, we first generate a sample X = {X1,…,X N}, and construct the likelihood function based on this, as shown in the following two formulas:.

[0060]

[0061] Then, an iterative optimization is performed with the maximum likelihood function as the optimization goal, and the optimization result parameters are the parameters that need to be estimated by the Gaussian Mixture Model (GMM).

[0062] 1.2. Gaussian mixture model parameter solution based on improved particle swarm algorithm:

[0063] In the process of solving the parameters of the traditional Gaussian mixture model GMMD, the expectation maximization (EM) algorithm is often used. Considering that the accuracy of the algorithm depends on the selection of initial values, inappropriate initial values ​​may cause the iterative optimization to obtain a local optimal solution, which will have a negative impact on the subsequent settlement results.

[0064] In contrast, the improved particle swarm algorithm can greatly reduce the impact of initial values ​​and achieve a balance between global optimization and local optimization, making it easier to obtain the global optimal solution.

[0065] Therefore, this technical solution adopts an improved particle swarm algorithm to estimate the parameters of the Gaussian mixture model GMM. The main steps are as follows:

[0066] 1) First determine x = [ω m ,μ m ,Σ m ] Form a particle group to be determined and initialize the particle position and velocity;

[0067] 2) Then the position and velocity of the particle are updated in each iteration, as shown in the following two formulas:

[0068]

[0069] x i =x i +v i (6)

[0070] Where c1 and c2 are acceleration constants, usually c1=c2∈[0,4], v i and x i is the particle's velocity and current position, p i and g i Corresponding to the optimal position and global optimal position searched by the current particle, random(0,1) represents a random number in the interval [0,1]. λ is the inertia factor. Studies have shown that a dynamic λ value can achieve better optimization results than a fixed value. Here, a linearly decreasing weight (LDW) strategy can be used, namely:

[0071]

[0072] In the formula, λ start and λ end are the initial weight and the final weight respectively, k is the current iteration number, T max is the maximum number of iterations.

[0073] 3) Calculate the updated particles according to the set objective function, update and record the current optimal value and global optimal value of the particle;

[0074] 4) Check whether the upper limit of the number of cycles is reached or whether the difference between the generations meets the error constraint condition. If so, the calculation ends and the result is output; otherwise, go to step 2) to continue the calculation.

[0075] 2. Analytical probabilistic power flow analysis method based on linearized power flow equation:

[0076] For situations that require repeated large-scale power flow calculations, such as distribution network reliability assessment and distribution network probabilistic power flow, the use of linearized power flow models can improve analysis efficiency and there is no convergence problem.

[0077] 2.1、Probabilistic power flow analytical expression:

[0078] The structural configuration of the traditional distribution network is that there is only one generator as a balancing node, and the rest are all PQ nodes, excluding PV nodes. The research on the linearized power flow equations of this type of system is relatively mature.

[0079] However, due to the characteristics of distributed renewable energy (DG) connected to the new distribution network, it is sometimes necessary to treat it as a PV node in the power flow calculation, and ZIP loads are increasingly valued in distribution network analysis. Therefore, it is necessary to establish a distribution network power flow calculation method with wider applicability.

[0080] In this technical solution, the distributed renewable energy (DG) access node is regarded as a PV node, the load is set as a ZIP model, the node injection power equation is used as the original power flow equation, and v i Incorporated into the summation operation, as shown in the following formula:

[0081]

[0082] In the formula, i and j are node numbers; n is the number of nodes; P i , Q i Inject active and reactive power into the node; v i 、v j is the voltage amplitude of nodes i and j; θ ij is the phase angle difference between nodes i and j; G ij , Bij are the elements in the admittance matrix.

[0083] The ZIP model of load describes the relationship between load and voltage amplitude, as shown in the following formula:

[0084]

[0085] Where: C Z , C I , C P They are the constant impedance, constant current and constant power coefficient of active power respectively; the superscript “′” indicates the coefficient related to reactive power; P0 and Q0 are the active load and reactive load under the nominal voltage respectively; V is the per-unit value of the voltage amplitude.

[0086] Substituting equation (9) into (8), the AC power flow equation taking into account the ZIP load model is as follows:

[0087]

[0088] With v i and θ ij As an independent variable, Taylor expansion is performed on the right side of the equation (taking the active power term as an example), and the linearized power flow equation shown in the following two formulas can be obtained:

[0089]

[0090] The AC power flow equation in branch power form is:

[0091]

[0092] With v and θ ij Taylor expansion is performed for the independent variables, and the linearized power flow equation can be obtained as follows:

[0093]

[0094] The branch power equation does not involve frequency modulation characteristics and load models, and can be reflected in the expressions of node voltage amplitude, phase angle and branch power coupling relationship.

[0095] The above formulas (12) and (14) are the node voltage amplitude V and line flow P considering the PV node and ZIP load model. ij The linear relationship with the output power X of the photovoltaic power station can be abbreviated as follows:

[0096]

[0097] Where X represents the output power of the photovoltaic power station; A and B represent the linearization coefficients respectively.

[0098] 2.2、Description of node voltage and section power flow probability distribution:

[0099] If the random variable obeys a Gaussian distribution, then it will still obey a Gaussian distribution after a linear transformation.

[0100] Therefore, based on the multivariate Gaussian distribution N m (X) The output power X of the photovoltaic power station obtained by modeling, after the linear transformation of the following formula, the node voltage U and the line flow S also obey the Gaussian distribution. Here, the node voltage is selected for derivation and verification. The amplitude of the node voltage obeys the Gaussian distribution, where the mean phasor is A1μ m +B1, the covariance matrix is ​​A1Σ m A1 T , so the joint probability density function (PDF) of the node voltage can be expressed as:

[0101]

[0102] In the formula, K represents the number of nodes.

[0103] By performing multiple integrations on the above formula, we can obtain the joint cumulative distribution function (CDF) of the node voltage V, as shown in the following formula:

[0104] F m (V) = ∫…∫N m (V)dV1…dV K (17)

[0105] When the photovoltaic power X is modeled by a multi-weight multi-variable GMM, the node voltage joint cumulative distribution function CDF can be expressed as:

[0106]

[0107] Accordingly, by differentiating the above formula, the joint probability density function PDF of the node voltage V can be obtained:

[0108]

[0109] Formulas (18) and (19) are the joint cumulative distribution function CDF and joint probability density function PDF of node voltage when photovoltaic output power is modeled using GMM.

[0110] Example:

[0111] In order to verify the accuracy of the Gaussian mixture model based on the improved particle swarm algorithm described in this technical solution, and the rationality of the distribution network planning evaluation system under distributed photovoltaic access formed based on this model, the typical IEEE 33 node simulation system and the actual distribution network system of a certain region are used for verification. The structure diagram of the IEEE 33 node simulation system is shown in the figure. Figure 2 shown.

[0112] Among them, the photovoltaic field is connected at nodes 7, 16, 20, 24 and 29, and the output power adopts the photovoltaic power station output active power sample X established based on the Nataf method.

[0113] In order to verify the superiority of the Gaussian mixture model based on the improved particle swarm proposed in this technical solution, the traditional expectation maximum algorithm (EM) was selected as the control group for comparison in terms of node voltage and line power flow joint probability solution.

[0114] (1) Node voltage probability density function:

[0115] The advantage of the Monte Carlo method (MCS) is its high calculation accuracy, but it is limited by its calculation speed. Therefore, it is often used as a benchmark to compare the advantages and disadvantages of other methods.

[0116] This technical solution obtains the probabilistic power flow distribution of node voltage and line power flow in the system based on the EM algorithm and the proposed method, respectively, and compares them with the MCS algorithm to illustrate the accuracy of the two.

[0117] The accuracy of the proposed improved algorithm is verified by calculation based on the probability distribution of node voltage and line flow.

[0118] Node 5 is selected as the observation point to obtain the corresponding voltage probability density distribution, and line 4-5 and line 30-31 are selected as power flow observation lines for calculation and comparison.

[0119] As shown in FIG. 3( a ) and FIG. 3( b ), compared with the traditional EM algorithm, the node voltage amplitude probability distribution obtained by the algorithm proposed in this technical solution is more consistent with the MCS result.

[0120] Specifically, based on the EM algorithm calculation Figure 2 The average root mean square error and maximum absolute error

[25] of the probability distribution of the voltage amplitude of the middle node 5 are 0.044 and 0.305 respectively. Based on the method proposed in this technical solution, the two error indexes are reduced to 0.000295 and 0.0472 respectively, which verifies the superiority of the method proposed in this technical solution in improving the calculation accuracy.

[0121] (2) Line power flow joint probability density function:

[0122] The joint probability density of line flows can be used to describe the probability that multiple random variables are in the same state. Here, it is used to evaluate the probability that two or more line flows exceed the limit at the same time.

[0123] As shown in Figures 4(a), 4(b) and 4(c), the white area represents the area where the joint probability density of the line flows of lines 4-5 and 30-31 is the largest, the black area represents the area with the smallest probability density, and the other gray areas are between them.

[0124] It can be found from the figure that for the area with the largest joint probability density of line flows in Figure 4(a), the corresponding flow range of line 4-5 is [2.35, 2.57], and the flow range of line 30-31 is [5.05, 5.15]. The two corresponding flow ranges in Figure 4(b) are [2.38, 2.5] and [5.08, 5.13] respectively, while the corresponding flow sections in Figure 4(c) are [2.36, 2.51] and [5.07, 5.14]. The similarity between Figures 4(b) and (c) is higher.

[0125] In addition, the gray areas in Figures 4(b) and (c) correspond to joint probabilities of 1 to 1.2, while the corresponding probability range in Figure 4(a) is 0.7 to 0.8, with a difference of 0.3 to 0.4. This further proves that the method described in the present technical solution has higher accuracy than the EM algorithm.

[0126] The technical solution of the present invention takes into account the randomness and correlation of the output of photovoltaic power stations during the process of power flow calculation, establishes an uncertainty model for distributed photovoltaic access, and obtains a description of the probability distribution of node voltage and cross-section power flow; improves the accuracy of related line power flow calculation by using the node voltage joint cumulative distribution function and the joint probability density function under the GMM modeling of photovoltaic output power; takes into account the randomness and correlation of renewable energy output, establishes a random power flow calculation method, and uses the line power flow joint probability density as the probability of evaluating the simultaneous over-limit of two or more line power flows, which has higher precision and accuracy than the EM algorithm.

[0127] The present invention can be widely used in the field of planning and evaluation of distribution networks.

Claims

1. A probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output, including power flow calculation, characterized by During the power flow calculation process, the following steps are performed: 1) First, establish an uncertainty model for distributed photovoltaic access; 2) The access nodes of distributed photovoltaic power stations are regarded as PV nodes, the load is set as ZIP model, and the node injection power equation is used as the original power flow equation; 3) Using improved particle swarm algorithm to estimate Gaussian mixture model parameters; 4) Based on the actual collected active power output of the photovoltaic power station, a sample X={X1,…,X N }, and based on this, construct the likelihood function, and iterate and optimize the result parameters with the maximum likelihood function as the optimization goal; 5) Obtain the linear relationship between the node voltage amplitude V and line flow Pij taking into account the PV node and ZIP load model and the output active power X of the PV power station; 6) The active power output X of the photovoltaic power station is obtained based on the multivariate Gaussian distribution Nm(X) model; 7) The probability distribution description of the node voltage and section power flow is obtained; 8) Obtaining the joint cumulative distribution function CDF and joint probability density function PDF of node voltage when the photovoltaic output power is modeled using the Gaussian mixture model GMM; The node voltage joint cumulative distribution function CDF of its node voltage V is expressed as: The joint probability density function PDF of its node voltage V is expressed as: 9) On this basis, a random power flow calculation method is established which takes into account the randomness and correlation of the output of photovoltaic power stations.

2. The probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output according to claim 1, characterized in that The probabilistic power flow calculation method considering the randomness and correlation of the output of the photovoltaic power station, when performing power flow calculation, considers the randomness and correlation of the output of the photovoltaic power station, and uses the improved particle swarm algorithm to estimate the parameters of the Gaussian mixture model GMM. The specific steps are as follows: 2.

1. First, determine x = [ω m ,μ m ,Σ m ] Form a particle group to be determined and initialize the particle position and velocity; 2.

2. Then update the position and velocity of the particle in each iteration; 2.

3. Calculate the updated particles according to the set objective function, update and record the current optimal value and global optimal value of the particle; 2.

4. Check whether the upper limit of the number of cycles is reached or whether the difference between the generations meets the error constraint condition. If so, the calculation ends and the result is output; Otherwise, go to step 2.2 to continue the calculation.

3. The probabilistic power flow calculation method considering the randomness and correlation of the output of a photovoltaic power station according to claim 2, characterized in that The specific steps for updating are as follows: Each iteration updates the particle's position and velocity according to the following two formulas: x i =x i +v i ; Where λ is the inertia factor, c1 and c2 are acceleration constants, and v i and x i is the particle's velocity and current position, p i and g i Corresponding to the optimal position searched by the current particle and the global optimal position, random(0,1) represents a random number in the interval [0,1].

4. The probabilistic power flow calculation method considering the randomness and correlation of the output of a photovoltaic power station according to claim 3 is characterized in that The inertia factor adopts a linear decreasing weight strategy, namely: In the formula, λ start and λ end are the initial weight and the final weight respectively, k is the current iteration number, T max is the maximum number of iterations.

5. The probabilistic power flow calculation method considering the randomness and correlation of the output of a photovoltaic power station according to claim 3 is characterized in that When the position and velocity of the particle are updated in each iteration, the inertia factor adopts a dynamic λ value to obtain a better optimization effect compared with a fixed value.

6. The probabilistic power flow calculation method considering the randomness and correlation of the output of a photovoltaic power station according to claim 1, characterized in that The node voltage amplitude V and line flow P considering the PV node and ZIP load model ij The linear relationship with the active power X output by the photovoltaic power station is expressed by the following formula: Where X represents the active power output of the photovoltaic power station; A and B represent the linearization coefficients respectively.

7. The probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output according to claim 1, characterized in that The probabilistic power flow calculation method considering the randomness and correlation of the output of the photovoltaic power station adopts an analytical probabilistic power flow analysis method based on the linearized power flow equation, considers the randomness and correlation of the output of the photovoltaic power station, and solves the parameters of the Gaussian mixture model based on the improved particle swarm algorithm by establishing an uncertainty model for distributed photovoltaic access, obtains the node voltage and the cross-section power flow probability distribution description, and then obtains the node voltage joint cumulative distribution function and the joint probability density function of the photovoltaic output power under the GMM modeling condition, improves the accuracy of the power flow calculation of the relevant lines, achieves the balance between the global optimization and the local optimization capabilities, and makes it easier to obtain the global optimal solution.

8. The probabilistic power flow calculation method considering the randomness and correlation of photovoltaic power station output according to claim 1, characterized in that The probabilistic power flow calculation method considering the randomness and correlation of the output of the photovoltaic power station adopts the line power flow joint probability density as the probability of evaluating the simultaneous over-limit of two or more line power flows.

Images