Probabilistic load flow calculation method and system based on charging station node fault analysis

Abstract

The invention discloses a probabilistic load flow calculation method and system based on charging station node fault analysis, and the method comprises the steps: firstly, building a road network model and a power distribution network model based on a graph theory, considering the coupling relation between the road network model and the power distribution network model, and building a road-electricity coupling model; and then, constructing an electric vehicle trip chain model based on a Markov chain, and on this basis, combining a road-electricity coupling model and considering the charging psychology of a user, providing an electric vehicle load real-time dynamic response description method under an N-1 fault, especially a probabilistic load flow calculation method based on charging station node fault analysis. The method is used for simulating an electric vehicle load curve after a charging station node fault occurs in a power distribution network. And finally, carrying out probabilistic load flow calculation by adopting a Monte Carlo method, analyzing the secondary influence of the time-space characteristics of the electric vehicle on the load flow of the power distribution network under an N-1 fault, particularly a charging station node fault, by utilizing a load flow calculation result, and constructing a voltage evaluation index to evaluate the importance of the charging station node.

Description

Technical Field

[0001] The present application belongs to the technical field of probabilistic power flow calculation, and in particular, relates to a probabilistic power flow calculation method and system based on charging station node fault analysis. Background Art

[0002] With the rapid development of new energy technologies, the penetration rate of distributed generation (DG) and electric vehicles (EV) in distribution networks is gradually increasing. These uncertainties will affect the stable operation of distribution networks. As a special type of load in distribution networks, EVs have dual uncertainties in time and space. If a large number of EVs with disordered charging are connected to the grid, it will inevitably bring huge challenges to the reliable operation of distribution networks. As the penetration rate of new energy in distribution networks gradually increases, the operating conditions of distribution networks become more complex, the probability of node failures gradually increases, and thus the probability of charging station node failures also gradually increases. However, in the analysis and research of the operating conditions of existing distribution networks, there is almost no existing technology to analyze the stability and reliability of the entire system in this case through the failure of charging station nodes.

[0003] For example, the invention patent with publication number CN116485014 A discloses a method for predicting the load of orderly charging of electric vehicles under information coupling. It provides its own real-time status parameters by building a "vehicle-road-network-station" information coupling model to accurately calculate the mileage, time consumed, charging equipment utilization rate and distribution network load fluctuation required for the electric vehicle to go to each charging station when fast charging is needed at the current moment. The car owner goes to the charging station for fast charging through an orderly charging strategy, thereby alleviating the poor charging experience of users, traffic network congestion, low utilization rate of charging station equipment and increased peak-to-valley difference of distribution network load caused by disorderly charging of electric vehicles; the load of the electric vehicle on the day of orderly charging is obtained through the load prediction model, and the predicted load is more realistic and available. This method is of great significance to the reasonable scheduling and planning of orderly charging of electric vehicles and access to the distribution network for electric vehicle loads, but it still does not take into account the impact of charging station node failures on the distribution network trend. Summary of the invention

[0004] Purpose of the invention: In view of the problem that there is insufficient analysis on the impact of EV transfer characteristics on distribution network power flow after a charging station node failure in the current distribution system, the present invention provides a probabilistic power flow calculation method based on charging station node failure analysis. The present invention also provides a probabilistic power flow calculation system based on charging station node failure analysis.

[0005] Technical solution: According to a first aspect of the present invention, a probabilistic power flow calculation method based on charging station node fault analysis is provided, the method comprising:

[0006] Based on graph theory, a road network model and a distribution network model are established. Taking into account the coupling relationship between the two, a road-electricity coupling model is established to determine each charging station node in the distribution network model.

[0007] Establish a travel demand model that takes into account the temporal and spatial distribution characteristics of electric vehicles and construct a charging demand model, thereby constructing a charging load model that takes into account the temporal and spatial distribution characteristics of electric vehicles;

[0008] Calculating the probability power flow based on the charging load model, the basic load model and the distributed power supply probability model, thereby obtaining the charging station node voltage value and the line power value;

[0009] The static voltage stability index is used to measure the impact of charging station node failure on the distribution network flow.

[0010] Further, including:

[0011] The distribution network model is a graphical representation model of the topological structure and parameters of the power system. The distribution network structure includes a collection of nodes and branches, and the nodes only include load charging stations and generators. The distribution network structure is represented as follows:

[0012]

[0013] Among them, G g Represents the distribution network topology; V g represents the node set of charging stations in the distribution network; E g represents the set of distribution network lines; ψ g represents the adjacency matrix of the distribution network; v i g 、v j g represents the charging station node of the distribution network; m represents the number of charging station nodes; i and j represent variables, with a value range of 1 to m; d ij g Lines between charging station nodes in the distribution network; i 、x i 、c i , P i They respectively represent the resistance, reactance, susceptance and line limit transmission power in the distribution network.

[0014] Further, including:

[0015] The construction of a charging load model for electric vehicles with temporal and spatial distribution characteristics includes:

[0016] The electric vehicle charging load model consists of a travel demand model and a charging behavior model. The travel demand model is the basic module, which calculates the temporal and spatial distribution of the driving path and charging demand of electric vehicles by combining the road network model, the travel chain of electric vehicles, the time chain characteristic quantity and the space chain characteristic quantity.

[0017] The core of the travel demand model is to plan the shortest path of electric vehicles according to Dijkstra theory, and combine the travel chain characteristics of electric vehicles to describe their dynamic changes in time and spatial transfer within a day. In the above, the travel chain refers to the complete process of an electric vehicle starting from the starting point, passing through each travel destination one by one in sequence according to the itinerary, and finally arriving at the end of the trip; the time chain characteristic quantity refers to the dynamic changes in the time of electric vehicles traveling within a day, including the time of arrival and departure from the destination, the driving time and the parking time at various destinations; the space chain characteristic quantity refers to the spatial transfer of electric vehicles traveling within a day, including the purpose of travel and the mileage of a single trip of the electric vehicle.

[0018] The charging behavior model is based on the user's charging psychology and the state of charge of the electric vehicle to obtain the charging demand of a single electric vehicle at the charging station node connected to the distribution network. The user's charging psychology is used to reflect the charging decision of each electric vehicle: at the beginning of each trip, the user will judge the final capacity of the trip, and the remaining battery capacity when the trip is completed can be described as:

[0019] S(t end )=S(t ini )-Dis·ω car ;

[0020] Among them, S(t end ) and S(t ini ) represent the capacity values ​​of the electric vehicle at the time of reaching the destination and the time of departure respectively; Dis represents the distance formed; w car Indicates the power consumption per unit distance traveled by an electric vehicle.

[0021] At this time, we use a random number r between 0 and 1 to represent the user's charging psychology. If the random number r is greater than S(t end ) or S(t end ) is less than 20% of the maximum battery capacity, the user chooses to charge; if the random number r is less than S(t end ), and S(t end ) is greater than 20% of the maximum battery capacity, charging is not selected. It can be described as follows:

[0022]

[0023] Wherein, μ represents the user's charging status. If μ=1, it means charging is selected; if μ=0, it means not charging; S max Indicates the maximum capacity of an electric vehicle battery.

[0024] Further, including:

[0025] If the user chooses to charge, the travel route will be replanned, that is, through the road-electricity coupling model, the Dijkstra theory is used to generate a new path through the charging station node. The charging demand of a single electric vehicle at the grid access node b is expressed as:

[0026]

[0027] Where t represents the time; P(t) represents the charging demand of a single electric vehicle at time t, t0 represents the charging start time; T c Indicates charging time; P b (t) represents the charging demand of the electric vehicle connected to the charging station node b in the distribution network at time t; b represents the charging station node.

[0028] In the charging vehicle probability load model, the electric vehicle charging load approximately follows the normal distribution. c The probability density function of the normal distribution is shown in the formula.

[0029]

[0030] In the formula, p(P c ) represents the probability density function of the electric vehicle charging load; P c Indicates the charging load of electric vehicles; μ c and σ c are the expected value and standard deviation of the active power demand of a single electric vehicle at a certain moment; N(·) represents the normal distribution function.

[0031] By integrating the travel demand model and the charging behavior model, an overall framework for charging load prediction is formed.

[0032] Further, including:

[0033] If a charging station node fails, the corresponding road network node will not be considered when the Dijkstra theory is used to generate a new path through the charging station node. For example, assuming that both grid nodes 3' and 5' are grid charging station nodes, their corresponding road network nodes are 2 and 4 respectively, and the original user's optimal road network charging path is 1→2→6, 1 is the road network node at the departure time, and 6 is the trip end node. If the grid node 3' fails, the road network node 2 will not be considered when the Dijkstra theory is used to generate a new path through the charging station node, and the new path will be 1→4→6. If the faulty charging station node is repaired, it will be considered.

[0034] Further, including:

[0035] The method is based on calculating the probability power flow based on the charging load model, the basic load model and the distributed power supply probability model.

[0036] The basic load refers to the long-term stable load in the distribution network, and the corresponding load model is expressed as follows:

[0037]

[0038] Among them, f(P L ) represents the base load active power P L The probability density distribution function of L ) is the basic load reactive power Q L The probability density distribution function of L and Q L are respectively the active power and reactive power of the base load; μ PL and μ QL They are active power P L and reactive power Q L The mean of PL and σ QL are the standard deviations of active power and reactive power respectively; σ 2 PL and σ 2 QL are the variances of active power and reactive power, respectively; exp[·] represents an exponential function with the natural constant e as the base.

[0039] The distributed power supply probability model includes a wind power output probability model and a photovoltaic output probability model. The wind power output model in the distributed power supply model is characterized by Weibull distribution, and the photovoltaic output model is characterized by Beta distribution. The wind power and photovoltaic random variables are respectively characterized by K distribution. wt and K pv The probability density functions of the two are shown as follows:

[0040]

[0041] in, represents the probability density function of wind power output that conforms to the Weibull distribution; υ represents the wind speed; k is the scale parameter and shape parameter of the Weibull distribution; exp[·] represents the exponential function with the natural constant e as the base;

[0042]

[0043] Among them, F pv (s, α, β) represents the probability density function of photovoltaic output that conforms to the Beta distribution; s represents the light intensity; α and β are the shape parameters of the Beta distribution; Γ(·) represents the Gamma function.

[0044] The static voltage stability index is used to measure the impact of a charging station node failure on the power flow of the distribution network, including: using the obtained node voltage value to represent the static voltage stability index, which is expressed as:

[0045]

[0046] Among them, L j It represents the static voltage stability index corresponding to node j, using L j The voltage stability is judged by the distance relative to the critical point 0. The farther the distance, the better the voltage stability. Z ij Represents the impedance of branch ij, R ij Represents the resistance of branch ij, R ij represents the resistance of branch ij, X ij Represents the reactance of branch ij; S j represents the apparent power of node j, P j represents the active power of node j, Q j represents the reactive power of node j; U i It represents the voltage amplitude of node i. The importance of the electric vehicle charging station node refers to the impact of the failure of the electric vehicle charging station node on the distribution network flow quantified by the evaluation index. The greater the impact, the more important the electric vehicle charging station node is.

[0047] Probabilistic power flow calculation refers to bringing the basic load values ​​of each node at each time output by the probabilistic load model of charging vehicles, the basic load model and the probabilistic model of distributed power sources, the charging load values ​​of the charging station nodes and the output values ​​of the distributed power sources into the power system power flow calculation to obtain the voltage and power values ​​of each node. The present invention uses the Newton-Raphson power flow calculation method in Matpower to perform probabilistic power flow calculation. Matpower is a toolkit for calculating power flow calculation. The Newton-Raphson power flow calculation method is a mathematical method.

[0048] Another aspect of the present invention provides a probabilistic power flow calculation system based on distribution network N-1 fault analysis, the system comprising:

[0049] The distribution network model building module is used to establish the road network model and the distribution network model based on graph theory, and consider the coupling relationship between the two to establish the road-power coupling model and determine the various charging station nodes in the distribution network model;

[0050] The charging load model building module is used to establish a travel demand model and a charging demand model that take into account the temporal and spatial distribution characteristics of electric vehicles, thereby building a charging load model that takes into account the temporal and spatial distribution characteristics of electric vehicles;

[0051] A probability power flow calculation module, used to calculate the probability power flow based on the charging load model, the basic load model and the distributed power supply probability model, so as to obtain the charging station node voltage value and the line power value;

[0052] The characterization module is used to use a static voltage stability index to measure the impact of a charging station node failure on the distribution network flow.

[0053] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0054] This application constructs an electric vehicle travel chain model based on the Markov chain. On this basis, combined with the road-electricity coupling model and considering the user's charging psychology, a real-time dynamic response characterization method of electric vehicle load under charging station node failure is proposed. The proposed method studies the spatiotemporal transfer characteristics of electric vehicles under charging station node failure. In terms of the user's psychological choice of charging, if the user chooses to charge, the travel route will be replanned, that is, through the previously constructed road-electricity coupling model, the Dijkstra theory is used to generate a new path through the charging station node, so that the user can charge the electric vehicle in time.

[0055] The present invention considers the electric vehicle load model, distributed power supply model and basic load model input system of charging station node failure to perform probabilistic power flow calculation, proposes a graph theory-based probabilistic power flow calculation method for charging station nodes of distribution network containing large-scale electric vehicles, and uses the power flow calculation results to analyze the secondary impact of the spatiotemporal characteristics of electric vehicles on the distribution network power flow under charging station node failure.

[0056] The present application constructs a static voltage stability index and compares the voltage stability indexes of different charging station nodes after failures to evaluate the importance of electric vehicle charging station nodes. The importance of electric vehicle charging station nodes refers to the impact of failures of electric vehicle charging station nodes on the distribution network flow quantified by evaluation indicators. The greater the impact, the more important the electric vehicle charging station node. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the conventional technology, the drawings required for use in the embodiments or the conventional technology descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0058] Figure 1 This is a flow chart of power flow calculation according to an embodiment of the present invention;

[0059] Figure 2 A user charging psychological simulation process according to an embodiment of the present invention;

[0060] Figure 3 A road network topology diagram according to an embodiment of the present invention;

[0061] Figure 4 A system structure diagram according to an embodiment of the present invention;

[0062] Figure 5 The electric vehicle load simulation result according to the embodiment of the present invention;

[0063] Figure 6 It is the voltage data of the node after the N-1 fault according to the embodiment of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] First, in order to facilitate understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. Embodiments of the present application are given in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used herein includes any and all couplings of one or more related listed items.

[0067] It can be understood that the terms "first", "second", etc. used in the present application can be used in this article to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0068] Some terms involved in this application are explained below to facilitate understanding of this application:

[0069] Step 1: Based on graph theory, a road network model and a distribution network model are established, and the coupling relationship between the two is considered to establish a road-electricity coupling model. An electric vehicle travel chain model is constructed based on the Markov chain. On this basis, a real-time dynamic response characterization method of electric vehicle load under N-1 fault is proposed by combining the road-electricity coupling model and considering the user's charging psychology. It is used to simulate the transfer characteristics of electric vehicles under N-1 faults, and then the electric vehicle load curve is obtained;

[0070] Step 2: Monte Carlo method is used to perform probabilistic power flow calculation. A graph-theoretic N-1 probabilistic power flow calculation method for distribution networks containing large-scale electric vehicles is proposed. The power flow calculation results are used to analyze the secondary impact of the spatiotemporal characteristics of electric vehicles on the power flow of the distribution network under N-1 faults.

[0071] Step 3: Construct a static voltage stability index based on the voltage quantity under N-1 faults to evaluate the importance of the EV charging station node.

[0072] The flow calculation process of the present invention is as follows: Figure 1 As shown. Specifically including the following steps:

[0073] 1. Real-time dynamic response characterization method of electric vehicle load under N-1 fault

[0074] Graph theory is a branch of combinatorics, which is usually used to analyze the pairwise relationship between research objects. The basic principle of graph theory can be described as using points and connecting lines to show the relationship and structure between research objects. The present invention uses graph theory to model the road network and the power grid respectively.

[0075] (1) Road network model

[0076] The road network model refers to the connection state model between each road network node in the traffic road network topology structure. In the road network topology diagram, the line segment represents the road and the point represents the road network node. The road network model is shown in formula (1). Figure 3 shown.

[0077]

[0078] In formula (1), G r Represents the road network topology; where V r Represents a road node set; v i r 、v j r represents the road node; n represents the number of road nodes; E r Represents a set of road segments; d ij r Indicates the distance between road nodes; i and j represent variables, with a value range of 1 to n; D r represents the adjacency matrix of the road network, e ij r It represents the length of the road section, and its calculation formula is shown in formula (2).

[0079]

[0080] In formula (2), e ij r 、e ji r Indicates the length of the road section; d ij r Represents the distance between road nodes; v i r 、v j r represents a road node; i and j represent variables, ranging from 1 to n; inf represents infinity. According to formula (1)-formula (2), D r It can be expressed as shown in formula (3).

[0081]

[0082] In formula (3), D r Represents the adjacency matrix of the road network; inf is infinity; d 12 and d 21are the distances between road nodes 1 and 2 in the road network; d 14 and d 41 are the distances between road nodes 1 and 4 in the road network; d 23 and d 32 are the distances between road nodes 2 and 3 in the road network; d 25 and d 52 are the distances between road node 2 and road node 5 in the road network; d 35 and d 53 are the distances between road nodes 3 and 5 in the road network; d 45 and d 54 Both are the distances between road node 4 and road node 5 in the road network.

[0083] (2) Distribution network model

[0084] The distribution network model is a graphical representation model of the topological structure and parameters of the power system. The present invention does not include switches when modeling the distribution network, and simplifies the distribution network structure into a collection of nodes and branches, and the nodes only include loads and generators. The distribution network structure is shown in formula (4).

[0085]

[0086] In formula (4), G g Represents the distribution network topology; V g represents the set of distribution network nodes; E g represents the set of distribution network lines; ψ g represents the adjacency matrix of the distribution network; v i g 、v j g represents the distribution network node; m represents the number of grid nodes; i and j represent variables, with a value range of 1 to m; d ij g Lines between distribution network nodes; i 、x i 、c i , P i They respectively represent the resistance, reactance, susceptance and line limit transmission power in the distribution network.

[0087] (3) Circuit-electric coupling model

[0088] A coupling relationship refers to an interaction and mutual influence between two things. The road-electricity coupling model is a model that represents the interaction between the road network and the distribution network. The road-electricity coupling model mainly includes three parts: the road network model, the distribution network model, and the coupling relationship between the topological structures of the two, as shown in formula (5).

[0089]

[0090] In formula (5), G represents the circuit-to-electric coupling topology; G r Represents the road network topology; G g Represents the distribution network topology; E r-g represents the network coupling edge set; d ij represents the network coupling edge; v i r Represents a road node; v j g Grid node; V r Represents a road node set; V g Represents a collection of distribution network nodes.

[0091] Markov theory is used to describe the random process of EVs transferring from the previous destination to the next destination. According to Markov theory, the transfer process of random events has no aftereffect, and the transfer probability is only determined by the state at the previous moment. The expression of Markov chain is shown in formula (6).

[0092] P(E i →E j )=P(E i |E j )=P ij (6)

[0093] In formula (6), E i and E j Respectively represent the state of the random event at the current moment and the next moment; P(E i →E j )、P(E i |E j ) and P ij All are random events from E i Transfer to E j probability.

[0094] The electric vehicle charging load model consists of two parts: the first part is the EV travel demand model; the second part is the EV charging demand model.

[0095] The spatiotemporal characteristics of EV are described by travel chains. Based on graph theory, a travel demand model considering the spatiotemporal distribution characteristics of EV is established. A travel chain refers to the complete process of an electric vehicle starting from a starting point, passing through each travel destination one by one in a time sequence according to the itinerary, and finally arriving at the end of the trip. The travel chain contains several characteristic quantities, which can be divided into two categories: time chain characteristic quantities and space chain characteristic quantities. The two can jointly describe the spatiotemporal distribution characteristics of EV. The time chain characteristic quantity refers to the time dynamic change of EV travel within a day, which is mainly composed of the time of arrival and departure from the destination, the driving time and the parking time at various destinations; the space chain characteristic quantity refers to the spatial transfer of EV travel within a day, which is mainly composed of the travel purpose and the mileage of a single EV trip.

[0096] Assume that EVs have five travel purposes: residential area (H), business office area (W), social and entertainment area (SR), commercial shopping and dining area (SE) and area with other uses (O). The Markov theory is used to construct the travel chain. Based on graph theory, considering road network constraints and factors such as road congestion, the Dijsktra theory is used to plan the shortest driving path. Some known quantities can be obtained based on statistical data, and the unknown quantities can be obtained based on the relationship between the spatial chain feature quantity and the time chain feature quantity. Finally, all the feature quantities in the travel chain can be obtained.

[0097] The degree of road congestion is described by the number of cars traveling on the path between two adjacent road network nodes in the road network, which is inversely proportional to the speed of cars on this road. The specific expression is shown in formula (7):

[0098]

[0099] In formula (7), V i,j (t) represents the formal speed of the i-th EV on the j-th path at time t; L path,j represents the length of the jth path; N car,j (t) represents the number of EVs on the jth path at time t.

[0100] Dijsktra algorithm is a commonly used algorithm for calculating the shortest path between nodes in a graph. According to Dijsktra algorithm, the distance between road nodes can be determined. The algorithm process combines graph theory ideas, as follows.

[0101] In a weighted graph, each vertex is assigned a specific "identifier". The "temporary identifier" T represents the upper limit of the currently estimated shortest path length from the starting vertex to the vertex; the "fixed identifier" P represents the confirmed shortest path length from the starting vertex to the vertex, which is a finalized value.

[0102] (1) Vertex v1 is labeled P, label d(v1) = 0, vertex vj (j=2,3,…,n) is labeled T, and the label d(v j )=l 1,j ;

[0103] (2) Find the minimum value among all the numbers T, that is: if d(v j0 )=l 1,j0 , v j0 The label of the vertex v is changed from T to P, and then the T labels of other nodes labeled T are recalculated; select vertex v j T label d(v j0 )+l j0,j The smaller of j The new T label.

[0104] (3) Repeat the above process until the label of the target vertex becomes P.

[0105] Figure 2 The user's charging psychology simulation process diagram, the charging behavior model is based on the user's charging psychology combined with the electric vehicle's state of charge to obtain the charging demand of a single electric vehicle at the charging station node connected to the distribution network. The user's charging psychology is used to reflect the charging decision of each electric vehicle: at the beginning of each trip, the user will judge the final capacity of the trip, and the remaining battery capacity when the trip is completed can be described as:

[0106] S(t end )=S(t ini )-Dis·ω car (8)

[0107] Among them, S(t end ) and S(t ini ) represent the capacity values ​​of the electric vehicle at the time of reaching the destination and the time of departure respectively; Dis represents the distance formed; w car Indicates the power consumption per unit distance traveled by an electric vehicle.

[0108] At this time, a random number r between 0 and 1 is used to represent the user's charging psychology. If the random number r is greater than S(t end ) or S(t end ) is less than 20% of the maximum battery capacity, the user chooses to charge; if the random number r is less than S(t end ), and S(t end ) is greater than 20% of the maximum battery capacity, charging is not selected. It can be described as follows:

[0109]

[0110] Wherein, μ represents the user's charging status. If μ=1, it means charging is selected; if μ=0, it means not charging; S maxIndicates the maximum capacity of an electric vehicle battery.

[0111] The time when the electric vehicle arrives at the destination is t end The calculation formula is as follows:

[0112]

[0113] In formula (10), t end and t ini They represent the time when the electric vehicle reaches the destination and the time when it departs; J represents the set of paths that the charging vehicle i passes through; L path,j represents the length of path j; V i,j Represents the driving speed of charging car i on path j.

[0114] In addition, if a charging station node fails, the corresponding road network node will not be considered when the Dijkstra theory is used to generate a new path through the charging station node. For example, assuming that both grid nodes 3' and 5' are grid charging station nodes, their corresponding road network nodes are 2 and 4, respectively. The original user's optimal road network charging path is 1→2→6, 1 is the road network node at the departure time, and 6 is the trip end node. If the grid node 3' fails, the road network node 2 will not be considered when the Dijkstra theory is used to generate a new path through the charging station node, and the new path will be 1→4→6. If the faulty charging station node is repaired, it will be considered.

[0115] N-1 failure means that when a component fails in the normal operation of the power system, it should be ensured that other equipment will not be implicated and fail, and the system as a whole can still maintain stable operation and normal power supply. The present invention refers to a failure at a node in the distribution network. In the present invention, N-1 failure is set to a failure and disconnection of a charging station node.

[0116] After completing the above simulation, the charging demand of a single EV at the grid access node b can be obtained as shown in formula (11).

[0117]

[0118] In formula (11), t represents the time; P c (t) represents the charging demand of a single EV at time t, t0 represents the charging start time; T c Indicates charging time; P b (t) represents the charging demand of EV connected to grid node b at time t; b represents the grid node.

[0119] After completing the simulation of EV travel demand and charging demand, a charging load model that considers the temporal and spatial distribution characteristics of electric vehicles can be constructed.

[0120] In the charging vehicle probability load model, the electric vehicle charging load approximately follows the normal distribution. c The probability density function of the normal distribution is shown in formula (12).

[0121]

[0122] In formula (12), p(P c ) represents the probability density function of the electric vehicle charging load; P c Indicates the charging load of electric vehicles; μ c and σ c are the expected value and standard deviation of the active power demand of a single electric vehicle at a certain moment; N(·) represents the normal distribution function.

[0123] 2. N-1 probabilistic power flow calculation method for distribution network with large-scale electric vehicles based on graph theory

[0124] Probabilistic power flow calculation generally refers to the process of inputting random variables such as wind power output, photovoltaic output, and random load into the power system and then performing deterministic power flow calculation as shown in formula (13), and finally calculating the probability distribution characteristics of output variables such as node voltage and line power.

[0125] K(U,δ,…)=G(P,Q) (13)

[0126] In formula (13), K(·) represents the output variable matrix; G(·) represents the power flow equation; U represents the node voltage amplitude; δ represents the node voltage phase angle; P represents active power; and Q represents reactive power.

[0127] Monte Carlo method is a kind of numerical solution method based on mathematical statistics and probability theory, and is one of the basic methods for probabilistic power flow calculation. Monte Carlo method is widely used in probabilistic power flow calculation because of its easy-to-understand principle, simple programming and high calculation accuracy.

[0128] The probabilistic power flow calculation model consists of a probabilistic load model for charging vehicles, a basic load model, and distributed power sources. The probabilistic load model for charging vehicles is characterized by a normal distribution as shown in formula (12). Probabilistic power flow calculation refers to bringing the basic load values ​​of each node at each time output by the probabilistic load model for charging vehicles, the basic load model, and the probabilistic model for distributed power sources, the charging load values ​​of the charging station nodes, and the output values ​​of the distributed power sources into the power system power flow calculation to obtain the voltage and power values ​​of each node. In the present invention, the Newton-Raphson power flow calculation method in Matpower is used to perform probabilistic power flow calculation. Matpower is a toolkit for calculating power flow calculations. The Newton-Raphson power flow calculation method is a mathematical method.

[0129] The basic load refers to the long-term stable load in the distribution network, and is characterized by the normal distribution shown in formula (14).

[0130]

[0131] In formula (14), f(P L ) represents the base load active power P L The probability density distribution function of L ) is the basic load reactive power Q L The probability density distribution function of L and Q L are respectively the active power and reactive power of the base load; μ PL and μ QL They are active power P L and reactive power Q L The mean of PL and σ QL are the standard deviations of active power and reactive power respectively; σ 2 PL and σ 2 QL are the variances of active power and reactive power, respectively; exp[·] represents an exponential function with the natural constant e as the base.

[0132] The distributed power generation probability model includes the wind power output probability model and the photovoltaic output probability model. The wind power output model in the distributed power generation model is characterized by Weibull distribution, and the photovoltaic output model is characterized by Beta distribution. The wind power and photovoltaic random variables are respectively characterized by K distribution. wt and K pv The probability density functions of the two are expressed as shown in formula (15) and formula (16) respectively.

[0133]

[0134] In formula (15), represents the probability density function of wind power output that conforms to the Weibull distribution; υ represents the wind speed; k is the scale parameter and shape parameter of the Weibull distribution; exp[·] represents an exponential function with the natural constant e as the base.

[0135]

[0136] In formula (16), F pv (s, α, β) represents the probability density function of photovoltaic output that conforms to the Beta distribution; s represents the light intensity; α and β are the shape parameters of the Beta distribution; Γ(·) represents the Gamma function.

[0137] The probabilistic power flow calculation method based on charging station node fault analysis of the present invention is actually a N-1 probabilistic power flow calculation method for a distribution network containing large-scale electric vehicles based on graph theory. It combines graph theory, probability theory and power flow calculation methods, and aims to evaluate the power flow distribution and uncertainty of a distribution network containing large-scale electric vehicles under N-1 fault conditions.

[0138] The power flow calculation results include node voltage values ​​and line power values.

[0139] The secondary impact of the spatiotemporal characteristics of electric vehicles on the distribution network flow under N-1 faults refers to the premise that the spatiotemporal uncertainty of electric vehicle loads has an impact on the distribution network flow. After an N-1 fault occurs in the system, the electric vehicles experience spatiotemporal transfer again due to the fault, which in turn causes a secondary impact on the distribution network flow.

[0140] 3. Determine the importance of charging station nodes

[0141] The static voltage stability index is used to measure the degree of voltage collapse and characterize the severity of the system instability. The calculation formula of the static voltage stability index is shown in formula (15).

[0142]

[0143] In formula (15), L j It represents the static voltage stability index corresponding to node j, using L j The voltage stability is judged by the distance relative to the critical point 0. The farther the distance, the better the voltage stability. Z ij Represents the impedance of branch ij, R ij Represents the resistance of branch ij, R ij represents the resistance of branch ij, X ij Represents the reactance of branch ij; S j represents the apparent power of node j, P j represents the active power of node j, Q j represents the reactive power of node j; U i Represents the voltage amplitude of node i.

[0144] The importance of an electric vehicle charging station node refers to the impact of a failure of an electric vehicle charging station node on the power flow of the distribution network quantified through evaluation indicators. The greater the impact, the more important the electric vehicle charging station node is.

[0145] The technical effects of the present invention will be further illustrated by examples below:

[0146] The present invention uses the IEEE-33 node distribution network as the basis for example analysis. Figure 4As shown in Figure 1. According to the different uses of loads, the road areas covered by the power grid are divided into five categories: residential areas (H), business office areas (W), social and entertainment areas (SR), commercial shopping and dining areas (SE), and areas with other uses (O). A charging station is set in each area. The corresponding relationship between the charging station and the power grid node is shown in Figure 1. Figure 4 shown.

[0147] The number of electric vehicles is 2500. The present invention names the grid node corresponding to the charging station as the charging station node. Assuming that the charging station node in area H fails during the daytime from 10:30 to 11:00, the transfer characteristics of the electric vehicle load are simulated, and the following is obtained: Figure 5 The EV load curve shown is given by Figure 5 It can be seen that during the fault period, the EV load fluctuation increased significantly, and at the moment of the node failure, the overall load showed a downward trend, proving that the point was instantly lost at this time. After the fault was repaired, the EV load showed an increasing trend, showing the characteristics of load transfer.

[0148] The Monte Carlo method is used to calculate the probability power flow, and the voltage data of each node at each time after N-1 fault is obtained as follows: Figure 6 As shown in Figure 1, the voltage data after the node fault is compared with the power flow calculation results obtained under normal system operation, and the power flow calculation results with or without fault are used to analyze the secondary impact of the spatiotemporal characteristics of electric vehicles on the distribution network power flow under N-1 fault. Then, it is assumed that the charging station loads in the W area, SR area, SE area, and O area fail in sequence, and then the static voltage stability index values ​​of the typical nodes of the system after the fault are calculated, as shown in Table 1.

[0149] Table 1 10: Static voltage stability indicators of some 45 nodes

[0150]

[0151] From the analysis of Table 1, it can be seen that when the charging station node in area W, that is, node 13, fails, the overall voltage stability index of the system node is closer to 0, indicating that the system node is more unstable in this case. Therefore, the charging station node in area W is relatively important, and the voltage weak point under the current working condition is obtained through the voltage stability index.

[0152] A second aspect of the present invention provides a probabilistic power flow calculation system based on charging station node fault analysis, the system comprising:

[0153] The distribution network model building module is used to establish the road network model and the distribution network model based on graph theory, and consider the coupling relationship between the two to establish the road-power coupling model and determine the various charging station nodes in the distribution network model;

[0154] The charging load model building module is used to establish a travel demand model and a charging demand model that take into account the temporal and spatial distribution characteristics of electric vehicles, thereby building a charging load model that takes into account the temporal and spatial distribution characteristics of electric vehicles;

[0155] A probability power flow calculation module, used to calculate the probability power flow based on the charging load model, the basic load model and the distributed power supply probability model, so as to obtain the charging station node voltage value and the line power value;

[0156] The characterization module is used to use a static voltage stability index to measure the impact of a charging station node failure on the distribution network flow.

[0157] A third aspect of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the aforementioned probabilistic power flow calculation method based on charging station node fault analysis when executing the computer program.

[0158] A fourth aspect of the present application provides a computer program, which, when executed by a processor, implements the steps of the aforementioned probabilistic power flow calculation method based on charging station node fault analysis.

[0159] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0160] The various embodiments in the present disclosure are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0161] The protection scope of the present disclosure is not limited to the above-mentioned embodiments. Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the scope and spirit of the present disclosure. If these changes and modifications fall within the scope of the claims of the present disclosure and their equivalents, the intention of the present disclosure also includes these changes and modifications.

Claims

1. A probabilistic power flow calculation method based on charging station node failure analysis, characterized in that: The method comprises: A road network model and a distribution network model are established based on graph theory, and the coupling relationship between the two is considered to establish a road-electricity coupling model, and each charging station node in the distribution network model is determined; the road-electricity coupling model is a model that represents the interaction relationship between the road network model and the distribution network model; Establish a travel demand model and charging behavior model that consider the temporal and spatial distribution characteristics of electric vehicles, so as to construct a charging load model of electric vehicles with temporal and spatial distribution characteristics; the travel demand model is to plan the shortest path of electric vehicles based on Dijkstra theory, and is combined with the travel chain characteristics of electric vehicles to describe their dynamic changes in time and spatial transfer within a day; the charging behavior model is to obtain the charging demand of a single electric vehicle at the charging station node connected to the distribution network based on the user's charging psychology and the charge state of the electric vehicle; The probabilistic power flow is calculated based on the charging load model, the basic load model and the distributed power source probability model, thereby obtaining the node voltage value and the line power value of the charging station; the basic load model refers to the stable load that exists in the distribution network for a long time; The static voltage stability index is used to measure the impact of charging station node failure on the distribution network flow.

2. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 1 is characterized in that: The distribution network model is a graphical representation model of the topological structure and parameters of the power system. The distribution network structure includes a collection of nodes and branches, and the nodes only include load charging stations and generators. The distribution network structure is represented as follows: Among them, G g Represents the distribution network topology; V g represents the node set of charging stations in the distribution network; E g represents the set of distribution network lines; ψ g represents the adjacency matrix of the distribution network; represents the charging station node of the distribution network; m represents the number of charging station nodes; i and j represent variables, with a value range of 1 to m; d ij g Lines between charging station nodes in the distribution network; i 、x i 、c i , P i They represent the resistance, reactance, susceptance and the line transmission limit power in the distribution network respectively; The road network model refers to the connection state model between each road network node in the traffic road network topology structure. In the road network topology diagram, the line segment represents the road and the point represents the road network node. The road network model is shown in formula (2): Among them, G r Represents the road network topology; where V r Represents a road node set; v i r 、v j r represents the road node; n represents the number of road nodes; E r Represents a set of road segments; d ij r Indicates the distance between road nodes; i and j represent variables, with a value range of 1 to n; D r represents the adjacency matrix of the road network, e ij r It represents the length of the road section, and its calculation formula is shown in formula (3): Among them, e ij r 、e ji r Indicates the length of the road section; d ij r Represents the distance between road nodes; v i r 、v j r represents a road node; i and j represent variables, with a value range of 1 to n; inf represents infinity; The road-electricity coupling model includes: a coupling relationship refers to a relationship of interaction and mutual influence between two things. The road-electricity coupling model is a model that represents the interaction relationship between the road network model and the distribution network model. The corresponding coupling relationship between topological structures is expressed as follows: Where G represents the circuit-electric coupling topology; G r Represents the road network topology; G g Represents the distribution network topology; E r-g represents the network coupling edge set; d ij represents the network coupling edge; Represents a road node; v j g Grid node; V r Represents a road node set; V g Represents a collection of distribution network nodes.

3. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 1 is characterized in that: The construction of a charging load model for electric vehicles with temporal and spatial distribution characteristics includes: The electric vehicle charging load model includes a travel demand model and a charging behavior model, wherein the travel demand model is used to calculate the time and space distribution of the driving path and charging demand of the electric vehicle by combining the road network model, the travel chain, the time chain characteristic quantity and the space chain characteristic quantity of the electric vehicle; The travel chain in the travel demand model refers to the complete process of an electric vehicle starting from a starting point, passing through each travel destination one by one in a time sequence according to the itinerary, and finally arriving at the end of the trip; the time chain characteristic quantity refers to the dynamic change of the time of the electric vehicle's travel within a day, including the time of arrival and departure at the destination, the driving time and the parking time at various destinations; the space chain characteristic quantity refers to the spatial transfer of the electric vehicle's travel within a day, including the travel purpose and the single travel mileage of the electric vehicle; The user's charging psychology is used to reflect the charging decision of each electric vehicle: at the beginning of each trip, the user will judge the final capacity of the trip, and the remaining battery capacity at the end of the trip is described as: S(t end )=S(t ini )-Dis·ω car ; (5) Among them, S(t end ) and S(t ini ) represent the capacity values ​​of the electric vehicle at the time of reaching the destination and the time of departure respectively; Dis represents the distance formed; w car Indicates the power consumption per unit trip of electric vehicles; At this time, a random number r between 0 and 1 is used to represent the user's charging psychology. If the random number r is greater than S(t end ) or S(t end ) is less than 20% of the maximum battery capacity, the user chooses to charge; If the random number r is less than S(t end ), and S(t end ) is greater than 20% of the maximum battery capacity, charging is not selected; It can be described as: Wherein, μ represents the user's charging status. If μ=1, it means charging is selected; if μ=0, it means not charging; S max Indicates the maximum capacity of an electric vehicle battery.

4. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 3 is characterized in that: If the user chooses to charge, the travel route will be replanned, that is, through the road-electricity coupling model, the Dijkstra theory is used to generate a new path through the charging station node. The charging demand of a single electric vehicle at the grid access node b is expressed as: Where, t represents the time; P c (t) represents the charging demand of a single electric vehicle at time t, t0 represents the charging start time; T c Indicates charging time; P b (t) represents the charging demand of the electric vehicle connected to the charging station node b in the distribution network at time t; b represents the charging station node; By integrating the travel demand model and the charging behavior model, an overall framework for charging load prediction is formed; In the charging vehicle probability load model, the electric vehicle charging load approximately follows the normal distribution. c The probability density function of the normal distribution is expressed as follows: Among them, p(P c ) represents the probability density function of the electric vehicle charging load; P c Indicates the charging load of electric vehicles; μ c and σ c are the expected value and standard deviation of the active power demand of a single electric vehicle at a certain moment; N(·) represents the normal distribution function.

5. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 3 is characterized in that: If a charging station node fails, the road network node corresponding to the failed charging station node in the road network model is not considered when generating a new path passing through the charging station node using Dijkstra theory.

6. The probabilistic power flow calculation method based on distribution network N-1 fault analysis according to claim 5 is characterized in that: The probability flow is calculated based on the charging load model, the basic load model and the distributed power supply probability model; the load model corresponding to the basic load is expressed as follows: Among them, f(P L ) represents the base load active power P L The probability density distribution function of L ) is the basic load reactive power Q L The probability density distribution function of L and Q L are respectively the active power and reactive power of the base load; μ PL and μ QL They are active power P L and reactive power Q L The mean of PL and σ QL are the standard deviations of active power and reactive power respectively; σ 2 PL and σ 2 QL are the variances of active power and reactive power, respectively; exp[·] represents an exponential function with the natural constant e as the base.

7. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 6 is characterized in that: The distributed power supply probability model includes a wind power output probability model and a photovoltaic output probability model. The wind power output model in the distributed power supply model is characterized by Weibull distribution, and the photovoltaic output model is characterized by Beta distribution. The wind power and photovoltaic random variables are respectively characterized by K distribution. wt and K pv The probability density functions of the two are shown as follows: in, represents the probability density function of wind power output that conforms to the Weibull distribution; υ represents the wind speed; k is the scale parameter and shape parameter of the Weibull distribution; exp[·] represents the exponential function with the natural constant e as the base; Among them, F pv (s, α, β) represents the probability density function of photovoltaic output that conforms to the Beta distribution; s represents the light intensity; α and β are the shape parameters of the Beta distribution; Γ(·) represents the Gamma function.

8. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 7 is characterized in that: Probabilistic power flow calculation refers to bringing the basic load values ​​of each node at each time output by the probabilistic load model of charging vehicles, the basic load model and the probabilistic model of distributed power sources, the charging load values ​​of the charging station nodes and the output values ​​of the distributed power sources into the power system power flow calculation to obtain the voltage and power values ​​of each node; specifically, the Newton-Raphson power flow calculation method in Matpower is used for probabilistic power flow calculation.

9. The probabilistic power flow calculation method based on charging station node failure analysis according to claim 1 is characterized in that: The static voltage stability index is used to measure the impact of a charging station node failure on the power flow of the distribution network, including: using the obtained node voltage value to represent the static voltage stability index, which is expressed as: In formula (12), L j It represents the static voltage stability index corresponding to node j, using L j The voltage stability is judged by the distance from the critical point 0. The farther the distance, the better the voltage stability. Z ij Represents the impedance of branch ij, R ij Represents the resistance of branch ij, R ij represents the resistance of branch ij, X ij Represents the reactance of branch ij; S j represents the apparent power of node j, P j represents the active power of node j, Q j represents the reactive power of node j; U i It represents the voltage amplitude of node i. The importance of the electric vehicle charging station node refers to the impact of the failure of the electric vehicle charging station node on the distribution network flow quantified by the evaluation index. The greater the impact, the more important the electric vehicle charging station node is.

10. A probabilistic power flow calculation system based on charging station node failure analysis, characterized in that: The system includes: The distribution network model building module is used to establish the road network model and the distribution network model based on graph theory, and consider the coupling relationship between the two to establish the road-power coupling model and determine the various charging station nodes in the distribution network model; The charging load model building module is used to establish a travel demand model and a charging demand model that take into account the temporal and spatial distribution characteristics of electric vehicles, thereby building a charging load model that takes into account the temporal and spatial distribution characteristics of electric vehicles; A probability power flow calculation module, used to calculate the probability power flow based on the charging load model, the basic load model and the distributed power supply probability model, so as to obtain the charging station node voltage value and the line power value; The characterization module is used to use a static voltage stability index to measure the impact of a charging station node failure on the distribution network flow.

Images