Methods and equipment for flexible DC interconnection in low-voltage power distribution

Abstract

This invention relates to the field of smart grid technology, and particularly to a method and device for flexible DC interconnection in low-voltage power distribution. The method designs a flexible DC line interconnection structure based on the load distribution and distributed energy access requirements of the distribution area; dynamically optimizes the line topology by calculating the electrical distance between adjacent load nodes to reduce transmission losses; introduces a flexible DC control module to monitor line voltage, current, and power data in real time, and achieves optimized power allocation and dynamic adjustment based on low-voltage power distribution system control algorithms; verifies the rationality of the interconnection path and the effectiveness of the control strategy through integrated testing to simulate actual operating scenarios, and finally deploys an adaptive flexible DC interconnection network. This method improves system energy efficiency, flexibility, and reliability, supports plug-and-play access for distributed energy, and adapts to the development needs of smart grids.

Description

Technical Field

[0001] This invention relates to the field of smart grid technology, and in particular to a method and device for flexible DC interconnection of low-voltage power distribution. Background Technology

[0002] With the development of smart grids and the large-scale integration of distributed energy resources, the limitations of traditional low-voltage AC power distribution systems in terms of flexibility, energy efficiency, and reliability are becoming increasingly apparent. Traditional AC power distribution networks typically employ radial or trunk-type fixed line structures, making it difficult to dynamically optimize interconnection paths based on load node distribution. This results in high line losses and insufficient plug-and-play compatibility with distributed energy resources. Furthermore, traditional power allocation relies on manual scheduling or preset strategies, lacking the ability to adaptively adjust to real-time load fluctuations and uncertainties in renewable energy output. This can easily lead to problems such as localized overloads and voltage fluctuations, affecting power quality and system stability.

[0003] Meanwhile, DC power distribution has become an important development direction in the low-voltage power distribution field due to its advantages in distributed energy access, efficient transmission, and support for new loads. However, existing DC power distribution systems lack quantitative analysis of the electrical distance between load nodes in their interconnection structure design, resulting in unreasonable line layouts. At the control level, there is a lack of intelligent management modules that integrate real-time monitoring, power optimization, and fault response, making it difficult to achieve flexible power interconnection and dynamic balance among multiple nodes.

[0004] Therefore, how to optimize the DC line structure based on the load node distribution characteristics, achieve flexible power allocation and efficient management through intelligent control, and improve the adaptability of low-voltage power distribution systems to complex scenarios has become an urgent technical problem to be solved.

[0005] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0006] The main objective of this invention is to provide a method and device for flexible DC interconnection of low-voltage power distribution, which aims to solve the technical problems of traditional low-voltage power distribution systems, which lack interconnection structure optimization based on the electrical distance of load nodes and intelligent dynamic power allocation mechanism, making it difficult to adapt to distributed energy access and load fluctuations, and resulting in high line losses, insufficient flexibility and stability.

[0007] To achieve the above objectives, the present invention provides a method for flexible DC interconnection of low-voltage power distribution, the method comprising: Design a flexible DC line structure for interconnection based on the needs of the power distribution area, calculate the electrical distance between adjacent load nodes, and optimize the line structure based on the electrical distance between adjacent load nodes; A flexible DC control module is introduced to monitor the optimized line structure, and a low-voltage power distribution system control algorithm is used to optimize the power allocation of the monitored line structure. The flexible DC line structure and flexible DC control module are integrated and tested to simulate actual operating scenarios. Based on the simulated scenarios, the flexible DC interconnection path for low-voltage power distribution is deployed to complete the flexible DC interconnection of low-voltage power distribution.

[0008] Optionally, the interconnected flexible DC line structure includes: Analyze the power consumption characteristics of load nodes, and design flexible DC distribution units by investigating the power demand and voltage level of the power consumption characteristics. The flexible DC distribution unit includes a DC bus, interconnection interface, energy storage interface, distributed power source interface and load interface. When the node requires additional interconnection capacity, the corresponding interface position is reserved in the design. Calculating the electrical distance between adjacent load nodes involves calculations based on the node coordinates and line parameters. The specific calculation formula is as follows:

[0009] in, The distance between adjacent load nodes is represented by (x1,y1) and (x2,y2), which represent the physical coordinates of the load nodes, and k is the line parameter coefficient. The length of the DC interconnection line is determined based on the calculation results. When the electrical distance between adjacent load nodes exceeds a predetermined threshold, a segmented interconnection structure is used for layout.

[0010] Optionally, the layout using a segmented interconnect structure includes: Design the intermediate interconnection interface; set a DC output interface on one side of the flexible DC distribution unit and a DC input interface matching the output interface on the other side. When the layout of the flexible DC distribution unit cannot meet the actual location of the load node, the interface position is adjusted by using the integrated structure of the multi-functional interconnection interface. The integration of multi-functional interconnect interfaces includes preprocessing for compatibility with different types of distributed power sources and loads. The specific preprocessing steps are as follows: The initial topology of the multi-functional interconnection interface is designed. Based on the requirements of the flexible DC power distribution unit, a segmented interconnection line interface of the multi-functional interconnection interface is constructed. The segmented interconnection line interface includes a plug, a socket and a communication protocol module. The socket includes specifications, rated current and insulation material. The sockets are subjected to electrical performance tests, including insulation resistance tests, current carrying capacity tests, and dynamic response tests. The results of the tests are used to classify the multi-functional interconnect interfaces. The classification includes storing information on different multi-functional interconnect interface types in an interface type database, and identifying the multi-functional interconnect interface type information stored in the interface type database; When interface type information is repeated during the identification process, it indicates that there is a conflict in the identified interface type information. In this case, the line structure is optimized based on the electrical distance between adjacent load nodes.

[0011] Optionally, the line structure optimized based on electrical distance between adjacent load nodes includes: An optimization model for flexible DC lines is constructed based on the electrical distance between adjacent load nodes. The line's voltage drop withstand capability is analyzed, and the line's constraint withstand capability is added. The specific calculation formula for constructing the optimization model of flexible DC lines is as follows:

[0012] Minimize represents the optimization objective of flexible DC lines. This indicates the intensity of the interconnection demand between nodes i and j. Indicates the priority of node interconnection. Indicates the electrical distance between adjacent load nodes. The line current is represented by n, the total number of load nodes in the distribution area is represented by i and j, and the load node numbers are represented by i and j. When the electrical distance between adjacent load nodes is detected to be greater than or equal to the line's constraint tolerance in the constructed flexible DC line optimization model, the segmented interconnection line interface is divided into several short interconnection line interfaces according to the distribution area conditions. The constraint environment variables of the short interconnection line interfaces are calculated, and the constraint tolerance is automatically adjusted based on the calculated constraint environment variables. The specific calculation formula for the constraint environment variables of the short interconnection line interfaces is as follows:

[0013] in, Indicates the electrical distance of short interconnect line interfaces. Indicates the electrical distance between adjacent load nodes. Indicates the number of short interconnect interface lines. Indicates the degree of tolerance to constraints; When the electrical distance between adjacent load nodes is detected to be less than the line's constraint tolerance in the constructed flexible DC line optimization model, the current configuration is recorded and the optimization process ends.

[0014] Optionally, monitoring the optimized line structure includes: The monitoring of power optimization allocation is completed using a low-voltage power distribution system control algorithm. Monitoring includes the system automatically collecting DC voltage and line current values ​​at the short-interconnection line interfaces. The voltage and current data are then transmitted to the central control system for data preprocessing. Data preprocessing includes calculating power characteristic values ​​based on the DC voltage and line current values. The specific formula for calculating the power characteristic values ​​is as follows:

[0015] in, Indicates the power characteristic value. Indicates the DC voltage value. Indicates the line current value; Based on the power characteristic value, the system detects whether there are voltage fluctuations, overcurrents, or power surges. If abnormal fluctuations are detected, the system will activate the fault diagnosis module to determine whether it is necessary to adjust the output of the distributed power supply, cut off non-critical loads, or switch to backup interconnection lines. If abnormal fluctuations still occur after the power is redistributed at the short interconnection line interface, the system will generate an alarm and arrange for maintenance personnel to check the interface connection status and equipment operating parameters. If the abnormal fluctuations affect the overall power distribution stability, the system will automatically trigger droop control or virtual synchronous machine control strategies to maintain the DC bus voltage stability.

[0016] Optionally, the integration test includes: The objective function of the power optimization model for both the flexible DC line structure and the flexible DC control module is calculated based on the power characteristic values, and the rules for simulating the actual operation scenario are based on the objective function. When the system detects that the real-time DC power of the short interconnection line interface is approaching the maximum allowable DC power, the system will automatically adjust the power distribution and transfer some of the distributed power output or load power to other short interconnection line interfaces. If the adjustment causes other short interconnection line interfaces to overload, the system will prioritize reducing the load on non-critical loads or temporarily cutting off the power supply to interruptible loads.

[0017] Optionally, the specific formula for calculating the objective function is as follows:

[0018] in, Indicates the maximum permissible DC power of the short interconnect line interface. This represents the real-time DC power at the short interconnect line interface, where λ represents the voltage stability weighting coefficient. Indicates the DC voltage deviation at the node. This represents the maximum allowable voltage deviation, i represents the interface index, and n represents the total number of interfaces.

[0019] Optionally, the flexible DC interconnection path for deploying low-voltage power distribution based on simulated scenarios includes: The current of the short interconnection line interface during the load reduction process is calculated by power allocation. Based on the current of the short interconnection line interface calculated during the load reduction process, the flexible DC interconnection path is adaptively learned to complete the flexible DC interconnection of low-voltage power distribution. After receiving feedback on the power allocation adjustment, the flexible DC control module will continue to monitor the operating status of the line structure, collect voltage, current and power data after the allocation adjustment, construct a power allocation strategy based on reinforcement learning, dynamically optimize the power flow distribution, and complete the power optimization allocation of the flexible DC interconnect.

[0020] Optionally, the specific formula for calculating the current at the short interconnection line interface during the load reduction process is as follows:

[0021] in, This indicates the current at the short interconnect interface during the load shedding process. This indicates the number of newly added interconnection lines. This indicates the total current before load reduction. This represents the current error term.

[0022] Furthermore, to achieve the above objectives, the present invention also provides a low-voltage power distribution flexible DC interconnection device, the device comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the steps of the low-voltage power distribution flexible DC interconnection method as described in any of the above descriptions.

[0023] This invention provides a method for flexible DC interconnection in low-voltage power distribution. The method calculates the electrical distance between load nodes and optimizes the line structure, reducing long-distance transmission losses and improving the energy efficiency and economy of the DC distribution network, while avoiding the redundancy problems of traditional fixed line layouts. It utilizes a flexible DC control module and intelligent algorithms to monitor power characteristics in real time, achieving dynamic power balance and adaptive adjustment among multiple nodes. This effectively addresses load fluctuations and the uncertainty of distributed energy output, enhancing system flexibility and anti-interference capabilities. The multi-functional interconnection interface design and segmented interconnection structure are compatible with various devices such as photovoltaics, energy storage, and charging piles, supporting plug-and-play functionality. The readily available and flexible expansion enhances the system's adaptability to new loads and distributed energy sources; integrated testing simulates actual operating scenarios, verifying the rationality of interconnection paths and the effectiveness of control strategies in advance, reducing on-site deployment risks; fault diagnosis and automatic power adjustment mechanisms can quickly respond to anomalies, maintain DC bus voltage stability, reduce the risk of local overload and voltage fluctuations, and ensure power supply quality; through flexible interconnection and intelligent control, it provides a dynamically reconfigurable topology and efficient management mode for low-voltage distribution networks, helping to build a low-carbon, efficient, and reliable new power system, meeting the technical requirements of smart grids for scenarios with high penetration rates of distributed energy sources. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating an embodiment of the low-voltage power distribution flexible DC interconnection method of the present invention.

[0025] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0026] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0027] Reference Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the flexible DC interconnection method for low-voltage power distribution according to the present invention.

[0028] In one embodiment, the low-voltage power distribution flexible DC interconnection method includes the following steps: Step S100: Design the interconnected flexible DC line structure according to the power distribution area requirements, calculate the electrical distance between adjacent load nodes, and optimize the line structure based on the electrical distance between adjacent load nodes.

[0029] Electrical distance can be a quantitative indicator reflecting the power transmission impedance characteristics between load nodes, taking into account the effects of resistance, inductance, and current distribution. It is understood that the equivalent electrical distance between nodes can be derived through power flow calculations or impedance matrix analysis based on the distribution network topology and the impedance parameters of each branch. In this embodiment, electrical distance can replace empirical wiring, guiding the optimization of DC line structures and reducing inefficient long-distance transmission losses. For example, the electrical distance can be one or more of the following: impedance-weighted electrical distance, power flow density-based electrical distance, and voltage drop accumulation-based electrical distance.

[0030] Calculating the electrical distance between adjacent load nodes and optimizing the wiring structure can be achieved by calculating the electrical distance between nodes based on network topology and impedance parameters, and then selecting low-impedance paths to form the optimal interconnection topology. Furthermore, calculating the electrical distance between adjacent load nodes and optimizing the wiring structure can be accomplished by using the minimum spanning tree algorithm to construct the optimal connection graph based on the electrical distance, and by using the particle swarm optimization algorithm to search for low-loss wiring combinations. This can eliminate redundancy in traditional radial wiring and reduce the average impedance of the transmission path.

[0031] In step S200, a flexible DC control module is introduced to monitor the optimized line structure, and a low-voltage power distribution system control algorithm is used to optimize the power allocation of the monitored line structure.

[0032] The flexible DC control module can be an intelligent control unit with power monitoring, dynamic adjustment, and fault response functions. It is understood that the flexible DC control module can collect voltage, current, and power data from each node through sensors, and generate control commands based on control algorithms to drive the power electronic converter to adjust the power flow direction. In this embodiment, the flexible DC control module can achieve real-time dynamic balance and adaptive adjustment of power among multiple nodes, improving the system's responsiveness to fluctuations. For example, the flexible DC control module can include, but is not limited to, one or more of the following: model predictive control-based modules, fuzzy logic-based modules, and reinforcement learning-based modules.

[0033] Utilizing low-voltage power distribution system control algorithms to optimize power allocation across monitored line structures can involve real-time acquisition of power data from each node and dynamic adjustment of converter outputs via control algorithms to achieve balanced power flow. Furthermore, this optimization can be achieved by employing distributed consensus algorithms for power negotiation between nodes and proportional-integral feedback control to regulate bus voltage deviations. This enables adaptive power allocation across multiple nodes and suppresses voltage drift caused by load fluctuations.

[0034] Step S300: Integrate and test the flexible DC line structure and flexible DC control module to simulate the actual operation scenario. Based on the simulated scenario, deploy the flexible DC interconnection path for low-voltage power distribution to complete the flexible DC interconnection of low-voltage power distribution.

[0035] The segmented interconnection structure can be a structure that divides the DC power distribution network into several independently controllable electrical sections. In this embodiment, the segmented interconnection structure can support local isolation and reconfiguration, improving system fault tolerance and scalability. For example, the segmented interconnection structure can adopt one or more of the following: a circuit breaker-based segmentation structure, a DC-DC converter-based segmentation structure, or a bus switchgear-based segmentation structure.

[0036] The multi-functional interconnect interface can be a standardized electrical interface that supports plug-and-play access for various types of distributed energy and load devices. In this embodiment, the multi-functional interconnect interface can be compatible with heterogeneous devices such as photovoltaics, energy storage, and charging piles, enabling rapid interconnection at the physical layer. For example, the multi-functional interconnect interface may include, but is not limited to, one or more of the following: interfaces supporting bidirectional power flow, interfaces supporting adaptive voltage matching, and interfaces supporting automatic identification of communication protocols.

[0037] Fault diagnosis and automatic power adjustment can detect abnormal bus voltage or branch overcurrent signals, triggering the control module to quickly redistribute power flow to maintain stability. Furthermore, fault diagnosis and automatic power adjustment can be achieved through rapid disconnection and redistribution based on threshold comparison, and predictive power migration based on impedance change trend prediction, thereby shortening fault response time and maintaining the DC bus voltage within the allowable fluctuation range.

[0038] Deploying flexible DC interconnection paths for low-voltage power distribution based on simulated scenarios allows for the construction of typical operating scenarios in a digital simulation environment, verifying the synergistic effectiveness of line structure and control strategies. Furthermore, this deployment can be achieved by building a multi-source load model (including photovoltaic, energy storage, and charging piles) using MATLAB / Simulink and conducting closed-loop testing by integrating a hardware-in-the-loop system with the actual control module. This approach allows for early identification of deployment risks and reduces the probability of on-site commissioning failures.

[0039] Taking the transformation of low-voltage DC power distribution in urban communities as an example, the flexible DC interconnection method for low-voltage power distribution in this embodiment can be used in communities equipped with rooftop photovoltaics, home energy storage, and electric vehicle charging piles. By calculating the electrical distance between each user node, the DC interconnection topology can be reconstructed to shorten the transmission path from photovoltaics to energy storage. The flexible DC control module monitors the power fluctuations of each node in real time. When the photovoltaic output increases sharply, the excess power is automatically directed to the nearest charging pile. The segmented structure isolates the faulty charging pile, and the multi-functional interface supports newly connected hydrogen fuel cell equipment. The integrated test verifies the stability of the control strategy in the combined scenario of cloudy days and peak electricity consumption. In case of a fault, the high-resistance path is automatically cut off and the power flow is reconstructed to ensure the stability of the bus voltage.

[0040] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By designing the interconnection flexible DC line structure according to the needs of the power distribution area and optimizing the line structure based on electrical distance, a flexible DC control module is introduced to monitor the optimized line structure and use control algorithms to achieve optimized power allocation. Through integrated testing and simulation scenario deployment of interconnection paths, the redundancy and high-loss paths of traditional wiring are eliminated, and dynamic redistribution of power at multiple nodes is achieved. It supports flexible access of heterogeneous devices, improves system fault tolerance and scalability, and can quickly reconstruct power flow under abnormal operating conditions to maintain voltage stability. It can achieve a comprehensive effect of improved system energy efficiency, enhanced anti-interference and flexible adaptation.

[0041] In some embodiments, designing an interconnected flexible DC line structure includes: analyzing the power consumption characteristics of load nodes, and designing flexible DC distribution units by investigating the power demand and voltage level of the power consumption characteristics. The flexible DC distribution unit includes a DC bus, interconnection interface, energy storage interface, distributed power source interface, and load interface. When a node requires additional interconnection capacity, the corresponding interface position is reserved in the design.

[0042] The flexible DC distribution unit can be a standardized distribution module integrating a DC bus and multiple functional interfaces, supporting the access and capacity expansion of various types of equipment. In this embodiment, the flexible DC distribution unit can construct a unified physical interface platform to realize modular interconnection of distributed power sources, energy storage, and load equipment. For example, the flexible DC distribution unit may include one or more of the following: unidirectional power flow flexible unit, bidirectional power flow flexible unit, and multi-voltage level compatible flexible unit. The DC bus can be a common DC conductor connecting all interfaces, undertaking the functions of power collection and distribution.

[0043] In this embodiment, the DC bus can serve as the power transmission backbone of the flexible DC distribution unit, supporting multi-interface collaborative operation. Furthermore, the DC bus can be connected in parallel with interconnect interfaces, distributed power source interfaces, load interfaces, and energy storage interfaces to form a unified power flow path. The interconnect interfaces can be standardized electrical connection ports used to connect adjacent load nodes or adjacent distribution units. In this embodiment, the interconnect interfaces enable flexible interconnection and topology reconfiguration of DC power between nodes.

[0044] For example, the interconnection interface can adopt a plug-and-play mechanical-electrical interface, an intelligent interface with a communication handshake protocol, or a hot-swappable protected interface. The distributed power interface can be a standardized DC interface specifically designed for connecting distributed power devices such as photovoltaic systems. In this embodiment, the distributed power interface can support plug-and-play for power supply devices and is compatible with MPPT output characteristics. For example, the distributed power interface can adopt a constant voltage power interface, a constant current power interface, or a wide-range input power interface. The load interface can be a standardized DC interface specifically designed for connecting terminal electrical equipment.

[0045] In this embodiment, the load interface can adapt to the voltage and power requirements of DC loads such as charging piles and lighting. For example, the load interface can be a constant power output type load interface, a dynamic voltage regulation type load interface, or a load identification type intelligent interface. The energy storage interface can be a standardized bidirectional power interface specifically designed for connecting to a battery energy storage system. In this embodiment, the energy storage interface can support the charging and discharging scheduling and voltage support functions of energy storage devices. For example, the energy storage interface can be a constant voltage charging type interface, a constant current discharging type interface, or a power bidirectional dynamic adjustment type interface. Designing a flexible DC power distribution unit and reserving interface positions to support capacity expansion can be done by reserving unused interface physical space and electrical paths during the power distribution unit design phase based on load forecasting and power consumption characteristic analysis. In this embodiment, designing a flexible DC power distribution unit and reserving interface positions to support capacity expansion can involve reserving redundant interface sockets and corresponding power level bus branches in the PCB or cabinet, or pre-setting interface identification protocols and power allocation strategies in the control system, which are automatically activated after equipment connection. This allows for future system expansion without reconstructing the main structure, reducing upgrade costs and downtime risks.

[0046] Calculating the electrical distance between adjacent load nodes involves calculations based on the node coordinates and line parameters. The specific calculation formula is as follows:

[0047] in, The distance between adjacent load nodes is represented by (x1,y1) and (x2,y2), which represent the physical coordinates of the load nodes, and k is the line parameter coefficient.

[0048] Electrical distance can be a quantitative indicator reflecting the power transmission impedance characteristics between load nodes, taking into account both physical coordinates and line parameters. In this embodiment, electrical distance can serve as a quantitative basis for line structure optimization, replacing empirical wiring and guiding the scientific layout of DC interconnection paths. Furthermore, electrical distance can be calculated using the Euclidean distance weighting formula based on the node physical coordinates (x1, y1), (x2, y2) and the line parameter coefficient k. Calculating the electrical distance between adjacent load nodes can be done by using the node physical coordinates and preset line parameter coefficients, calculated using the Euclidean distance weighting formula. Furthermore, the calculation of electrical distance between adjacent load nodes can be achieved by automatically extracting node coordinates using a GIS geographic information system and calling a parameter library, or by back-deriving an equivalent k value from on-site measured resistance-reactance data and substituting it into the formula. This allows for quantifiable and reproducible calculation of electrical distance, providing an objective basis for topology optimization.

[0049] The length of the DC interconnection line is determined based on the calculation results. When the electrical distance between adjacent load nodes exceeds a predetermined threshold, a segmented interconnection structure is used for layout.

[0050] The segmented interconnection structure can be a structure that divides a long-distance DC interconnection link into multiple independently controllable electrical segments. In this embodiment, the segmented interconnection structure can be automatically activated when the electrical distance exceeds a threshold, reducing voltage drop and power attenuation in a single link and improving transmission efficiency. Furthermore, the segmented interconnection structure can be activated when the electrical distance calculation result triggers a threshold, achieving segment isolation and voltage support by inserting DC-DC converters or switching units. For example, the segmented interconnection structure can adopt a segmented structure based on DC-DC converters, a segmented structure based on solid-state circuit breakers, or a segmented structure based on hybrid bus relay points. The segmented interconnection structure layout based on the electrical distance exceeding a predetermined threshold can be implemented by automatically triggering a segmented structure insertion mechanism when the calculated electrical distance exceeds the set threshold, inserting relay units in the long link. Furthermore, when the electrical distance exceeds a predetermined threshold, a segmented interconnection structure can be adopted for layout. This can be achieved by inserting DC-DC boost relay modules in the middle of long-distance paths to reduce voltage drop, or by setting solid-state circuit breakers and voltage support capacitors at the midpoint of the path to form segmented nodes. This can avoid the accumulation of voltage attenuation and power loss caused by a single long link and maintain the power supply quality at the end.

[0051] Taking a newly built low-carbon community DC power distribution system as an example, the low-voltage power distribution flexible DC interconnection method in this embodiment can be as follows: During the community planning stage, the location of each household's photovoltaic installation point and charging pile is obtained through GPS positioning, and the electrical distance between nodes is calculated; when the electrical distance from a photovoltaic node to the main bus exceeds the threshold, the system automatically inserts a DC-DC relay unit at the midpoint of the path to form a segmented structure; the flexible DC power distribution unit is uniformly deployed in the power distribution box of each building, integrating the DC bus and the reserved energy storage, load, and power interface; when adding hydrogen fuel cell equipment in the future, only the reserved interface needs to be inserted, and the system automatically identifies and connects the power flow without modifying the main line.

[0052] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By analyzing the power consumption characteristics of load nodes, it designs a flexible DC power distribution unit with multiple interfaces and reserves expansion interfaces. By calculating the electrical distance between adjacent nodes based on physical coordinates and line parameter coefficients, and by automatically activating a segmented interconnection structure for layout when the electrical distance exceeds a threshold, it can achieve quantitative design of power distribution topology, suppress voltage attenuation in long links, support modular access of equipment and seamless expansion, forming a complete technical closed loop from scientific wiring and dynamic response to elastic expansion, and supporting the long-term stable operation of the system in high-penetration scenarios.

[0053] In one embodiment, the use of a segmented interconnection structure for layout includes: designing intermediate interconnection interfaces; setting a DC output interface on one side of the flexible DC distribution unit and a DC input interface matching the output interface on the other side; when the layout of the flexible DC distribution unit cannot meet the actual location of the load node, the interface position is adjusted using the integrated structure of the multi-functional interconnection interface.

[0054] The segmented interconnection structure can be a structure that divides a DC interconnection link into multiple independently controllable electrical segments. This structure can support dynamic adjustment of interface positions and local power flow reconfiguration, enhancing the system's adaptability to physical layout deviations. In this embodiment, the segmented interconnection structure can work in conjunction with a multi-functional interconnection interface to maintain segment electrical continuity by inserting relay units when the interface position changes. For example, the segmented interconnection structure can adopt one or more of the following: a segmented structure based on a DC-DC converter, a segmented structure based on a solid-state circuit breaker, or a segmented structure based on a hybrid bus relay point.

[0055] The integration of multi-functional interconnect interfaces includes preprocessing for compatibility with different types of distributed power sources and loads. The specific preprocessing steps are as follows: The initial topology of the multi-functional interconnection interface is designed. Based on the requirements of the flexible DC power distribution unit, a segmented interconnection line interface of the multi-functional interconnection interface is constructed. The segmented interconnection line interface includes a plug, a socket and a communication protocol module. The socket includes specifications, rated current and insulation material. The sockets are subjected to electrical performance tests, including insulation resistance tests, current carrying capacity tests, and dynamic response tests. The results of the tests are used to classify the multi-functional interconnect interfaces. The classification includes storing information on different multi-functional interconnect interface types in an interface type database, and identifying the multi-functional interconnect interface type information stored in the interface type database; When interface type information is repeated during the identification process, it indicates that there is a conflict in the identified interface type information. In this case, the line structure is optimized based on the electrical distance between adjacent load nodes.

[0056] A multi-functional interconnect interface can be a standardized device access port integrating physical connection and communication identification functions. It supports plug-and-play heterogeneous distributed power supplies and loads, and can be used to achieve physical compatibility and protocol self-identification for device access, reducing system deployment complexity. In this embodiment, adjusting the interface position of the integrated structure of the multi-functional interconnect interface can be achieved by changing or relocating the plug-socket combination to adjust the access point based on the actual location of the load node and the layout deviation of the power distribution unit. For example, this adjustment can be achieved by using modular pluggable interface units to slide and reposition on the busbar, or by adding relay interface points to adapt to the offset node through an expanded segmented interconnect structure. This allows for flexible access under physical space constraints, and device deployment adjustments can be completed without rewiring.

[0057] Plugs and sockets can be interoperable mechanical and electrical components that constitute the physical connection part of a multifunctional interconnection interface, used to ensure the physical security and electrical current reliability of device access. In this embodiment, the socket has standardized specifications, rated current, and insulation material parameters, and the plug corresponds to and matches its mechanical structure and electrical contacts. For example, the socket can be one or more of the following: a single-core quick-connect socket, a multi-core locking socket, or a waterproof and dustproof socket.

[0058] The communication protocol module can be a digital identification and interaction unit embedded in a multi-functional interconnect interface, used to achieve automatic identification of access devices and matching of power allocation strategies. In this embodiment, the communication protocol module triggers a communication handshake when the plug is inserted into the socket, reading the device type, rated parameters, and control command set. For example, the communication protocol module can employ one or more of the following: a MODBUS-based protocol module, a CANopen-based protocol module, or a lightweight JSON-based protocol module.

[0059] An interface type database can be a digital information repository storing tested and certified multi-functional interconnect interface types and their electrical and communication characteristics. It can provide data for interface conflict detection and device compatibility matching. In this embodiment, testing the socket's insulation resistance, current carrying capacity, and dynamic response, and then classifying and storing the results in the interface type database, can be achieved by applying rated voltage and current to the socket, measuring the insulation resistance value, temperature rise, and transient response characteristics, outputting a standardized test report, and entering it into the database. Furthermore, this testing can be implemented using a high-voltage insulation tester and electronic load for static and dynamic combined testing, or by using an automated testing platform to collect data in batches and generate unique identifiers for storage in the database. This allows for the establishment of a traceable certification system for interface performance, supporting intelligent matching and security verification of connected devices.

[0060] When interface type information identifiers conflict, an electrical distance-driven line structure reconfiguration is triggered. This can be achieved by detecting duplicate or contradictory identifiers in the interface type database, automatically activating the electrical distance calculation module, and re-evaluating the optimal path between nodes. Furthermore, this reconfiguration can trigger a preset conflict resolution rule base, prioritizing the alternative path with the shortest electrical distance, or by activating a simulation engine to simulate the power distribution of conflicting paths and candidate paths, selecting the optimal solution. This creates a closed-loop feedback mechanism between interface compatibility and physical topology, ensuring the continuous and stable operation of the system under highly heterogeneous access conditions.

[0061] Electrical distance can be a quantitative indicator reflecting the power transmission impedance characteristics between load nodes. Incorporating the influence of physical coordinates and line parameters, it can serve as a trigger for line structure reconfiguration, driving dynamic topology adjustments in the event of interface conflicts. In this embodiment, the electrical distance is calculated using the formula above, based on a preset line parameter coefficient k of the node's physical coordinates. For example, the electrical distance can be one or more of the following: electrical distance weighted based on straight-line distance, electrical distance based on the actual path direction, or electrical distance corrected for soil / laying environment.

[0062] Taking the upgrade of DC power distribution in an old office area as an example, the low-voltage power distribution flexible DC interconnection method in this embodiment can be as follows: During the renovation, multiple photovoltaic and energy storage devices are added. Due to different manufacturers, their interface sockets cause communication protocols and rated current conflicts. The system detects duplicate identifiers in the database and immediately starts electrical distance calculation. It is found that the electrical distance of a certain node in the original connection path is significantly longer. The system automatically triggers the segmented structure to insert a relay unit and relocates the multi-functional interface to a nearby power distribution unit with a better electrical distance. The newly connected energy storage device is automatically identified and matched with the power strategy through the communication protocol module. The test record is updated to the database, the conflict is resolved, and the bus voltage remains stable.

[0063] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By designing intermediate interconnection interfaces and constructing segmented interconnection structures, the access positions are dynamically adjusted using the integrated characteristics of multi-functional interconnection interfaces. The electrical performance of the sockets is tested and stored in an interface type database. When interface type identifiers conflict, topology reconstruction is driven by electrical distance feedback, enabling the device access to have physical compatibility, protocol self-identification, and security verification capabilities. At the same time, in complex environments with limited space layout and heterogeneous equipment, path optimization and system stability are automatically achieved. This can achieve the technical effects of plug-and-play interfaces, topology self-adaptation, high operational reliability, and zero-contact expansion.

[0064] In one embodiment, optimizing the line structure based on the electrical distance between adjacent load nodes includes: constructing a flexible DC line optimization model based on the electrical distance between adjacent load nodes, and adding the line's constraint tolerance based on the voltage drop withstand capability analysis.

[0065] The specific calculation formula for constructing the optimization model of flexible DC lines is as follows:

[0066] Minimize represents the optimization objective of flexible DC lines. This indicates the intensity of the interconnection demand between nodes i and j. Indicates the priority of node interconnection. Indicates the electrical distance between adjacent load nodes. The line current is represented by n, the total number of load nodes in the distribution area is represented by i and j, and the load node numbers are represented by i and j. The flexible DC line optimization model can be a mathematical programming model that integrates node interconnection demand intensity, priority, and electrical distance with the objective of minimizing total transmission loss. It can be used to achieve quantitative optimization of the topology, replacing empirical wiring and reducing overall system power loss. In this embodiment, the flexible DC line optimization model can be constructed using an objective function. For example, the flexible DC line optimization model can include, but is not limited to, optimization models based on linear programming, optimization models based on integer programming, and optimization models based on multi-objective genetic algorithms.

[0067] Constructing an optimization model for flexible DC lines based on electrical distance and incorporating voltage drop constraints can be achieved by using node interconnection demand intensity, priority, and electrical distance as variables to build an objective function that minimizes total losses, and embedding voltage drop constraints to limit the product of current and distance. Furthermore, this optimization model can be implemented by using a linear programming solver to search for the globally optimal topology under node constraints, or by using heuristic algorithms to iteratively generate candidate connection graphs that meet the voltage drop upper limit. This allows for a shift in topology design from empirical judgment to mathematical optimization, improving the energy efficiency benchmark of the line structure.

[0068] When the electrical distance between adjacent load nodes is detected to be greater than or equal to the line's constraint tolerance in the constructed flexible DC line optimization model, the segmented interconnection line interface is divided into several short interconnection line interfaces according to the distribution area conditions. The constraint environment variables of the short interconnection line interfaces are calculated, and the constraint tolerance is automatically adjusted. The specific calculation formula for the constraint environment variables of the short interconnection line interfaces is as follows:

[0069] in, Indicates the electrical distance of short interconnect line interfaces. Indicates the electrical distance between adjacent load nodes. Indicates the number of short interconnect interface lines. Indicates the degree of tolerance to constraints; The segmented interconnection line interface can be a standardized interface unit that breaks down a long-distance DC interconnection link into multiple short-distance, independently deployable units. It can be used to segment paths under conditions of excessive electrical distance, reducing voltage drop and current density per segment. In this embodiment, the segmented interconnection line interface can be linked to constraint environmental variables, and its number and location are dynamically determined by these variables, forming a cascaded structure of variable length. For example, the segmented interconnection line interface can include, but is not limited to, short-segment interfaces based on DC-DC converters, short-segment interfaces based on solid-state circuit breakers, and short-segment interfaces based on bus relay capacitors.

[0070] The constraint tolerance can be the maximum electrical distance threshold allowed by the system for a single interconnected line segment. It is used to determine whether a segmentation mechanism is triggered and can serve as a trigger criterion for topology reconfiguration, ensuring that line voltage drop remains within a safe range. In this embodiment, the initial value of the constraint tolerance is set by the design specifications and dynamically adjusted during operation based on constraint environment variables. For example, the constraint tolerance may include, but is not limited to, tolerance based on temperature rise limits, tolerance based on current density limits, and tolerance based on voltage deviation tolerance.

[0071] Constraint environment variables can be comprehensive parameters reflecting the operating environment of the segmented short interconnect lines. They are used to dynamically adjust the tolerance of constraint conditions and can be used to make the constraint threshold adapt to the segmentation density and load distribution, thereby improving the adaptability of the optimization strategy to non-uniform scenarios. For example, constraint environment variables can include, but are not limited to, environment variables based on current distribution uniformity, environment variables based on temperature rise accumulation, and environment variables based on voltage gradient changes.

[0072] When the electrical distance is greater than or equal to the constraint tolerance, the line is decomposed into multiple short interconnect interfaces and constraint environment variables are calculated. This can be achieved by detecting that the electrical distance of a single segment exceeds the threshold, automatically splitting it into m short interconnect interfaces, and calculating new constraint environment variables based on the original distance and the number of segments. Furthermore, when the electrical distance is greater than or equal to the constraint tolerance, decomposing the line into multiple short interconnect interfaces and calculating constraint environment variables can be achieved by dividing the long link into m segments using an equidistant segmentation method and calculating the average electrical distance, or by using a non-uniform segmentation method based on load density to densify the segments in high-current areas and reduce local voltage drop. This can overcome the fixed threshold limitation and make the segmentation strategy respond to the actual operating load distribution.

[0073] When the electrical distance between adjacent load nodes is detected to be less than the line's constraint tolerance in the constructed flexible DC line optimization model, the current configuration is recorded and the optimization process ends.

[0074] Automatically adjusting constraint tolerance based on environmental constraints can be achieved by proportionally modifying the original tolerance using these constraints to generate a new threshold adapted to the current segmented structure. Furthermore, this automatic adjustment can be accomplished by using a linear regression model to fit the historical relationship between environmental variables and the optimal tolerance, dynamically predicting new thresholds, or by introducing a fuzzy inference system to adjust the tolerance based on current density and temperature rise trends. This allows for self-calibration of constraints, enabling the optimization model to possess environmental awareness and parameter evolution capabilities.

[0075] Taking dynamic DC power distribution in an industrial park as an example, the low-voltage power distribution flexible DC interconnection method in this embodiment can be as follows: In areas with dense photovoltaic arrays, uneven distribution of load nodes leads to a certain interconnection electrical distance far exceeding the initial tolerance; the system triggers a segmentation mechanism to split the long link into three short interfaces and calculates the increase in constraint environment variables caused by local current concentration; the system automatically adjusts the tolerance threshold accordingly to avoid excessive segmentation and resource redundancy; the new segment interface is connected to a DC-DC module to support voltage stability, and the optimal tolerance parameters under this environment are recorded for reuse in similar scenarios in the future; the entire process requires no manual intervention, and the topology continues to maintain a low-loss state.

[0076] This embodiment provides a low-voltage power distribution flexible DC interconnection method. It constructs a flexible DC line optimization model based on the electrical distance between adjacent load nodes and incorporates voltage drop constraints. When the electrical distance exceeds the limit, it automatically decomposes into multiple short interconnection interfaces and calculates constraint environment variables. The constraint tolerance is then dynamically adjusted based on these variables, causing the optimization model to terminate the process when it detects that the electrical distance has not exceeded the limit. This forms a closed-loop optimization chain of model-driven, over-limit detection, segmented reconstruction, and environmental feedback threshold self-adjustment. This allows the system to maintain the optimal transmission path and voltage stability even under non-uniform loads, dynamic output, and spatial constraints. The modular characteristics of the segmented interfaces naturally support expansion, ultimately constructing an intelligent power distribution architecture with parameter evolution capabilities and structural elasticity, achieving a three-in-one technical effect of improved energy efficiency, enhanced anti-interference, and adaptive evolution.

[0077] In one embodiment, monitoring the optimized line structure includes: monitoring power optimization allocation using a low-voltage power distribution system control algorithm. Monitoring includes the system automatically collecting DC voltage and line current values ​​at short interconnection line interfaces, transmitting the voltage and current data to a central control system for data preprocessing. Data preprocessing includes calculating power characteristic values ​​based on the DC voltage and line current values. The specific formula for calculating the power characteristic values ​​is as follows:

[0078] in, Indicates the power characteristic value. Indicates the DC voltage value. Indicates the line current value; Based on the power characteristic value, the system detects whether there are voltage fluctuations, overcurrents, or power surges. If abnormal fluctuations are detected, the system will activate the fault diagnosis module to determine whether it is necessary to adjust the output of the distributed power supply, cut off non-critical loads, or switch to backup interconnection lines. The power characteristic value can be an instantaneous power index calculated by real-time product of DC voltage and line current, reflecting the power transmission status of the node. In this embodiment, the power characteristic value can be directly calculated by collecting voltage and current values ​​at short interconnect interfaces using the instantaneous product formula, serving as a quantitative monitoring benchmark for power dynamic behavior and used to identify voltage fluctuations, overcurrent, and power surges. The central control system can be a centralized data processing unit responsible for collecting, preprocessing, and distributing power monitoring data. In this embodiment, the central control system can periodically acquire voltage and current data from each short interconnect interface via the communication bus, perform power characteristic value calculation and preliminary anomaly screening, serving as the central hub for monitoring data aggregation and preprocessing, supporting the input basis for fault diagnosis and control strategies. For example, the central control system can adopt one or more of the following: a central system based on SCADA architecture, a central system based on edge computing nodes, or a central system based on cloud platform remote monitoring.

[0079] The power characteristic value is calculated based on DC voltage and line current. This can be achieved by real-time acquisition of voltage and current signals from each short interconnection interface, using the instantaneous product formula. Calculate the power characteristic value. Furthermore, the power characteristic value can be calculated based on DC voltage and line current by using a high-speed ADC to synchronously sample the voltage and current waveforms and calculate the instantaneous power once per cycle, or by smoothing the sampled data with a digital filter and then calculating the average power characteristic value of the sliding window. This allows the establishment of a quantitative monitoring benchmark based on dynamic power changes, thereby improving the sensitivity to fluctuation events.

[0080] Detecting whether there are voltage fluctuations, overcurrents, or power abrupt changes in power characteristic values ​​can be achieved by comparing real-time power characteristic values ​​with preset safety thresholds or historical dynamic baselines to identify abnormal deviations. Furthermore, detecting voltage fluctuations, overcurrents, or power abrupt changes in power characteristic values ​​can be achieved by using the moving standard deviation method to detect whether the rate of power change exceeds the limit, or by using the exponentially weighted moving average method to identify persistent deviation trends. This allows for early identification of minor disturbances and sudden anomalies, providing a triggering basis for graded responses.

[0081] The fault diagnosis module can be a judgment unit that identifies potential fault sources based on abnormal patterns in power characteristic values ​​and generates local control commands. In this embodiment, the fault diagnosis module can identify abnormal patterns such as overcurrent, voltage surges / dips, or power jumps by comparing power characteristic values ​​with historical baselines or threshold ranges, and execute the first step of a graded response, prioritizing local adjustments to avoid global control intervention. For example, the fault diagnosis module can employ one or more of the following: a threshold comparison-based diagnosis module, a statistical deviation-based diagnosis module, or a machine learning pattern recognition-based diagnosis module. The fault diagnosis module can collaborate with the central control system, receiving the power data it collects and outputting a judgment result to the control strategy module regarding whether to trigger a voltage regulation strategy.

[0082] If abnormal fluctuations still occur after the power is redistributed at the short interconnection line interface, the system will generate an alarm and arrange for maintenance personnel to check the interface connection status and equipment operating parameters. If the abnormal fluctuations affect the overall power distribution stability, the system will automatically trigger droop control or virtual synchronous machine control strategies to maintain the DC bus voltage stability.

[0083] The droop control strategy can be a communication-free autonomous adjustment mechanism based on power-voltage or power-frequency negative feedback relationships, enabling distributed units to proportionally share power variations. In this embodiment, the droop control strategy can achieve current sharing by setting a slope characteristic where the voltage decreases linearly with increasing output power. Each node automatically adjusts its output voltage according to local power changes, achieving power self-balancing among multiple inverters under master station communication conditions and suppressing bus voltage drift. For example, the droop control strategy can employ one or more of the following: linear droop control strategy, nonlinear droop control strategy, or adaptive slope droop control strategy.

[0084] The virtual synchronous machine control strategy can be a power electronic control method that simulates the inertia and damping characteristics of a synchronous generator, stabilizing the voltage frequency through virtual impedance and virtual moment of inertia. In this embodiment, the virtual synchronous machine control strategy can construct a dynamic equation for the voltage frequency by introducing a virtual moment of inertia J and a damping coefficient D, enabling the inverter output to exhibit a disturbance rejection response similar to that of a synchronous machine, suppressing voltage oscillations caused by load changes or new energy fluctuations, and improving the transient stability of the system. For example, the virtual synchronous machine control strategy can employ one or more of the following: a virtual synchronous machine strategy based on a second-order inertial element, a virtual synchronous machine strategy based on damping compensation, or a virtual synchronous machine strategy based on phase-locked loop coordination.

[0085] When abnormal fluctuations persist, the system automatically triggers droop control or virtual synchronous machine control strategies. This means that if the fault diagnosis module fails to eliminate the fluctuations and the bus voltage continues to deviate from the allowable range, the system automatically activates the droop control or virtual synchronous machine control strategy. Furthermore, automatically triggering droop control or virtual synchronous machine control strategies when abnormal fluctuations persist can be achieved by prioritizing droop control to achieve communication-free current sharing, reducing control complexity, or by switching to the virtual synchronous machine strategy in high-inertia demand scenarios to enhance the system's damping characteristics. This provides a global voltage stabilization mechanism when local adjustment fails, ensuring that the bus voltage does not collapse.

[0086] Taking a residential DC microgrid with multiple photovoltaic (PV) connections as an example, the low-voltage distribution flexible DC interconnection method in this embodiment can be as follows: when the output of three PV units increases suddenly at the same time, causing a sudden rise in the power characteristic value of a short interconnection interface, the central control system detects overcurrent and triggers the fault diagnosis module; the diagnosis module prioritizes instructing low-priority loads to temporarily reduce their loads, but the voltage still fluctuates; the system automatically activates droop control, causing each inverter to reduce its output voltage proportionally to share the current; if the fluctuation continues, the system further switches to virtual synchronous machine mode, introducing virtual inertia to suppress voltage oscillation; the entire process requires no manual intervention, the bus voltage is maintained within ±3%, and normal control is automatically restored after the abnormality ends.

[0087] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By real-time acquisition of DC voltage and line current at the interface of short interconnection lines and calculation of power characteristic values, voltage fluctuations, overcurrents, or power surges are detected based on these indicators. Local regulation is performed by the fault diagnosis module. When the abnormality persists, droop control or virtual synchronous machine control strategies are automatically triggered. A complete closed-loop response chain from data acquisition, dynamic monitoring, local diagnosis to global voltage stabilization is constructed. This enables the system to collaboratively maintain the stability of the DC bus voltage without a master control center. It achieves adaptive suppression of complex disturbances such as drastic fluctuations in new energy output and multi-node load surges. This breaks through the traditional passive response mode that relies on manual intervention or fixed thresholds, and achieves the technical effect of highly robust and self-healing operation.

[0088] In one embodiment, integration testing includes: The objective function of the power optimization model for both the flexible DC line structure and the flexible DC control module is calculated based on the power characteristic values, and the rules for simulating the actual operation scenario are based on the objective function. When the system detects that the real-time DC power of the short interconnection line interface is approaching the maximum allowable DC power, the system will automatically adjust the power distribution and transfer some of the distributed power output or load power to other short interconnection line interfaces. If the adjustment causes other short interconnection line interfaces to overload, the system will prioritize reducing the load on non-critical loads or temporarily cutting off the power supply to interruptible loads.

[0089] The objective function of the power optimization model can be a mathematical optimization objective expression that takes power characteristic values ​​as input variables and integrates line losses, node priorities, and equipment capacity constraints. This expression can be used to drive the dynamic generation of power allocation strategies in simulation scenarios and support the quantitative verification of operating rules. In this embodiment, the objective function of the power optimization model can calculate power characteristic values ​​based on real-time collected voltage and current data, and substitute them into a preset objective function, such as a weighted combination of minimizing total losses, balancing load distribution, and maximizing renewable energy consumption, to achieve coordinated optimization of structure and control. For example, the objective function of the power optimization model can include one or more of the following: functions aiming to minimize total losses, functions aiming to balance load distribution, and functions aiming to maximize renewable energy consumption.

[0090] A short interconnect line interface can be a standardized, short-distance power transmission unit formed by breaking down a long-distance DC interconnect link. It can serve as the smallest execution unit for power monitoring and redistribution, carrying local power flow and capacity constraints. In this embodiment, the short interconnect line interface is linked to the objective function of the power optimization model, with its real-time power value serving as the input to the objective function; it is also linked to interruptible loads, receiving power-off commands in case of overload. For example, the short interconnect line interface can be a short interface based on a DC-DC converter, a short interface based on a solid-state circuit breaker, or a short interface based on a bus relay capacitor, etc.

[0091] Interruptible loads can be non-critical electrical equipment or load groups that can be temporarily disconnected without affecting core functions when the system is overloaded. They can serve as a system safety buffer mechanism, providing controllable load reduction when power redistribution triggers a cascading overload. In this embodiment, the interruptible load is preset with load type and interrupt priority via a communication interface, and the control module executes a power-off command under trigger conditions. For example, interruptible loads may include interruptible loads such as air conditioning and lighting, slow-charging charging piles, and non-real-time industrial equipment.

[0092] The objective function for calculating the power optimization model of the flexible DC line structure and control module based on power characteristic values ​​can be achieved by taking the real-time power characteristic values ​​of each short interconnection interface as input and substituting them into the joint optimization objective function to calculate the system's comprehensive optimization index under the current operating state. Furthermore, the objective function for calculating the power optimization model of the flexible DC line structure and control module based on power characteristic values ​​can be generated by using a linear weighted method to integrate the three indices of loss, load balancing, and voltage deviation, or by using model predictive control to continuously optimize the power allocation scheme for the next cycle. This enables the collaborative evaluation of the line structure and control strategy in the simulation scenario, supporting the dynamic generation of operating rules.

[0093] When the power of a short interconnection interface approaches its maximum allowable value, the distributed power source or load power is automatically transferred. This can be achieved by detecting that the power of an interface exceeds 85% of a set threshold, triggering a power redistribution algorithm to transfer some output or load to a nearby low-load interface. Furthermore, this automatic transfer of distributed power source or load power when the power of a short interconnection interface approaches its maximum allowable value can be achieved by prioritizing power transfer to the nearest available interface based on electrical distance, or by migrating high-value loads to redundant capacity interfaces based on node priority. This breaks through the fixed allocation mode, achieves dynamic load balancing, and delays the system from entering an overload state.

[0094] When overloads occur at other short interconnection interfaces, non-critical or interruptible loads are prioritized for disconnection. This can be achieved by disconnecting interruptible loads according to a preset priority order when power transfer causes other interfaces to exceed limits, thus avoiding triggering voltage regulation strategies or equipment protection. Furthermore, prioritizing the disconnection of non-critical or interruptible loads when overloads occur at other short interconnection interfaces can be achieved by prioritizing the disconnection of non-survivable loads based on user level or load type, or by selecting the load group with the least impact on users based on historical interruption records. This proactively releases system capacity without interrupting critical power supply, preventing cascading overloads and voltage collapse.

[0095] Taking a commercial complex DC power distribution system as an example, the low-voltage power distribution flexible DC interconnection method in this embodiment can be as follows: During the midday peak period, the rooftop photovoltaic power output increases dramatically, and multiple charging piles operate at full load, causing the power of a certain short interconnection interface to reach 92% of its rated capacity; the system judges the current state as high-risk based on the objective function and automatically transfers part of the photovoltaic power to the adjacent low-load interface; this operation causes the power of another interface to rise to 95%, and the system then activates the interruptible load shedding strategy, prioritizing the shutdown of non-emergency lighting and charging piles during off-peak hours; the system maintains continuous power supply to critical loads, the bus voltage is stable, and no drooping or virtual synchronous machine control is triggered, avoiding unnecessary dynamic responses.

[0096] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By calculating the objective function of the power optimization model of the flexible DC line structure and control module based on the power characteristic value, dynamic operation rules are generated. When the power of the short interconnection line interface approaches the limit, the distributed power source or load power is automatically transferred to delay overload. If the transfer causes a chain overload, interruptible loads are cut off first as a safety buffer. It can achieve a three-level closed-loop mechanism of objective function prediction, dynamic redistribution and graded load reduction. Under high-density distributed energy access and multi-point power coupling, it realizes a smart operation paradigm that flexibly responds to fluctuations, ensures uninterrupted core loads, and avoids voltage collapse. It forms an adaptive capability with capacity elasticity, risk gradient and priority awareness, breaks through the binary dilemma of traditional systems that are either overloaded or de-voltaged, and realizes the adaptive capability within the operating boundary of stable supply, no over-limit, and no collapse in high penetration scenarios.

[0097] In one embodiment, the specific formula for calculating the objective function is as follows:

[0098] in, Indicates the maximum permissible DC power of the short interconnect line interface. This represents the real-time DC power at the short interconnect line interface, where λ represents the voltage stability weighting coefficient. Indicates the DC voltage deviation at the node. This represents the maximum allowable voltage deviation, i represents the interface index, and n represents the total number of interfaces.

[0099] The objective function can be a multivariate optimization objective expression that integrates the power utilization level of short interconnect interfaces and node voltage deviation. This expression quantifies the conflict between power overload risk and voltage stability, serving as the basis for optimizing power redistribution strategies. In this embodiment, the objective function is constructed by weighted summation of the ratio of real-time power to maximum allowable power at each interface and the ratio of node voltage deviation to maximum allowable deviation. Its calculation process does not directly rely on a single variable but reflects the system's operating state through the collaborative action of multiple parameters. For example, the objective function can include, but is not limited to, one or more of the following: an optimization function primarily focused on power load balancing, an optimization function primarily focused on voltage stability, and a weighted optimization function with both power and voltage objectives.

[0100] The voltage stability weighting coefficient λ can be a dynamic parameter used to adjust the relative importance of the voltage deviation penalty term in the objective function, enabling adaptive switching of priority between power load and voltage stability, and avoiding exacerbating voltage instability due to current sharing. In one specific embodiment, the voltage stability weighting coefficient λ is dynamically adjusted based on the ratio of the node voltage deviation to the maximum allowable deviation. When this ratio exceeds 0.7, λ automatically increases to strengthen voltage stability constraints. Furthermore, the voltage stability weighting coefficient λ can be adjusted in three stages based on the deviation ratio using fuzzy logic, or an adaptive function can be trained based on historical operating data to achieve predictive adjustment for typical disturbance modes.

[0101] A short interconnection line interface can be a standardized, short-distance power transmission unit formed by breaking down a long-distance DC interconnection link. It serves as the smallest computational unit for the objective function, carrying real-time power and capacity constraint data. In this embodiment, the real-time power and maximum allowable power of the short interconnection line interface are directly input into the objective function, and the voltage deviation of its connection nodes participates in the dynamic adjustment calculation of the voltage stability weighting coefficient λ. For example, the short interconnection line interface can be a short interface based on a DC-DC converter, a short interface based on a solid-state circuit breaker, or a short interface based on a bus relay capacitor, etc.

[0102] The node DC voltage deviation can be the difference between the actual DC voltage of the node and the rated voltage, reflecting the degree of local voltage deviation. As a direct measure of voltage stability, it drives the dynamic adjustment of λ and the optimization direction of the objective function. In an exemplary embodiment, the node DC voltage deviation is calculated by acquiring the actual node voltage through a voltage sensor and comparing it with a preset rated value. Furthermore, the node DC voltage deviation can include positive voltage deviation, negative voltage deviation, and continuously accumulating voltage deviation, etc.

[0103] The objective function based on the real-time power and maximum allowable power of short interconnection line interfaces can be constructed by using the ratio of the real-time power to the maximum allowable power of each interface as a load risk term, which is then accumulated to form the basic optimization term. Furthermore, the objective function based on the real-time power and maximum allowable power of short interconnection line interfaces can be constructed by using a linear normalization method to calculate the power ratio, or by introducing a nonlinear penalty term such as a square term to strengthen the cost of overload risk. This allows for the quantification of power utilization risk at each interface, providing the system with overload warnings and a basis for triggering reallocation.

[0104] Introducing a voltage stability weighting coefficient λ to balance the optimization priority of power load and voltage deviation can be achieved by multiplying the ratio of the voltage deviation term to the maximum allowable deviation by λ and adding it to the objective function, making voltage stability an optimization dimension with equal weight to power load. Furthermore, the optimization priority of balancing power load and voltage deviation can be achieved by dynamically setting the λ value using fuzzy logic or by training an adaptive function based on historical operating data. This allows for a dynamic trade-off between power balance and voltage stability, avoiding exacerbating voltage fluctuations by reducing load.

[0105] Taking a DC microgrid in an office area with high photovoltaic penetration as an example, the low-voltage distribution flexible DC interconnection method in this embodiment can be as follows: In the evening, the photovoltaic output drops sharply, and multiple household loads start up simultaneously, causing the voltage deviation of a certain short interconnection interface to approach 80% of the maximum allowable deviation. The system automatically increases λ from 0.3 to 2.0, and the objective function prioritizes suppressing voltage drop, pausing the transfer of power to that interface. Even if its load rate is only 75%, the system still maintains the existing distribution and instead reduces non-critical loads from the adjacent low voltage deviation interface. The bus voltage is stabilized within ±2%, and droop control is not triggered, avoiding cascading voltage drops caused by forced current sharing.

[0106] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By constructing a comprehensive objective function that integrates power utilization ratio and voltage deviation, and introducing a dynamically adjustable voltage stability weight coefficient λ, the system achieves mathematically adjustable coordination between power load risk and voltage stability. When the voltage approaches the critical point, it automatically abandons current sharing priority and prioritizes voltage stabilization to prevent cascading instability caused by blind power transfer. Thus, under the severe fluctuations of distributed energy and multi-point load coupling, it ensures that the bus voltage does not exceed the limit and avoids local overload, achieving a composite technical effect that combines voltage stability, load controllability, and system robustness.

[0107] In some embodiments, deploying a flexible DC interconnection path for low-voltage power distribution based on a simulated scenario includes: calculating the current of the short interconnection line interface during the load shedding process through power allocation; based on the calculated short interconnection line interface current during the load shedding process, enabling the flexible DC interconnection path to perform adaptive learning to complete the flexible DC interconnection of low-voltage power distribution; after feedback of power allocation adjustment, the flexible DC control module will continue to monitor the operating status of the line structure, collect voltage, current and power data after allocation adjustment, construct a power allocation strategy based on reinforcement learning, dynamically optimize power flow distribution, and complete the power optimization allocation of the flexible DC interconnection.

[0108] The short interconnect line interface can be a standardized, short-distance power transmission unit formed by splitting a long-distance DC interconnect link. It can serve as the smallest execution unit and state observation point for power adjustment, and its current change is the input state signal for the reinforcement learning strategy. In this embodiment, the current data of the short interconnect line interface can be collected by a flexible DC control module to construct the state space of reinforcement learning; its response results constitute the output feedback for action execution. For example, the short interconnect line interface can include, but is not limited to, one or more of the following: short interfaces based on DC-DC converters, short interfaces based on solid-state circuit breakers, and short interfaces based on bus relay capacitors.

[0109] The flexible DC control module can be an intelligent control unit with power monitoring, dynamic adjustment, and learning data acquisition functions. It can serve as the executor of reinforcement learning strategies and a data acquisition center, supporting state perception, action implementation, and experience playback. In this embodiment, the flexible DC control module can collect voltage, current, and power data from each node through sensors, generate control commands based on control algorithms, and record the adjusted operating state for strategy training. For example, the flexible DC control module can be linked with a short interconnect line interface to obtain real-time operating data; forming a perception-execution-feedback-learning closed loop with the reinforcement learning-based power allocation strategy.

[0110] A power allocation strategy based on reinforcement learning can be an adaptive control model that learns optimal power allocation decisions through interaction with the environment. Using states, actions, and rewards as basic elements, it can be used to continuously optimize power flow distribution during dynamic operation, achieving autonomous decision-making and long-term strategy evolution in the face of unknown disturbances. In this embodiment, the power allocation strategy based on reinforcement learning can use voltage, current, and power data collected by a flexible DC control module as the state space, the action space as the power allocation adjustment amount for each interface, and the reward function designed based on voltage stability, load balancing, and loss minimization.

[0111] Adaptive learning of flexible DC interconnect paths based on current changes during load shedding can be achieved by monitoring the dynamic changes in current at each short interconnect interface after a load shedding operation, and using these changes as input signals for system state evolution into the learning model. Furthermore, adaptive learning of flexible DC interconnect paths based on current changes during load shedding can be achieved by encoding the magnitude and duration of current spikes as state features to update the Q-value table of the policy network, or by statistically analyzing current change trends through a sliding time window as environmental feedback signals for reinforcement learning. This allows the system to extract key operational experience from load shedding events and form a causal understanding of the consequences of power reallocation.

[0112] Collecting voltage, current, and power data after power adjustments allows for the construction of a power allocation strategy based on reinforcement learning. This can be achieved by continuously recording the system state and corresponding actions after each power adjustment, and building an experience replay pool for strategy training. Furthermore, collecting voltage, current, and power data after power adjustments and constructing a power allocation strategy based on reinforcement learning can be achieved by using a deep Q-network to optimize the discrete action space, or by using a proximal policy optimization algorithm to learn a stable policy for continuous power adjustments. This allows the strategy to automatically extract the optimal decision rules from historical operations, adapting to complex operating conditions without the need for a pre-set mathematical model.

[0113] Taking a community DC microgrid with multiple photovoltaic and energy storage collaborations as an example, the low-voltage distribution flexible DC interconnection method in this embodiment can be as follows: In the scenario of alternating photovoltaic fluctuations and peak charging of charging piles over several consecutive days, the system triggers the load reduction mechanism multiple times; after each load reduction, the flexible DC control module records the current changes, voltage response, and load adjustment results of each short interface; the reinforcement learning module learns gradually through experience playback that: "when the current of a certain interface rises by more than 15% within 10 minutes and the voltage of adjacent nodes drops, non-critical loads should be cut off first rather than photovoltaic power transfer"; after dozens of interactions, the strategy automatically avoids the previous erroneous actions that caused voltage drops, and predictively adjusts the power flow in advance under new scenarios to achieve stable operation with zero overload and no voltage limit exceedance.

[0114] This embodiment provides a low-voltage power distribution flexible DC interconnection method. It drives adaptive learning by monitoring current changes at short interconnection line interfaces during load shedding. It constructs a power allocation strategy based on reinforcement learning by collecting voltage, current, and power data after power adjustment. A flexible DC control module achieves closed-loop operation of state perception, action execution, and experience playback. This allows for continuous strategy evolution driven by a "state-action-reward" mechanism, overcoming the limitations of traditional algorithms' poor generalization ability to unknown disturbances and achieving a leap from passive response to proactive prediction. The system continuously optimizes decision logic without requiring manual modeling and parameter tuning, achieving "long-term self-optimization, scenario self-adaptation, and strategy self-evolution" capabilities in high-uncertainty, multi-source coupled environments. This injects a true intelligent core into the low-voltage DC distribution network, significantly improving the system's adaptability to renewable energy fluctuations, dynamic balance accuracy, and long-term operational reliability without increasing hardware costs.

[0115] In one embodiment, the specific formula for calculating the current of the short interconnect interface during the load reduction process is as follows:

[0116] in, This indicates the current at the short interconnect interface during the load shedding process. This indicates the number of newly added interconnection lines. This indicates the total current before load reduction. This represents the current error term.

[0117] In this embodiment, the current at the short interconnect interface during the load shedding process can be the instantaneous current value redistributed to each short interconnect interface after partial load removal or disconnection of redundant paths. This can be used to achieve quantified redistribution of power flow after load shedding, avoiding local overload caused by current sharing assumptions. In this embodiment, the current at the short interconnect interface during the load shedding process can be based on the total system current before load shedding, the number of newly added interconnects, and the current error term. The above calculations are used to determine the current. For example, the current of the short interconnect interface during the load shedding process may include, but is not limited to, one or more of the following: interface current based on equal current sharing, interface current based on impedance weighting, and interface current based on dynamic load response.

[0118] Among them, the current error term This can be a dynamic correction variable used to compensate for the deviation between the actual current distribution and the ideal model during load shedding, and can be used to enhance the robustness of the current calculation model to nonlinear impedance, power supply response hysteresis, and sensor noise. In a specific embodiment, the current error term ϵ can be derived comprehensively from non-ideal factors such as voltage fluctuations, equipment response delays, and measurement noise monitored in real time by the system, and is an implicit adaptive compensation term. For example, the current error term... Error compensation terms may include, but are not limited to, error compensation terms based on temperature drift, error compensation terms based on communication delay, and error compensation terms based on dynamic charging and discharging of energy storage.

[0119] Calculating the current at short interconnection interfaces during load shedding based on the number of newly added interconnections and the total current before load shedding can be achieved by dynamically calculating the reconfiguration current value carried by each interface based on the total system current before load shedding, the number of newly added short interconnections, and implicit error terms. Furthermore, calculating the current at short interconnection interfaces during load shedding based on the number of newly added interconnections and the total current before load shedding can be done using a fixed-ratio allocation method, where... Predicted using a historical error statistical model, or using an online learning method: updated online based on the measured current deviations from the most recent load shedding events. The weighting coefficients are implemented so that the current redistribution after load reduction has the ability to adapt to non-ideal conditions, avoiding local overload and voltage collapse caused by the ideal current sharing assumption.

[0120] Taking the sudden load shedding of a community DC power distribution system as an example, the low-voltage power distribution flexible DC interconnection method in this embodiment can be as follows: when a charging pile is forcibly disconnected due to a fault, the system triggers a load reduction process; the central control system identifies that the number of newly added interconnection paths is 2, and the original total current is 45A; according to the formula, the new current of each short interface is calculated to be 22.5A+. The system calculates the voltage drop and inverter response delay in real time. The current is set to +1.2A, and the final current of each interface is set to 23.7A. This value is used to adjust the droop control slope and the segment interface current limiting threshold to ensure that all short circuits operate within a safe range and to avoid fuse malfunctions caused by sudden current surges.

[0121] This embodiment provides a low-voltage power distribution flexible DC interconnection method. By calculating the interface current of short interconnection lines based on the total current before load reduction and the number of newly added interconnection lines, and introducing a current error term for dynamic correction, the current redistribution model can absorb uncertainties such as power supply response delay, impedance nonlinearity and measurement noise, and achieve autonomous power rebalancing without central dispatch. This maintains stable system operation during dynamic topology reconfiguration and significantly reduces voltage fluctuations and equipment overload risks.

[0122] Furthermore, this embodiment of the invention also proposes a storage medium storing a low-voltage power distribution flexible DC interconnection program, which, when executed by a processor, implements the steps of the low-voltage power distribution flexible DC interconnection method described above.

[0123] Furthermore, this invention also proposes a low-voltage power distribution flexible DC interconnection device, the device comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the steps of the low-voltage power distribution flexible DC interconnection method as described in any of the above descriptions.

[0124] Other embodiments or specific implementations of the low-voltage power distribution flexible DC interconnection equipment described in this invention can be referred to the above-described method embodiments, and will not be repeated here.

[0125] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for flexible DC interconnection in low-voltage power distribution, characterized in that, The method includes: Design a flexible DC line structure for interconnection based on the needs of the power distribution area, calculate the electrical distance between adjacent load nodes, and optimize the line structure based on the electrical distance between adjacent load nodes; A flexible DC control module is introduced to monitor the optimized line structure, and a low-voltage power distribution system control algorithm is used to optimize the power allocation of the monitored line structure. The flexible DC line structure and flexible DC control module are integrated and tested to simulate actual operating scenarios. Based on the simulated scenarios, the flexible DC interconnection path for low-voltage power distribution is deployed to complete the flexible DC interconnection of low-voltage power distribution.

2. The low-voltage power distribution flexible DC interconnection method as described in claim 1, characterized in that, The interconnected flexible DC line structure includes: Analyze the power consumption characteristics of load nodes, and design flexible DC distribution units by investigating the power demand and voltage level of the power consumption characteristics. The flexible DC distribution unit includes a DC bus, interconnection interface, energy storage interface, distributed power source interface and load interface. When the node requires additional interconnection capacity, the corresponding interface position is reserved in the design. Calculating the electrical distance between adjacent load nodes involves calculations based on the node coordinates and line parameters. The specific calculation formula is as follows: in, The distance between adjacent load nodes is represented by (x1,y1) and (x2,y2), which represent the physical coordinates of the load nodes, and k is the line parameter coefficient. The length of the DC interconnection line is determined based on the calculation results. When the electrical distance between adjacent load nodes exceeds a predetermined threshold, a segmented interconnection structure is used for layout.

3. The low-voltage power distribution flexible DC interconnection method as described in claim 2, characterized in that, The use of a segmented interconnect structure for layout includes: Design the intermediate interconnection interface; set a DC output interface on one side of the flexible DC distribution unit and a DC input interface matching the output interface on the other side. When the layout of the flexible DC distribution unit cannot meet the actual location of the load node, the interface position is adjusted by using the integrated structure of the multi-functional interconnection interface. The integration of multi-functional interconnect interfaces includes preprocessing for compatibility with different types of distributed power sources and loads. The specific preprocessing steps are as follows: The initial topology of the multi-functional interconnection interface is designed. Based on the requirements of the flexible DC power distribution unit, a segmented interconnection line interface of the multi-functional interconnection interface is constructed. The segmented interconnection line interface includes a plug, a socket and a communication protocol module. The socket includes specifications, rated current and insulation material. The sockets are subjected to electrical performance tests, including insulation resistance tests, current carrying capacity tests, and dynamic response tests. The results of the tests are used to classify the multi-functional interconnect interfaces. The classification includes storing information on different multi-functional interconnect interface types in an interface type database, and identifying the multi-functional interconnect interface type information stored in the interface type database; When interface type information is repeated during the identification process, it indicates that there is a conflict in the identified interface type information. In this case, the line structure is optimized based on the electrical distance between adjacent load nodes.

4. The low-voltage power distribution flexible DC interconnection method as described in claim 3, characterized in that, The line structure optimized based on the electrical distance between adjacent load nodes includes: An optimization model for flexible DC lines is constructed based on the electrical distance between adjacent load nodes. The line's voltage drop withstand capability is analyzed, and the line's constraint withstand capability is added. The specific calculation formula for constructing the optimization model of flexible DC lines is as follows: Minimize represents the optimization objective of flexible DC lines. This indicates the intensity of the interconnection demand between nodes i and j. Indicates the priority of node interconnection. Indicates the electrical distance between adjacent load nodes. The line current is represented by n, the total number of load nodes in the distribution area is represented by i and j, and the load node numbers are represented by i and j. When the electrical distance between adjacent load nodes is detected to be greater than or equal to the line's constraint tolerance in the constructed flexible DC line optimization model, the segmented interconnection line interface is divided into several short interconnection line interfaces according to the distribution area conditions. The constraint environment variables of the short interconnection line interfaces are calculated, and the constraint tolerance is automatically adjusted based on the calculated constraint environment variables. The specific calculation formula for the constraint environment variables of the short interconnection line interfaces is as follows: in, Indicates the electrical distance of short interconnect line interfaces. Indicates the electrical distance between adjacent load nodes. Indicates the number of short interconnect interface lines. Indicates the degree of tolerance to constraints; When the electrical distance between adjacent load nodes is detected to be less than the line's constraint tolerance in the constructed flexible DC line optimization model, the current configuration is recorded and the optimization process ends.

5. The low-voltage power distribution flexible DC interconnection method as described in claim 4, characterized in that, The monitoring of the optimized line structure includes: The monitoring of power optimization allocation is completed using a low-voltage power distribution system control algorithm. Monitoring includes the system automatically collecting DC voltage and line current values ​​at the short-interconnection line interfaces. The voltage and current data are then transmitted to the central control system for data preprocessing. Data preprocessing includes calculating power characteristic values ​​based on the DC voltage and line current values. The specific formula for calculating the power characteristic values ​​is as follows: in, Indicates the power characteristic value. Indicates the DC voltage value. Indicates the line current value; Based on the power characteristic value, the system detects whether there are voltage fluctuations, overcurrents, or power surges. If abnormal fluctuations are detected, the system will activate the fault diagnosis module to determine whether it is necessary to adjust the output of the distributed power supply, cut off non-critical loads, or switch to backup interconnection lines. If abnormal fluctuations still occur after the power is redistributed at the short interconnection line interface, the system will generate an alarm and arrange for maintenance personnel to check the interface connection status and equipment operating parameters. If the abnormal fluctuations affect the overall power distribution stability, the system will automatically trigger droop control or virtual synchronous machine control strategies to maintain the DC bus voltage stability.

6. The low-voltage power distribution flexible DC interconnection method as described in claim 5, characterized in that, The integration test includes: The objective function of the power optimization model for both the flexible DC line structure and the flexible DC control module is calculated based on the power characteristic values, and the rules for simulating the actual operation scenario are based on the objective function. When the system detects that the real-time DC power of the short interconnection line interface is approaching the maximum allowable DC power, the system will automatically adjust the power distribution and transfer some of the distributed power output or load power to other short interconnection line interfaces. If the adjustment causes other short interconnection line interfaces to overload, the system will prioritize reducing the load on non-critical loads or temporarily cutting off the power supply to interruptible loads.

7. The low-voltage power distribution flexible DC interconnection method as described in claim 6, characterized in that, The specific formula for calculating the objective function is as follows: in, Indicates the maximum permissible DC power of the short interconnect line interface. This represents the real-time DC power at the short interconnect line interface, where λ represents the voltage stability weighting coefficient. Indicates the DC voltage deviation at the node. This represents the maximum allowable voltage deviation, i represents the interface index, and n represents the total number of interfaces.

8. The low-voltage power distribution flexible DC interconnection method as described in claim 7, characterized in that, The flexible DC interconnection path for deploying low-voltage power distribution based on simulated scenarios includes: The current of the short interconnection line interface during the load reduction process is calculated by power allocation. Based on the current of the short interconnection line interface calculated during the load reduction process, the flexible DC interconnection path is adaptively learned to complete the flexible DC interconnection of low-voltage power distribution. After receiving feedback on the power allocation adjustment, the flexible DC control module will continue to monitor the operating status of the line structure, collect voltage, current and power data after the allocation adjustment, construct a power allocation strategy based on reinforcement learning, dynamically optimize the power flow distribution, and complete the power optimization allocation of the flexible DC interconnect.

9. The low-voltage power distribution flexible DC interconnection method as described in claim 5, characterized in that, The specific formula for calculating the current at the short interconnection line interface during the load shedding process is as follows: in, This indicates the current at the short interconnect interface during the load shedding process. This indicates the number of newly added interconnection lines. This indicates the total current before load reduction. This represents the current error term.

10. A low-voltage power distribution flexible DC interconnection device, characterized in that, The device includes: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the steps of the low-voltage power distribution flexible DC interconnection method as described in any one of claims 1-9.

Images