A power distribution network voltage coordination control method for high proportion of new energy access

Abstract

The application relates to the technical field of power distribution network operation control, and discloses a power distribution network voltage cooperative control method for high-proportion new energy access, which comprises the following steps: collecting multi-source operation data of a power distribution network; selecting a disturbance source device from fast-response equipment and outputting a small-amplitude reactive power disturbance, and collecting voltage response data; generating voltage response fingerprints according to the voltage response data, comparing the voltage response fingerprints with expected response characteristics generated based on account topology data, and obtaining a topology consistency verification result; generating a virtual electrical topology according to the topology consistency verification result, the account topology data and the voltage response fingerprints, and constructing a voltage sensitivity atlas; identifying a weak support area based on the voltage sensitivity atlas, determining a network-type energy storage converter as a virtual voltage anchor point; dividing a voltage control area with the virtual voltage anchor point as the center, calculating an adjustment priority, and executing layered cooperative adjustment of the fast-response equipment and slow-response equipment. The method can improve the accuracy and adaptability of power distribution network voltage control.

Description

Technical Field

[0001] This invention relates to the field of distribution network operation and control technology, and in particular to a distribution network voltage coordinated control method for high proportion of new energy access. Background Technology

[0002] With the increasing proportion of distributed photovoltaic (PV), decentralized wind power, user-side energy storage, electric vehicle charging loads, and flexible adjustable loads in distribution networks, the operating mode of distribution networks is gradually transforming from a traditional unidirectional power flow network to an active distribution network characterized by multi-source access, multidirectional power flow, and strong fluctuations. Under conditions of high-proportion renewable energy access, renewable energy output is significantly affected by changes in sunlight, wind speed, and user-side loads. The operating status of the distribution network exhibits strong uncertainty and time-varying characteristics, which can easily lead to problems such as voltage rise at the end of lines, local voltage exceeding limits, reverse power flow, three-phase imbalance, frequent voltage fluctuations, and increased frequency of voltage regulation equipment operation.

[0003] Currently, voltage control in distribution networks typically relies on equipment such as renewable energy inverters, energy storage converters, dynamic reactive power compensation devices, voltage regulating transformers equipped with OLTC (On-Line Voltage Regulator), capacitor banks, and line voltage regulators. While these devices can improve the voltage operation of the distribution network to some extent, in scenarios with high renewable energy penetration, rapid load changes, and weak local grid support capabilities, the regulation capabilities, response speeds, and impact ranges of various voltage regulation resources differ. Without effective coordination, problems such as inaccurate resource allocation, insufficient local voltage regulation effects, or uncoordinated regulation actions between different devices can easily occur.

[0004] Meanwhile, discrepancies may exist between the actual operating status of the distribution network and the records. For example, the wiring relationships in low-voltage distribution areas are complex, the operating mode of tie switches may change, and line parameters and equipment connection locations may be outdated. In such cases, if the voltage control process relies excessively on existing record data or static models, it may lead to inconsistencies between the controlled objects, regulation resources, and actual electrical influences, thereby affecting the accuracy and reliability of voltage control.

[0005] Furthermore, in scenarios with a high proportion of renewable energy integration, the voltage support capabilities vary significantly across different areas. Some feeder ends, low-voltage distribution areas, or areas with concentrated renewable energy integration may experience issues such as low short-circuit capacity, insufficient reactive power support, or sensitivity to voltage fluctuations. When the voltage support capability in a localized area is insufficient, relying solely on conventional voltage regulation methods is insufficient to meet voltage control requirements in a timely and accurate manner, and may even lead to mutual interference between local regulation and the operating status of adjacent areas.

[0006] Traditional mechanical voltage regulators are typically suitable for handling voltage deviations on slower timescales, while fluctuations in renewable energy output and charging loads can occur on shorter timescales. Frequent involvement of mechanical voltage regulators in rapid fluctuation regulation can increase the number of operations and reduce their lifespan. Conversely, relying solely on fast-acting power electronic devices may struggle to maintain long-term voltage compliance during periods of persistent voltage deviation or insufficient regulation capacity. Therefore, coordinating controllable devices with varying response speeds, regulation capabilities, and electrical influence ranges is a crucial issue that needs to be addressed in voltage control for high-proportion renewable energy integration into distribution networks.

[0007] Therefore, it is necessary to provide a distribution network voltage coordination control method that can adapt to high proportion of new energy access, changes in distribution network operating status, and differences in local voltage support capabilities, so as to improve the accuracy, stability, and equipment coordination level of distribution network voltage control. Summary of the Invention

[0008] To address the shortcomings of existing technologies, the present invention aims to provide a voltage coordination control method for distribution networks with high proportions of renewable energy access. This method improves voltage control accuracy, coordination regulation stability, and voltage regulation equipment reliability in situations where the distribution network topology is uncertain, renewable energy output fluctuates, and local voltage support capacity is insufficient.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a distribution network voltage coordinated control method for high-proportion renewable energy access, comprising: S1 collects multi-source operation data of the distribution network, including node voltage, new energy output, energy storage status, voltage regulating equipment status, and ledger topology data, to determine the status of controllable equipment; controllable equipment includes fast-response equipment and slow-response equipment. S2, Select a disturbance source device from the fast response device, control its output to meet the preset voltage safety constraints for small reactive disturbances, and collect voltage response data from each voltage monitoring node; S3, generate a voltage response fingerprint based on the voltage response data, generate the expected response characteristics of each voltage monitoring node relative to the disturbance source device based on the ledger topology data, and compare them to obtain the topology consistency verification result; S4. Based on the topology consistency verification result, ledger topology data and voltage response fingerprint, generate a virtual electrical topology and construct a voltage sensitivity map containing the voltage influence edge weights between controllable devices and voltage monitoring nodes. S5. Identify weak support areas based on the voltage sensitivity spectrum and multi-source operation data of the distribution network, and determine energy storage converters with network control capabilities as virtual voltage anchors from the weak support areas. S6, with the virtual voltage anchor point as the center, divide the voltage control area according to the voltage response fingerprint, virtual electrical topology and voltage sensitivity spectrum; S7. When there is a voltage over-limit node in the voltage control zone, calculate the adjustment priority of the controllable equipment according to the voltage sensitivity spectrum and the multi-source operation data of the distribution network, and allocate voltage adjustment tasks according to the adjustment priority. S8 performs hierarchical collaborative regulation of fast-response and slow-response devices according to the voltage regulation task, and updates the voltage response fingerprint, virtual electrical topology, voltage sensitivity spectrum, virtual voltage anchor point, voltage control area and regulation priority according to the regulated node voltage and controllable device status.

[0010] Furthermore, the fast-response equipment includes a new energy inverter, an energy storage converter, and a dynamic reactive power compensation device; the slow-response equipment includes a voltage regulating transformer equipped with an OLTC, a capacitor bank, and a line voltage regulator; the controllable equipment status includes the remaining regulation capacity, communication status, protection status, lockout status, and historical regulation burden of the controllable equipment.

[0011] Further, in step S2, the disturbance source device is determined from the fast response device according to the controllable device status, and the disturbance source device satisfies the following conditions: the node voltage of its node does not exceed the preset severe over-limit threshold, the remaining reactive capacity is greater than the preset reactive capacity threshold, the communication status is normal, the protection status is normal and not in a blocked state. The amount of the small reactive power disturbance shall not exceed the minimum value among the rated reactive power capacity, the remaining reactive power capacity, and the maximum safe disturbance amount that meets the preset voltage safety constraints of the disturbance source equipment.

[0012] Further, in step S3, the voltage response data includes voltage response amplitude, voltage response direction, voltage response delay, and voltage recovery slope; The voltage response fingerprint is generated based on the voltage response amplitude, voltage response direction, voltage response delay, and voltage recovery slope, and is used to characterize the actual electrical coupling relationship between the voltage monitoring node and the disturbance source device.

[0013] Further, according to the disturbance source device selection conditions in step S2, multiple disturbance source devices are sequentially determined from the fast response devices, and each disturbance source device is controlled to output the small reactive disturbance to generate a corresponding voltage response fingerprint. The voltage response fingerprints generated by the same voltage monitoring node under multiple disturbance source devices are compared with the corresponding expected response features. When the difference between the voltage response fingerprint and the corresponding expected response feature exceeds a preset topology verification threshold, and the difference occurs in no less than a preset number of comparisons, a topology anomaly result is generated, and the electrical connection relationship used to generate the virtual electrical topology is corrected based on the topology anomaly result.

[0014] Further, in step S4, the ledger topology data includes switch status and line impedance. The virtual electrical topology is generated after correcting the electrical connection relationships in the ledger topology data based on the topology consistency verification result, the switch status, the line impedance, and the voltage response fingerprint. The voltage influence weights are determined based on the reactive voltage sensitivity, active voltage sensitivity, response coupling strength, and historical regulation burden of the controllable device to the voltage monitoring node, wherein the response coupling strength is determined by the voltage response fingerprint.

[0015] Further, in step S5, multiple voltage monitoring nodes are divided into candidate regions based on the electrical connection relationships and voltage response fingerprint similarity in the virtual electrical topology; When two candidate regions have the same voltage monitoring node, the same voltage monitoring node is assigned to the candidate region with the largest sum of edge weights of the voltage influence of the controllable device on the same voltage monitoring node in the region, so as to obtain the candidate region after deduplication. Based on the fact that at least two of the following parameters in the candidate region after deduplication—new energy penetration rate, node voltage deviation, reverse power flow duration, available reactive power capacity, short-circuit capacity, voltage response amplitude, and number of slow-response device actions—meet the corresponding preset threshold conditions, the weak support region is identified so that there are no identical voltage monitoring nodes among the weak support regions.

[0016] Further, in step S5, determining the virtual voltage anchor point for each weak support region includes: From the weak support region, select energy storage converters that have grid control capabilities and whose energy storage status meets preset safety constraints as candidate energy storage converters. The preset safety constraints include state of charge constraints, health state constraints, temperature state constraints, communication state constraints, and protection state constraints. The anchor evaluation value is calculated based on the voltage support leverage value, remaining available power, state-of-charge safety margin, health status, historical regulation burden, and virtual electrical distance determined by the virtual electrical topology and voltage response fingerprint of the candidate energy storage converter for the weakly supported region. The voltage support lever value is determined based on the edge weight of the voltage influence of the candidate energy storage converter on the voltage monitoring node in the weak support area. The candidate energy storage converter with the highest anchor evaluation value was selected as the main virtual voltage anchor point.

[0017] Furthermore, in step S6, voltage monitoring nodes that meet at least two zoning conditions are categorized into a voltage control zone centered on the virtual voltage anchor point; The zoning conditions include: the response strength of the voltage monitoring node to the virtual voltage anchor point determined by the voltage response fingerprint is greater than a preset response strength threshold; the edge weight of the voltage influence of the virtual voltage anchor point on the voltage monitoring node is greater than a preset edge weight threshold; the virtual electrical distance between the voltage monitoring node and the virtual voltage anchor point is less than a preset distance threshold; and the similarity between the voltage response fingerprint of the voltage monitoring node and the comprehensive voltage response fingerprint of the node corresponding to the virtual voltage anchor point is greater than a preset similarity threshold. When there is no already assigned node, the voltage monitoring node corresponding to the virtual voltage anchor point grid-connected node is used as the initial assigned node; When the same voltage monitoring node meets the zoning conditions of multiple voltage control zones, the same voltage monitoring node is assigned to the voltage control zone with the largest edge weight of the voltage influence of the corresponding virtual voltage anchor point on the same voltage monitoring node, so that there are no identical voltage monitoring nodes between the voltage control zones.

[0018] Furthermore, in steps S7 and S8, the adjustment priority is determined based on the voltage influence weight of the controllable device on the voltage over-limit node, the remaining adjustment capacity, the active power fluctuation contribution, and the historical adjustment burden. When the controllable device does not contribute to active power fluctuations, the active power fluctuation contribution is set to a preset value; when the voltage deviation of the voltage over-limit node is not greater than a preset first voltage deviation threshold, the fast response device is given priority to perform the voltage regulation task; when the duration of the voltage deviation of the voltage over-limit node being greater than a preset second voltage deviation threshold exceeds a preset duration threshold, the slow response device is activated to perform the voltage regulation task, and the preset second voltage deviation threshold is greater than the preset first voltage deviation threshold. Before the slow-response device performs a voltage regulation task, the voltage deviation change of key voltage monitoring nodes in adjacent voltage control areas is predicted based on the virtual electrical topology and voltage sensitivity map, and the voltage regulation task is performed when the voltage deviation change does not exceed a preset deterioration threshold.

[0019] The beneficial effects of this invention are: (1) The present invention obtains voltage response data by small reactive disturbance and generates voltage response fingerprints, which are used to verify the consistency between the ledger topology and the actual electrical coupling relationship, which is beneficial to improving the accuracy of the distribution network voltage control model.

[0020] (2) The present invention generates a virtual electrical topology based on the topology consistency verification results, ledger topology data and voltage response fingerprint, and constructs a voltage sensitivity spectrum so that the regulation of controllable equipment can match the actual operating state, thereby reducing ineffective regulation.

[0021] (3) Based on voltage sensitivity spectrum and multi-source operation data of distribution network, this invention identifies weak support areas and determines the grid-type energy storage converter as virtual voltage anchor point, which is beneficial to improve the voltage support capability of local weak support areas.

[0022] (4) The present invention dynamically divides the voltage control zone with the virtual voltage anchor point as the center, so that the voltage control zone can be adjusted according to the changes in new energy output, switching status and electrical coupling relationship, thereby improving the adaptability of voltage coordinated control.

[0023] (5) The present invention calculates the adjustment priority of controllable equipment based on voltage influence side weight and multi-source operation data of distribution network, and performs hierarchical coordinated adjustment of fast response equipment and slow response equipment, which is conducive to improving voltage regulation efficiency, reducing the frequent operation of traditional voltage regulation equipment, and improving equipment operation reliability. Attached Figure Description

[0024] Figure 1 This is a flowchart of the steps in the distribution network voltage collaborative control method for high-proportion renewable energy access in this invention. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0026] Example 1, as Figure 1 As shown, this embodiment provides a distribution network voltage coordination control method for high-proportion renewable energy access, applied to a 10kV distribution feeder and its downstream multiple 0.4kV transformer substations. The 10kV distribution feeder connects to six 0.4kV transformer substations, three of which have distributed photovoltaic (PV) installed capacity reaching 80% of their transformer capacity. A user-side energy storage system is connected at the end of the feeder, comprising lithium iron phosphate battery clusters and an energy storage converter with grid-connected control capabilities; a dynamic reactive power compensation device (SVG) is configured in the middle of the feeder; and the distribution transformer is equipped with an on-load tap changer (OLTC) voltage regulating transformer. Due to multiple modifications to the low-voltage transformer substation wiring, there are inconsistencies between the registered locations of some renewable energy access points and the actual wiring relationships.

[0027] The hardware used in this embodiment includes a voltage and current acquisition terminal, a smart distribution area fusion terminal, a photovoltaic inverter communication module, an energy storage converter controller, a dynamic reactive power compensation equipment controller, an on-load tap changer (OLTC) controller, an edge computing control host, and a distribution automation master station. The voltage and current acquisition terminal uses a Class 0.2 power quality acquisition module with a sampling period of 1 second. The photovoltaic inverter and energy storage converter upload status data via Modbus TCP or IEC 104 communication protocol with a sampling period of 2 seconds. The status sampling period for voltage regulating transformers, line switches, and capacitor banks equipped with OLTC is 30 seconds. The edge computing control host uses an industrial computer with at least an 8-core processor, 16GB of RAM, and a 512GB solid-state drive, running a Linux real-time system and a distribution network voltage collaborative control program. The energy storage system has a rated power of 500kW and a rated capacity of 1MWh. The energy storage converter has a rated capacity of 500kVA and supports V / f control, droop control and reactive power regulation. The dynamic reactive power compensation equipment has a rated capacity of 300kvar. The voltage regulating transformer equipped with OLTC has a capacity of 1000kVA and the OLTC voltage regulation range is ±5×2.5%.

[0028] The operating conditions for this embodiment are: From 12:00 to 13:30, the ambient temperature was 32℃, and the photovoltaic irradiance was 900W / m². 2 Up to 1000W / m 2 The user load is approximately 30% of the transformer capacity in the distribution area, and reverse power flow occurs at the end of the feeder. The allowable operating range of node voltage is set to 0.93 times the rated voltage to 1.07 times the rated voltage, and the severe over-limit threshold is set to be higher than 1.10 times the rated voltage or lower than 0.90 times the rated voltage; the proportion of small reactive power disturbance is set to 3% of the rated reactive power capacity of the disturbance source equipment, the disturbance duration is set to 5s, and the disturbance interval is set to 120s. The preset voltage safety constraints are as follows: For voltage monitoring nodes that have not exceeded limits before the disturbance, the node voltage during and after the disturbance should be maintained between 0.93 times and 1.07 times the rated voltage; for voltage monitoring nodes that have exceeded limits before the disturbance but have not yet reached the severe limit threshold, the disturbance must not cause their voltage deviation to increase further relative to before the disturbance, and must not cause the node voltage to exceed the severe limit threshold. The topology verification threshold is set to a comprehensive difference value of 0.35, the topology consistency confidence threshold is set to 0.60, and the number of consecutive confirmations is set to 3.

[0029] In the weak support area identification, the new energy penetration rate threshold is set to 60%, the node voltage deviation threshold is set to 5% of the rated voltage, the reverse power flow duration threshold is set to 5 minutes, the available reactive power capacity threshold is set to 20% of the region's rated reactive power demand, the short-circuit capacity threshold is set to 3 times the new energy installed capacity, the voltage response amplitude threshold is set to 0.5%, and the slow response equipment action frequency threshold is set to 80% of the daily allowable action frequency. In the energy storage safety constraints, the state of charge range is set to 20% to 90%, the health status threshold is set to 80%, the upper temperature limit is set to 45℃, the upper communication delay limit is set to 200ms, and the protection status requirement is no alarms. In dynamic zoning, the response intensity threshold is set to 0.60, the voltage influence edge weight threshold is set to 0.60, the virtual electrical distance threshold is set to 0.40, and the response fingerprint similarity threshold is set to 0.75. In hierarchical control, the first voltage deviation threshold is set to 2% of the rated voltage, the second voltage deviation threshold is set to 5% of the rated voltage, the duration threshold is set to 5 minutes, and the adjacent control area deterioration threshold is set to 1% of the rated voltage.

[0030] The core of this embodiment lies in: Voltage response fingerprints are generated using small reactive power disturbances, and topology self-verification is performed based on these fingerprints. A virtual electrical topology is generated based on the self-verification results, and a voltage sensitivity map is constructed based on the virtual electrical topology. Weak support regions are identified based on the voltage sensitivity map and operating status. Within the weak support regions, energy storage with grid control capabilities is selected as virtual voltage anchor points. Voltage control zones are dynamically divided with the virtual voltage anchor points as the center. Within the voltage control zones, regulation tasks are allocated based on voltage influence weights, remaining regulation capacity, historical regulation burden, and contribution from renewable energy fluctuations. When fast-response devices are insufficient in regulation, slow-response devices are triggered, and the topology, map, anchor points, and control zones are updated based on control feedback.

[0031] Specifically, the edge computing control host collects multi-source operational data from the distribution network. This multi-source operational data includes node voltages, renewable energy output, energy storage status, voltage regulating equipment status, and ledger topology data, used to determine the status of controllable equipment. Node voltages include the voltages at the beginning, middle, and end of 10kV feeders, as well as the voltages of key nodes in each 0.4kV distribution area. New energy output includes the active power output, reactive power output, rated capacity, and remaining reactive power capacity of distributed photovoltaic inverters; energy storage status includes the energy storage's state of charge, health status, temperature, remaining active power regulation capacity, remaining reactive power regulation capacity, communication status, and protection status; voltage regulation equipment status includes the output status of dynamic reactive power compensation equipment, the tap position of voltage regulating transformers equipped with OLTCs, capacitor bank switching status, and line voltage regulator status; ledger topology data includes feeder connection relationships, transformer substation affiliation relationships, switch status, and line impedance. Controllable equipment includes fast-response equipment and slow-response equipment. Fast-response equipment includes new energy inverters, energy storage converters, and dynamic reactive power compensation equipment; slow-response equipment includes voltage regulating transformers equipped with OLTCs, capacitor banks, and line voltage regulators. Controllable equipment status includes the remaining regulation capacity, communication status, protection status, lockout status, and historical regulation load of controllable equipment.

[0032] The edge computing control host selects a disturbance source device from the fast-response devices based on multi-source operation data of the distribution network. It controls the disturbance source device to output a small reactive power disturbance that meets preset voltage safety constraints and collects voltage response data from each voltage monitoring node. In this embodiment, the current voltage of the node where the energy storage converter at the feeder end is located is 1.08 times the rated voltage, which has not yet exceeded the severe over-limit threshold of 1.10 times the rated voltage. The energy storage's state of charge is 58%, its healthy state is 95%, the temperature is 36℃, the communication delay is 80ms, the protection status is normal, and the energy storage converter has network control capabilities. Therefore, the system selects the energy storage converter at the feeder end as the disturbance source device. The small reactive power disturbance is determined according to the minimum value among the rated reactive power capacity, the remaining reactive power capacity, and the maximum safe disturbance that meets voltage safety constraints. In this embodiment, it is taken as 3% of the rated reactive power capacity of the energy storage converter, and the disturbance lasts for 5 seconds. The power flow forecast results show that during the disturbance, the voltage of the non-over-limit node will not exceed the safe range of 0.93 times to 1.07 times the rated voltage, and the voltage deviation of the over-limit feeder end node will not increase further relative to before the disturbance. Furthermore, the voltage of the feeder end node will not exceed the severe over-limit threshold of 1.10 times the rated voltage. Therefore, the disturbance is permitted.

[0033] During periods of small reactive power disturbances from the source equipment, the system collects the voltage response amplitude, voltage response direction, voltage response delay, and voltage recovery slope of each voltage monitoring node, and generates a voltage response fingerprint based on this data. The voltage response fingerprint characterizes the actual electrical coupling relationship between the voltage monitoring node and the source equipment. Simultaneously, the system generates expected response characteristics of each voltage monitoring node relative to the source equipment based on the ledger topology data. These expected response characteristics include expected response amplitude, expected response direction, and expected response delay. Subsequently, the system compares the voltage response fingerprint with the expected response characteristics to obtain the topology consistency verification result. In actual operation, some nodes recorded as adjacent areas in the ledger exhibit larger voltage response amplitudes and shorter response delays to end-point energy storage disturbances, while some nodes recorded as belonging to the same area in the ledger show smaller response amplitudes and longer response delays, indicating a discrepancy between the ledger topology and the actual electrical coupling relationship.

[0034] The system generates a virtual electrical topology based on topology consistency verification results, ledger topology data, and voltage response fingerprints, and constructs a voltage sensitivity map containing voltage influence weights between controllable devices and voltage monitoring nodes. In this embodiment, the system sequentially applies small disturbances to multiple fast-response devices. If, for three consecutive occurrences at the same voltage monitoring node, the combined difference between the voltage response fingerprint and the expected response characteristics is greater than 0.35 and the topology consistency confidence value is less than 0.60, the system determines that the node has a ledger topology anomaly. The system combines switch status, line impedance, and voltage response fingerprints to correct the electrical connections in the ledger topology, generating a virtual electrical topology. Subsequently, the system calculates voltage influence weights based on reactive voltage sensitivity, active voltage sensitivity, response coupling strength, and historical regulation burden. A larger voltage influence weight indicates a stronger voltage regulation effect of the controllable device on the voltage monitoring node. The calculation results show that the energy storage converter at the end of the feeder and several photovoltaic inverters near the end have higher voltage influence weights for nodes exceeding limits, while some inverters originally recorded as belonging to the same distribution area in the ledger but with weaker actual coupling have lower voltage influence weights.

[0035] The system identifies weak support areas based on voltage sensitivity maps and multi-source operation data of the distribution network, and determines energy storage converters with network control capabilities within these weak support areas as virtual voltage anchor points. In this embodiment, the system divides multiple voltage monitoring nodes into candidate regions based on the electrical connection relationships in the virtual electrical topology and the similarity of voltage response fingerprints. When two candidate regions have the same voltage monitoring node, the same voltage monitoring node is assigned to the candidate region with the largest sum of edge weights of the voltage influence of controllable devices on that node, thus obtaining the deduplicated candidate region. The candidate region at the end of the feeder simultaneously meets the conditions of a new energy penetration rate of over 60%, a key node voltage deviation of over 5%, a reverse power flow duration of over 5 minutes, and a voltage response amplitude of over 0.5%, and is therefore identified as a weak support region.

[0036] For this weakly supported area, the system selects energy storage converters with grid control capabilities and energy storage status that meet safety constraints as candidate energy storage converters. The anchoring value is calculated based on voltage support leverage, remaining available power, state-of-charge safety margin, health status, historical regulation burden, and virtual electrical distance determined by virtual electrical topology and voltage response fingerprint. The energy storage converter at the end of the feeder has the highest anchoring value and is therefore identified as the primary virtual voltage anchor; the backup energy storage converter in the middle of the feeder is identified as the backup virtual voltage anchor.

[0037] The system uses a virtual voltage anchor point as the center and divides the voltage control zone based on voltage response fingerprint, virtual electrical topology, and voltage sensitivity spectrum. During the division, the system assigns voltage monitoring nodes that meet at least two of the zone division conditions to the voltage control zone centered on the virtual voltage anchor point.

[0038] The criteria for zoning include: the response strength of the voltage monitoring node determined by the voltage response fingerprint to the virtual voltage anchor point is greater than 0.60; the edge weight of the voltage influence of the virtual voltage anchor point on the voltage monitoring node is greater than 0.60; the virtual electrical distance between the voltage monitoring node and the virtual voltage anchor point is less than 0.40; and the similarity between the voltage response fingerprint of the voltage monitoring node and the comprehensive voltage response fingerprint of the node corresponding to the virtual voltage anchor point is greater than 0.75.

[0039] When no node has yet been assigned, the voltage monitoring node corresponding to the virtual voltage anchor point grid-connected node is used as the initial assigned node. When the same voltage monitoring node meets the zoning conditions for multiple voltage control zones, the voltage monitoring node is assigned to the voltage control zone with the largest edge weight of the virtual voltage anchor point's influence on the node's voltage, thus ensuring that there are no identical voltage monitoring nodes between different voltage control zones. After the division, end-point energy storage converters, photovoltaic inverters located near the end and with strong response coupling, dynamic reactive power compensation equipment, and related voltage monitoring nodes are assigned to the same voltage control zone, while nodes with weak response coupling and inconsistent adjustment directions are not assigned to this control zone.

[0040] When there are voltage over-limit nodes within the voltage control zone, the system calculates the adjustment priority of each controllable device based on the voltage sensitivity spectrum and multi-source operation data of the distribution network, and allocates voltage adjustment tasks according to the adjustment priority.

[0041] In this embodiment, the voltage at the feeder end node reaches 1.08 times the rated voltage, which is considered a high-voltage over-limit node. The system calculates the regulation priority based on the voltage impact weight of controllable equipment on the over-limit node, the remaining regulation capacity, the active power fluctuation contribution, and the historical regulation burden.

[0042] The energy storage converter at the end of the grid has a high voltage influence weight, a large remaining regulation capacity, and a low historical regulation burden, thus receiving a high regulation priority. Photovoltaic inverters located near the end of the grid and with large output fluctuations also receive a high regulation priority. Inverters with weak actual coupling do not participate in regulation first due to their low voltage influence weight. The system first controls the energy storage converter corresponding to the main virtual voltage anchor point to provide a local reference voltage in grid mode and absorb active power and a suitable amount of reactive power. Then, it controls the photovoltaic inverters with high regulation priority to absorb reactive power; finally, it controls the dynamic reactive power compensation equipment to absorb reactive power.

[0043] If the voltage still does not return to the allowable range after the above adjustment, the system will only perform minimum active power reduction on a small number of photovoltaic inverters with high active power fluctuation contribution and large voltage impact weight, so as to avoid uniform power restriction on the entire photovoltaic area.

[0044] The system performs hierarchical collaborative regulation of fast-response and slow-response devices according to the voltage regulation task, and updates the voltage response fingerprint, virtual electrical topology, voltage sensitivity spectrum, weak support region, virtual voltage anchor point, voltage control area and regulation priority according to the regulated node voltage and controllable device status.

[0045] In this embodiment, when the voltage deviation of the voltage-over-limit node is no greater than 2%, the photovoltaic inverter, energy storage converter, and dynamic reactive power compensation device are given priority to perform voltage regulation tasks. When the voltage deviation of the voltage-over-limit node exceeds 5% for more than 5 minutes, the system initiates the slow-response device adjustment judgment. Before the slow-response device performs the voltage regulation task, the system predicts the voltage deviation change of key voltage monitoring nodes in adjacent voltage control areas based on the virtual electrical topology and voltage sensitivity map. When the predicted voltage deviation change does not exceed 1%, the voltage regulating transformer or capacitor bank configured with OLTC is allowed to perform the action. When the predicted voltage deviation change exceeds 1%, the action of the slow-response device is delayed, and the fast-response device is called for compensation and regulation.

[0046] The working principle of this embodiment is as follows: The system does not directly rely on static ledger topology for voltage regulation, but instead uses fast-response equipment to output small reactive disturbances within a safe range, so that the actual distribution network can produce an observable voltage response to the disturbance. Then, a voltage response fingerprint is formed based on the voltage response amplitude, direction, delay, and recovery slope. The actual electrical coupling relationship is identified by the difference between the voltage response fingerprint and the expected response characteristics of the ledger topology.

[0047] Subsequently, the system uses the topology consistency verification results to correct the ledger topology, obtaining a virtual electrical topology that is closer to the actual operating state. The virtual electrical topology is then further transformed into a voltage sensitivity map, so that the actual adjustment effect of the controllable equipment on each voltage monitoring node is expressed in the form of voltage influence side weights.

[0048] Weak support area identification is based on voltage sensitivity maps and operating status, enabling the location of areas with high renewable energy penetration, significant voltage fluctuations, and insufficient support capacity. Grid-type energy storage converters operate as virtual voltage anchors within these areas, providing a local reference voltage.

[0049] The voltage control zone is no longer divided according to fixed substation or fixed feeder boundaries, but is dynamically generated around virtual voltage anchor points, combining response strength, edge weight, virtual electrical distance and response fingerprint similarity, so that nodes and devices in the same control zone have stronger actual electrical coupling relationships.

[0050] The regulation task is then allocated based on voltage impact weights, remaining regulation capacity, historical regulation burden, and contribution from renewable energy fluctuations. This allows for the priority use of controllable equipment that is most effective and has the lowest burden on nodes exceeding limits. Fast-response equipment handles rapid fluctuations first, while slow-response equipment is activated only when persistent deviations are large and will not cause deterioration of adjacent voltage control zones. This forms a closed-loop control system that includes topology self-verification, virtual electrical topology, voltage sensitivity maps, virtual voltage anchor points, dynamic control zones, and hierarchical collaborative regulation.

[0051] After the aforementioned coordinated control, the voltage at the feeder end node decreased from 1.08 times the rated voltage to within 1.03 times the rated voltage. The voltage regulating transformer equipped with OLTC did not experience unnecessary operation, and the photovoltaic active power reduction was lower than that of the direct curtailment scheme. The system uses the controlled node voltage, energy storage output, inverter reactive power output, dynamic reactive power compensation equipment output, and changes in the control zone boundary as feedback data to update the voltage response fingerprint, virtual electrical topology, voltage sensitivity spectrum, virtual voltage anchor point, voltage control zone, and regulation priority. When subsequent detections include tie switch switching, sudden changes in photovoltaic output, energy storage state of charge falling below 20%, energy storage temperature exceeding 45°C, communication anomalies, or the main virtual voltage anchor point exiting, the system re-executes the topology self-verification, virtual voltage anchor point selection, and dynamic zoning process. If necessary, the backup energy storage converter is switched to the new virtual voltage anchor point.

[0052] As can be seen from this embodiment, this method can achieve dynamic coordinated voltage control based on actual electrical response relationships under conditions of inaccurate ledger topology, high proportion of photovoltaic reverse power flow, and insufficient voltage support at the end of weak grids. This improves the accuracy and adaptability of voltage control, reduces ineffective regulation and frequent operation of slow-response equipment, and enhances the capacity for renewable energy absorption.

[0053] Example 2, based on Example 1, further illustrates the specific implementation methods for disturbance source device screening, small-amplitude reactive power disturbance application, voltage response fingerprint generation, topology self-verification, virtual electrical topology generation, and voltage sensitivity map construction. Unless otherwise stated in this example, the multi-source operating data of the distribution network, controllable device types, hardware environment, and threshold settings can adopt the configuration in Example 1.

[0054] To avoid confusion regarding the meaning of formula characters, the following formulas use a unified character rule: 'a' represents the disturbance source device number, 'n' represents the voltage monitoring node number, 'c' represents the controllable device number, 'g' represents the weak support area number, 'h' represents the candidate energy storage converter number, 'm' represents the voltage over-limit node number, and 'r' only represents the set summation index; 'Ω' always represents the voltage influence weight, 'B' always represents the normalized value of historical regulation burden, and 'Ξ' always represents the normalized value of virtual electrical distance. All function inputs in the formulas are dimensionless, and unless otherwise specified, the values ​​of each normalization parameter range from 0 to 1.

[0055] In step S2, the disturbance source device is determined from the fast-response devices based on the status of controllable devices. Fast-response devices include controllable renewable energy inverters, energy storage converters, or SVG (Static Var Generator). The disturbance source device must meet the following conditions: the current voltage of the node where the disturbance source device is located is not in a severely over-limit state; the remaining reactive power capacity of the disturbance source device is greater than the preset reactive power capacity threshold; the communication status of the disturbance source device is normal; no protection alarm has occurred in the disturbance source device within the most recent control cycle; when the disturbance source device is an energy storage converter, the energy storage SOC is within a safe range; the predicted disturbance will not cause the voltage of adjacent nodes to exceed the allowable range; and the disturbance source device is not under maintenance, locked, or restricted operation.

[0056] In this embodiment, the preset severe over-limit threshold is specifically set to be higher than 1.10 times the rated voltage or lower than 0.90 times the rated voltage; the preset reactive capacity threshold is specifically set to 10% of the rated reactive capacity of the disturbance source device; the energy storage SOC safety range is specifically set to 20% to 90%; normal communication status corresponds to a communication delay of no more than 200ms and no more than 3 consecutive communication packet losses; normal protection status corresponds to no overvoltage, undervoltage, overcurrent, overtemperature or islanding protection alarms in the most recent control cycle. The preset voltage safety constraints are as follows: for voltage monitoring nodes that have not exceeded the limit before the disturbance, the voltage after the disturbance shall be maintained between 0.93 times the rated voltage and 1.07 times the rated voltage; for voltage monitoring nodes that have exceeded the limit before the disturbance, the disturbance shall not cause the voltage deviation to increase further relative to the pre-disturbance state, and shall not cause the node voltage to exceed the severe over-limit threshold.

[0057] Small reactive power disturbances are calculated using the following formula: ; in, This represents the small reactive power disturbance output by the a-th disturbance source device at time t after the disturbance begins, in kvar. This represents the disturbance direction coefficient, which is -1 when absorbing reactive power and 1 when outputting reactive power. Represents the hyperbolic tangent function; This indicates the time since the disturbance began, expressed in seconds (s). This represents the time reference for smoothing the start of the disturbance, in seconds (1 second is acceptable). This represents the disturbance proportionality coefficient, with a value ranging from 0.02 to 0.05; This represents the rated reactive power capacity of the a-th disturbance source device, in kvar. This represents the current remaining reactive power capacity of the a-th disturbance source device, in kvar. This represents the maximum safe disturbance amount that satisfies the preset voltage safety constraint, in kvar. This represents the duration of the disturbance from the a-th disturbance source device, in seconds. This indicates taking the minimum value among the multiple values ​​within the parentheses.

[0058] In this formula, , and The units of all are kvar. and All of them are dimensionless quantities, therefore The dimension is kvar. (Through) The function limits the disturbance amplitude by... To achieve smooth loading during the disturbance start-up phase, by Achieving a smooth exit at the end of the disturbance phase and restoring the disturbance amount to 0 outside the disturbance time range can avoid new voltage over-limits or voltage surges caused by the disturbance test itself.

[0059] The duration of small reactive power disturbances can be set from 1 to 10 seconds, and the disturbance interval can be set from 30 to 300 seconds. In one implementation, the disturbance duration is set to 5 seconds, and the disturbance interval is set to 120 seconds. The disturbance direction is determined based on the current voltage state: when the regional voltage is high, disturbances in the reactive power absorption direction are preferred; when the regional voltage is low, disturbances in the reactive power output direction are preferred; when used for topology self-verification, alternating positive and negative disturbances are used without affecting the power quality of users. If the distribution network is in a state of severe voltage over-limit, fault protection, communication abnormality, or dispatch lockout, the disturbance test is suspended.

[0060] In step S3, after the disturbance source device outputs a small reactive power disturbance, the system collects response data from each voltage monitoring node. The response data includes voltage response amplitude, voltage response direction, voltage response delay, voltage response duration, voltage recovery slope, voltage change trend before and after the disturbance, and the degree of similarity to the response characteristics of adjacent nodes.

[0061] The voltage response amplitude reflects the strength of the impact of the disturbance source equipment on the voltage of the monitored node; the voltage response direction reflects whether the node voltage rises or falls after the disturbance; the voltage response delay reflects the disturbance propagation and control link delay; the voltage recovery slope reflects the speed at which the node voltage recovers after the disturbance ends; the voltage response duration reflects the time the disturbance effect remains on the node side; the voltage change trend before and after the disturbance is used to distinguish between voltage changes caused by disturbances and voltage changes caused by natural fluctuations in new energy output; the similarity of the response characteristics with adjacent nodes is used to determine whether the nodes belong to the same actual electrical influence area.

[0062] The system generates a voltage response fingerprint based on the voltage response amplitude, direction, delay, and recovery slope. The duration of the voltage response, the voltage change trend before and after the disturbance, and the similarity to the response characteristics of adjacent nodes are used as auxiliary parameters in the voltage response fingerprint. The voltage response fingerprint characterizes the actual electrical coupling relationship between the voltage monitoring node and the disturbance source device.

[0063] To reduce the risk of misjudgment caused by a single disturbance, the system uses multiple disturbance source devices for cross-validation. Specifically, according to the disturbance source device selection criteria in step S2, multiple disturbance source devices are sequentially selected from the fast-response devices, and each disturbance source device is controlled to output a small reactive power disturbance to generate a corresponding voltage response fingerprint. The system compares the voltage response fingerprints generated by the same voltage monitoring node under multiple disturbance source devices with the corresponding expected response characteristics. When the same voltage monitoring node exhibits response characteristics inconsistent with the ledger topology in three consecutive disturbance tests, a topology correction judgment is triggered. Preferably, when the response amplitude ranking and response delay ranking of the same voltage monitoring node for multiple disturbance source devices are inconsistent with the electrical distance ranking in the ledger topology, it is determined that the voltage monitoring node may have an incorrect ledger attribution, incorrect switch status, or inaccurate line parameters.

[0064] In step S4, the system calculates the expected response characteristics of the voltage monitoring node relative to the disturbance source device based on the ledger topology data and line impedance, and compares the expected response characteristics with the actual voltage response fingerprint. The expected response characteristics include the expected voltage response amplitude, expected response direction, and expected response delay. The system calculates the topology consistency confidence value according to the following formula: ; in, Indicates the first The topology consistency confidence value of each voltage monitoring node relative to the a-th disturbance source device; (*) represents the Logistic function. e represents the natural constant; This represents the dimensionless bias coefficient, which can be set to 0.20. , and All of these represent dimensionless weighting coefficients, which can be 0.35, 0.30, and 0.35 respectively. Represent the natural logarithm function; This represents the actual voltage response amplitude of the nth voltage monitoring node to the ath disturbance source device, expressed in per-unit value; This represents the expected voltage response amplitude calculated based on the ledger topology data, expressed in per-unit values. This represents the response amplitude smoothing factor, using a per-unit value, which can be 0.001; This represents the actual response delay of the nth voltage monitoring node, in seconds. This represents the expected response delay calculated based on the ledger topology data, expressed in seconds (s). This represents the normalized baseline for response delay, in seconds (1 second is acceptable). Indicates the actual response direction: 1 for boost, -1 for buck, and 0 for no obvious response; This indicates the expected response direction: 1 for boost, -1 for buck, and 0 for no significant response.

[0065] In this formula, the amplitude term uses a ratio of per-unit values, the delay term uses a ratio of seconds to seconds, and the direction term is a dimensionless product; therefore, the input to the Logistic function is dimensionless. This formula simultaneously considers differences in response amplitude, response delay, and response direction consistency, avoiding reliance on a single response metric to determine topology anomalies and improving the reliability of topology self-checking. The comprehensive topology difference value can be calculated from 1- This indicates that when the overall topology difference value is greater than the preset topology verification threshold, and this result appears in three consecutive comparisons, a topology anomaly result is generated.

[0066] In this embodiment, the topology consistency confidence threshold can be set to 0.60, the preset topology verification threshold can be set to 0.35, and the preset number of times can be set to 3. The actual voltage response amplitude and the expected voltage response amplitude are in the same unit, as are the response delay and the expected response delay. All terms entering the Logistic function are ratio-normalized or normalized, making them dimensionless. This formula incorporates response amplitude, response delay, and response direction simultaneously into the topology self-verification judgment, avoiding topology judgment based solely on a single response indicator.

[0067] When the difference between the voltage response fingerprint and the corresponding expected response feature exceeds 0.35, and this difference occurs in three consecutive comparisons, the system generates a topology anomaly result and corrects the electrical connection relationships used to generate the virtual electrical topology based on the topology anomaly result.

[0068] The specific judgment rules include: If a voltage monitoring node is close to the disturbance source device in the ledger topology, but the actual response amplitude is less than 0.3% of the rated voltage and the response delay is greater than 2s, then the actual electrical coupling between the voltage monitoring node and the disturbance source device is determined to be weaker than the ledger record; if a voltage monitoring node is far from the disturbance source device in the ledger topology, but the actual response amplitude is greater than 0.5% of the rated voltage and the response delay is less than 1s, then the voltage monitoring node and the disturbance source device are determined to have a strong electrical coupling relationship. If the difference in response amplitude of multiple voltage monitoring nodes to the same disturbance source device is no greater than 0.2% of the rated voltage, the response direction is consistent, and the difference in response delay is less than 0.5s, then the multiple voltage monitoring nodes are determined to belong to the same actual electrical influence area. If the response characteristics of the same voltage monitoring node under different disturbance source devices do not match the ledger topology, then the confidence level of the ledger topology anomaly of that voltage monitoring node is increased. If the anomaly judgment only occurs in a single disturbance and does not recur in subsequent disturbances, then the topology is not corrected immediately, but only recorded as a low-confidence anomaly.

[0069] The virtual electrical topology is generated after correcting the electrical connections in the ledger topology data based on topology consistency verification results, switch states, line impedances, and voltage response fingerprints from the ledger topology data. This virtual electrical topology integrates the ledger topology, switch states, line impedances, and voltage response fingerprints, and is used to replace or correct the physical ledger topology in subsequent voltage control. In this way, the voltage control model can more closely approximate the actual operating state when the ledger topology is inaccurate, the status of tie switches is not updated in a timely manner, or the ledger of new energy access points is incorrect.

[0070] Following step S4, the system constructs a voltage sensitivity map based on the virtual electrical topology and voltage response fingerprint. The voltage sensitivity map includes the voltage influence edge weights between controllable devices and voltage monitoring nodes. Controllable devices include renewable energy inverters, energy storage converters, SVG (Static Var Generator), voltage regulating transformers equipped with OLTC (Optical Voltage Regulator), capacitor banks, and line voltage regulators. The system calculates the voltage influence edge weights of the controllable devices on the node voltage using the following formula: ; in, Let represent the edge weight of the voltage influence of the c-th controllable device on the n-th voltage monitoring node; (*) represents the Softplus function. e represents the natural constant; , and This represents the dimensionless weighting coefficient, which can be 0.40, 0.45, and 0.15 respectively. The normalized reactive voltage sensitivity represents the degree of influence of the reactive power change of the c-th controllable device on the voltage of the n-th voltage monitoring node. The normalized active voltage sensitivity represents the degree of influence of the active power change of the c-th controllable device on the voltage of the n-th voltage monitoring node. This represents the response coupling strength between the nth voltage monitoring node and the cth controllable device, determined by the voltage response fingerprint. This represents the normalized value of the virtual electrical distance between the nth voltage monitoring node and the cth controllable device; This represents the normalized value of the historical adjustment burden of the c-th controllable device.

[0071] In this formula, , , , and All are dimensionless quantities. The input to the Softplus function is dimensionless, and the output is dimensionless. It is a dimensionless boundary weight. Used to integrate the effects of active and reactive power regulation on node voltage; Used for controllable devices with strong response coupling and short virtual electrical distance; This formula is used to reduce the tendency of equipment with high historical regulation load to be called up. It enables the voltage sensitivity spectrum to simultaneously reflect the actual response relationship, control sensitivity, and equipment operating load.

[0072] Reactive voltage sensitivity and active voltage sensitivity can be estimated using distribution network power flow models, historical control records, or small disturbance responses. Response coupling strength is determined by voltage response fingerprints, which can be calculated comprehensively based on voltage response amplitude, response delay, and response direction.

[0073] Historical regulation burden can be determined based on the cumulative reactive power output, cumulative active power charging and discharging, number of actions, or temperature rise of the equipment within a preset time window; the preset time window can be set to 24 hours. For equipment without active power regulation capability, the active power voltage sensitivity can be 0 or a preset value; for equipment without reactive power regulation capability, the reactive power voltage sensitivity can be 0 or a preset value to ensure that the calculation object of the formula is clear. The larger the voltage influence weight, the stronger the regulation effect of the corresponding controllable equipment on the corresponding voltage monitoring node.

[0074] The working principle of this embodiment is as follows: The system first determines the operating status of various fast-response and slow-response devices through multi-source operational data. Then, it selects the disturbance source device from the fast-response devices that meet the conditions for voltage safety, capacity, communication, protection, and interlocking. The small-amplitude reactive power disturbances applied by the disturbance source device are jointly limited by the rated reactive power capacity, the remaining reactive power capacity, and the safe voltage constraint. Furthermore, the disturbance initiation and termination processes are processed using smoothing functions, so the disturbance process will not cause new voltage overruns or significant voltage surges.

[0075] The amplitude, direction, delay, and recovery slope of the response generated by each voltage monitoring node under disturbance reflect the actual electrical coupling relationship between the node and the disturbance source device, and the system forms a voltage response fingerprint based on this. Through cross-validation with multiple disturbance source devices, misjudgments caused by single measurement noise, random load fluctuations, and short-term fluctuations in new energy sources can be filtered out. Subsequently, the system compares the voltage response fingerprint with the expected response characteristics obtained based on the ledger topology and line impedance, generates topology consistency confidence values ​​and topology anomaly results, and corrects the electrical connection relationships used for control accordingly, forming a virtual electrical topology. Finally, the system constructs a voltage sensitivity map based on the virtual electrical topology and voltage response fingerprint, so that subsequent weak support area identification, virtual voltage anchor point selection, and voltage control zone division no longer rely solely on the static ledger topology, but are based on the actual operational response relationship.

[0076] This embodiment can improve the reliability of topology verification results and the accuracy of voltage control models when there are inaccurate topology records, untimely updates to switch status changes, or deviations in the location of new energy access points in the ledger. Cross-validation using voltage response fingerprints from multiple disturbance sources can reduce the risk of misjudgment of a single disturbance. Furthermore, virtual electrical topology and voltage sensitivity maps can clarify the regulatory influence relationship between controllable equipment and voltage monitoring nodes, thus providing an accurate basis for subsequent identification of weak support areas, determination of virtual voltage anchor points, division of dynamic control zones, and allocation of regulation tasks.

[0077] Example 3, based on Examples 1 and 2, further illustrates the specific implementation methods for weak support region identification, virtual voltage anchor point determination for grid-type energy storage, dynamic voltage control zone division, adjustment priority calculation, fast-slow hierarchical coordinated control, and feedback update of control parameters. Unless otherwise stated in this example, the relevant hardware environment, data acquisition methods, disturbance source screening methods, voltage response fingerprints, virtual electrical topology, and voltage sensitivity maps can adopt the configurations in Examples 1 and 2. In this example, the meanings of formula characters follow the unified character rules of Example 2, and the meanings of characters not repeatedly explained are consistent with those in Example 2.

[0078] In this embodiment, the system identifies weak support regions based on voltage sensitivity maps. A weak support region refers to an area with high renewable energy penetration, low short-circuit capacity, frequent reverse power flow, prolonged voltage exceedance duration, insufficient available reactive power capacity, or voltage sensitivity to small reactive power disturbances. A weak support region can be a low-voltage distribution area, a feeder terminal area, an actual electrical influence area determined by a virtual electrical topology, or a local control area composed of multiple adjacent distribution areas.

[0079] When identifying weakly supported regions, the system first divides multiple voltage monitoring nodes into candidate regions based on the electrical connections in the virtual electrical topology and the similarity of the voltage response fingerprint. The voltage response fingerprint similarity can be calculated based on the normalized differences in voltage response amplitude, response direction, response delay, and voltage recovery slope, and the similarity threshold can be set to 0.75.

[0080] If two candidate regions have the same voltage monitoring node, the system calculates the sum of the edge weights of the voltage influence of controllable devices in each candidate region on that same voltage monitoring node, and assigns the same voltage monitoring node to the candidate region with the largest sum of edge weights, thus obtaining the deduplicated candidate regions. Through this unique assignment rule, there are no identical voltage monitoring nodes between candidate regions, and there are also no identical voltage monitoring nodes between the subsequently identified weak support regions, avoiding the same node being repeatedly controlled by multiple regions at the same time.

[0081] The system identifies weak support areas based on whether at least two of the following parameters within the deduplicated candidate area—new energy penetration rate, node voltage deviation, reverse power flow duration, available reactive power capacity, short-circuit capacity, voltage response amplitude, and number of slow-response device actions—meet the corresponding preset threshold conditions.

[0082] Specifically, the new energy penetration rate threshold can be set as follows: the proportion of new energy installed capacity to transformer capacity in the region exceeds 60%; the node voltage deviation threshold can be set as the voltage deviation of key nodes exceeds ±5% of the rated voltage; the reverse power flow duration threshold can be set as exceeding 5 minutes; the available reactive power capacity threshold can be set as being less than 20% of the region's reactive power demand; the voltage response amplitude threshold can be set as the voltage response amplitude of multiple nodes to small reactive power disturbances exceeds 0.5% of the rated voltage; the slow response equipment operation frequency threshold can be set as reaching 80% of the daily allowable operation frequency; the line end voltage fluctuation amplitude threshold can be set as exceeding 2% of the rated voltage; and the short-circuit capacity threshold can be set as being less than 3 times the region's new energy installed capacity. When any candidate region simultaneously meets at least two of the above conditions, the system identifies the candidate region as a weak support region. This identification method can distinguish regions requiring key voltage support from general operating regions, providing a regional basis for subsequent grid-based energy storage as a virtual voltage anchor point.

[0083] For each weakly supported region, the system selects energy storage converters with grid-connection control capabilities and whose energy storage status meets preset safety constraints as candidate energy storage converters. Preset safety constraints include state of charge (SOC) constraints, health status constraints, temperature status constraints, communication status constraints, and protection status constraints. Specifically, the SOC safety range can be set to 20% to 90%; the health status threshold can be set to no less than 80%; the temperature status threshold can be set to no more than 45℃; the communication status constraint can be set to a communication delay of no more than 200ms and no more than 3 consecutive packet losses; the protection status constraint can be set to no overvoltage, undervoltage, overcurrent, overtemperature, islanding, or insulation abnormality alarms; the remaining charging and discharging power must meet the region's voltage support requirements, and the power margin threshold can be set to no less than 20% of the region's expected regulating power demand; the energy storage equipment must not be under maintenance, locked, or in a restricted operating state. Candidate energy storage converters that meet the grid-connection capability, safety status, and communication status requirements are given an admission flag of 1; otherwise, the admission flag is 0.

[0084] The system calculates the anchoring value of candidate energy storage converters as virtual voltage anchors using the following formula: ; in, This represents the evaluation value of the h-th candidate energy storage converter as the virtual voltage anchor point of the g-th weak support region; This represents the admission flag for the h-th candidate energy storage converter. It is set to 1 if the grid construction capability, security status and communication status are met, and 0 otherwise. (*) represents the Logistic function; , , , and This represents a dimensionless weighting coefficient, which can be 0.30, 0.20, 0.15, 0.15, and 0.20 respectively. This represents the voltage support lever value of the h-th candidate energy storage converter for the g-th weak support region; This represents the normalized virtual electrical distance from the h-th candidate energy storage converter to the g-th critical voltage monitoring node in the weakly supported area; This represents the normalized value of the remaining available power of the h-th candidate energy storage converter; This represents the normalized power smoothing factor, which can be 0.10; This represents the normalized value of the state-of-charge safety margin of the h-th candidate energy storage converter; This represents the normalized health status value of the h-th candidate energy storage converter. This represents the normalized value of the historical regulation load of the h-th candidate energy storage converter.

[0085] Among them, voltage support lever value Determine according to the following formula: ; in, This represents the set of voltage monitoring nodes within the g-th weakly supported region; It represents the importance weight of the nth voltage monitoring node in the gth weak support area, which can be determined based on the degree of voltage exceedance, load importance, or new energy access capacity. This represents the voltage influence weight of the h-th candidate energy storage converter on the n-th voltage monitoring node, and the h-th candidate energy storage converter belongs to the energy storage converter category in the controllable device number c. and All of them are dimensionless quantities, therefore It is a dimensionless quantity.

[0086] In this formula, all parameters entering the Logistic function are dimensionless quantities. Used for security access screening; Used to simultaneously evaluate the regional support capability and virtual electrical distance of candidate energy storage converters; Evaluation values ​​used to suppress devices with excessively low remaining power; and Used to reflect the energy storage status and health status; This formula is used to avoid long-term reliance on energy storage devices with high historical regulation burdens. It can identify a more suitable virtual voltage anchor point from multiple candidate energy storage converters as a local voltage reference source.

[0087] The system identifies the candidate energy storage converter with the highest anchor value and that meets safety constraints as the primary virtual voltage anchor point, and identifies the candidate energy storage converter with the second highest anchor value and that meets safety constraints as the backup virtual voltage anchor point.

[0088] The virtual voltage anchor point operates in grid-connected mode and provides a local reference voltage to its voltage control area. The local reference voltage can include a reference voltage amplitude and a reference phase angle. The reference voltage amplitude can be set to a target value within the range of 1.00 to 1.03 times the rated voltage, and the reference phase angle can be generated based on the droop control or virtual synchronization control strategy of the grid-connected controller. The grid-connected renewable energy inverter performs reactive power regulation or active power reduction based on the deviation between its own grid connection point voltage and the local reference voltage. In this way, energy storage is no longer merely used as a regular charging and discharging device or a regular reactive power compensation resource, but rather as a local voltage reference source in weak grid areas, improving the voltage support capability of these areas.

[0089] When the SOC of the primary virtual voltage anchor point is lower than the first SOC threshold, the temperature is higher than the first temperature threshold, the supporting power margin is lower than the first power threshold, or the voltage deviation within the voltage control zone corresponding to the primary virtual voltage anchor point continues to increase, the system switches to the backup virtual voltage anchor point or re-executes the virtual voltage anchor point selection process. In one embodiment, the first SOC threshold is set to 25%, the first temperature threshold is set to 45°C, the first power threshold is set to 10% of the expected regulating power demand of the region, and the judgment time for the continuous increase in voltage deviation is set to 60 seconds. This switching mechanism can prevent a single energy storage device from affecting the voltage stability of the weakly supported area due to insufficient energy, abnormal temperature, or decreased supporting capacity.

[0090] The system uses a virtual voltage anchor point as the center and dynamically divides voltage control zones based on factors such as the node's response strength to the virtual voltage anchor point, voltage influence weight, virtual electrical distance, response fingerprint similarity, consistency of adjustment direction, and risk of cross-zone impact. During zone division, the system assigns voltage monitoring nodes that meet at least two of the zone division conditions to the voltage control zone centered on the virtual voltage anchor point.

[0091] The zoning conditions include: the response strength of the voltage monitoring node determined by the voltage response fingerprint to the virtual voltage anchor point is greater than a preset response strength threshold, which can be set to 0.60; the edge weight of the voltage influence of the virtual voltage anchor point on the voltage monitoring node is greater than a preset edge weight threshold, which can be set to 0.60; the virtual electrical distance between the voltage monitoring node and the virtual voltage anchor point is less than a preset distance threshold, which can be set to 0.40; and the similarity between the voltage response fingerprint of the voltage monitoring node and the comprehensive voltage response fingerprint of the node already included in the zoning corresponding to the virtual voltage anchor point is greater than a preset similarity threshold, which can be set to 0.75. The comprehensive voltage response fingerprint can be obtained by averaging, weighting, or selecting representative nodes from the voltage response fingerprints of the already included nodes. When there are no already included nodes, the voltage monitoring node corresponding to the grid-connected node of the virtual voltage anchor point is used as the initial included node.

[0092] When dividing the voltage control area, the system also determines whether including a voltage monitoring node in the current voltage control area will cause the voltage of the adjacent voltage control area to deteriorate, and determines whether the branch where the voltage monitoring node is located and the branch where the virtual voltage anchor point is located have a direct or indirect strong coupling relationship in the virtual electrical topology.

[0093] If there are a large number of identical regulating devices between two voltage control zones and the regulating direction is consistent, the system allows the voltage control zones to be merged; if there are opposite regulating directions within the same voltage control zone, or if cross-zone regulation causes voltage deterioration in adjacent voltage control zones, the system splits the voltage control zone.

[0094] When the same voltage monitoring node meets the zoning conditions for multiple voltage control zones, the system assigns this same voltage monitoring node to the voltage control zone with the largest voltage influence weight of the corresponding virtual voltage anchor point for that same voltage monitoring node, ensuring that there are no identical voltage monitoring nodes between different voltage control zones. When multiple voltage control zones compete for the same controllable device, the controllable device is preferentially assigned to the voltage control zone with the largest voltage influence weight; if the voltage influence weights of two voltage control zones are close, the priority is determined according to the severity of voltage exceedance, the duration of exceedance, and the loss of renewable energy absorption. The voltage control zones obtained by the above dynamic division are not fixed feeder areas or fixed transformer areas, but rather dynamic control areas updated in real time or periodically based on the virtual electrical topology and the actual voltage response relationship.

[0095] Within each voltage control zone, when a voltage monitoring node exceeds its limit, the system calculates the adjustment priority of each controllable device for that node. The adjustment priority is determined based on the controllable device's voltage impact weight on the node, remaining regulation capacity, active power fluctuation contribution, and historical regulation burden. When a controllable device has no active power fluctuation contribution, the active power fluctuation contribution is set to a preset value, which can be set to 0. The system can calculate the adjustment priority using the following formula: ; ; in, This indicates the adjustment priority of the c-th controllable device for the m-th voltage over-limit node; This represents the priority intermediate quantity of the c-th controllable device for the m-th voltage over-limit node; (*) represents an exponential function; r represents the set of controllable devices within the voltage control zone where the m-th voltage over-limit node is located; r represents The controllable device summation index; This represents the priority intermediate quantity of the r-th controllable device for the m-th voltage over-limit node, and its calculation method is the same as... same; Let represent the edge weight of the voltage influence of the c-th controllable device on the m-th voltage-over-limit node; This represents the normalized value of the remaining regulating capacity of the c-th controllable device; This represents the normalized value of the active power fluctuation contribution of the c-th controllable device. This is true when the c-th controllable device has no active power fluctuation contribution. Set to 0; This represents the smoothing factor contributing to volatility, and can be set to 0.10. This represents the normalized value of the historical adjustment load of the c-th controllable device; , , and All represent dimensionless weighting coefficients, which can be 0.40, 0.25, 0.20 and 0.15 respectively.

[0096] In this formula, , , , and All of them are dimensionless quantities, therefore As a dimensionless quantity, the input quantity of the exponential function meets the dimensional requirements. The exponentially normalized form ensures that the sum of the adjustment priorities of all controllable devices within the same voltage control zone is 1, facilitating the proportional allocation of voltage regulation tasks. Used to prioritize devices that have a greater impact on voltage over-limit nodes; Used to prioritize the use of devices with more remaining adjustment capacity; This is used to assign corresponding regulation responsibilities to new energy equipment that contributes significantly to voltage fluctuations; This formula is used to avoid continuously calling up equipment with historically high regulation loads. It reduces ineffective regulation and localized equipment overload, making regulation task allocation more consistent with actual electrical influences and equipment conditions.

[0097] When a voltage-over-limit node is in a state of high voltage, the system performs regulation according to the regulation priority: first, it controls the grid-type energy storage corresponding to the virtual voltage anchor point to absorb active power; then it controls the new energy inverter with high regulation priority to absorb reactive power; then it controls the SVG to absorb reactive power; if the voltage still cannot be restored, it performs the minimum reduction of active power of new energy; when the high voltage persists, it triggers the slow response device to regulate.

[0098] When a voltage-over-limit node is in a low-voltage state, the system first controls the grid-type energy storage corresponding to the virtual voltage anchor point to release active power; then it controls the high-priority renewable energy inverter to output reactive power; subsequently, it controls the SVG to output reactive power; if the voltage still cannot be restored, it triggers the voltage regulator transformer or line regulator equipped with OLTC to step up the voltage. The adjustment of any controllable device must meet the constraints of device capacity, power factor, SOC, temperature, voltage limit, line current carrying capacity, communication status, and protection status.

[0099] When reactive power regulation capacity is insufficient, the system prioritizes the active power absorption or release capacity of energy storage; only when both energy storage and reactive power compensation capacity are insufficient will the system perform minimum active power reduction on the renewable energy inverter, thereby improving the renewable energy absorption capacity.

[0100] This embodiment divides controllable devices into fast-response devices and slow-response devices. Fast-response devices include new energy inverters, energy storage converters, and SVG, with control cycles set from 1 second to 30 seconds. Slow-response devices include voltage regulating transformers equipped with OLTCs, capacitor banks, and line voltage regulators, with control cycles set from 1 minute to 15 minutes. When the voltage deviation of a voltage over-limit node is not greater than a preset first voltage deviation threshold, the system prioritizes the fast-response devices to perform voltage regulation tasks. The first voltage deviation threshold can be set to 2% of the rated voltage.

[0101] When the voltage deviation exceeds a first voltage deviation threshold and the duration exceeds a first duration threshold, the system strengthens the virtual voltage anchor point support and reallocates the reactive power regulation task of the new energy inverter. The first duration threshold can be set to 10s to 60s, and in one embodiment, it is set to 30s. When the voltage deviation of the voltage over-limit node exceeds a preset second voltage deviation threshold for a duration exceeding a preset duration threshold, the system activates the slow response device to perform the voltage regulation task. The second voltage deviation threshold can be set to 5% of the rated voltage, and the second voltage deviation threshold is greater than the first voltage deviation threshold. The preset duration threshold can be set to 3min to 10min, and in one embodiment, it is set to 5min.

[0102] Before the slow-response device performs voltage regulation, the system predicts the voltage deviation changes of key voltage monitoring nodes in adjacent voltage control zones based on the virtual electrical topology and voltage sensitivity map. The slow-response device performs its voltage regulation task only if the voltage deviation change does not exceed a preset deterioration threshold. The preset deterioration threshold can be set to 1% of the rated voltage. If the slow-response device's action is predicted to cause voltage deterioration in adjacent voltage control zones, or if a reverse action may be needed in the short term, the slow-response device's action is delayed or canceled. When a voltage regulating transformer, capacitor bank, or line regulator equipped with an OLTC has reached its maximum number of actions, is in a locked state, or the predicted benefit of the action is lower than the cost, the system prioritizes using a fast-response device for compensation regulation. This hierarchical fast-slow coordination approach avoids frequent actions of traditional slow-response devices due to short-term fluctuations in renewable energy, thus improving equipment lifespan.

[0103] After implementing voltage coordinated control, the system collects data on node voltage, equipment output, energy storage status, traditional voltage regulating equipment operation status, renewable energy absorption, and line losses. Based on the control effect, it updates the voltage response fingerprint, virtual electrical topology, voltage sensitivity spectrum, weak support area identification results, main and backup virtual voltage anchors, voltage control zone boundaries, historical regulation burden of controllable equipment, regulation priority parameters, fast and slow equipment operation thresholds, and reactive power regulation strategies for renewable energy inverters. When changes in switch status, sudden changes in renewable energy output, sudden changes in load, energy storage SOC exceeding limits, communication anomalies, main virtual voltage anchor exit, voltage deterioration in the voltage control zone, or a renewed inconsistency between the ledger topology and the actual response are detected, the system re-executes the topology self-verification and dynamic partitioning process. Through feedback updates, the system can adapt to continuous changes in the distribution network's operating status.

[0104] The working principle of this embodiment is as follows: The system first uses virtual electrical topology and voltage response fingerprinting to divide voltage monitoring nodes into non-overlapping candidate regions. Then, it identifies weak support regions based on indicators such as renewable energy penetration rate, voltage deviation, reverse power flow, reactive power capacity, short-circuit capacity, voltage disturbance response, and the number of slow-response device operations. Subsequently, within each weak support region, the system selects an energy storage converter with grid control capabilities and whose energy storage status meets safety constraints as a virtual voltage anchor point, enabling the weak support region to obtain a local reference voltage.

[0105] The system then dynamically divides voltage control zones centered on virtual voltage anchor points, combining response strength, voltage influence weights, virtual electrical distance, and response fingerprint similarity. A unique attribution rule prevents voltage monitoring nodes or regulation resources from being redundantly allocated across multiple control zones. Voltage regulation tasks are prioritized based on voltage influence weights, remaining regulation capacity, active power fluctuation contribution, and historical regulation burden. Fast-response devices prioritize handling rapid voltage fluctuations, while slow-response devices only activate when there are persistent voltage deviations that will not worsen adjacent voltage control zones. Finally, the system continuously updates weak support areas, virtual voltage anchor points, voltage control zones, and regulation priorities through control feedback, forming a closed-loop control system that adapts to changes in operating conditions.

[0106] This embodiment enables accurate identification of weak support areas, avoiding errors in resource allocation caused by fixed feeder zoning or fixed substation zoning; it allows grid-based energy storage to serve as a local voltage reference source, improving voltage support capabilities in weak grid areas; it reduces cross-regional regulation interference through dynamic voltage control zone division; it reduces ineffective regulation and local equipment overload through regulation priority allocation; and it reduces frequent actions of slow-response equipment through fast and slow hierarchical coordination, and minimizes active power reduction from new energy sources when reactive power and energy storage regulation capabilities are insufficient, thereby balancing voltage qualification rate, new energy absorption capacity, and equipment lifespan.

[0107] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A voltage coordination control method for distribution networks with a high proportion of renewable energy access, characterized in that, include: S1 collects multi-source operation data of the distribution network, including node voltage, new energy output, energy storage status, voltage regulating equipment status, and ledger topology data, to determine the status of controllable equipment; controllable equipment includes fast-response equipment and slow-response equipment. S2, Select a disturbance source device from the fast response device, control its output to meet the preset voltage safety constraints for small reactive disturbances, and collect voltage response data from each voltage monitoring node; S3, generate a voltage response fingerprint based on the voltage response data, generate the expected response characteristics of each voltage monitoring node relative to the disturbance source device based on the ledger topology data, and compare them to obtain the topology consistency verification result; The voltage response data includes voltage response amplitude, voltage response direction, voltage response delay, and voltage recovery slope. The voltage response fingerprint is generated based on the voltage response amplitude, voltage response direction, voltage response delay, and voltage recovery slope, and is used to characterize the actual electrical coupling relationship between the voltage monitoring node and the disturbance source device. According to the disturbance source device selection conditions in step S2, multiple disturbance source devices are sequentially determined from the fast response devices, and each disturbance source device is controlled to output the small reactive disturbance to generate a corresponding voltage response fingerprint. The voltage response fingerprints generated by the same voltage monitoring node under multiple disturbance source devices are compared with the corresponding expected response features. When the difference between the voltage response fingerprint and the corresponding expected response feature exceeds a preset topology verification threshold, and the difference occurs in no less than a preset number of comparisons, a topology anomaly result is generated, and the electrical connection relationship used to generate the virtual electrical topology is corrected based on the topology anomaly result. S4. Based on the topology consistency verification result, ledger topology data and voltage response fingerprint, generate a virtual electrical topology and construct a voltage sensitivity map containing the voltage influence edge weights between controllable devices and voltage monitoring nodes. S5. Identify weak support areas based on the voltage sensitivity spectrum and multi-source operation data of the distribution network, and determine energy storage converters with network control capabilities as virtual voltage anchors from the weak support areas. S6, with the virtual voltage anchor point as the center, divide the voltage control area according to the voltage response fingerprint, virtual electrical topology and voltage sensitivity spectrum; S7. When there is a voltage over-limit node in the voltage control zone, calculate the adjustment priority of the controllable equipment according to the voltage sensitivity spectrum and the multi-source operation data of the distribution network, and allocate voltage adjustment tasks according to the adjustment priority. S8 performs hierarchical collaborative regulation of fast-response and slow-response devices according to the voltage regulation task, and updates the voltage response fingerprint, virtual electrical topology, voltage sensitivity spectrum, virtual voltage anchor point, voltage control area and regulation priority according to the regulated node voltage and controllable device status.

2. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 1, characterized in that, The fast-response equipment includes a new energy inverter, an energy storage converter, and a dynamic reactive power compensation device; the slow-response equipment includes a voltage regulating transformer equipped with an OLTC, a capacitor bank, and a line voltage regulator; the controllable equipment status includes the remaining regulation capacity, communication status, protection status, lockout status, and historical regulation burden of the controllable equipment.

3. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 2, characterized in that, In step S2, the disturbance source device is determined from the fast response device according to the controllable device status, and the disturbance source device satisfies the following conditions: the node voltage of its node does not exceed the preset severe over-limit threshold, the remaining reactive capacity is greater than the preset reactive capacity threshold, the communication status is normal, the protection status is normal and not in a blocked state. The amount of the small reactive power disturbance shall not exceed the minimum value among the rated reactive power capacity, the remaining reactive power capacity, and the maximum safe disturbance amount that meets the preset voltage safety constraints of the disturbance source equipment.

4. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 1, characterized in that, In step S4, the ledger topology data includes switch status and line impedance. The virtual electrical topology is generated after correcting the electrical connection relationships in the ledger topology data based on the topology consistency verification result, the switch status, the line impedance, and the voltage response fingerprint. The voltage influence weights are determined based on the reactive voltage sensitivity, active voltage sensitivity, response coupling strength, and historical regulation burden of the controllable device to the voltage monitoring node, wherein the response coupling strength is determined by the voltage response fingerprint.

5. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 1, characterized in that, In step S5, multiple voltage monitoring nodes are divided into candidate regions based on the electrical connection relationships and voltage response fingerprint similarity in the virtual electrical topology; When two candidate regions have the same voltage monitoring node, the same voltage monitoring node is assigned to the candidate region with the largest sum of edge weights of the voltage influence of the controllable device on the same voltage monitoring node in the region, so as to obtain the candidate region after deduplication. Based on the fact that at least two of the following parameters in the candidate region after deduplication—new energy penetration rate, node voltage deviation, reverse power flow duration, available reactive power capacity, short-circuit capacity, voltage response amplitude, and number of slow-response device actions—meet the corresponding preset threshold conditions, the weak support region is identified so that there are no identical voltage monitoring nodes among the weak support regions.

6. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 1, characterized in that, In step S5, determining the virtual voltage anchor point for each weak support region includes: From the weak support region, select energy storage converters that have grid control capabilities and whose energy storage status meets preset safety constraints as candidate energy storage converters. The preset safety constraints include state of charge constraints, health state constraints, temperature state constraints, communication state constraints, and protection state constraints. The anchor evaluation value is calculated based on the voltage support leverage value, remaining available power, state-of-charge safety margin, health status, historical regulation burden, and virtual electrical distance determined by the virtual electrical topology and voltage response fingerprint of the candidate energy storage converter for the weakly supported region. The voltage support lever value is determined based on the edge weight of the voltage influence of the candidate energy storage converter on the voltage monitoring node in the weak support area. The candidate energy storage converter with the highest anchor evaluation value was selected as the main virtual voltage anchor point.

7. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 1, characterized in that, In step S6, voltage monitoring nodes that meet at least two zoning conditions are assigned to the voltage control zone centered on the virtual voltage anchor point; The zoning conditions include: the response strength of the voltage monitoring node to the virtual voltage anchor point determined by the voltage response fingerprint is greater than a preset response strength threshold; the edge weight of the voltage influence of the virtual voltage anchor point on the voltage monitoring node is greater than a preset edge weight threshold; the virtual electrical distance between the voltage monitoring node and the virtual voltage anchor point is less than a preset distance threshold; and the similarity between the voltage response fingerprint of the voltage monitoring node and the comprehensive voltage response fingerprint of the node corresponding to the virtual voltage anchor point is greater than a preset similarity threshold. When there is no already assigned node, the voltage monitoring node corresponding to the virtual voltage anchor point grid-connected node is used as the initial assigned node; When the same voltage monitoring node meets the zoning conditions of multiple voltage control zones, the same voltage monitoring node is assigned to the voltage control zone with the largest edge weight of the voltage influence of the corresponding virtual voltage anchor point on the same voltage monitoring node, so that there are no identical voltage monitoring nodes between the voltage control zones.

8. The distribution network voltage coordinated control method for high-proportion renewable energy access according to claim 1, characterized in that, In steps S7 and S8, the adjustment priority is determined based on the voltage influence weight of the controllable device on the voltage over-limit node, the remaining adjustment capacity, the active power fluctuation contribution, and the historical adjustment burden. When the controllable device does not contribute to active power fluctuations, the active power fluctuation contribution is taken as a preset value; When the voltage deviation of the voltage over-limit node is not greater than a preset first voltage deviation threshold, the fast response device shall perform the voltage regulation task first. When the duration of the voltage deviation of the voltage over-limit node being greater than a preset second voltage deviation threshold exceeds a preset duration threshold, the slow response device shall be activated to perform the voltage regulation task. The preset second voltage deviation threshold is greater than the preset first voltage deviation threshold. Before the slow-response device performs a voltage regulation task, the voltage deviation change of key voltage monitoring nodes in adjacent voltage control areas is predicted based on the virtual electrical topology and voltage sensitivity map, and the voltage regulation task is performed when the voltage deviation change does not exceed a preset deterioration threshold.

Images