A switch exchange method for distribution network topology reconstruction based on sensitivity

Abstract

The present invention discloses a switch exchange method for topology reconstruction of a distribution network based on sensitivity, including: S1, obtaining a sensitivity matrix in the original topology of the distribution network; S2, closing the interconnecting switch to obtain the active power loss of a new ring topology; S3, opening the segmented switch to obtain the active power loss of a new radial topology; S4, sequentially opening other segmented switches to obtain the active power loss of other new radial topologies; until all segmented switches in the ring topology are opened, k minimum active power losses and corresponding topological information are obtained; S5, closing all interconnecting switches to obtain m×k topologies; S6, finding the minimum objective function and the corresponding optimal topology in the m×k topologies; S7, calculating the active power and reactive power output of the optimal distributed power source under the optimal topology. The present invention is safe and reliable, can reduce the active power shortage, and keep the bus voltage within a feasible range.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of distribution network, and in particular to a switch exchange method for topology reconstruction of a distribution network based on sensitivity. Background Art

[0002] Generally speaking, the goal of topology reconstruction is to find a topology with minimal network loss while ensuring the radial topology. Currently, there are three schemes widely used in the distribution network topology reconstruction process, such as mathematical programming, heuristic algorithm, and switch opening and switching method. Among them, the mathematical programming method is characterized by being able to obtain the global optimal solution, but due to the large number of unknown variables and constraints, the computational cost of this method is high; while the heuristic algorithm has a fast solution speed, but cannot guarantee the global optimization.

[0003] The switch opening and exchange method is faster than the mathematical programming method and generally has three steps:

[0004] Step 1: Close all switches on the distribution network to form a ring network;

[0005] Step 2: Select and turn on a switch, and repeat this step until a radial network is formed;

[0006] Step 3: Exchange the switch states of the topology obtained in the previous step to obtain a more optimal radial network.

[0007] However, the traditional switch opening and switching methods are based on the results of power flow calculation to obtain the topological structure corresponding to the minimum network loss, which requires multi-step iterative update or inversion of the Jacobian matrix, and the calculation time is long. With the expansion of the scale of the distribution network and the increase in the penetration rate of renewable energy, the difficulty and demand of topological reconstruction solutions have further increased, and it is urgent to improve the calculation speed while ensuring the calculation accuracy. Summary of the invention

[0008] The technical problem to be solved by the present invention is: in response to the technical problems existing in the prior art, the present invention provides a sensitivity-based distribution network topology reconstruction switch exchange method which is safe and reliable, reduces active power shortage, and keeps the bus voltage within a feasible range.

[0009] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0010] A sensitivity-based distribution network topology reconstruction switch exchange method comprises the following steps:

[0011] S1, obtain the sensitivity matrix in the original topology of the distribution network;

[0012] S2, close the tie switch in the distribution network to obtain a new ring topology; if the circulating current of the new ring topology is less than a preset value, calculate the sensitivity matrix, bus voltage and compensation power of the new ring topology according to the sensitivity matrix in the original topology of the distribution network, and then obtain the active power loss of the new ring topology;

[0013] S3, turning on the segmented switches in the distribution network to obtain a new radial topology; if the circulating current of the new radial topology is less than a preset value, calculating the sensitivity matrix, bus voltage and compensation power of the new radial topology, and then obtaining the active power loss of the new radial topology;

[0014] S4, turn on other segment switches in sequence to obtain other new radial topologies; if the circulating current is less than the set maximum value, calculate the sensitivity matrix, bus voltage and compensation power of each new radial topology, and then obtain the corresponding active power loss;

[0015] Repeat this step until all segment switches in the ring topology are turned on, and record the active power losses of all radial topologies under the ring topology, and obtain k minimum active power losses and corresponding topology information;

[0016] S5, closing the switches of other tie lines to obtain the corresponding ring topology, and calculating the active power loss of each radial topology under the ring topology according to steps S2-S4, and obtaining k minimum active power losses and corresponding topology information, until all tie switches in the original topology are closed to obtain m×k topologies; where m is the number of tie switches;

[0017] S6. Based on the first constraint condition, find the minimum objective function and the topology corresponding to the objective function among the m×k topologies, and use the topology as the optimal topology;

[0018] S7. Based on the second constraint, calculate the active power and reactive power output of the optimal distributed power source under the optimal topology.

[0019] Preferably, in step S1, the calculation process of the sensitivity matrix is:

[0020] Voltage relative to active power and reactive power The partial derivative of is:

[0021]

[0022]

[0023]

[0024]

[0025] In the formula, is the node admittance matrix i Line j Column elements;

[0026] , are the sensitivity matrices of voltage amplitude with respect to active power and reactive power injection, respectively;

[0027] Indicates busbar i The voltage amplitude relative to the bus l Sensitivity of injected active power;

[0028] Indicates busbar i The voltage amplitude relative to the bus l Sensitivity to injected reactive power;

[0029] and Indicates busbar i The voltage amplitude and phase angle;

[0030] , They are the sensitivity matrices of voltage phase angle relative to active power and reactive power injection respectively;

[0031] Indicates busbar i The phase angle relative to the busbar l Sensitivity of injected active power;

[0032] Indicates busbar i The sensitivity of the phase angle of relative to the reactive power injected into bus i;

[0033] represents a plural number, represents the conjugate of a complex number;

[0034] Indicates busbar j The voltage relative to the bus l Sensitivity to injected active power;

[0035] Indicates busbar j The voltage relative to the bus l Sensitivity to injected reactive power;

[0036] In the distribution network, the original topology is considered to consist of 1 relaxed bus and N PQ buses A network consisting of , ;

[0037] The relationship between power and bus voltage is: ; Representation Node i The apparent power of the output; , Respectively represent nodes i Conjugate and Node Voltages j Voltage; is the node admittance matrix i Line j Column elements.

[0038] Preferably, in step S2, the specific process of calculating the bus voltage of the new ring topology is:

[0039]

[0040]

[0041] In the formula, , They are respectively the bus voltage amplitude and bus phase angle vector at the next moment; , are the voltage amplitude and phase angle vector at the current moment respectively; , They are active power increment and reactive power increment respectively; is a collection of switch nodes.

[0042] Preferably, in step S2, the process of calculating the circulating current and compensating the apparent power is:

[0043]

[0044] In the formula, It is the voltage difference of the feeders on both sides before the tie switch is closed; , They are the network nodes before the tie switch is closed. i , j Voltage;

[0045] According to Kirchhoff's law and Ohm's law, the relationship between voltage and current is expressed by the following equation; node i and nodes j is the node where the switch is closed;

[0046]

[0047]

[0048]

[0049]

[0050] In the formula, , , They are the compensation network nodes during the closing process of the tie switch. i, j , T Voltage;

[0051] is the compensation network node i , j The current flowing through the circuit;

[0052] , , They are i side, j Side and i , j Impedance on the line;

[0053] , , They are the nodes when the tie switch is closed. i Injected apparent power, active power and reactive power increments;

[0054] , , They are the nodes when the tie switch is closed. j Injected apparent power, active power and reactive power increments;

[0055] , , , They are right , , , Sensitivity;

[0056] Represents the circulating current when the tie switch is closed.

[0057] Preferably, the active power loss is calculated The process is:

[0058] ;

[0059] is the active network loss at the current moment; and They are the voltage amplitude vectors No. i and j elements; is the voltage phase angle vector No. i and j elements; is the set of all nodes.

[0060] Preferably, in step S6, the process of finding the minimum objective function is:

[0061]

[0062] is the active network loss of the original network;

[0063] is a binary variable of the branch status;

[0064] is a binary variable representing the branch status of the original network;

[0065] and are the weight vectors used to minimize the active power loss and the number of operations of the switch, respectively.

[0066] Preferably, the first constraint condition is:

[0067] Voltage upper and lower bound constraints:

[0068] The open branches must be constrained on the ring:

[0069] The circulation must be less than the maximum circulation constraint:

[0070] in , Indicates the upper and lower bounds of a number; Indicates the branch that is opened; Indicates a ring; Indicates circulation.

[0071] Preferably, in step S7, the specific process of calculating the active power and reactive power output of the optimal distributed power source under the optimal topology is:

[0072]

[0073] In the formula, and are the weight coefficients of the minimum voltage average deviation target and the minimum power shortage target respectively;

[0074] and is the set of all renewable energy nodes and the set of all PQ nodes;

[0075] is the maximum active power available from renewable energy at node d;

[0076] is the active power output of renewable energy at node d;

[0077] is the reference voltage of all nodes;

[0078] and is the number of all renewable energy nodes and all PQ nodes.

[0079] Preferably, the second constraint condition is:

[0080]

[0081]

[0082]

[0083] and are the active power and reactive power vectors output by renewable energy sources, respectively;

[0084] is the maximum reactive power vector of renewable energy output;

[0085] and are the minimum and maximum reactive power vectors of renewable energy output, respectively;

[0086] is the set of all nodes;

[0087] is the magnitude of the voltage at node i;

[0088] and is the minimum and maximum value of the voltage amplitude at node i.

[0089] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are executed.

[0090] Compared with the prior art, the advantages of the present invention are:

[0091] The present invention sets the maximum circulating current as a constraint condition when solving the optimal network topology to ensure the safety of the switch operation process; optimizes the active and reactive power outputs of the distributed power source to reduce the active power shortage while keeping the bus voltage within a feasible range. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 It is a schematic diagram of a power compensation model of the present invention in an embodiment.

[0093] Figure 2 It is a schematic diagram of an equivalent model of the closing ring action of the present invention in an embodiment.

[0094] Figure 3 It is a schematic diagram of an open-loop action equivalent model of the present invention in an embodiment.

[0095] Figure 4 The present invention is a flow chart of a switch exchange method for reconfiguring a distribution network topology in an embodiment.

[0096] Figure 5 This is a diagram showing an embodiment of the switch exchange method for reconfiguring the topology of a distribution network of the present invention in a specific application. DETAILED DESCRIPTION

[0097] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0098] like Figure 4-Figure 5 As shown, the sensitivity-based distribution network topology reconstruction switch switching method according to an embodiment of the present invention is used to achieve power optimization, and specifically includes the following steps:

[0099] S1, obtain the bus voltage, branch impedance and topology information in the original topology of the distribution network, and calculate the sensitivity matrix;

[0100] S2. Close the tie switch of the distribution network to obtain a new ring topology; if the circulating current of the new ring topology is less than the set maximum value, calculate the sensitivity matrix, bus voltage and compensation power of the new ring topology; at the same time, calculate the bus voltage of the new ring topology according to the sensitivity matrix, bus voltage and compensation power in the original topology of the distribution network, and then obtain the corresponding active power loss;

[0101] S3, turn on the segmented switch to obtain a new radial topology; if the circulating current is less than the set maximum value, calculate the sensitivity matrix, bus voltage and compensation power of the new radial topology; at the same time, calculate the bus voltage of the new radial topology according to the sensitivity matrix, bus voltage and compensation power of the new ring topology in step S2, and then obtain the corresponding active power loss;

[0102] S4, turn on other segment switches in sequence to obtain other new radial topologies; if the circulating current is less than the set maximum value, calculate the sensitivity matrix, bus voltage and compensation power of this topology; at the same time, calculate the bus voltage of the new radial topology according to the sensitivity matrix, bus voltage and compensation power of the previous new radial topology, and then obtain the corresponding active power loss;

[0103] Repeat this step until all segment switches in the ring topology are turned on, and the active power losses of all radial topologies under the ring topology are recorded, and k minimum active power losses and corresponding topology information are obtained; k is defined as 3, and the topology information includes the connection relationship between the busbar and the branch, as well as the busbar voltage and the branch impedance, etc.;

[0104] S5, closing the switches of other tie lines to obtain the corresponding ring topology, and calculating the active power loss of each radial topology under the ring topology according to step S2-4, and obtaining k minimum active power losses and corresponding topology information, until all tie switches in the original topology are closed; and m×k corresponding topologies are obtained; m is the number of tie switches;

[0105] S6, considering the constraints, finding the minimum objective function and the topology corresponding to the objective function among the m×k topologies obtained in step S5, and taking this topology as the optimal topology;

[0106] S7. After obtaining the optimal topology, the active and reactive outputs of the optimal distributed generation under the optimal topology are calculated taking into account the constraints.

[0107] Specifically, the process of calculating the sensitivity matrix of the original topology in step S1 is:

[0108] In the distribution network, the original topology is considered to consist of 1 relaxed bus and a collection of N PQ buses A network consisting of , ;

[0109] The relationship between power and bus voltage is: ;

[0110] Representation Node i The apparent power of the output; , Representation Node i Conjugate and Node Voltages j Voltage; is the node admittance matrix i Line j Column elements;

[0111] Voltage relative to active power and reactive power The partial derivative of is:

[0112]

[0113]

[0114]

[0115]

[0116] In the formula, is the node admittance matrix i Line j Column elements;

[0117] , are the sensitivity matrices of voltage amplitude with respect to active power and reactive power injection, respectively;

[0118] Indicates busbar i The voltage amplitude relative to the bus l Sensitivity to injected active power;

[0119] Indicates busbar i The voltage amplitude relative to the bus l Sensitivity to injected reactive power;

[0120] and Indicates busbar i The voltage amplitude and phase angle;

[0121] , They are the sensitivity matrices of voltage phase angle relative to active power and reactive power injection respectively;

[0122] Indicates busbar i The phase angle relative to the busbar l Sensitivity to injected active power;

[0123] Indicates busbar i The sensitivity of the phase angle of relative to the reactive power injected into bus i;

[0124] represents a plural number, represents the conjugate of a complex number;

[0125] Indicates busbar jThe voltage relative to the bus l Sensitivity to injected active power;

[0126] Indicates busbar j The voltage relative to the bus l Sensitivity to injected reactive power.

[0127] Specifically, the process of calculating the sensitivity matrix in step S2 is the same as step S1;

[0128] The bus voltage amplitude is calculated and bus voltage phase angle The method is:

[0129]

[0130]

[0131] In the formula, , is the active power increment and reactive power increment;

[0132] , is the bus voltage amplitude and bus phase angle vector at the next moment;

[0133] , is the voltage amplitude and phase angle vector at the current moment;

[0134] is a collection of switch nodes.

[0135] According to the superposition theorem, the voltage and current after closing the tie switch are divided into two parts. One is the voltage and current of the network before closing the tie switch, and the other is The voltage and current in the compensation network are generated. It is the voltage difference between the feeders on both sides before the tie switch is closed.

[0136] The method for calculating circulating current and compensation power is:

[0137]

[0138] Where: , They are the network nodes before the tie switch is closed. i , j voltage.

[0139] According to Kirchhoff's law and Ohm's law, the relationship between voltage and current is expressed by the following equation. i and nodes jis the node where the switch is closed. i and nodes j In addition, any node on the ring can be selected as T node.

[0140]

[0141]

[0142]

[0143]

[0144] In the formula, , , They are the compensation network nodes during the closing process of the tie switch. i, j , T Voltage;

[0145] is the compensation network node i , j The current flowing through the circuit;

[0146] , , They are i side, j Side and i , j Impedance on the line;

[0147] , , They are the nodes when the tie switch is closed. i Injected apparent power, active power and reactive power increments;

[0148] , , They are the nodes when the tie switch is closed. j Injected apparent power, active power and reactive power increments;

[0149] , , , They are right , , , Sensitivity;

[0150] Represents the circulating current when the tie switch is closed.

[0151] Specifically, the process of calculating the sensitivity matrix in step S3 is the same as step S1; the process of calculating the bus voltage is the same as step S2;

[0152] The calculation process of circulating current and compensation power is:

[0153]

[0154]

[0155]

[0156]

[0157] Indicates the circulating current when the section switch is open; is the compensation network node i , j The current flowing through the circuit;

[0158] , , They are the nodes in the compensation network during the opening of the segmented switch. i , j.T Voltage;

[0159] , They are the network nodes before the segment switch is turned on. i , j Voltage;

[0160] , They are the segment switch opening process nodes i and j Compensated apparent power increase;

[0161] , They are the nodes when the analog switch is turned on. i The injected active power increment and reactive power increment;

[0162] , They are the nodes when the analog switch is turned on. j Injected active and reactive power increments;

[0163] , , , They are right , , , sensitivity.

[0164] Specifically, in each step, the calculation process of active power loss is:

[0165] ;

[0166] is the active network loss at the current moment; and They are the voltage amplitude vectors No. i and j elements; is the voltage phase angle vector No. i and j elements; is the set of all nodes.

[0167] Specifically, the process of obtaining the minimum value of the objective function in step S6 is:

[0168]

[0169] is the active network loss of the original network;

[0170] is a binary variable of the branch status;

[0171] is a binary variable representing the branch status of the original network;

[0172] and are the weight vectors used to minimize the active power loss and the number of operations of the switch, respectively.

[0173] Specifically, the first constraint condition in step S6 is:

[0174] Voltage upper and lower bound constraints:

[0175] The open branches must be constrained on the ring:

[0176] The circulation must be less than the maximum circulation constraint:

[0177] in , Indicates the upper and lower bounds of a number; Indicates the branch that is opened; Indicates a ring; Indicates circulation.

[0178] Specifically, the process of calculating the active power and reactive power of the optimal distributed power source in step S7 is:

[0179]

[0180] In the formula, and are the weight coefficients of the minimum voltage average deviation target and the minimum power shortage target respectively;

[0181] and is the set of all renewable energy nodes and the set of all PQ nodes;

[0182] is the maximum active power available from renewable energy at node d;

[0183] is the active power output of renewable energy at node d;

[0184] is the reference voltage of all nodes;

[0185] and is the number of all renewable energy nodes and all PQ nodes.

[0186] The second constraint in step S7 is:

[0187] ;

[0188] ;

[0189] .

[0190] and are the active power and reactive power vectors output by renewable energy sources, respectively;

[0191] is the maximum reactive power vector of renewable energy output;

[0192] and are the minimum and maximum reactive power vectors of renewable energy output, respectively;

[0193] is the set of all nodes;

[0194] is the magnitude of the voltage at node i;

[0195] and is the minimum and maximum value of the voltage amplitude at node i.

[0196] The present invention sets the maximum circulating current as a constraint condition when solving the optimal network topology to ensure the safety of the switch operation process; optimizes the active and reactive power outputs of the distributed power source to reduce the active power shortage while keeping the bus voltage within a feasible range.

[0197] Figure 1 The power compensation model proposed in this invention simulates the switching action by the power compensation method. The switching operation can be regarded as a dual-port opening and closing process. The rest of the network is unknown, but constant. As long as the port node voltage is equal, the two networks can be regarded as equivalent networks;

[0198] A power compensation strategy in which one node absorbs power and another node injects power is used to simulate the switching operation process so that the equation is satisfied before and after the switching action:

[0199] The equation that needs to be satisfied during the switch opening process:

[0200] The equations that need to be satisfied during the switch closing process are: .

[0201] Where: is the set of all switch nodes; , is the voltage of the actual model node i, j after the switch is turned on; , is the voltage of the compensation model node i, j after the analog switch is turned on; , is the voltage of the actual model node i, j after the switch is closed; , is the voltage at the compensation model node i, j after the analog switch is closed.

[0202] Figure 2 The equivalent model of the closed-loop action proposed in this invention does not require iterative calculations and can directly solve the power required to compensate for the closed-loop action. According to the superposition theorem, the voltage and current of the network after the closed-loop operation are divided into two parts. One is the voltage and current of the network before the closed-loop operation, and the other is The voltage and current in the resulting compensation network; Equal to the voltage difference of the feeders on both sides before closing the loop:

[0203]

[0204] In the formula, , They are the network nodes before the switch is closed. i , j voltage.

[0205] According to Kirchhoff's law and Ohm's law, the relationship between voltage and current is expressed by the following equation; node i and nodes j is the node where the switch is closed; except for the node i and nodes j In addition, any node on the ring can be selected as T node;

[0206]

[0207]

[0208] in and is known; the above equation can eliminate the variable , and establish a system with 4 unknown quantities and a linear system of 4 equations;

[0209] Once you get , and can be represented by the formula Calculated.

[0210] Figure 3 The open-loop action equivalent model proposed in the present invention does not require iterative calculation and can directly solve the power required to compensate for the open-loop action:

[0211]

[0212]

[0213] The open-loop operation can be considered as the reverse process of the closed-loop operation. Can be obtained from the network before the switch is turned on.

[0214] In order to ensure that the network remains radial after topology reconfiguration and to minimize the number of loops in the process, the present invention adopts a switch exchange method consisting of multiple switch opening and switch closing operation pairs, eliminating all switch closing and switch sequential disconnection steps to obtain a specific switch exchange sequence from the original network to the target network; each time a switch opening and switch closing operation pair is performed, the connecting switch is first closed and then the segmented switches in the loop are disconnected until all the segmented switches in the loop have been operated; in order to prevent the algorithm from falling into a local optimum, the k active power network loss minimum solutions of the previous switch opening and switch closing operation pair are recorded, and the next switch exchange is continued.

[0215] The sensitivity-based distribution network topology reconstruction switch exchange method of the present invention improves the speed of solving the distribution network reconstruction while minimizing the network power loss and the number of switches. In order to quantify the impact of switch exchange on the bus voltage, a power compensation strategy solved by closed-loop and open-loop action models is invented to simulate the switch action process. Each switch action problem can be expressed as an integer quadratic programming problem without iterative calculation. After determining the optimal network, the present invention optimizes the active and reactive power output of the distributed power source to reduce the active power deficit while keeping the bus voltage within a feasible range.

[0216] Table 1 is a comparison table of network loss, number of switch actions and switch action timing in a 33-node system under three different situations and two different targets after implementing the method of the present invention.

[0217] Table 1

[0218]

[0219] Two different goals are:

[0220] Objective I: Minimize active power loss;

[0221] Objective II: Minimize network reconfiguration costs;

[0222] Energy costs W L Set to 0.03 yuan / kWh, the switching cost W N Set to 0.1 yuan.

[0223] The three different situations are:

[0224] Case I: There is no DG in the system. The load is equivalent to the system power demand of the 33-bus system; the load level is 100%.

[0225] Case II: There are two DGs on Buses 11 and 26; one provides an output of 500kW and 120kVar, and the other provides an output of 350kW and 60kVar; the load is equal to the system power demand of the 33-bus system; the load level is 100%.

[0226] Case III: There is a 600-kW wind turbine on bus 11 and a 400-kW photovoltaic generator on bus 26; the output levels of wind power generation and photovoltaic power generation are 39% and 0% respectively; the load level is 80%.

[0227] Table 1 shows the active power loss, number of switching actions, and switching action sequence from the original network topology to the target network topology. Three conclusions can be drawn from the results in Table 1:

[0228] i) Reasonable configuration of distributed power sources and reduction of power load will reduce active power loss. After topology reconstruction, the active power loss of case I is 140.71kW and 150.06kW respectively; the active power loss of case II is 97.69kW and 104.88kW respectively, and the active power loss of case III is 75.01kW and 81.96kW respectively;

[0229] ii) Reasonable configuration of wind turbines and reduction of power load will reduce the number of switching times from the original network topology to the target network topology.

[0230] iii) The topology with the smallest network loss is not always the best choice. Take option III as an example. The active power loss of option II is only 6.95 kW more than that of option I, but the number of switching times is reduced by 3.

[0231] Table 2 is a table showing the contact line results of the sensitivity-based switch exchange method of the present invention in 33-node, 84-node and 136-node systems.

[0232] Table 2

[0233]

[0234] Table 3 is a comparison table of network loss and solution time of the switch exchange method based on sensitivity in 33-node, 84-node and 136-node systems of the present invention;

[0235] Table 3

[0236]

[0237] As can be seen from Table 2-3, in the 33-node, 84-node and 136-node systems, the proposed method is only 0.43kW, 0.69kW and 0.11kW more than the SOE (switch opening and exchanging) method, respectively. However, in the 33-node, 84-node and 136-node systems, the calculation time of the proposed method is 1 second, 5.85 seconds and 36.45 seconds less than the SOE method, respectively. Therefore, compared with the SOE method, the proposed method can solve the distribution network topology reconstruction problem faster while ensuring the accuracy of the solution.

[0238] Table 4 is a comparison table of voltage deviation and active power shortage of two different targets of the switch exchange method based on sensitivity of the present invention in a 118-node system;

[0239] Table 4

[0240]

[0241] There are two 6,000-kW photovoltaic generators on busbars 35 and 71, and two 8,000-kW wind turbines on busbars 101 and 111. The output levels of wind and photovoltaic power generation are 93% and 17% respectively; the load level is 100%.

[0242] Two different sets of weights are set to minimize the average deviation of bus voltage and the active power deficit of DG.

[0243] Objective I: Minimize the average bus voltage deviation.

[0244] Objective II: Minimize power deficit cost. Voltage deviation cost W V Set to 0.5 yuan, the active power shortage cost W P Set to 0.5 yuan / kWh.

[0245] The cost of the power deficit is not considered in Target I. Although the average bus voltage deviation of Scheme I is 0.008 less than that of Scheme II, and the active power loss is 973.92kW less than that of Scheme II, the active power deficit of Scheme I is 6810.81kW less than that of Scheme II. However, if we rely too much on the active power output of DG to compensate for the loss, the gain cannot make up for the loss.

[0246] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A switch exchange method for topology reconstruction of distribution network based on sensitivity, characterized in that: Includes steps: S1, obtain the sensitivity matrix in the original topology of the distribution network; S2, close the tie switch in the distribution network to obtain a new ring topology; if the circulating current of the new ring topology is less than a preset value, calculate the sensitivity matrix, bus voltage and compensation power of the new ring topology according to the sensitivity matrix in the original topology of the distribution network, and then obtain the active power loss of the new ring topology; S3, turning on the segmented switches in the distribution network to obtain a new radial topology; if the circulating current of the new radial topology is less than a preset value, calculating the sensitivity matrix, bus voltage and compensation power of the new radial topology, and then obtaining the active power loss of the new radial topology; S4, turn on other segment switches in sequence to obtain other new radial topologies; if the circulating current is less than the set maximum value, calculate the sensitivity matrix, bus voltage and compensation power of each new radial topology, and then obtain the corresponding active power loss; Repeat this step until all segment switches in the ring topology are turned on, and record the active power losses of all radial topologies under the ring topology, and obtain k minimum active power losses and corresponding topology information; S5, closing the switches of other tie lines to obtain the corresponding ring topology, and calculating the active power loss of each radial topology under the ring topology according to steps S2-S4, and obtaining k minimum active power losses and corresponding topology information, until all tie switches in the original topology are closed to obtain m×k topologies; where m is the number of tie switches; S6. Based on the first constraint condition, find the minimum objective function and the topology corresponding to the objective function among the m×k topologies, and use the topology as the optimal topology; S7. Based on the second constraint, calculate the active power and reactive power output of the optimal distributed power source under the optimal topology; In step S2, the specific process of calculating the bus voltage of the new ring topology is: Where V and θ are the bus voltage amplitude and bus phase angle vector at the next moment, respectively; V(0) and θ(0) are the voltage amplitude and phase angle vector at the current moment, respectively; ΔP i , ΔQ i are the active power increment and reactive power increment respectively; φ S is the set of switch nodes; are the sensitivity matrices of voltage amplitude with respect to active power and reactive power injection, respectively; They are the sensitivity matrices of voltage phase angle relative to active power and reactive power injection respectively; In step S2, the process of calculating the circulating current and compensating the apparent power is: In the formula, It is the voltage difference between the feeders on both sides before the tie switch is closed; are the voltages of network nodes i and j before the tie switch is closed; According to Kirchhoff's law and Ohm's law, the relationship between voltage and current is expressed by the following equation; Node i and node j are the nodes where the switch is closed; In the formula, They are the voltages of the compensation network nodes i, j, and T during the closing process of the tie switch; is the current flowing through the line between nodes i and j of the compensation network; They are the impedances on the i side, j side and the line between i and j; ΔP i sc , are the apparent power, active power and reactive power increments injected into node i when the tie switch is closed; are the apparent power, active power and reactive power increments injected into node j when the tie switch is closed; They are ΔP i sc , Sensitivity; The calculation of bus voltage in step S3 is the same as step S2; the calculation process of circulating current and compensation power is: Indicates the circulating current when the section switch is open; They are the voltages of nodes i, j, and T in the compensation network during the opening of the segmented switch; are the voltages of network nodes i and j before the segment switch is turned on; are the apparent power increments of nodes i and j compensated during the opening process of the segmented switch; ΔP i so , They are respectively the active power increment and reactive power increment injected into node i when the simulated switch is turned on; They are respectively the active power and reactive power increments injected into node j when the simulated switch is turned on; They are ΔP i so , sensitivity.

2. The sensitivity-based distribution network topology reconstruction switch exchange method according to claim 1 is characterized in that: In step S1, the calculation process of the sensitivity matrix is: Voltage relative to active power P l and reactive power Q l The partial derivative of is: In the formula, is the element in the i-th row and j-th column of the node admittance matrix; It represents the sensitivity of the voltage amplitude of bus i to the active power injected into bus l; It represents the sensitivity of the voltage amplitude of bus i to the reactive power injected by bus l; V i and θ i represents the voltage amplitude and phase angle of bus i; It represents the sensitivity of the phase angle of bus i to the active power injected into bus l; represents the sensitivity of the phase angle of bus i to the reactive power injected by bus i; represents a complex number, · represents the conjugate of the complex number; represents the sensitivity of the voltage of bus j to the active power injected into bus l; represents the sensitivity of the voltage of bus j to the reactive power injected by bus l; In the distribution network, the original topology is considered to consist of 1 relaxed bus and N PQ buses A network consisting of The relationship between power and bus voltage is: S i represents the apparent power output by node i; , denote the conjugate of the voltage at node i and the voltage at node j respectively; is the element in the i-th row and j-th column of the node admittance matrix.

3. The sensitivity-based distribution network topology reconstruction switch switching method according to claim 1 is characterized in that: The process of calculating active power loss Loss is: Loss is the active network loss at the current moment; V i and V j are the i-th and j-th elements of the voltage amplitude vector V respectively; θ ij are the i-th and j-th elements of the voltage phase angle vector θ; φ B is the set of all nodes; is the element in the i-th row and j-th column of the node admittance matrix.

4. The sensitivity-based distribution network topology reconstruction switch exchange method according to claim 3 is characterized in that: In step S6, the process of finding the minimum objective function is: Loss 0 is the active network loss of the original network; s ij is a binary variable of the branch status; is a binary variable representing the branch status of the original network; W L and W N are the weight vectors used to minimize the active power loss and the number of operations of the switch, respectively.

5. The sensitivity-based distribution network topology reconstruction switch exchange method according to claim 3 is characterized in that: The first constraint condition is: Voltage upper and lower bound constraints: The open branches must be constrained on the ring: The circulation must be less than the maximum circulation constraint: in Indicates the upper and lower bounds of a number; Indicates an open branch; l C Indicates ring; I c Indicates circulation.

6. The sensitivity-based distribution network topology reconstruction switch exchange method according to claim 4 is characterized in that: In step S7, the specific process of calculating the active power and reactive power output of the optimal distributed power source under the optimal topology is: Where W V and W P are the weight coefficients of the minimum voltage average deviation target and the minimum power shortage target respectively; φ D and is the set of all renewable energy nodes and the set of all PQ nodes; is the maximum active power available from renewable energy at node d; P d is the active power output of renewable energy at node d; V ref is the reference voltage of all nodes; n DG and N is the number of all renewable energy nodes and all PQ nodes.

7. The sensitivity-based distribution network topology reconstruction switch exchange method according to claim 6, characterized in that: The second constraint is: P D and Q D are the active power and reactive power vectors output by renewable energy sources, respectively; is the maximum reactive power vector of renewable energy output; and are the minimum and maximum reactive power vectors of renewable energy output, respectively; φ B is the set of all nodes; V i is the magnitude of the voltage at node i; and is the minimum and maximum value of the voltage amplitude at node i.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 7.

Images