Node reliability optimization evaluation method based on power flow constraint model

Abstract

The node reliability optimization evaluation method based on the power flow constraint model comprises the following steps: establishing node reliability evaluation indexes, and taking node fault frequency and node interruption duration as main consideration indexes; establishing a linear programming model according to the node reliability evaluation indexes, and establishing a power flow constraint model according to power flow constraints; and carrying out model expansion according to the established power flow constraint model, carrying out improvement according to the power supply condition of the distributed power supply DG, and calculating a reliability evaluation index of the load point i. The method not only can calculate the reliability indexes of the power distribution network under different structures, but also can specifically reflect the logic relationship among the nodes and reflect the influence factors among the nodes; and key factors influencing the power supply reliability of the power distribution network are identified. And subsequent power distribution network planning is further guided through a calculation result and a linear model.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution network power supply reliability evaluation, and particularly relates to a node reliability optimization evaluation method based on a power flow constraint model. Background Art

[0002] In recent years, with the access of distributed photovoltaics on the power supply side and the increasing electricity demand on the load side, the complexity of the distribution network has gradually increased, bringing a certain impact to the stability of the distribution network. The reliability evaluation of the distribution network urgently needs further development and improvement. The development of distributed photovoltaics and energy storage systems has improved the power supply reliability of the distribution network to a certain extent. However, it has also brought new challenges to the operation and management of the power grid, and more advanced control technologies and coordination strategies are required to ensure the stable and efficient operation of the power system. The reliability evaluation of the distribution network power supply is crucial for ensuring the safe and stable operation of the power grid.

[0003] The power supply reliability of the distribution network is usually evaluated by simulation methods and analytical methods. Among them, the Monte Carlo simulation method predicts the fault frequency by randomly generating system states, which is applicable to various complex and random systems and can handle different uncertain factors. However, it has a large amount of calculation, a long time consumption, and cannot reflect the logical relationship between faults. By analyzing the topological structure between components, the influencing factors between components can be obtained, which can specifically reflect the logical relationship between each component and the power supply reliability index.

[0004] Traditional reliability evaluation of distribution networks relies too much on simulation models and cannot intuitively reflect the logical relationship and topological structure of the mutual influence between fault nodes. Summary of the Invention

[0005] With the increasing complexity of the distribution network due to the access of distributed power sources, in order to improve the efficiency of reliability evaluation in the distribution network. The present invention provides a node reliability optimization evaluation method based on a power flow constraint model. This method can not only calculate the reliability indexes of the distribution network under different structures, but also specifically reflect the logical relationship between each node and the influencing factors between nodes; thereby identifying the key factors affecting the power supply reliability of the distribution network. Further guide the subsequent distribution network planning through the calculation results and the linear model.

[0006] The technical solution adopted by the present invention is as follows:

[0007] A node reliability optimization evaluation method based on a power flow constraint model, comprising the following steps:

[0008] Step 1: Establish node reliability evaluation indexes, with the node fault frequency and the node interruption duration as the main consideration indexes; Step 2: Based on the node reliability evaluation indexes in Step 1, establish a linear programming model and establish a power flow constraint model according to the power flow constraint;

[0009] Step 3: Expand the model according to the power flow constraint model established in Step 2, and calculate the reliability evaluation index of load point i according to the power supply situation of distributed generation DG.

[0010] The said Step 1 includes the following steps:

[0011] Step 1.1: Establish basic reliability evaluation indexes:

[0012] The basic reliability evaluation indexes adopt traditional distribution network reliability indexes as primary evaluation indexes, including System Average Interruption Frequency Index (SAIFI); System Average Interruption Duration Index (SAIDI); Average System Availability Index (ASAI); Expected Energy Not Supplied (EENS);

[0013] Meanwhile, define load node s. Among all the feeders connected to node s, the upstream branches will cause node s to be powered off due to switch operation and fault repair; for other branches except the upstream branches, when a fault occurs, node s will be powered off due to switch operation;

[0014] Step 1.2: Conduct simplified analysis on the basis of traditional distribution network reliability evaluation indexes, and make the following settings for the distribution network:

[0015] The establishment of standard node reliability evaluation indexes depends on the distribution network topology model and load conditions. The actual topology and operation state of the distribution network are complex. For the sake of simplified processing and analysis, the present invention proposes the following practical definitions:

[0016] (1) Consider the continuous interruption caused by a single branch fault. The fault on the branch is represented by two indexes: fault frequency λ ij and power outage duration τ ij The two indexes are used to represent.

[0017] (2) Consider that the distribution network system is radial, and each branch connected to the power source is equipped with a circuit breaker and has an automatic reclosing device;

[0018] (3) All branches are equipped with disconnect switches to ensure that faulty components can be isolated in time during a fault, reducing the power outage time of other non-faulty components.

[0019] (4) When a fault occurs, the upstream circuit breaker will trip, and network reconfiguration is carried out by operating the disconnect switch and the circuit breaker.

[0020] (5) Isolate the faulty component in time, keep the non-faulty components continuously powered, and restore power supply after the faulty component is repaired;

[0021] According to the above definitions, the node interruption duration includes the following two aspects:

[0022] 1). When there is no component failure, the power outage of the component is only caused by switch switching;

[0023] 2). When a component fails, the duration of the interruption is the time for switch switching and fault repair;

[0024] According to the above definitions, the failure frequency of node s affected by switch actions is proposed The failure frequency of node s affected by switch actions and fault repair The power outage time of node s caused by switch actions The power outage time of node s caused by switch actions and fault repair The total system failure rate That is, the tripping frequency of the system circuit breaker The failure frequency of node s affected by switch actions Can be expressed as the total failure rate And the failure frequency of node s affected by switch actions and fault repair The difference, and the expression is as follows:

[0025]

[0026] The power outage time of node s only caused by switch actions For all feeders connected to node s, except for the upstream branch, when a fault occurs in the branch, it is the time when node s is powered off due to switch actions;

[0027] The power outage time D of node s caused by switch actions and fault repair s RP Is equal to the upstream branch repair time And the switch interruption time The sum, and the expression is as follows:

[0028]

[0029] Step 1.3: Define the basic reliability evaluation index in Step 1.1 using the relevant indexes in Step 1.2 as follows:

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] In the above formula: CIF s represents the user interruption frequency, which refers to the number of power outages experienced by a certain user or user group within a certain period of time. It reflects the frequency of user power outage events in the distribution network. CID s represents the user interruption duration, which refers to the power outage duration experienced by a certain user or user group during a power outage event. Ψ LN represents the set of all nodes, specifically referring to the set of nodes affected by power outages here; N represents the total number of users during the statistical period, and calculates the average power outage time experienced by each user; C s represents the unsupplied energy of node s. △ b represents the power outage duration of branch b, in hours; μ b represents the failure frequency of branch b; L s represents the load importance or load factor of node s, which is used to adjust or reflect the priority or sensitivity of power demand. B represents the set of all power supply areas or power supply units. 8760 represents the total number of hours in a year, which is used to convert the annual average load to the average load per hour.

[0037] In step 2, a linear programming model is established, including the following processes:

[0038] Define the objective function through the power flow analysis of the distribution network and formulate the standard calculation formula. Define linear inequalities as the constraint range of the standard calculation formula, and limit the value range of the decision variables. The set of all constraint conditions can form a feasible region. Find the value of the linear inequality that satisfies the objective function within this feasible region to obtain the optimal solution that meets the conditions. The principle flow chart of the linear programming model is as Figure 1 shown.

[0039] Before conducting power system analysis, it is first necessary to define relevant decision variables, which include the output power of generators, load demands, and line flows. Then, conduct power flow analysis to calculate the voltages of each node and the power flows of each line in the power system, which is used to analyze the operating state of the system.

[0040] Subsequently, ensure the power balance between generation and load in the system, that is, the total generation must meet the load demand plus the system losses. Through line capacity constraints, ensure that the power flow of each line in the power system does not exceed its thermal limit or capacity limit to avoid line overload. Through generator output constraints, ensure that the output power of each generator is within the range allowed by its technology, including minimum and maximum output limits. The constraint formulas are as follows:

[0041] (1) Power balance constraint:

[0042] P g =P l +P b(9);

[0043] Where: P g is the output power of the g-th generator; P l is the power demand of the l-th load; P b is the power loss through the b-th branch, which is proportional to the square of the resistance and current of the line.

[0044] (2) Line capacity constraint:

[0045] |F ij | ≤ F rate,ij (10);

[0046] Where: |F ij | represents the absolute value of the power flow on line ij; F rate,ij represents the rated capacity of line ij, that is, the maximum power flow that this line can safely transmit.

[0047] (3) Generator output constraint:

[0048] P min,g ≤ P g ≤ P max.g (11);

[0049] Where P g is the actual output of the generator; P min,g is the minimum output limit of the generator; P max.g is the maximum output limit of the generator.

[0050] Considering the non-linear characteristics of the power system, direct solution may be very complex. Through linearization, the non-linear problem can be transformed into a linear problem to simplify the calculation. Thus, a linear programming model will be constructed, as shown in formula (12). By solving the linear programming model, the optimal solution that satisfies the constraint conditions can be calculated.

[0051] Minimize(Z) = c 1 x 1 + c 2 x 2 + … + c n x n (12);

[0052] Where: Z is the objective value; Minimize(Z) is to find the minimum objective value; c 1 、c 2 … c n are coefficients; x 1 、x 2 … x n are decision variables.

[0053] In the reliability assessment of a distribution network, according to binary discrete decision variables, decision variables are set to simulate the power supply interruption conditions of different nodes, the fault conditions of branches, and the durations of these faults. Based on the mutual influence relationships among these three factors, the key indicators for the reliability assessment of the distribution network power supply are calculated. According to the power flow constraint model of the distribution network, the optimal values of these power flow variables are equivalent to algebraic expressions.

[0054] In step 2, analyze the optimal power flow distribution of a certain node, and calculate the operating level of this node under the corresponding load conditions according to the calculation constraint formula. Simulate the load operation conditions in the virtual system based on the power flow distribution. Set the load of a certain node to 1 pu, and at the same time, ignore the influence of the remaining nodes on the system and set them to 0 pu. Thus, the following power flow constraint model is proposed:

[0055]

[0056] In Equation (7): and are two continuous variables, where is the power flow of the virtual system connecting branches i and j under the operating condition corresponding to node s. When the power flow is from i to j, is 1, otherwise it is 0. is a continuous variable, representing the power input of power generation station node i under the operating condition corresponding to node s in the virtual system, and can only take 1 or 0. Y represents the set of all lines in the system, i represents node i, and j represents node j.

[0057] The power flow constraint model includes the following conditions:

[0058] Power balance condition:

[0059]

[0060] In Equation (8): is a binary parameter. When i = s, takes the value of 1, and at the same time, ensure that the parameter taken can only be 0 or 1.

[0061] Branch-related variable restrictions:

[0062]

[0063] Power generation station-related variable restrictions:

[0064]

[0065] Among them, the binary parameter has the following values:

[0066]

[0067] In Equation (11): Ψ LN represents the set of all users or nodes;

[0068] Among them, the conditions included in the power flow constraint model have a unimodular structure, that is, the unimodular matrix is a square matrix, each element on its main diagonal is 1, indicating that the coefficient of the basis vector in the basis is 1, and the other elements of the matrix are 0.

[0069] The above power flow constraint model excludes and the situation where both are equal to 1 at the same time.

[0070] The reliability evaluation index proposed in Step 1.2 can be represented by the proposed variables and respectively, and can be specifically represented according to Equation (22), Equation (18), Equation (28), and Equation (23); further reducing the dependence on historical data and enhancing the logical relationship between nodes.

[0071] (1): When or branch ij supplies power to node s, according to the assumptions in the first section, the variables and can be used to determine whether the interruption of branch ij will cause the repair and switching actions of node s. Therefore, the interruption frequencies due to fault repair and switching actions are as follows:

[0072]

[0073] Among them: λ ij is the failure rate of branch ij, Y represents the set of all lines in the system, i represents node i, and j represents node j.

[0074] (2): For each load node, the difference between the total interruption frequency and is the interruption frequency due to switching actions as follows:

[0075]

[0076] and affects the total interruption frequency of node s, and is the sum of the failure frequencies of all branches in the feeder where s is located

[0077]

[0078] Among them: Ψ SS represents the set of all nodes; Ψi Represents the set of nodes connected to the i-node; Is the power flow of the virtual system connecting branches i and j under the operating condition corresponding to node s, when the power flow is from i to j; Is the power flow of the virtual system connecting branches i and j under the operating condition corresponding to node s, when the power flow is from j to i; Y represents the set of all lines in the system.

[0079] And the failure frequency of all branches is the interruption frequency of the circuit breaker That is, there is the following expression:

[0080]

[0081] Where: r represents the circuit breaker affecting the power supply of branch ij; λ rs Represents the failure rate of branch ij affected by the circuit breaker;

[0082] The interruption frequency caused by the switch operation can be obtained The calculation formula is:

[0083]

[0084] Where: Ψ SS Represents the set of all nodes; Ψ i Represents the set of nodes connected to the i-node; Ψ LN Represents the set of all nodes, specifically referring to the set of nodes affected by power outages here; Ψ s Represents the set of nodes connected to node s; Y represents the set of all lines in the system.

[0085] (3): Interruption frequency due to fault repair and switch operation Is:

[0086]

[0087] Where: λ ij Represents the failure rate of branch ij; Represents the total interruption time of branch ij; Y represents the set of all lines in the system.

[0088] (4): Interruption duration due to switch operation Is:

[0089]

[0090] Among them, the repair time of the upstream branch Is:

[0091]

[0092] Total interruption duration Can also be expressed as:

[0093]

[0094] Interruption duration affected by the circuit breaker Is:

[0095]

[0096] Then the interruption duration due to the switch operation is expressed as:

[0097]

[0098] Where: Ψ SS Represents the set of all nodes; Ψ i Represents the set of nodes connected to node i. Ψ LN Represents the set of all nodes, specifically referring to the set of nodes affected by the power outage here; Ψ s Represents the set of nodes connected to node s; Y represents the set of all lines in the system. Represents the interruption time of branch ij only due to switch switching; λ ij Represents the failure rate of branch ij; Represents the total interruption time of branch ij; Y represents the set of all lines in the system.

[0099] In step 3, considering the influence of branch line protection, disconnector, and sectional circuit breaker, and making improvements according to the power supply situation of distributed generation (DG), when calculating the reliability evaluation index of load point i, the double faults of main feeder DG need to be considered, where: interruption frequency λ i , average fault interruption time γ i , annual average power outage time Y i The calculation is as follows:

[0100]

[0101]

[0102] Y i =λ i γ i (31);

[0103] In the formula: λ D And γ D Are respectively the interruption frequency and average power outage duration due to faults of DG; λ S,k And γ S,k Are respectively the interruption frequency and average power outage duration due to faults of the k - th section of the feeder; N Dis the number of main feeders connected to DG and load points.

[0104] A node reliability optimization evaluation method based on a power flow constraint model according to the present invention has the following technical effects:

[0105] 1) Step 1 of the present invention proposes a reliability logic calculation method, which calculates by linking reliability indicators with interruption frequency and interruption time, simplifies the calculation process of the complex topological structure of the distribution network, and improves the calculation efficiency of the reliability evaluation of the distribution network;

[0106] 2) In step 2 of the present invention, by using the power flow constraint model of the distribution network, the proposed optimization evaluation model reduces the dependence on data. When the historical data of the distribution network is incomplete, the distribution network can also be reliably evaluated, increasing the applicability of the model.

[0107] 3) Step 3 of the present invention further examines the applicability of the model and proposes a method to improve the power supply reliability of the distribution network. After adding distributed power sources, various indicators of the power supply reliability of the distribution network can be improved to a certain extent, significantly improving the power supply reliability of the distribution network. Description of the Drawings

[0108] Figure 1 is the flowchart of power flow analysis based on a linear programming model.

[0109] Figure 2 is the topological structure diagram of a 23-node test system.

[0110] Figure 3 is the node failure rate diagram.

[0111] Figure 4 is the node interruption time diagram.

[0112] Figure 5 is the interruption frequency and interruption duration diagram of switch switching and maintenance.

[0113] Figure 6 is the annual interruption frequency and interruption duration diagram of switch switching.

[0114] Figure 7 is the user-side parameter index diagram.

[0115] Figure 8 is the system reliability evaluation result. Detailed Embodiments

[0116] A method for evaluating the reliability of distribution network nodes based on a power flow constraint model. First, based on the operation of the distribution network, simple assumptions for fault analysis of the distribution network are established, and reliability indicators are calculated by relating them to the interruption frequency and interruption time. Then, according to the power flow constraint model of the distribution network and the analysis of the influence relationships among components, the proposed reliability indicators of the distribution network are logically transformed. Finally, considering the access of distributed power sources, the applicability of the model is extended. Case studies are carried out in 23-node, 76-node, 126-node, 377-node, and 957-node systems, verifying that the calculation efficiency of this method is increased by nearly 70% in complex distribution networks, and the power supply reliability of the distribution network can be greatly improved by the access of distributed power sources.

[0117] including the following steps:

[0118] Step 1: Establish node reliability evaluation indicators, with the node fault frequency and node interruption duration as the main consideration indicators; Step 2: Based on the node reliability evaluation indicators in Step 1, establish a linear programming model and establish a power flow constraint model according to the power flow constraints;

[0119] Step 3: Expand the model according to the power flow constraint model established in Step 2, improve it according to the power supply situation of the distributed power source DG, and calculate the reliability evaluation indicators of load point i

[0120] Step 4: Calculate the reliability evaluation indicators in a multi-node test system, as follows:

[0121] (1): Take IEEE-RBTS Bus 6 as the test system. This distribution system includes 30 lines, 23 load segments, 23 transformers, 23 disconnectors, and 5 circuit breakers. All branches are equipped with disconnectors to ensure that faults can be quickly cleared and the power outage time of non-faulty branches can be reduced. Each branch connected to the power station is equipped with a circuit breaker. The topology of the test system is as Figure 2 shown.

[0122] The failure rate, fault repair interruption duration, and switch switching interruption duration parameters of this system are as Figures 3 to 4 shown, and simulation analysis is carried out by this test system.

[0123] When assuming the node failure rate of the distribution network, considering influencing factors such as historical parameters, node complexity, load distribution, and fault propagation, the node failure rate is assumed. Different branchings and positions of nodes will affect the node failure rate. The failure rates of different nodes are different. When there are more branch nodes or they are close to the end of the branch, the node failure rate is higher and the interruption time is longer.

[0124] Since the time for fault repair far exceeds the time for switch switching, the interruption duration caused by node faults is much higher than that caused only by switch switching. The node test system is simulated in the software MATLAB through the above model and solved by CPLEX calculation. The interruption frequency and interruption duration are calculated, as Figure 5 shown. Whether the interruption is caused by repair and switch switching or only by switch switching, the node failure rate and the node interruption time have the same trend. Therefore, the development trend of another index can be roughly predicted through the development trend of the failure rate or the interruption duration, the logical relationship between the faulty nodes can be inferred, and further guidance for the distribution network planning can be provided to improve the power supply reliability of the distribution network.

[0125] When the load node is only affected by the interruption of switch switching, the annual power outage time and annual power outage rate of the load node are as Figure 6 shown. From Figure 6 the simulation results, whether the interruption is caused by repair and switch switching or only by switch switching, the node failure rate and the node interruption time have the same trend.

[0126] (2): Based on the interruption time and interruption frequency calculated in (1), the system reliability indexes are further calculated as shown in Table 1 below.

[0127] Table 1 Calculated values of reliability indexes

[0128]

[0129] At the same time, the customer interruption frequency (CID s ) and customer interruption duration (CIF s ) of the 23-node system are obtained through simulation calculation, and the results are as Figure 7 shown. The interruption duration on the user side is the same as the annual total interruption duration. The closer to the end of the distribution network, the longer the interruption duration caused by fault repair or switch switching.

[0130] (3): To further verify the improvement of the model on the reliability evaluation efficiency, simulation comparison is carried out with the Monte Carlo method under various node conditions.

[0131] The test system will continue to be simulated and calculated in distribution network systems with different numbers of nodes such as 76 nodes, 126 nodes, 377 nodes, and 957 nodes. It is verified that this method not only has a high accuracy in the calculation of distribution network reliability indexes, but also greatly improves the calculation speed when dealing with large-scale distribution network systems. For example, in the simulation calculation of 957 nodes, the simulation based on the traditional Monte Carlo method requires 10458.23 s, while the time required for the topological logic calculation simulation proposed in the present invention is only 3008.96 s, and the running time is greatly shortened.

[0132] Table 2 Comparison Table of Simulation Durations

[0133]

[0134] As can be seen from Table 2, the simulation time is greatly shortened, significantly improving the computational efficiency of reliability assessment. The simulation efficiency is increased by nearly 70% in the multi-node distribution system.

[0135] (4): Conduct a simulation analysis of the power supply reliability of the distribution system after adding distributed power sources:

[0136] Add two DGs with rated powers of 0.75 MW and 1 MW respectively at the feeders 20 and 29 nodes of the IEEE-RBTS bus 6 test system of the simulation system, and keep the reference parameters of other nodes unchanged, and conduct a reliability assessment of the test system.

[0137] Through the access of distributed power sources, the reliability index is improved to a certain extent, and the power supply reliability of the distribution network is significantly improved. When a fault occurs in the distribution network, the DG can form an island with the surrounding load points, and the surrounding load points can resume power supply, improving the power supply reliability of the distribution system.

Claims

1. A node reliability optimization evaluation method based on a power flow constraint model is characterized by The following steps are involved: Step 1: Establish node reliability evaluation indicators, with node failure frequency and node interruption duration as the main considerations; Step 2: Establish a linear programming model based on the node reliability evaluation index of step 1, and establish a power flow constraint model based on the power flow constraint; Step 3: Expand the model based on the power flow constraint model established in step 2, and calculate the reliability evaluation index of load point i according to the power supply situation of distributed generation DG.

2. The node reliability optimization evaluation method based on the power flow constraint model according to claim 1 is characterized in that: The step 1 comprises the following steps: Step 1.1: Establish basic reliability evaluation indicators: The basic reliability evaluation index adopts the traditional distribution network reliability index as the first-level evaluation index, including the system average interruption frequency index SAIFI; the system average interruption duration index SAIDI; the average system availability index ASAI; the expected unprovided energy EENS; At the same time, define the load node s. Among all feeders connected to the node s, the upstream branch will cause the node s to be powered off due to switch action and fault maintenance; if other branches except the upstream branch fail, the node s will be powered off due to switch action; Step 1.2: Simplify the analysis based on the traditional distribution network reliability evaluation indicators. The node outage duration includes the following two aspects: (1) When there is no component failure, the component only causes power outage due to switch switching; (2) When a component fails, the interruption duration is the time it takes to switch and repair the fault; According to the above definition, the failure frequency of node s affected by the switch action is proposed Failure frequency of nodes s affected by switch actions and fault repair Power outage time of node s due to switch action Power outage time of node s due to switch action and fault repair Total system failure rate The tripping frequency of the system circuit breaker Failure frequency of node s affected by the switch action Can be expressed as the total failure rate The failure frequency of nodes s affected by switching actions and fault repair The difference is expressed as follows: The power outage time of node s caused by switch action only The time when a branch other than the upstream branch fails among all feeders connected to the node s, resulting in power outage at the node s due to switch action; Power outage time of node s due to switch action and fault repair Equal to the upstream branch repair time and switch interruption time The sum, expression is as follows: Step 1.3: Define the basic reliability evaluation indicators in step 1.1 using the relevant indicators in step 1.2 as follows: In the above formula: CIF s Indicates user interruption frequency, which refers to the number of power outages experienced by a user or user group within a certain period of time; it reflects the frequency of user power outage events in the distribution network; CID s Indicates the duration of user interruption, which refers to the duration of power outage experienced by a certain user or user group during a power outage event; LN Refers to the set of nodes affected by the power outage; N represents the total number of users during the statistical period, and the average power outage time experienced by each user is calculated; C s represents the unsupplied energy of node s; b Indicates the duration of power outage in branch b, in hours; μ b Indicates the frequency of failure of branch b; L s It represents the load importance or load factor of node s, which is used to adjust or reflect the priority or sensitivity of power demand; B represents the set of all power supply areas or power supply units.

3. The node reliability optimization evaluation method based on the power flow constraint model according to claim 2 is characterized in that: In step 2, a linear programming model is established, including the following process: First, define the relevant decision variables, including generator output power, load demand, and line flow; then, perform power flow analysis to calculate the voltage of each node in the power system and the power flow of each line to analyze the operating status of the system; Subsequently, line capacity constraints are used to ensure that the power flow of each line in the power system does not exceed its thermal limit or capacity limitation, and generator output constraints are used to ensure that the output power of each generator set is within the allowable range.

4. The node reliability optimization evaluation method based on the power flow constraint model according to claim 3 is characterized in that: The constraints are as follows: (1) Power balance constraints: P g =P l +P b (9); Where: P g is the output power of the g-th generator; P l is the power demand of the lth load; P b is the power loss through the bth branch, which is proportional to the resistance of the line and the square of the current; (2) Line capacity constraints: |F ij |≤F rate,ij (10); Where: |F ij | represents the absolute value of the power flow on line ij; F rate,ij represents the rated capacity of line ij, i.e., the maximum power flow that the line can safely transmit; (3) Generator output constraints: P min,g ≤P g ≤P max.g (11); Where P g is the actual output of the generator; P min,g is the minimum output limit of the generator; P max.g is the maximum output limit of the generator; Considering the nonlinear characteristics of the power system, the nonlinear problem is transformed into a linear problem to simplify the calculation. Thus, a linear programming model is constructed, as shown in formula (12). By solving the linear programming model, the optimal solution that meets the constraints is calculated. Minimize(Z)=c1x1+c2x2+…+c n x n (12); Where: Z is the target value; Minimize(Z) is to find the minimum target value; c1, c2…c n are coefficients; x1, x2…x n is the decision variable.

5. The node reliability optimization evaluation method based on the power flow constraint model according to claim 4 is characterized in that: In step 2, the power flow constraint model is as follows: In formula (7): and are two continuous variables, where is the flow of the virtual system connecting branches i and j under the operating condition corresponding to node s. When the flow flows from i to j, is 1, otherwise it is 0; is a continuous variable, representing the power input of node i of the power plant under the working conditions corresponding to the virtual system, and can only be 1 or 0; Y represents the set of all lines in the system, i represents node i, and j represents node j.

6. The node reliability optimization evaluation method based on the power flow constraint model according to claim 5 is characterized by: The power flow constraint model includes the following conditions: Power balance conditions: In formula (8): is a binary parameter. When i=s, The value is 1, and it guarantees The parameter can only be 0 or 1; Branch related variable restrictions: Power plant related variable restrictions: Among them, the binary parameters The possible values ​​are: In formula (11): LN Represents the set of all users or nodes; The above power flow constraint model excludes and At the same time, it is equal to 1.

7. The node reliability optimization evaluation method based on the power flow constraint model according to claim 2 is characterized in that: The reliability evaluation index proposed in step 1.2 Ability to adopt proposed variables and Specifically, it can be expressed according to formula (22), formula (18), formula (28), and formula (23) respectively; (1): When or Branch ij supplies power to node s. According to the assumptions in Section 1, the variable and It can be used to determine whether the interruption of branch ij will lead to the repair and switching action of node s. Therefore, the interruption frequency due to fault repair and switching action is as follows: Where: ij is the failure rate of branch ij, Y represents the set of all lines in the system, i represents node i, and j represents node j; (2): For each load node, the total interruption frequency and The difference is the interruption frequency caused by the switching action As follows: And the total interruption frequency affecting node s is the sum of the fault frequencies of all branches in the feeder where s is located So we have the following expression: Among them: SS represents the set of all nodes; Ψ i Represents the set of nodes connected to the i-node; is the flow of the virtual system connecting branches i and j under the corresponding operating condition of node s, when the flow flows from i to j; is the flow of the virtual system connecting branches i and j under the operating condition corresponding to node s, when the flow flows from j to i; Y represents the set of all lines in the system; And the fault frequency of all branches is the interruption frequency of the circuit breaker That is, there is the following expression: Where: r represents the circuit breaker that affects the power supply of the ij branch; λ rs It is expressed as the failure rate of branch ij affected by the circuit breaker; The interruption frequency caused by switching action can be obtained The calculation formula is: Among them: SS represents the set of all nodes; Ψ i Represents the set of nodes connected to node i; Ψ LN Represents the set of all nodes, specifically the set of nodes affected by the power outage; s represents the set of nodes connected to node s; Y represents the set of all lines in the system; (3) Interruption frequency due to fault repair and switching action for: Where: ij Expressed as the failure rate of branch ij; represents the total interruption time of branch ij; Y represents the set of all lines in the system; (4): Duration of interruption due to switch action for: Among them, the upstream branch repair time for: Total outage duration It can also be expressed as: Duration of outages affected by circuit breakers for: The interruption duration due to the switch action is expressed as: Among them: SS represents the set of all nodes; Ψ i Represents the set of nodes connected to node i; Ψ LN Represents the set of all nodes, specifically the set of nodes affected by the power outage; s represents the set of nodes connected to node s; Y represents the set of all lines in the system; represents the interruption time of branch ij due to switch switching; ij Expressed as the failure rate of branch ij; represents the total interruption time of branch ij; Y represents the collection of all lines in the system.

8. The node reliability optimization evaluation method based on the power flow constraint model according to claim 1 is characterized in that: In step 3, the influence of branch line protection, disconnector, and section circuit breaker is considered, and improvements are made according to the power supply situation of distributed power sources. The reliability evaluation index of load point i is calculated, and the double fault of the main feeder DG needs to be considered at the same time, where: the interruption frequency λ i , mean fault interruption time γ i , Annual average power outage time Y i The calculation of is as follows: Y i =λ i c i (31); Where: D and γ D are the interruption frequency of DG and the average duration of power outage due to faults; S,k and γ S,k are the interruption frequency and average power outage duration of k-section feeder respectively; N D is the number of main feeders connected to the DG and load points.

Images