A distributed voltage control method for distribution network considering dynamic voltage regulation cost

Abstract

The present invention discloses a distributed control method for distribution network voltage taking into account dynamic voltage regulation cost, including: analyzing the mechanism of voltage regulation of distribution network by distributed photovoltaic inverters and energy storage systems in distribution network, and calculating active and reactive voltage sensitivity coefficient matrix of each node; establishing active unit voltage regulation cost model of energy storage system, and establishing reactive unit cost model of photovoltaic inverter; obtaining unit voltage regulation cost model, and deploying the unit voltage regulation cost model in local controllers of photovoltaic inverter and energy storage system; obtaining consistency parameters through consistency algorithm, and passing them to local controller, and local controller performs cost-optimized distributed control through consistency parameters-unit voltage regulation cost. The present invention can coordinate and plan energy storage and photovoltaic inverters in distribution network under limited communication conditions, and suppress voltage over-limit of distribution network in the lowest cost manner while fully considering the life damage caused by the use of batteries and inverters.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network voltage control, and in particular relates to a distribution network voltage distributed control method taking into account dynamic voltage regulation costs. Background Art

[0002] With the promulgation and implementation of the whole-county photovoltaic policy, the number of distributed photovoltaic grid-connected in China's distribution network is accelerating. Due to the uneven power generation of photovoltaic power generation systems during the day, distribution networks equipped with a high proportion of distributed photovoltaics will have intermittent voltage limit problems, which seriously affects the safe operation of the distribution network. Therefore, it is of great significance to study the voltage control strategy of distribution networks with a high proportion of distributed photovoltaics.

[0003] Since the redundant capacity of the inverter of the photovoltaic system can be used for reactive voltage regulation, and with the advancement of photovoltaic storage integration, the voltage regulation capacity of the photovoltaic power generation-energy storage system needs to be explored. Compared with the traditional equipment transformation method, the method of combining reactive regulation based on photovoltaic inverters with active regulation based on energy storage has the advantage of lower construction and operation costs, which can avoid the high cost of grid transformation. However, the construction and operation of energy storage is still relatively high, and the power charging and discharging under different conditions have a greater impact on the life of energy storage. At the same time, there are many studies and applications based on the advantages of photovoltaic inverter voltage regulation, but its cost is often overlooked. And as more and more photovoltaic generators are connected to the distribution network, the distributed characteristics of more and more network active components put forward high requirements on the data communication and computing capabilities of the distribution management system, and traditional centralized control will no longer be applicable. Summary of the invention

[0004] Purpose of the invention: In order to overcome the deficiencies in the prior art, a distributed control method for distribution network voltage taking into account the dynamic voltage regulation cost is provided, which can coordinate the planning of energy storage and photovoltaic inverters in the distribution network under limited communication conditions, and suppress the voltage exceeding the limit of the distribution network in a lowest-cost manner while fully considering the life damage caused by the use of batteries and inverters.

[0005] Technical solution: To achieve the above-mentioned purpose, the present invention provides a distribution network voltage distributed control method taking into account the dynamic voltage regulation cost, comprising the following steps:

[0006] S1: Analyze the mechanism of voltage regulation of the distribution network by the distributed photovoltaic inverter and energy storage system, and calculate the active and reactive voltage sensitivity coefficient matrix of each node using the impedance value of the distribution network;

[0007] S2: Establish an active unit voltage regulation cost model for the energy storage system based on the battery cumulative damage life model, and establish a reactive unit cost model for the photovoltaic inverter using the inverter life reduction model;

[0008] S3: The unit voltage regulation cost model is obtained by combining the active unit voltage regulation cost model of the energy storage system and the reactive unit cost model of the photovoltaic inverter with the voltage sensitivity coefficient, and the unit voltage regulation cost model is deployed in the local controller of the photovoltaic inverter and the energy storage system;

[0009] S4: The consistency parameters are obtained through the designed consistency algorithm and passed to the local controller. The local controller performs cost-optimal distributed control based on the consistency parameters-unit voltage regulation cost.

[0010] Furthermore, the analysis of the mechanism by which the distributed photovoltaic inverter and the energy storage system of the distribution network regulate the voltage of the distribution network in step S1 is as follows:

[0011] The distributed photovoltaic inverter in the distribution network only adjusts the reactive power output when regulating the voltage, and does not affect the photovoltaic active power output; the energy storage system only adjusts the active power output; the two do not share the controller and grid-connected inverter.

[0012] When the difference between photovoltaic power generation and load power is large enough, the voltage will be higher than the previous node. If the accumulated power difference before the node is positive, the voltage will be higher than the access point voltage. When it reaches a certain level, the voltage will exceed the limit. Therefore, for the line overvoltage caused by high penetration photovoltaics, it can be achieved by reducing the active power reverse transmission on the line and increasing the inductive reactive power flow without affecting the load and photovoltaic active output. In the distribution network, the resistance and reactance are not much different, so active and reactive voltage regulation have the same status.

[0013] Furthermore, the calculation method of the active and reactive voltage sensitivity coefficients in step S1 is:

[0014] The approximate voltage sensitivity is derived based on the Distflow power flow equation. Considering the low voltage level of the distribution network, the charging capacitance, line mutual inductance and leakage reactance of the distribution line are ignored during the calculation. The final voltage sensitivity is:

[0015]

[0016] Where: is the active influence coefficient of the change of active power injected by the active load or inverter at node i on the voltage of node j; is the reactive influence coefficient of the reactive load or inverter reactive output change at node i on the voltage at node j, R N , X N They represent the resistance and reactance values ​​of the line between the upstream node and node N respectively.

[0017] Furthermore, the battery cumulative damage life model in step S2 takes into account that the unreasonable frequent charging and discharging, high-power charging and discharging, and deep discharge of the energy storage system will greatly shorten the service life of the energy storage module, and obtains the unit power cost model of the energy storage system based on the combination of the other part of the energy storage cost, the continuous operation and maintenance cost, and the charging and discharging power loss cost;

[0018] The expression of the active unit voltage regulation cost model of the energy storage system is:

[0019]

[0020] Where: c i BSS is the unit power cost of the energy storage system; is the effective discharge ampere-hours equivalent to the rated condition during the i-th irregular discharge process under non-rated conditions; ΓR is the total effective discharge capacity in the full life cycle under rated conditions; is the maintenance cost per unit power of the energy storage system; is the active loss cost per unit of power generated.

[0021] Furthermore, the inverter life reduction model in step S2 is based on the life short board of the inverter equipment, takes the life of the capacitor as the benchmark of the inverter life, and is based on the life formula of the capacitor under normal working conditions. For the convenience of calculation, the unit cost of reactive voltage regulation of the photovoltaic inverter caused by the shortened inverter life is approximately calculated, thereby obtaining the reactive unit cost of the i-th irregular discharge of the inverter, which is expressed as:

[0022]

[0023] Where: Q PV is the inverter output active power; η is the fitting constant, which is given by the inverter manufacturer’s test; is the maintenance cost per unit power of the PV inverter; It is the active power loss cost per unit power generated by the PV inverter.

[0024] Furthermore, the expression of the unit voltage regulation cost model in step S3 is:

[0025]

[0026] Where: The unit voltage regulation cost of the energy storage system at node i for regulating the voltage at node j; The unit voltage regulation cost of the photovoltaic inverter at node i to regulate the voltage at node j; are the unit power costs of the energy storage system and the PV inverter respectively; is the active voltage sensitivity; is the reactive voltage sensitivity; ΔP ESS The active power added or subtracted when the energy storage system adjusts the voltage; ΔQ INV Reactive power that is added or subtracted when a PV inverter regulates voltage.

[0027] Furthermore, the specific process of the distributed control in step S4 is as follows:

[0028] A1: Use the ring election mechanism to determine the dominant node based on the node voltage exceeding the limit: Node voltage detection is performed. If the voltage exceeds the limit, the ring election algorithm is triggered to perform asynchronous ring leader election to determine the dominant node.

[0029] A2: The dominant node iteratively calculates the consistency parameter - unit voltage regulation cost according to its own voltage over-limit value, and transmits the consistency parameter to the local controller of the photovoltaic inverter and energy storage system through the point-to-point communication network established between nodes;

[0030] A3: The local controller of the photovoltaic inverter and energy storage system calculates the active or reactive power reference output value based on the designed dynamic unit voltage regulation cost model and the received consistency parameters. The calculation method is:

[0031] The reactive power output required by the photovoltaic inverter controller to adjust the voltage of node i to exceed the limit at time t is:

[0032]

[0033] in:

[0034] Where: Only positive values ​​are taken. When the unit voltage regulation cost is lower than the maintenance cost, it is taken as zero. At this time, the inverter does not output additional reactive power.

[0035] Photovoltaic inverters must meet power and power factor constraints:

[0036]

[0037] Where: is the power factor angle of photovoltaic grid-connected point i;

[0038] The energy storage adjusts the active power that needs to be output at time t according to the unit adjustment cost variable:

[0039]

[0040] Where: P ESS,i,t is the active power of energy storage at node i at time t, where P ESS,i,t >0 means energy storage charging; C ESS,iis the energy storage capacity at node i; Δt is the control time interval.

[0041] A4: Based on the active or reactive power reference output value, the local controller of the PV inverter and energy storage system controls the grid-connected inverter to output power.

[0042] Step A1 is as follows: the processor P that detects that the voltage exceeds the limit sends a msg containing node voltage information (its own identifier) ​​to its left neighbor, and then waits for the msg from its right neighbor; when the node processor receives a msg, it checks the received msg. If the voltage value of the msg is greater than the recorded value, it updates the recorded value and forwards the msg, otherwise it does not take any action; until a processor receives a msg containing its own identifier, it declares itself as the leader and sends a termination msg to its left neighbor; when other processors receive a termination msg, they forward this message to their left neighbor, and then terminate as a follower node. After the termination msg is transmitted back to the processor P, the election algorithm officially ends, and the node where the processor P is located is determined to be the dominant node.

[0043] Furthermore, the iterative calculation formula of the consistency parameter in step A2 is:

[0044]

[0045] Where: is the unit voltage regulation cost coefficient of the dominant node at time t; V N,t is the voltage of the dominant node N at time t; a is the parameter that adjusts the convergence speed and accuracy of the control method.

[0046] The present invention utilizes the distributed photovoltaic inverter and energy storage system in the distribution network to regulate the voltage of the distribution network; and uses the unit voltage regulation cost as the consistency parameter, and utilizes the consistency algorithm to coordinate and control the photovoltaic power generation system and the energy storage system. First, the method determines the dominant node based on the ring election algorithm; then the dominant node iteratively calculates the consistency parameter based on the local voltage over-limit value, and transmits the consistency parameter to each energy storage and photovoltaic local controller through the point-to-point communication network established between the nodes; then, each energy storage and photovoltaic local controller calculates the active or reactive power reference output value based on the designed dynamic unit voltage regulation cost model and the received consistency parameter; finally, the energy storage and photovoltaic local controller controls the grid-connected inverter to output power; during this period, the output power of each node is continuously adjusted until the node voltage no longer exceeds the limit.

[0047] The present invention aims to coordinate the control of energy storage and photovoltaic inverters in the distribution network under limited communication conditions, consider the influence of power output and charge state on the life attenuation of batteries and inverters, solve the voltage over-limit problem of the distribution network in the most cost-effective way, take into account the service life of the voltage regulation equipment and the cost of voltage regulation, and have good application value in high-penetration photovoltaic distribution networks.

[0048] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0049] 1. Using a distributed control strategy based on a consistency algorithm, communication is only carried out on adjacent nodes containing inverters or energy storage, greatly reducing the amount of communication data; and the calculation is no longer performed on the centralized controller of the distribution network, but is completed by the local controller of each node, with less computing pressure; and since there is no need for a centralized controller for unified scheduling, the reliability of the entire system is also improved.

[0050] 2. The present invention applies the proposed control algorithm to deploy a unit voltage regulation cost model based on the life reduction model on the local controller of each node photovoltaic inverter and energy storage, and finely considers the dynamic voltage regulation cost of the photovoltaic inverter and energy storage. The unit voltage regulation cost is used as a consistency parameter to ensure that all devices always operate at the lowest unit voltage regulation cost, thereby reducing the overall voltage regulation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of the present invention;

[0052] Figure 2 is a schematic diagram of the network topology of the present invention;

[0053] Figure 3 is a control structure diagram of a photovoltaic grid-connected inverter of the present invention;

[0054] Figure 4 It is a control structure diagram of the energy storage grid-connected inverter of the present invention;

[0055] Figure 5 This is the voltage specific control flow chart of this method;

[0056] Figure 6 is the dominant node voltage diagram under different control strategies;

[0057] Figure 7 It is a comparison chart of control results under three strategies;

[0058] Figure 8 It is the cumulative adjustment cost diagram under three control strategies. DETAILED DESCRIPTION

[0059] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0060] The present invention provides a distributed voltage control method for a distribution network taking into account the dynamic voltage regulation cost, such as Figure 1 As shown, it includes the following steps:

[0061] S1: Analyze the mechanism of voltage regulation of the distribution network by the distributed photovoltaic inverter and energy storage system, and use the impedance value of the distribution network to calculate the active and reactive voltage sensitivity coefficient matrix of each node:

[0062] The distributed photovoltaic inverter in the distribution network only adjusts the reactive power output when regulating the voltage, and does not affect the photovoltaic active power output; the energy storage system only adjusts the active power output; the two do not share the controller and grid-connected inverter;

[0063] When the difference between photovoltaic power generation and load power is large enough, the voltage will be higher than the previous node. If the accumulated power difference before the node is positive, the voltage will be higher than the access point voltage. When it reaches a certain level, the voltage will exceed the limit. Therefore, for the line overvoltage caused by high penetration photovoltaics, it can be achieved by reducing the active power reverse transmission on the line and increasing the inductive reactive power flow without affecting the load and photovoltaic active output. In the distribution network, the resistance and reactance are not much different, so active and reactive voltage regulation have the same status.

[0064] The calculation method of active and reactive voltage sensitivity coefficient is:

[0065] The approximate voltage sensitivity is derived based on the Distflow power flow equation. Considering the low voltage level of the distribution network, the charging capacitance, line mutual inductance and leakage reactance of the distribution line are ignored during the calculation. The final voltage sensitivity is:

[0066]

[0067] Where: is the active influence coefficient of the change of active power injected by the active load or inverter at node i on the voltage of node j; is the reactive influence coefficient of the reactive load or inverter reactive output change at node i on the voltage at node j, R N , X N They represent the resistance and reactance values ​​of the line between the upstream node and node N respectively.

[0068] S2: Based on the battery cumulative damage life model, the active unit voltage regulation cost model of the energy storage system is established, and the reactive unit cost model of the photovoltaic inverter is established using the inverter life reduction model:

[0069] The battery cumulative damage life model takes into account that the unreasonable frequent charging and discharging, high-power charging and discharging, and deep discharge of the energy storage system will greatly shorten the service life of the energy storage module, and the unit power cost model of the energy storage system is obtained based on the combination of the other part of the energy storage cost, the continuous operation and maintenance cost, and the charging and discharging power loss cost;

[0070] The expression of the active unit voltage regulation cost model of the energy storage system is:

[0071]

[0072] Where: c i BSS is the unit power cost of the energy storage system; is the effective discharge ampere-hours equivalent to the rated condition during the i-th irregular discharge process under non-rated conditions; ΓR is the total effective discharge capacity in the full life cycle under rated conditions; is the maintenance cost per unit power of the energy storage system; is the active loss cost per unit of power generated.

[0073] The inverter life reduction model is based on the short life of the inverter equipment. It takes the life of the capacitor as the benchmark of the inverter life. Based on the life formula of the capacitor under normal working conditions, and for the convenience of calculation, the unit cost of the reactive voltage regulation of the photovoltaic inverter caused by the shortened inverter life is approximately calculated, so as to obtain the reactive unit cost of the inverter's i-th irregular discharge. The expression of the reactive unit cost model of the photovoltaic inverter is:

[0074]

[0075] Where: Q PV is the inverter output active power; η is the fitting constant, which is given by the inverter manufacturer’s test; is the maintenance cost per unit power of the PV inverter; It is the active power loss cost per unit power generated by the PV inverter.

[0076] S3: The unit voltage regulation cost model is obtained by combining the active unit voltage regulation cost model of the energy storage system and the reactive unit cost model of the photovoltaic inverter with the voltage sensitivity coefficient, and the unit voltage regulation cost model is deployed in the local controller of the photovoltaic inverter and the energy storage system:

[0077] The expression of the unit voltage regulation cost model is:

[0078]

[0079] Where: The unit voltage regulation cost of the energy storage system at node i for regulating the voltage at node j; The unit voltage regulation cost of the photovoltaic inverter at node i to regulate the voltage at node j; are the unit power costs of the energy storage system and the PV inverter respectively; is the active voltage sensitivity; is the reactive voltage sensitivity; ΔP ESS The active power added or subtracted when the energy storage system adjusts the voltage; ΔQ INV Reactive power that is added or subtracted when a PV inverter regulates voltage.

[0080] S4: The consistency parameters are obtained through the designed consistency algorithm and passed to the local controller. The local controller performs cost-optimal distributed control based on the consistency parameters-unit voltage regulation cost:

[0081] like Figure 5 As shown in Figure 2, the specific process of distributed control is:

[0082] A1: Use the ring election mechanism to determine the dominant node based on the node voltage limit. The network topology is as follows: Figure 2 As shown, the steps are as follows: perform node voltage detection. If the voltage exceeds the limit, trigger the ring election algorithm to perform asynchronous ring leader election and determine the dominant node;

[0083] The processor P that detects that the voltage exceeds the limit sends a msg containing the node voltage information (its own identifier) ​​to its left neighbor, and then waits for the msg from its right neighbor; when the node processor receives a msg, it checks the received msg. If the voltage value of the msg is greater than the recorded value, it updates the recorded value and forwards the msg, otherwise it does not take any action; until a processor receives a msg containing its own identifier, it declares itself as the leader and sends a termination msg to its left neighbor; when other processors receive a termination msg, they forward this message to their left neighbor, and then terminate as a follower node. After the termination msg is transmitted back to processor P, the election algorithm officially ends, and the node where processor P is located is determined to be the dominant node.

[0084] A2: The dominant node iteratively calculates the consistency parameter - unit voltage regulation cost according to its own voltage over-limit value, and transmits the consistency parameter to the local controller of the photovoltaic inverter and energy storage system through the point-to-point communication network established between nodes;

[0085] The iterative calculation formula for the consistency parameter is:

[0086]

[0087] Where: is the unit voltage regulation cost coefficient of the dominant node at time t; V N,t is the voltage of the dominant node N at time t; a is the parameter that adjusts the convergence speed and accuracy of the control method.

[0088] A3: The local controller of the PV inverter and the energy storage system calculates the active or reactive power reference output value based on the designed dynamic unit voltage regulation cost model and the received consistency parameters;

[0089] The calculation method of active or reactive power reference output value is:

[0090] The reactive power output required by the photovoltaic inverter controller to adjust the voltage of node i to exceed the limit at time t is:

[0091]

[0092] in:

[0093] Where: Only positive values ​​are taken. When the unit voltage regulation cost is lower than the maintenance cost, it is taken as zero. At this time, the inverter does not output additional reactive power.

[0094] Photovoltaic inverters must meet power and power factor constraints:

[0095]

[0096] Where: is the power factor angle of photovoltaic grid-connected point i;

[0097] The energy storage adjusts the active power that needs to be output at time t according to the unit adjustment cost variable:

[0098]

[0099] Where: P ESS,i,t is the active power of energy storage at node i at time t, where P ESS,i,t >0 means energy storage charging; C ESS,i is the energy storage capacity at node i; Δt is the control time interval.

[0100] A4: According to the active or reactive power reference output value, the local controller of the photovoltaic inverter and the energy storage system controls the grid-connected inverter to output power. The control structure of the photovoltaic grid-connected inverter is as follows: Figure 3As shown in the figure, it is a two-stage grid-connected inverter. The photovoltaic cell is connected to the grid through the Boost chopper circuit and the grid-connected inverter. The Boost stage realizes the MPPT and power limit switching control functions. The grid-connected inverter realizes the fixed reactive power control function and stabilizes the DC voltage to ensure that the Boost output active power is sent to the grid. The control structure of the energy storage grid-connected inverter is as follows: Figure 4 As shown, the rotor motion equation is embedded to realize virtual synchronous machine control, combined with voltage and current dual-loop control to ensure that the full-bridge output active power is sent to the grid.

[0101] In order to verify the effect of the method of the present invention, the method of the present invention is compared with the existing method in this example:

[0102] Based on the IEEE33 distribution network system, Matlab was used to simulate and analyze the suppression effect of the present invention on the distribution network. Assuming Δt = 5min, the capacity penetration rate of the system photovoltaic is 113.6%. During the period of 9:55-13:50, the node has a voltage limit problem. The present invention is compared with the existing local control and two-stage control. Figure 6 The 18-node voltage curves under the three control strategies are shown in the figure. It can be seen that the three control strategies use photovoltaic inverters and energy storage to effectively suppress the voltage over-limit problem. The main difference in their control effects lies in the control cost. Figure 7 It can be seen that the decentralized local control method only calls the energy storage and inverter of nodes 16, 17, and 18 for voltage regulation, while the other two-stage voltage control method calls the energy storage and inverter of all nodes for voltage regulation. Although the strategy designed by the present invention also calls the energy storage and inverter of most nodes for voltage regulation, compared with the two-stage method based on utilization, the power allocation of the strategy of the present invention takes into account the voltage regulation cost of different devices and the voltage sensitivity of different nodes.

[0103] In terms of inverter control, taking into account the characteristic that the higher the output power, the greater the service life loss, the present invention draws on the advantages of two-stage control, distributes the reactive power demand more evenly to each inverter, and enables the energy storage and inverter at nodes with higher voltage sensitivity to participate more in voltage regulation, and vice versa, reduce their participation in voltage regulation, and the shortfall is mainly made up by the energy storage system at the node with higher voltage sensitivity, avoiding an increase in the average unit power cost of the inverter.

[0104] In terms of energy storage control, since the inverter is called first for voltage regulation when the voltage exceeds the limit, the demand for active voltage regulation of energy storage is relatively small. The energy storage system near the voltage-exceeding node with high voltage sensitivity is mainly called for voltage regulation. At the same time, the cost increase caused by the increase in energy storage SOC is taken into account. Compared with decentralized on-site control, more nodes of energy storage are called, and at the same time, 82kW of active power is called less than the two-stage control strategy, thereby fully reducing the voltage regulation cost.

[0105] from Figure 8 The cumulative voltage regulation costs of different control strategies are shown in the figure. It can be seen that the cumulative curve trends of the three control costs are basically the same, but the unit voltage regulation cost of the control strategy designed by the present invention is always lower than that of other methods. The control strategy of the present invention has a significant advantage in total cost. By using the unit voltage regulation cost as the consistency parameter of the consistency algorithm, the cost control effect of the strategy designed in this paper is significantly better than that of other methods.

[0106] Table 1 Comparison of different strategy parameters

[0107]

[0108] As shown in Table 1, by comparing the voltage regulation effects of the three control strategies, it can be seen that: compared with local control, the total reactive power regulation of the inverter increases by 553kVar, and the average unit reactive cost is close, so the total reactive voltage regulation cost is 56.4 yuan higher; but the active output power of the energy storage is reduced by 114kW, and the average unit active cost is only 32.9% compared with local control, which greatly reduces the active voltage regulation cost to only 30 yuan, making the total voltage regulation cost of the control strategy designed by the present invention only 51.1% of the local control. Compared with the two-stage control, the total active and reactive unit costs increase by 12.5% ​​and 5.7% respectively compared with the two-stage control, but the required power decreases by 87.3% and 44.8%, thereby reducing the final total cost by 43.5%. In summary, it can be seen that the control method designed by the present invention can effectively solve the voltage limit problem caused by distributed photovoltaic grid connection, and has greater economic advantages than other control methods.

Claims

1. A distributed voltage control method for a distribution network taking into account the dynamic voltage regulation cost, characterized in that: The steps include: S1: Analyze the mechanism of voltage regulation of the distribution network by the distributed photovoltaic inverter and energy storage system, and calculate the active and reactive voltage sensitivity coefficient matrix of each node using the impedance value of the distribution network; S2: Establish an active unit voltage regulation cost model for the energy storage system based on the battery cumulative damage life model, and establish a reactive unit cost model for the photovoltaic inverter using the inverter life reduction model; S3: The unit voltage regulation cost model is obtained by combining the active unit voltage regulation cost model of the energy storage system and the reactive unit cost model of the photovoltaic inverter with the voltage sensitivity coefficient, and the unit voltage regulation cost model is deployed in the local controller of the photovoltaic inverter and the energy storage system; S4: Obtain consistency parameters through the designed consistency algorithm and pass them to the local controller. The local controller performs cost-optimal distributed control based on the consistency parameter-unit voltage regulation cost. The calculation method of active and reactive voltage sensitivity coefficient in step S1 is: The approximate voltage sensitivity is derived based on the Distflow power flow equation. Considering the low voltage level of the distribution network, the charging capacitance, line mutual inductance and leakage reactance of the distribution line are ignored during the calculation. The final voltage sensitivity is: Where: is the active influence coefficient of the change of active power injected by the active load or inverter at node i on the voltage of node j; is the reactive influence coefficient of the reactive load or inverter reactive output change at node i on the voltage at node j, R N , X N Respectively represent the resistance and reactance values ​​of the line between the upstream node and node N; The expression of the active unit voltage regulation cost model of the energy storage system in step S2 is: Where: c i BSS is the unit power cost of the energy storage system; is the effective discharge ampere-hours equivalent to the rated condition during the i-th irregular discharge process under non-rated conditions; ΓR is the total effective discharge capacity in the full life cycle under rated conditions; is the maintenance cost per unit power of the energy storage system; is the cost of active losses per unit of power generated; The expression of the reactive power unit cost model of the photovoltaic inverter in step S2 is: Where: Q PV is the inverter output reactive power; η is the fitting constant; is the maintenance cost per unit power of the PV inverter; is the reactive power loss cost per unit power generated by the PV inverter; The expression of the unit voltage regulation cost model in step S3 is: Where: The unit voltage regulation cost of the energy storage system at node i for regulating the voltage at node j; The unit voltage regulation cost of the photovoltaic inverter at node i to regulate the voltage at node j; are the unit power costs of the energy storage system and the PV inverter respectively; is the active voltage sensitivity; is the reactive voltage sensitivity; ΔP ESS The active power added or subtracted when the energy storage system adjusts the voltage; ΔQ INV Reactive power added or subtracted when the photovoltaic inverter regulates voltage; The specific process of distributed control in step S4 is: A1: Use the ring election mechanism to determine the dominant node based on the node voltage exceeding the limit: Node voltage detection is performed. If the voltage exceeds the limit, the ring election algorithm is triggered to perform asynchronous ring leader election to determine the dominant node. A2: The dominant node iteratively calculates the consistency parameter - unit voltage regulation cost according to its own voltage over-limit value, and transmits the consistency parameter to the local controller of the photovoltaic inverter and energy storage system through the point-to-point communication network established between nodes; A3: The local controller of the PV inverter and the energy storage system calculates the active or reactive power reference output value based on the designed dynamic unit voltage regulation cost model and the received consistency parameters; A4: According to the active or reactive power reference output value, the local controller of the photovoltaic inverter and the energy storage system controls the grid-connected inverter to output power; The iterative calculation formula of the consistency parameter in step A2 is: Where: is the unit voltage regulation cost coefficient of the dominant node at time t; V N,t is the voltage of the dominant node N at time t; a is the parameter for adjusting the convergence speed and accuracy of the control method; Step A1 is specifically as follows: The processor P that detects that the voltage exceeds the limit sends a msg containing the node voltage information to its left neighbor, and then waits for the msg from its right neighbor; when the node processor receives a msg, it checks the received msg. If the voltage value of the msg is greater than the recorded value, it updates the recorded value and forwards the msg, otherwise it does not take any action; until a processor receives a msg containing its own identifier, it declares itself as the leader and sends a termination msg to its left neighbor; when other processors receive a termination msg, they forward this message to their left neighbor and then terminate as a follower node. After the termination msg is transmitted back to processor P, the election algorithm officially ends, and the node where processor P is located is determined to be the dominant node; Step A3 The calculation method of the active or reactive power reference output value is: The reactive power output required by the photovoltaic inverter controller to adjust the voltage of node i to exceed the limit at time t is: in: Where: Only positive values ​​are taken. When the unit voltage regulation cost is lower than the maintenance cost, it is taken as zero. At this time, the inverter does not output additional reactive power. Photovoltaic inverters must meet power and power factor constraints: Where: is the power factor angle of photovoltaic grid-connected point i; The energy storage adjusts the active power that needs to be output at time t according to the unit adjustment cost variable: Where: P ESS,i,t is the active power of energy storage at node i at time t, where P ESS,i,t >0 means energy storage charging; C ESS,i is the energy storage capacity at node i; Δt is the control time interval.

2. A method for distributed voltage control of a distribution network taking into account dynamic voltage regulation costs according to claim 1, characterized in that: The analysis of the mechanism by which the distributed photovoltaic inverter and the energy storage system of the distribution network regulate the voltage of the distribution network in step S1 is as follows: The distributed photovoltaic inverter in the distribution network only adjusts the reactive power output when regulating the voltage, and does not affect the photovoltaic active power output; the energy storage system only adjusts the active power output; the two do not share the controller and grid-connected inverter.

Images