Asymmetric medium voltage three-phase distribution network probabilistic power flow calculation method with single-phase photovoltaic access

Abstract

The application discloses an asymmetric medium-voltage three-phase distribution network probability power flow calculation method containing single-phase photovoltaic access, and comprises the following steps: 1, based on historical power data of photovoltaic, weight values are obtained by an improved entropy weight method, and then the output probability density function of distributed photovoltaic is obtained by LightGBM prediction; 2, for the medium-voltage three-phase distribution network with single-phase photovoltaic access, the uncertainty of photovoltaic output and access position is considered, the neutral point voltage is alternately iterated with Newton method, and a three-phase distribution network power flow model is constructed; 3, integral points and weights of point estimation method are determined, space transformation is carried out by Nataf inverse transformation, and power flow distribution is obtained from the three-phase distribution network power flow model; each order statistical moment is calculated, Gram-Charlier series fitting is adopted, and the probability distribution of the distribution network system is obtained. The application considers the influence of uncertain single-phase photovoltaic access to obtain the probability power flow distribution, and has positive significance for the study of power flow, probability optimization problem, system operation and safety evaluation of the neutral point not directly grounded distribution network system.

Description

Technical Field

[0001] This invention belongs to the fields of photovoltaic prediction, three-phase distribution network power flow and probabilistic power flow, specifically a probabilistic power flow calculation method for an asymmetrical medium-voltage three-phase distribution network with single-phase photovoltaic access. Background Technology

[0002] With the widespread adoption of new energy sources, the controllability of new energy power generation is receiving increasing attention. The power output of photovoltaic (PV) power generation is affected by numerous factors such as sunlight intensity and temperature, resulting in uncertainty. Under the trend of diversified energy consumption demands, distribution network loads have shifted from a single electrical load to multiple types of loads with uncertainties, including electricity, gas, and heat.

[0003] With the integration of new energy sources such as wind and solar power into the grid, the uncertainty of the grid's operational status has increased significantly. Furthermore, over 80% of the neutral points in medium-voltage distribution networks in my country employ non-effective grounding. These factors pose risks to distribution network planning and operational status control; therefore, assessing the probability distribution of these factors is crucial for evaluating the system's operational status.

[0004] Existing probabilistic power flow algorithms do not adequately consider the implementation effects of renewable energy output, relying solely on fuzzy logic, which leads to errors in the results. Furthermore, the neutral point of my country's medium-voltage distribution network is generally not effectively grounded, accounting for over 80% of the total grid, with most of these being ungrounded or grounded via arc suppression coils. Existing power flow algorithms largely ignore neutral point parameters or assume ideal operation, failing to consider the impact of neutral point voltage deviations on distribution network power flow. Summary of the Invention

[0005] This invention addresses the shortcomings of existing technologies by proposing a probabilistic power flow calculation method for asymmetrical medium-voltage three-phase distribution networks with single-phase photovoltaic access. The aim is to obtain the probabilistic power flow distribution of the entire system by calculating the power flow of the medium-voltage distribution network under photovoltaic access, thereby assessing and predicting the operating status of the distribution network and evaluating the security of the distribution network under renewable energy access.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: The present invention provides a probabilistic power flow calculation method for an asymmetrical medium-voltage three-phase distribution network with single-phase photovoltaic access, characterized by the following steps: Step 1: Obtain the normalized historical dataset of photovoltaic power plant power. z t | t =1, 2, … , T},in, z t Indicates the firstt Power data for individual photovoltaic power plants T Indicates the number of data items; Step 2, obtain the dataset of influencing factors { h g | g =1, 2, … , t},in, h g Indicates the first g One influencing factor, t Indicates the number of influencing factors; Step 2.1, using equation (1) to obtain h g weight : (1) In equation (1), H g For the first g The information entropy of each influencing factor is obtained through equation (2). Indicates the first g’ Information entropy of each influencing factor; This is the average of all information entropies; express 35th power; (2) In equation (2), f g For the first g The proportion of each influencing factor is obtained from equation (3); (3) Step 2.2, use equation (4) to obtain the weighted result of the first step. g Influencing factors s g : (4) Step 3, use the LightGBM algorithm to... s g Make a prediction and obtain the first g One influencing factor s g Predicted value S g ; Step 4, use the KDE method to... S g After fitting, the probability density function of photovoltaic power output is obtained. f ( P PI );in, PPI This indicates the active power output of the photovoltaic system. Step 5, using the probability density function f ( P PI The active power of T photovoltaic outputs on the grid. x t | t =1, 2, … , T} constitute the original space variables; where, x t Indicates the first t The active power output of each photovoltaic system; Step 6, x t As the active load of any phase in a node of a three-phase distribution network, a power flow model of a three-phase distribution network with single-phase photovoltaic access is constructed. Step 7: Based on the voltage and network parameters in the three-phase distribution network power flow model, obtain the integrand using equation (5). f ( g , β ): (5) In equation (5), g Indicates the integral position of the estimated point. λ represents the location of the single-phase photovoltaic grid connection point, and λ represents any one of phases a, b, and c. Represents a node i λ-phase voltage; express The conjugate of ; q represents phases a, b, and c; Represents a node j The conjugate of the q-phase voltage; Indicates the location of the photovoltaic grid connection point β The voltage conjugate; Represents a node i The conjugate of the load equivalent admittance between the q-phase and the neutral point; Represents a node i With nodes j The conjugate of the admittance matrices between them; The location of the estimated point is obtained using equation (6). g : (6) In equation (6), g k Indicates the first k The integral position of each estimated point k =0, 1, … , l ; l Indicates the estimated number of points; Indicates the first k One estimated point g k The weights; f ( g k , β ) indicates the first k One estimated point g k The integrand; Step 8, obtain using Nataf inverse transform g k The mapping matrix on the original spatial variables { x t,k | t =1, 2, …, T};in, x t,k express g k exist x t Mapping on; Step 9, using the power flow model to... x t,k Conduct the first k The second deterministic power flow calculation yields the... t The active power output of the photovoltaic system x t The k Secondary tidal current distribution X ( t,k ); Step 10: Calculate the expected power flow distribution using equation (7). m X and υ-th order moment : (7) In equation (7), E μ Represents the mapping matrix { x t,k | t =1, 2, … , T The expected value of}; E ( X ( t,k )) indicates the first k Secondary power flow distribution X ( t,k The mean of ) for x t Expected value; Step 11: Fit the υth moment using Gram-Charlier series. The power flow probability distribution of the three-phase distribution network is obtained.

[0007] The method for calculating the probabilistic power flow of an asymmetrical medium-voltage three-phase distribution network with single-phase photovoltaic access, as described in this invention, is also characterized in that step 6 includes: Step 6.1, use equation (8) to obtain the node. i Access the t The active power output of the photovoltaic system x t neutral point voltage U iN : (8) In equation (8), p represents a phase corresponding to a single-phase photovoltaic system; m represents any phase other than phase p. Represents a node i The initial value of the m-phase voltage; Y ipN Represents a node i The equivalent load admittance between phase p and the neutral point; Y imN Represents a node i The equivalent load admittance between phase m and the neutral point; Y iN Represents a node i Neutral point-to-ground admittance; i =1, 2, … , n ,in, n Indicates the number of nodes; Step 6.2, use equation (9) to obtain the node. i Unconnected x t The initial value of the neutral point voltage : (9) In equation (9), Represents a node i The initial value of the q-phase voltage; Represents a node i The equivalent load admittance between phase q and the neutral point; Step 6.3, use equation (10) to obtain the node. i Active power imbalance of p-phase connected single-phase photovoltaic 、 Reactive power imbalance : (10) In equation (10), Represents a node i The p-phase voltage value; 1i is the imaginary unit; ϕ Indicates the power factor angle; i p The phase angle represents the voltage of phase p. i N This represents the voltage phase angle at the neutral point; Step 6.4, use equation (11) to obtain the node. i Not connected x t Active power imbalance 、 Reactive power imbalance : (11) In equation (11), Represents a node i q phase and node j The electrical conductance between phases a; Represents a node i q phase and node j The susceptance between phases a; Represents a node i q phase and node j The electrical conductance between phase b; Represents a node i q phase and node j The susceptance between phase b; Represents a node i q phase and node j The electrical conductance between phases c; Represents a node i q phase and node j The susceptance between the c phases; sin i iqja Represents a node i q phase and node j The sine of the phase angle difference between phases a and b; cos i iqja Represents a node i q phase and node j The cosine of the phase angle difference between phases a; sin i iqjb Represents a node i q phase and node j The sine of the phase angle difference between phases b; cos i iqjb Represents a node i q phase and node j The cosine of the phase angle difference between phases b; sin i iqjc Represents a node i q phase and node j The sine of the phase angle difference between phases c; cos i iqjc Represents a node i q phase and node j The cosine of the phase angle difference between phases c; Represents a node i The electrical conductance between the q-phase and the neutral point; Represents a node i The susceptance between the q phase and the neutral point; sin i iqN Represents a node i The sine of the phase angle difference between phase q and the neutral point; cos i iqN Represents a node i The cosine of the phase angle difference between phase q and the neutral point; Represents a node i The q-phase voltage; Represents a node j Phase a voltage; Represents a node j The voltage of phase b; Represents a node j c-phase voltage; U iN Represents a node i The neutral point voltage; Step 6.5, correct the nodes using equation (12). i neutral point voltage U iN In the e Calculated values ​​under secondary power flow iteration : (12) In equation (12), Indicates the first e -1 Nodes under Trend Iteration i The load admittance between phase q and the neutral point; Step 6.6: Use equation (13) to obtain the location of the single-phase photovoltaic access point. p-phase voltage : (13) In equation (13), G iβ Represents a node i and β The conductivity between; B iβ Represents a node i and β The susceptance values ​​between; Represents a node i The amount of active power injected into the p-phase; Represents a nodei The amount of reactive power injected into the p-phase; Indicates the location of the access point β The p-phase voltage value of the previous node; Step 6.7: Use equation (14) to obtain the location of the access point. β Three-phase load imbalance : (14) In equation (14), U βN Access point location β The neutral point voltage is obtained from equation (15); (15).

[0008] The present invention provides an electronic device, comprising a memory and a processor, wherein the memory is used to store a program that supports the processor in executing the probabilistic power flow calculation method for the asymmetrical medium-voltage three-phase distribution network, and the processor is configured to execute the program stored in the memory.

[0009] The present invention discloses a computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, performs the steps of the probabilistic power flow calculation method for the asymmetrical medium-voltage three-phase distribution network.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The probabilistic power flow method for asymmetrical medium-voltage three-phase distribution networks with single-phase photovoltaic access provided by this invention can retain the volatility and intermittency of photovoltaic power generation, consider the impact of relevant factors of photovoltaic power generation on power output, and thus establish an accurate photovoltaic power output prediction model. It can quantitatively evaluate the uncertainties of photovoltaics, reduce the degree of danger to the system after grid connection, and improve the stability of grid connection.

[0011] 2. The probabilistic power flow method for asymmetrical medium-voltage three-phase distribution networks with single-phase photovoltaic access provided by this invention can consider the impact of neutral point voltage offset on power flow results, thereby avoiding the impact of neutral point voltage phase angle changes on power flow without increasing system memory or affecting power flow convergence results; at the same time, it considers the uncertain definition of photovoltaic access location to obtain the three-phase voltage imbalance curve, quantifies the influencing factors of three-phase load asymmetry, thereby improving the impact of three-phase imbalance on power flow.

[0012] 3. The probabilistic power flow method for asymmetrical medium-voltage three-phase distribution networks with single-phase photovoltaic access provided by this invention has practical effects, is suitable for most photovoltaic output models and is applicable to most distribution networks where the neutral point is not directly grounded, thereby improving the impact of the neutral point not being directly grounded on the power flow of the distribution network system and the risk of photovoltaic access system, and improving the stability and security of the system. Attached Figure Description

[0013] Figure 1 This is a block diagram illustrating the photovoltaic historical data prediction method of the present invention. Figure 2 This is a probability flow diagram of the point estimation method of the present invention. Detailed Implementation

[0014] In this embodiment, a probabilistic power flow flowchart for an asymmetrical medium-voltage three-phase distribution network with single-phase photovoltaic access is shown below. Figure 2 As shown. Based on load data from random variables, probabilistic power flow calculations are performed using point estimation. Based on the Gauss-Hermite algorithm, the number of calculation points and their corresponding weights are calculated in the independent standard normal variable space. The estimated points in the original random variable space are obtained using the inverse Nataf transform. For each estimated point, a set of deterministic load data is obtained, and the deterministic power flow at each estimated point is calculated using an improved Newton's method to obtain the power flow distribution. Based on the power flow distribution, the overall expectation and corresponding statistical moments are calculated. The overall probability distribution is obtained by fitting the moments using Gram-Charlier series. Photovoltaic output is predicted using historical data, and the relevant weights are obtained using the entropy weight method. The probability density function is obtained through LightGBM training and KDE fitting. Specifically, as... Figure 1 As shown, the photovoltaic forecasting method includes the following steps: Step 1: Obtain the normalized historical dataset of photovoltaic power plant power. z t | t =1, 2, … , T},in, z t Indicates the first t Power data for individual photovoltaic power plants T Indicates the number of data items; Step 2, obtain the dataset of influencing factors { h g | g =1, 2, … , t},in, h g Indicates the first g One influencing factor, t This indicates the number of influencing factors.

[0015] Step 2.1, using equation (1) to obtain h g weight : (1) In equation (1), H g For the first g The information entropy of each influencing factor is obtained through equation (2). Indicates the first g’ Information entropy of each influencing factor; This is the average of all information entropies; express 35th power; (2) In equation (2), f g For the first g The proportion of each influencing factor is obtained from equation (3); (3) Step 2.2, use equation (4) to obtain the weighted result of the first step. g Influencing factors s g : (4) Step 3, use the LightGBM algorithm to... s g Make a prediction and obtain the first g One influencing factor s g Predicted value S g .

[0016] Step 3.1: Obtain the delay error value using equation (5). f r ( u ): (5) In equation (5), T ( s g , ) is a decision tree; These are the decision tree parameters; R is the number of decision trees; Step 3.2, use equation (6) to obtain the predicted value. S g : (6) In equation (6), y This is the output value; L This is the loss function.

[0017] Step 4, using equation (7) S g After fitting, the probability density function of photovoltaic power output is obtained. f ( P PI );in, P PI This indicates the active power output of the photovoltaic system. (7) In equation (7), α Indicates the number of quantiles; z This represents the dataset before normalization. D The bandwidth can be obtained through equation (8); K The kernel function can be obtained through equation (9); (8) In equation (8), This represents the standard deviation of the estimate; (9) Based on the probability density function of photovoltaics, data for deterministic power flow are determined, and deterministic power flow results are calculated; the specific power flow diagram is as follows. Figure 2 As shown.

[0018] Taking into account the effective grounding methods of the neutral point in medium-voltage distribution networks (where the neutral point is not directly grounded), a Newton-Raphson power flow constraint equation considering the neutral point voltage is constructed. An improved Newton-Raphson method is then calculated by modifying the neutral point voltage, and the locations and weights of the corresponding integration points are used for further analysis. k The calculation of the subdeterministic power flow is performed, and the power flow calculation results and unbalance curves are obtained.

[0019] Step 5, using the probability density function f ( P PI The active power of T photovoltaic outputs on the grid. x t | t =1, 2, … , T} forms the original space variables; where, x t Indicates the first t The active power output of each photovoltaic system.

[0020] Step 6, x tAs the active load of any phase in a node of a three-phase distribution network, a power flow model of a three-phase distribution network with single-phase photovoltaic access is constructed. Step 6.1: Obtain the line node admittance matrix using equation (10); (10) In equation (10), Y Represents the nodal admittance matrix; Represents a node i With nodes j The reciprocal of the impedance matrix between them; Y igN Represented as nodes i The equivalent load admittance between each phase and the neutral point is obtained through equation (11); Y jgN Represented as nodes j Equivalent load admittance between each phase and the neutral point; Y j0 Represents a node j Earth admittance matrix; Y i0 Represents a node i The ground admittance matrix can be obtained through equation (12); (11) In equation (11), Represents a node i The admittance between phase a and the neutral point; Represents a node i The admittance between phase b and the neutral point; Represents a node i The admittance between the c-phase and the neutral point; (12) In equation (12), Represents a node i The relative admittance of a; Represents a node i The relative admittance of b; Represents a node i The c is relatively admittance.

[0021] Step 6.2, use equation (13) to obtain the node. i Access the t The active power output of the photovoltaic system x t neutral point voltage U iN : (13) In equation (13), p represents a phase corresponding to a single-phase photovoltaic system; m represents any phase other than phase p. Represents a node i The initial value of the m-phase voltage; Y ipN Represents a node i The equivalent load admittance between phase p and the neutral point; Y imN Represents a node i The equivalent load admittance between phase m and the neutral point; Y iN Represents a node i Neutral point-to-ground admittance; i =1, 2, … , n ,in, n Indicates the number of nodes.

[0022] Step 6.3, use equation (14) to obtain the node. i Unconnected x t The initial value of the neutral point voltage : (14) In equation (14), Represents a node i The initial value of the q-phase voltage; Represents a node i The equivalent load admittance between the q-phase and the neutral point.

[0023] Step 6.4, use equation (15) to obtain the node. i Active power imbalance of p-phase connected single-phase photovoltaic 、 Reactive power imbalance : (15) In equation (15), Represents a node i The p-phase voltage value; 1i is the imaginary unit; ϕ Indicates the power factor angle; i p The phase angle represents the voltage of phase p. i N This represents the voltage phase angle at the neutral point.

[0024] Step 6.5, use equation (16) to obtain the node. i Not connected x t Active power imbalance 、 Reactive power imbalance : (16) In equation (16), Represents a node i q phase and node j The electrical conductance between phases a; Represents a node i q phase and node j The susceptance between phases a; Represents a node i q phase and node j The electrical conductance between phase b; Represents a node i q phase and node j The susceptance between phase b; Represents a node i q phase and node j The electrical conductance between phases c; Represents a node i q phase and node j The susceptance between the c phases; sin i iqja Represents a node i q phase and node j The sine of the phase angle difference between phases a and b; cos i iqja Represents a node i q phase and node j The cosine of the phase angle difference between phases a; sin i iqjb Represents a node i q phase and node j The sine of the phase angle difference between phases b; cos i iqjb Represents a node i q phase and node j The cosine of the phase angle difference between phases b; sin i iqjc Represents a node i q phase and node j The sine of the phase angle difference between phases c; cos i iqjc Represents a node i q phase and node j The cosine of the phase angle difference between phases c; Represents a node i The electrical conductance between the q-phase and the neutral point; Represents a node i The susceptance between the q phase and the neutral point; sin i iqN Represents a nodei The sine of the phase angle difference between phase q and the neutral point; cos i iqN Represents a node i The cosine of the phase angle difference between phase q and the neutral point; Represents a node i The q-phase voltage; Represents a node j Phase a voltage; Represents a node j The voltage of phase b; Represents a node j c-phase voltage; U iN Represents a node i The neutral point voltage.

[0025] Step 6.6, correct the nodes using equation (17). i neutral point voltage U iN In the e Calculated values ​​under secondary power flow iteration : (17) In equation (17), Indicates the first e -1 Nodes under Trend Iteration i The load admittance value between phase q and the neutral point.

[0026] Step 6.7: Use equation (18) to obtain the location of the single-phase photovoltaic access point. p-phase voltage : (18) In equation (18), G iβ Represents a node i and β The conductivity values ​​between; B iβ Represents a node i and β The susceptance values ​​between; Represents a node i The amount of active power injected into the p-phase; Represents a node i The amount of reactive power injected in the p-phase; Indicates the location of the access point β The p-phase voltage value of the previous node.

[0027] Step 6.8: Use equation (19) to obtain the location of the access point. β Three-phase load imbalance : (19) In equation (19), U βN Access point location β The neutral point voltage is obtained from equation (20); (20) Step 7: Based on the voltage and network parameters in the three-phase distribution network power flow model, obtain the integrand using equation (21). f ( g , β ): (twenty one) In equation (21), g Indicates the integral position of the estimated point. λ represents the location of the single-phase photovoltaic grid connection point, and λ represents any one of phases a, b, and c. Represents a node i λ-phase voltage; express The conjugate of ; q represents phases a, b, and c; Represents a node j The conjugate of the q-phase voltage; Indicates the location of the photovoltaic grid connection point β The voltage conjugate; Represents a node i The conjugate of the load equivalent admittance between the q-phase and the neutral point; Represents a node i With nodes j The conjugate of the admittance matrices between them.

[0028] The location of the estimated point is obtained using equation (22). g : (twenty two) In equation (22), g k Indicates the first k The integral position of each estimated point k =0, 1, … , l ; l Indicates the estimated number of points; Indicates the first k One estimated point g k The weights; f ( g k , β ) indicates the first k One estimated point g k The integrand.

[0029] Step 8, obtain using Nataf inverse transform g k The mapping matrix on the original spatial variables { x t,k | t =1, 2, …, T};in, x t,k express g k exist x t Mapping on; Step 8.1: Use equation (23) to obtain the elements of the standard normal random vector. y t ; (twenty three) In equation (23), F t ( x t )for x t The cumulative distribution function; P and P -1 These are the standard normal cumulative distribution function and the inverse cumulative distribution function, respectively.

[0030] Step 8.2: Use equation (24) to obtain the original random variable matrix through the inverse Nataf transformation. X : (twenty four) In equation (24), V Represented as an independent standard normal vector; Represented as the inverse cumulative distribution function; P Represented as the standard normal cumulative distribution function; L 0 yes r 0 The lower triangular matrix obtained by Choleskey decomposition r 0 For a standard normal random vector Y=L 0 V The correlation coefficient matrix can be determined by solving the nonlinear equation using equation (25). r 0 ; This represents the inverse Nataf transform.

[0031] Step 8.3: Obtain the input random vector using equation (25). X Correlation coefficient matrix r ts : (25) In equation (25), s =1, 2, … , T ; r 0ts for r 0 Elements in; ( x t , x s )express x t and x t Joint probability density function; Φ2( y t , y s , r 0ts ) represents the correlation coefficient r 0ts Two-dimensional standard normal random variable y t and y s The joint distribution function; m t , s t express x t Expectation and variance; m s , s s express x s Expectation and variance; Represented as x t Inverse cumulative distribution function; Represented as x s Inverse cumulative distribution function.

[0032] Step 9, use the power flow model to... x t,k Conduct the first k The second deterministic power flow calculation yields the... t The active power output of the photovoltaic system x t The k Secondary tidal current distribution X ( t,k ); Step 10: Calculate the expected power flow distribution using equation (26). m X and υ-th order moment : (26) In equation (26), E μ Represents the mapping matrix { x t,k | t =1, 2, … , T The expected value of}; E ( X ( t,k )) indicates the first k Secondary tidal current distribution X ( t,k The mean of ) for x t The expected value.

[0033] Step 11: Fit the υth moment using Gram-Charlier series. Thus, the power flow probability distribution of the three-phase distribution network is obtained; Step 11.1, using the power flow distribution obtained from equation (27) X Standardization : (27) Step 11.2: Use equation (28) to find the probability density function of the power flow probability distribution of the three-phase distribution network. : (28) In equation (28), f ( x The probability density function that satisfies the standard normal distribution is... c γ The coefficients of the Gram-Charlier series are... The probability density function of the standard normal distribution c First derivative, c =0, 1, … , t .

[0034] In this embodiment, an electronic device includes a memory and a processor. The memory stores a program that supports the processor in executing the methods described above, and the processor is configured to execute the program stored in the memory.

[0035] In this embodiment, a computer-readable storage medium stores a computer program, which is executed by a processor to perform the steps of the above method.

Claims

1. A method for calculating probabilistic power flow in an asymmetrical medium-voltage three-phase distribution network with single-phase photovoltaic access, characterized in that, Includes the following steps: Step 1: Obtain the normalized historical dataset of photovoltaic power plant power. z t | t =1, 2, … , T },in, z t Indicates the first t Power data for individual photovoltaic power plants T Indicates the number of data items; Step 2, obtain the dataset of influencing factors { h g | g =1, 2, … , τ },in, h g Indicates the first g One influencing factor, τ Indicates the number of influencing factors; Step 2.1, using equation (1) to obtain h g weight : (1) In equation (1), H g For the first g The information entropy of each influencing factor is obtained through equation (2). Indicates the first g’ Information entropy of each influencing factor; This is the average of all information entropies; express 35th power; (2) In equation (2), f g For the first g The proportion of each influencing factor is obtained from equation (3); (3) Step 2.2, use equation (4) to obtain the weighted result of the first step. g Influencing factors s g : (4) Step 3, use the LightGBM algorithm to... s g Make a prediction and obtain the first g Influencing factors s g Predicted value S g ; Step 4, use the KDE method to... S g After fitting, the probability density function of photovoltaic power output is obtained. f ( P PI );in, P PI This indicates the active power output of the photovoltaic system. Step 5, using the probability density function f ( P PI The active power of T photovoltaic outputs on the grid. x t | t =1, 2, … , T } forms the original space variables; where, x t Indicates the first t The active power output of each photovoltaic system; Step 6, x t As the active load of any phase in a node of a three-phase distribution network, a power flow model of a three-phase distribution network with single-phase photovoltaic access is constructed. Step 7: Based on the voltage and network parameters in the three-phase distribution network power flow model, obtain the integrand using equation (5). f ( ζ , β ): (5) In equation (5), ζ Indicates the integral position of the estimated point. λ represents the location of the single-phase photovoltaic grid connection point, and λ represents any one of phases a, b, and c. Represents a node i λ-phase voltage; express The conjugate of ; q represents phases a, b, and c; Represents a node j The conjugate of the q-phase voltage; Indicates the location of the photovoltaic grid connection point β The voltage conjugate; Represents a node i The conjugate of the load equivalent admittance between the q-phase and the neutral point; Represents a node i With nodes j The conjugate of the admittance matrices between them; The location of the estimated point is obtained using equation (6). ζ : (6) In equation (6), ζ k Indicates the first k The integral position of each estimated point k =0, 1, … , l ; l Indicates the estimated number of points; Indicates the first k One estimated point ζ k The weights; f ( ζ k , β ) indicates the first k One estimated point ζ k The integrand; Step 8, obtain using Nataf inverse transform ζ k The mapping matrix on the original spatial variables { x t,k | t =1, 2, … , T };in, x t,k express ζ k exist x t Mapping on; Step 9, using the power flow model to... x t,k Proceed to the first k The second deterministic power flow calculation yields the... t The active power output of the photovoltaic system x t The k Secondary tidal current distribution X ( t,k ); Step 10: Calculate the expected power flow distribution using equation (7). μ X and υ-th order moment : (7) In equation (7), E μ Represents the mapping matrix { x t,k | t =1, 2, … , T The expected value of}; E ( X ( t,k )) indicates the first k Secondary tidal current distribution X ( t,k The mean of ) for x t Expected value; Step 11: Fit the υth moment using Gram-Charlier series. The power flow probability distribution of the three-phase distribution network is obtained.

2. The method for calculating probabilistic power flow in an asymmetrical medium-voltage three-phase distribution network with single-phase photovoltaic access according to claim 1, characterized in that, Step 6 includes: Step 6.1, use equation (8) to obtain the node. i Access the t The active power output of the photovoltaic system x t neutral point voltage U iN : (8) In equation (8), p represents a phase corresponding to a single-phase photovoltaic system; m represents any phase other than phase p. Represents a node i The initial value of the m-phase voltage; Y ipN Represents a node i The equivalent load admittance between phase p and the neutral point; Y imN Represents a node i The equivalent load admittance between phase m and the neutral point; Y iN Represents a node i Neutral point-to-ground admittance; i =1, 2, … , n ,in, n Indicates the number of nodes; Step 6.2, use equation (9) to obtain the node. i Unconnected x t The initial value of the neutral point voltage : (9) In equation (9), Represents a node i The initial value of the q-phase voltage; Represents a node i The equivalent load admittance between phase q and the neutral point; Step 6.3, obtain the node using equation (10). i Active power imbalance of p-phase connected single-phase photovoltaic 、 Reactive power imbalance : (10) In equation (10), Represents a node i The p-phase voltage value; 1i is the imaginary unit; ϕ Indicates the power factor angle; θ p The phase angle represents the voltage of phase p. θ N This represents the voltage phase angle at the neutral point; Step 6.4, use equation (11) to obtain the node. i Not connected x t Active power imbalance 、 Reactive power imbalance : (11) In equation (11), Represents a node i q phase and node j The electrical conductance between phases a; Represents a node i q phase and node j The susceptance between phases a; Represents a node i q phase and node j The electrical conductance between phase b; Represents a node i q phase and node j The susceptance between phase b; Represents a node i q phase and node j The electrical conductance between phases c; Represents a node i q phase and node j The susceptance between the c phases; sin θ iqja Represents a node i q phase and node j The sine of the phase angle difference between phases a and b; cos θ iqja Represents a node i q phase and node j The cosine of the phase angle difference between phases a; sin θ iqjb Represents a node i q phase and node j The sine of the phase angle difference between phases b; cos θ iqjb Represents a node i q phase and node j The cosine of the phase angle difference between phases b; sin θ iqjc Represents a node i q phase and node j The sine of the phase angle difference between phases c; cos θ iqjc Represents a node i q phase and node j The cosine of the phase angle difference between phases c; Represents a node i The electrical conductance between the q-phase and the neutral point; Represents a node i The susceptance between the q phase and the neutral point; sin θ iqN Represents a node i The sine of the phase angle difference between phase q and the neutral point; cos θ iqN Represents a node i The cosine of the phase angle difference between phase q and the neutral point; Represents a node i The q-phase voltage; Represents a node j Phase a voltage; Represents a node j The voltage of phase b; Represents a node j c-phase voltage; U iN Represents a node i The neutral point voltage; Step 6.5, correct the nodes using equation (12). i neutral point voltage U iN In the ε Calculated values ​​under secondary power flow iteration : (12) In equation (12), Indicates the first ε -1 Nodes under Trend Iteration i The load admittance between phase q and the neutral point; Step 6.6: Use equation (13) to obtain the location of the single-phase photovoltaic access point. p-phase voltage : (13) In equation (13), G iβ Represents a node i and β The conductivity values ​​between; B iβ Represents a node i and β The susceptance values ​​between; Represents a node i The amount of active power injected into the p-phase; Represents a node i The amount of reactive power injected into the p-phase; Indicates the location of the access point β The p-phase voltage value of the previous node; Step 6.7: Use equation (14) to obtain the location of the access point. β Three-phase load imbalance : (14) In equation (14), U βN Access point location β The neutral point voltage is obtained from equation (15); (15)。 3. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store a program that supports the processor in executing the probabilistic power flow calculation method for asymmetrical medium-voltage three-phase distribution networks as described in claim 1 or 2, and the processor is configured to execute the program stored in the memory.

4. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is run by the processor, it executes the steps of the probabilistic power flow calculation method for asymmetrical medium-voltage three-phase distribution networks as described in claim 1 or 2.

Images