Park-level flexible micro-grid system and control method

Abstract

The invention provides a park-level flexible micro-grid system and a control method, and the system is based on an AC-DC hybrid power grid architecture, and comprises an electric energy router, a flexible interconnection device, a full-DC building and a DC bus. Wherein the system is connected with a plurality of power distribution areas through the electric energy router and the flexible interconnection device; the direct current bus is directly connected with a photovoltaic module, an energy storage unit and a direct current load of a power distribution area, and at least part of the direct current load is configured in a full direct current building; the system is also provided with an energy management system and a virtual power plant. According to the method, a system comprising cross-transformer-area flexible direct current interconnection and energy routing is established, dynamic routing and real-time sharing of power between transformer areas are achieved in combination with a corresponding control method, output is achieved through an energy storage reverse support main bus in the load peak period, power utilization continuity and reliability are ensured, and further, the power utilization efficiency is improved. The interconnection structure breaks through the island operation limitation of a traditional micro-grid, improves the flexibility and redundancy of the system, and helps to alleviate the overload problem of a distribution transformer.

Description

Technical Field

[0001] This invention relates to the field of energy technology, specifically to a campus-level flexible microgrid system and its control method. Background Technology

[0002] The development of the microgrid industry is currently driven by both technological iteration and upgrading market demand. In recent years, the domestic microgrid market has experienced significant growth, and it is expected to maintain a high growth rate in the coming years, with its market potential continuing to be released. From a technological perspective, mainstream technologies include centralized AC microgrids and localized DC microgrids. Centralized AC microgrids, primarily based on AC power distribution, offer advantages such as strong grid compatibility and mature modular design, but their reliance on multiple energy conversion processes leads to higher energy losses. Localized DC microgrids, on the other hand, focus on integrated energy management within a DC architecture, featuring high DC absorption efficiency, but they have limitations in cross-regional coordinated dispatch and grid interconnection.

[0003] Analysis of similar technologies on the market reveals significant limitations in existing solutions. For example, some smart microgrids possess integrated energy conversion and dispatch functions, but rely on traditional AC architectures, resulting in limited cross-regional power exchange capabilities; some microgrid solutions exhibit strong grid compatibility and modular design, but lack end-to-end DC ecosystem construction; while certain types of energy storage systems possess large capacity and rapid response capabilities, they lack multi-energy collaborative control mechanisms; some companies have launched integrated energy solutions covering multiple aspects, but they have not achieved deep coupling between the hybrid grid framework and virtual power plants, still relying on traditional power distribution methods; although some park-level energy management platforms are being piloted in local scenarios, overall, existing solutions mostly focus on single technical fields or regional demonstrations, lacking integrated comprehensive solutions.

[0004] Traditional microgrid systems face several technical bottlenecks in practical applications. For example, in terms of energy conversion efficiency, traditional microgrids often employ centralized AC architectures, relying on multiple energy conversion processes, leading to significant conversion losses and hindering the efficient utilization of renewable energy sources. Regarding multi-energy coordination, current integrated energy systems lack a unified dispatch platform, making it difficult to achieve full-process optimization of multiple energy forms and limiting cross-temporal and spatial coordinated dispatch capabilities, thus failing to fully leverage the advantages of each resource. Furthermore, in terms of system response and prediction capabilities, existing energy management systems often rely on long-cycle dispatch strategies, with insufficient coordination between edge-side control and cloud-based prediction, resulting in low real-time performance and prediction accuracy, failing to meet the rapid adjustment needs of park-level microgrids. In terms of building heating and cooling energy management, the lack of adequate heating and cooling dispatch forces demand-side energy to rely on electricity conversion, exacerbating peak-hour electricity shortages and hindering its widespread application in park-level energy systems. Finally, in terms of market participation, there are obstacles to the large-scale participation of distributed energy resources in the electricity market; business models and operating standards are not yet mature, affecting the overall economic efficiency and reliability of the system.

[0005] In summary, the shortcomings of existing microgrid systems in terms of energy conversion efficiency, cross-regional coordination, multi-energy complementarity, and market participation are becoming increasingly prominent. There is an urgent need for a comprehensive solution that can integrate hybrid grid architecture, realize multi-energy coordinated dispatch and intelligent management, so as to improve the overall energy efficiency and power supply reliability of the park, and promote green and low-carbon transformation. Summary of the Invention

[0006] This invention addresses the problems existing in the prior art by providing a park-level flexible microgrid system and control method that takes into account the requirements of speed and stability and can achieve energy mutual assistance through cross-regional dispatch.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] On the one hand, the present invention provides a campus-level flexible microgrid system, which is based on an AC / DC hybrid power grid architecture and includes a power router, a flexible interconnection device, an all-DC building and a DC bus;

[0009] The system is connected to multiple distribution substations via the power router and the flexible interconnection device;

[0010] The DC bus is directly connected to the photovoltaic modules, energy storage units and DC loads of the distribution substation, and at least some of the DC loads are configured in the all-DC building;

[0011] The system is also equipped with an energy management system and a virtual power plant.

[0012] Optionally, the power router and the flexible interconnection device form a DC flexible interconnection network;

[0013] The flexible DC interconnection network is directly connected to the DC port of the distribution station area via the DC bus.

[0014] Optionally, the DC bus is divided into at least three specifications, including DC750V, DC375V and DC48V;

[0015] The DC 750V bus connects the photovoltaic module, the energy storage unit, and the DC load.

[0016] Optionally, the DC load is equipped with a full DC photovoltaic-storage-supercharging system;

[0017] The all-DC photovoltaic-storage supercharging system is connected to the virtual power plant based on bidirectional DC fast charging technology.

[0018] Optionally, the energy storage unit is equipped with a phase change energy storage device, and the phase change energy storage device has a modular structure.

[0019] On the other hand, based on the above system, the present invention also provides a control method for a campus-level flexible microgrid system, comprising the following steps:

[0020] Based on the system's AC / DC hybrid power grid architecture, a global scheduling strategy is generated through the energy management system; the global scheduling strategy includes at least power mutual assistance parameters and multi-energy coordinated scheduling instructions.

[0021] The power router dynamically adjusts the power balance between the AC bus and the DC bus based on power balance parameters.

[0022] Based on multi-energy collaborative scheduling instructions, energy cross-time shift scheduling is performed through energy storage units and DC loads.

[0023] Optionally, the multi-energy cooperative scheduling instructions include the following instructions:

[0024] When the local area is in a period of low electricity demand and the photovoltaic output exceeds the local load, the surplus electricity is transferred to other distribution areas and / or energy storage units through flexible interconnection devices.

[0025] When the local area is in peak electricity consumption period and the photovoltaic output is lower than the local load, the energy storage unit and the adjacent DC bus are triggered to provide reverse power to the local area.

[0026] Optionally, the energy storage unit performs energy storage operations during off-peak electricity periods;

[0027] The energy storage unit is equipped with a phase change energy storage device for converting electrical energy into cold or hot energy for storage;

[0028] The phase change energy storage device releases energy during peak power periods to supply energy to hot and cold terminals.

[0029] Optionally, the global scheduling policy generation includes the following steps:

[0030] The energy management system uses meteorological data and historical load curves to make predictions through artificial intelligence models to obtain power output and load forecast results.

[0031] Based on the output and load prediction results, a global scheduling strategy is generated in rolling increments from minutes to hours using a multi-timescale optimization algorithm.

[0032] Optionally, the energy management system embeds a carbon flow tracking algorithm to measure the carbon footprint in real time and generate emission reduction paths.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] This invention establishes a system that includes flexible DC interconnection and energy routing across distribution areas. Combined with corresponding control methods, it realizes dynamic routing and real-time sharing of power between distribution areas. During peak load periods, energy storage supports the output of the main bus, ensuring power continuity and reliability. Furthermore, the interconnection structure breaks through the islanded operation limitations of traditional microgrids, improves the system's flexibility and redundancy, and helps alleviate the overload problem of distribution transformers. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a system diagram in a specific embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram illustrating the working principle of the power router in a specific embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of the AC / DC conversion of the power router in a specific embodiment of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0041] It is worth noting that, unless otherwise specified, the methods used in this invention are all conventional methods; and the raw materials and equipment used are all conventional commercially available products, and their sources are not specifically limited.

[0042] It should also be noted that, for ease of understanding, the method steps in the specific embodiments of the present invention are described in a certain order, but those skilled in the art can change the order of the steps according to actual needs, so this should not be used as a limiting condition.

[0043] like Figure 1 As shown, taking a conventional industrial park flexible microgrid as an example, it mainly includes six application scenarios, including AC / DC hybrid flexible power system, V2G all-DC photovoltaic-storage-supercharging, phase change energy storage and thermal energy storage, energy management system, and virtual power plant.

[0044] Therefore, this embodiment provides a campus-level flexible microgrid system based on an AC / DC hybrid grid architecture, including a power router, flexible interconnection devices, all-DC buildings, and DC buses. The system is also equipped with an energy management system (EMS) and a virtual power plant (VPP).

[0045] The power router in this embodiment adopts a modular, bidirectional active bridge power unit topology, which has multi-port collaborative operation capability and supports multiple voltage level interfaces such as 10kVAC / 10kVDC / 750VDC / 380VAC. The internal structure of the power router includes: a high-frequency bidirectional conversion module, a high-frequency transparent isolation transformer, silicon carbide power device units, integrated drive protection circuitry, and a composite fault current control module. For example... Figure 2 and Figure 3 As shown, in the bilateral structure of the power router, the system on one side adopts a three-stage "AC-DC-AC" architecture stacked, with a high-frequency isolation transformer in the middle for regulation, supplemented by a common high-frequency bus interconnection bilateral structure, to achieve flexible flow and isolation control of energy between ports of different voltage levels, significantly improving the system's power density and scalability. Compared with the traditional common DC bus structure, it reduces the number of power conversion stages by more than 30%, simplifies circuit complexity, improves system efficiency, and reduces equipment failure rate by more than 90%.

[0046] The high-frequency transformer employs nanocrystalline iron core material and a layered winding structure, matched with a 1700V / 200A silicon carbide MOSFET module. Optimized parasitic parameters effectively suppress high-frequency oscillations, improving the device's safety and reliability. This structure supports independent port control and decoupling, enabling automatic reconfiguration and fault isolation in the event of load fluctuations or single-port anomalies, ensuring safe operation under complex conditions in various scenarios. The entire unit supports parallel operation of multiple units via a cluster controller, allowing for arbitrary combinations of port capacities and providing module-level bypass and switching capabilities, further enhancing the system's flexibility and fault tolerance.

[0047] The system comprises a DC flexible interconnection network constructed using power routers and flexible interconnection devices. This network directly connects to the DC ports of distribution substations via DC buses. Multiple distribution substations are connected through the power routers and flexible interconnection devices. The AC / DC hybrid power grid architecture, centered on multi-node power routers and relying on rooftop photovoltaic arrays, combines real-time power consumption and generation data to construct a DC flexible interconnection network between substations via low-voltage flexible interconnection devices. This enables dynamic power transfer and resource sharing between distribution rooms and adjacent substations on the DC bus. When photovoltaic (PV) power generation exceeds the local load, surplus power can be transmitted to other distribution areas or integrated into the energy storage system via flexible interconnection devices. During peak electricity demand, energy storage and adjacent DC buses can provide reverse power supplementation, significantly increasing the PV self-consumption rate to over 90%, effectively alleviating transformer overload and reducing line losses. Specifically, low-voltage flexible interconnection devices are deployed on the DC bus to achieve dynamic routing and real-time sharing of power between distribution areas. When one area's power generation is saturated, energy can be transmitted to other areas or the energy storage system. During peak load periods, energy storage supports the main bus output, ensuring power continuity and reliability. This dynamic routing capability overcomes the islanding limitations of traditional microgrids, improving system flexibility and redundancy, and helping to alleviate distribution transformer overload problems. This not only increases the PV self-consumption rate but also significantly enhances the overall power supply reliability of the park.

[0048] The DC bus directly connects to the photovoltaic modules, energy storage units, and DC loads of the distribution substation, with at least some of the DC loads configured in all-DC buildings. The DC bus is available in three specifications: DC750V, DC375V, and DC48V. The DC750V bus connects the photovoltaic modules, energy storage units, and DC loads. Furthermore, the DC ports of the aforementioned interconnection devices extend to a 750V DC bus, uniformly supporting rooftop photovoltaic systems, ground-based energy storage units, and DC fast-charging piles for basketball courts and electric vehicle sheds, enabling centralized dispatch and emergency power supply. It can also flexibly connect to buildings, industrial equipment, 5G base stations, and phase-change thermal energy storage modules according to actual load requirements, constructing long-distance DC transmission scenarios and completing the integrated "generation-storage-use" DC ecosystem deployment. This solution features a modular and interface-based design, meeting the needs of multi-source mutual support and high-reliability power supply at the park level, and also possessing the ability to be quickly replicated and promoted in other similar scenarios.

[0049] The all-DC building encompasses an employee activity center and a zero-carbon office building, employing an all-DC power supply architecture connected to the DC bus and its system. Photovoltaic modules are distributed across the building's roof and exterior walls, with the power generation side connected to the 750V main bus via an all-DC aggregation method. Energy storage units are configured for energy regulation and peak shaving, ensuring stable system operation. Through voltage divider outputs from DC / DC and AC / DC modules, direct power is provided to DC air conditioning, LED lighting, computers, televisions, and communication equipment, shortening the energy conversion path and avoiding energy consumption and power quality issues caused by multiple inverters. This all-DC building solution supports segmented photovoltaic access and flexible configuration, effectively improving building energy utilization and achieving efficient synergy between photovoltaics, energy storage, and various DC loads. The overall energy efficiency is more than 15% higher than traditional AC systems. In actual operation, it reduces dependence on the public power grid, lowers electricity costs, possesses excellent peak-valley response and load adaptability, and enables seamless switching between on-grid and off-grid operation, providing reliable power supply. Meanwhile, by constructing a standardized and modular DC power supply and distribution network, it provides technical support for the integration of high-penetration renewable energy sources, multi-source power exchange, and diversified load power supply. This solution improves system energy efficiency and controllability while effectively reducing carbon emissions, and is suitable for replication and promotion in various parks, office buildings, and public buildings.

[0050] This embodiment employs a DC 750 V main bus and two levels of DC 375 V and DC 48 V sub-buses to directly connect photovoltaic modules, energy storage units, and various DC terminals. This eliminates the multiple energy conversion paths of "DC → inverter → AC → rectification," significantly reducing system conversion losses and improving energy utilization efficiency. As a result, system energy conversion losses are reduced by 15%–20%, and overall energy efficiency is improved by 10%. This modular multi-voltage-level DC power distribution method provides underlying support for the efficient integrated access of distributed energy resources and diverse loads, making the power distribution structure simpler and the access more flexible.

[0051] The DC load is equipped with a full DC photovoltaic-storage-supercharging system, which is a V2G full DC photovoltaic-storage-supercharging system. This system directly connects the photovoltaic modules and energy storage units via a 750V DC main bus. Specifically, the V2G full DC photovoltaic-storage-supercharging system is a fully DC coupled system between photovoltaic, energy storage, and charging facilities. The 750V DC main bus directly connects the photovoltaic modules and energy storage units, avoiding the losses caused by multiple energy conversions (DC → inverter → AC → rectification), minimizing the energy transmission path, improving power quality, and enhancing system controllability. Field measurements show that compared to traditional AC microgrids, DC microgrids can improve photovoltaic utilization efficiency by 6%–8% and increase overall energy efficiency by approximately 10%. Furthermore, taking a certain type of DC fast charging pile as an example, its cable loss is only 37.7% of that of an equivalent AC cable, further verifying the significant advantage of the full DC architecture in reducing line losses. This solution reduces system losses by 15%–20% compared to traditional AC solutions without increasing overall hardware investment, significantly reducing the total electricity consumption of the park and improving energy utilization efficiency. In terms of bidirectional interaction, the system integrates V2G bidirectional DC fast charging technology, enabling electric vehicles to become dispatchable mobile energy storage units. During off-peak hours, they prioritize energy storage and discharge in reverse during peak hours, providing auxiliary support for the park's load. Edge computing enables millisecond-level switching and orderly charging, thereby mitigating charging load fluctuations and grid impacts. The all-DC photovoltaic-storage supercharging system connects to a virtual power plant based on bidirectional DC fast charging technology, deeply integrating V2G functionality with virtual power plant technology. By dynamically aggregating multiple vehicles and energy storage resources, it participates in the electricity spot market, ancillary services, and demand response market, obtaining peak-shaving subsidies and capacity trading revenue. Verification shows that the VPP participation model not only improves grid stability but also brings significant market benefits to participating entities, including ancillary service fees and demand response incentives.

[0052] Existing technologies typically employ AC unidirectional charging designs, which cannot participate in grid peak shaving and often impact the power distribution system due to the need for multiple energy conversions. This embodiment, however, directly introduces DC supercharging piles and integrates V2G bidirectional functionality under a full DC aggregation architecture, achieving integrated DC coupling of photovoltaic / energy storage / charging facilities. Through EMS intelligent scheduling, electric vehicles are charged in an orderly manner during off-peak hours and discharged in reverse during peak hours. The achieved effect is the elimination of multi-stage conversion losses on the AC side, improving overall energy efficiency, and effectively shaving off peaks and filling valleys through orderly charging and discharging, significantly mitigating the impact of high-power charging on the power distribution system.

[0053] The energy storage unit is equipped with a phase change energy storage device, which adopts a modular design for easy expansion and deployment on demand, making it particularly suitable for areas with limited space resources such as building rooftops and underground equipment rooms. Specifically, the phase change energy storage device is constructed using a thermal storage cube as its basic unit structure, employing inorganic salt phase change materials internally. Its energy storage density is several times that of traditional water storage, achieving both cold and heat storage through solid-liquid-gas phase change at -5℃ or 80℃. Addressing the cooling and heating needs of buildings in the park, the phase change energy storage device leverages the efficient thermophysical properties of phase change materials to achieve cross-time utilization of clean energy through "electricity-to-heat" and "electricity-to-cooling" conversions, effectively solving energy waste and supply-demand mismatch issues in thermal energy use. The system is driven by renewable energy sources such as off-peak electricity, photovoltaics, and wind power, charging the phase change materials through cooling or heating devices. After energy storage is completed, the heat can be released during peak hours for use in air conditioning terminal loads, improving overall system efficiency by more than 10% and reducing comprehensive energy costs by more than 55%.

[0054] The energy storage unit operates without combustion, producing no waste gas or wastewater, achieving zero-carbon operation and aligning with the development trends of green and near-zero energy buildings. Its all-weather operation, combined with the EMS intelligent dispatch platform, effectively absorbs surplus electricity from distributed energy sources such as photovoltaics, reducing curtailment rates and enhancing the local consumption capacity of new energy. Simultaneously, it alleviates peak load pressure through off-peak electricity storage for cooling and heating.

[0055] The system adopts a full DC power supply scheme and is seamlessly integrated with the existing DC bus in the park to build a multi-energy complementary energy system integrating "photovoltaics, electricity, cooling and heating". It is a key supporting technology for promoting the electrification transformation of thermal energy and low-carbon operation in the park.

[0056] Optionally, the EMS in this embodiment constructs a three-level collaborative architecture of cloud-edge-device: the cloud is responsible for short-term and ultra-short-term forecasts based on meteorological data and historical load curves, and generates global scheduling strategies on a rolling basis from minute to hour through a multi-timescale optimization module; the edge side achieves local rapid matching, monitors the status of photovoltaic power generation, diversified energy storage, and flexible loads in real time, and executes optimization instructions issued by the cloud; the terminal layer collects various equipment operation data to form closed-loop control, ensuring that the system operates according to the optimal plan. The dynamic aggregation module integrates dispersed photovoltaic, energy storage, phase change thermal and cold storage units, and charging piles into a dispatchable flexible capacity pool through virtual power plant technology, improving the overall energy efficiency by more than 15% through multi-energy complementarity, and enhancing the microgrid's ability to cope with load fluctuations and grid accidents. Under the strategy of "horizontal multi-energy complementarity and vertical spatiotemporal coordination," the EMS achieves deep integration of the electricity-heat-cooling multi-energy coupled network: horizontally, it connects phase change thermal and cold storage, flexible interconnection of transformer substations, and energy storage peak shaving and orderly charging and discharging channels for charging piles, improving the overall energy utilization efficiency of the park through combined scheduling; vertically, it constructs a three-level optimization mechanism of day-ahead, intraday, and real-time, combining minute-level forecasting and demand response to guide high-energy-consuming equipment to operate off-peak, smoothing the system's peak-valley difference and reducing operation and maintenance costs. Simultaneously, the system innovatively embeds a carbon flow tracking algorithm to measure the carbon footprint of each link in real time and generate emission reduction paths, helping the park achieve the goal of becoming a national-level green industrial park and providing key technical support for the construction of distributed energy and smart parks.

[0057] Optionally, in this embodiment, the VPP constructs a virtual power plant platform within the park. This platform unifies and aggregates diverse resources within the park, including distributed photovoltaic systems, energy storage systems, DC fast charging piles, phase change energy storage, and microgrids. Through a highly integrated spot trading and control decision-making model, it achieves integrated automation of trading and scheduling, replacing traditional manual processes and significantly improving the system's automation and intelligence levels. Based on this, the platform employs a multi-scale source-load price prediction algorithm, integrating artificial intelligence technology, to increase load prediction accuracy to 90% and new energy output prediction accuracy to over 95%, providing precise data support for trading strategies and scheduling schemes, thereby deeply exploring operational economics and optimizing energy utilization efficiency. Furthermore, under the control of the virtual power plant, the park can integrate dispersed distributed energy resources and adjustable loads, such as air conditioners, charging piles, and industrial production lines, into dispatchable "virtual machine groups," breaking down resource silos and achieving dynamic responses at the minute or even second level. The system monitors user-side electricity consumption data in real time and combines it with market electricity prices or grid dispatch signals to automatically adjust energy storage charging and discharging and load start-up and shutdown strategies. It can participate in the electricity spot market and ancillary services, such as peak shaving and frequency regulation, to obtain multiple benefits, and can also quickly smooth out power fluctuations when photovoltaic output fluctuates or grid frequency is abnormal.

[0058] On the other hand, this embodiment also provides a control method for a campus-level flexible microgrid system, used to control the aforementioned system, including the following steps:

[0059] Based on the AC / DC hybrid power grid architecture, a global scheduling strategy is generated through the energy management system.

[0060] The steps for generating a global scheduling policy are as follows:

[0061] The energy management system, based on meteorological data and historical load curves, uses artificial intelligence models to predict load with an accuracy rate of over 90% and renewable energy output with an accuracy rate of over 95%. Specifically, by introducing artificial intelligence models (such as LSTM load forecasting and photovoltaic power forecasting models), the system establishes high-precision load and output forecasting models based on multi-source heterogeneous data including meteorological information, historical load curves, and electricity price curves, achieving a prediction accuracy rate of over 90%. The forecast results drive multi-timescale optimization algorithms, continuously generating optimal energy dispatching schemes at multiple time scales, including minute-level, hourly-level, and day-ahead-level, thus realizing a prediction-driven dispatching closed loop.

[0062] Therefore, based on the prediction results, a global scheduling strategy is generated in rolling increments from minutes to hours using a multi-timescale optimization algorithm. The strategy includes power mutual assistance parameters and multi-energy collaborative scheduling instructions.

[0063] The multi-energy coordinated dispatch instructions include: when local electricity consumption is low and photovoltaic output exceeds the load, the surplus electricity is transferred to other distribution areas or energy storage units through flexible interconnection devices; when local electricity consumption is high and photovoltaic output is insufficient, the energy storage units and adjacent DC bus are triggered to provide reverse power.

[0064] The power router dynamically adjusts the power balance between the AC and DC buses based on power balance parameters; based on scheduling instructions, the energy storage unit and DC load perform energy cross-time shift scheduling: during off-peak hours, the energy storage unit converts electrical energy into -5℃ cold energy or 80℃ hot energy through phase change energy storage devices for storage, and releases energy to supply cold and hot terminals during peak hours.

[0065] It should be noted that power mutual assistance means that the power router dynamically allocates energy between multiple transformer stations based on the system's internal load, energy storage status, and power supply boundary conditions, realizing a real-time power mutual assistance strategy of transferring surplus energy to supply and compensating for energy shortages. Specifically, the system supports cross-transformer flow of electricity, cross-equipment coordination, and cross-voltage level mutual assistance, effectively alleviating local supply and demand imbalances. At the same time, the energy cross-time-shift scheduling mechanism combines energy storage units and phase change thermal and cold storage devices to store electricity during periods of low electricity demand or surplus renewable energy output, and release it during peak load periods or peak electricity prices to support flexible loads such as heating and cooling terminals and electric vehicles, achieving complementarity of electricity, cooling, and heating energy and time mismatch between "source and load".

[0066] Optionally, in this embodiment, to enhance the collaborative operation capability of multiple power routers in the system, the control system further introduces a cluster control strategy. The cluster control is based on a unified control platform, using a cluster controller to centrally monitor, coordinate, schedule, and monitor the status of each power router, thereby achieving information sharing and coordinated energy allocation among the devices.

[0067] Specifically as follows:

[0068] The cluster control adopts a control architecture that combines master-slave collaboration with distributed autonomy. The master node is responsible for generating global scheduling policies and issuing control commands, while slave nodes have independent operating capabilities and can dynamically adjust their operating status according to the master's commands. When the master node malfunctions, other nodes can automatically reconstruct the control link to ensure continuous and stable system operation. This control strategy supports multi-machine parallel operation and features automatic load distribution, optimal power optimization, module-level hot standby switching, and fault complementarity capabilities.

[0069] Specifically, the cluster controller collects real-time operating data from each router, including voltage, current, load status, and device temperature. It uses edge computing to determine the operating status of each module and dynamically configures energy flow paths based on power balance requirements. The system can complete load reconfiguration and power switching in milliseconds, improving the overall system's response speed and power supply reliability under complex operating conditions.

[0070] Through the above-mentioned cluster control strategy, the microgrid system in this embodiment breaks through the limitation of isolated operation of multiple devices in traditional microgrid systems, realizes the clustered operation and intelligent scheduling of multiple power routers in a park-level scenario, and provides key technical support for realizing integrated collaborative control of power generation, grid, load and storage.

[0071] Based on the above control strategies, the energy management system also embeds a carbon flow tracking algorithm, and while controlling the system, it measures the carbon footprint in real time and generates emission reduction paths. Combined with the cloud-edge-device three-level collaborative architecture, it achieves deep integration of the electricity-heat-cooling multi-energy coupled network, which can improve the overall energy efficiency by more than 15%.

[0072] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, and is not intended to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions made by those skilled in the art to the technical solution of the present invention do not depart from the essence and scope of the technical solution of the present invention.

Claims

1. A campus-level flexible microgrid system, characterized in that: The system is based on an AC / DC hybrid power grid architecture, including power routers, flexible interconnection devices, all-DC buildings, and DC buses; The system is connected to multiple distribution substations via the power router and the flexible interconnection device; The DC bus is directly connected to the photovoltaic modules, energy storage units and DC loads of the distribution substation, and at least some of the DC loads are configured in the all-DC building; The system is also equipped with an energy management system and a virtual power plant.

2. The campus-level flexible microgrid system according to claim 1, characterized in that: The power router and the flexible interconnection device form a DC flexible interconnection network; The flexible DC interconnection network is directly connected to the DC port of the distribution station area via the DC bus.

3. The campus-level flexible microgrid system according to claim 1, characterized in that: The DC bus is divided into at least three specifications, including DC750V, DC375V and DC48V; The DC 750V bus connects the photovoltaic module, the energy storage unit, and the DC load.

4. The campus-level flexible microgrid system according to claim 1, characterized in that: The DC load is equipped with a full DC photovoltaic-storage-supercharging system; The all-DC photovoltaic-storage supercharging system is connected to the virtual power plant based on bidirectional DC fast charging technology.

5. The campus-level flexible microgrid system according to claim 1, characterized in that: The energy storage unit is equipped with a phase change energy storage device, and the phase change energy storage device has a modular structure.

6. A control method for a campus-level flexible microgrid system, characterized in that: For controlling the campus-level flexible microgrid system according to any one of claims 1-5; comprising the following steps: Based on the system's AC / DC hybrid power grid architecture, a global scheduling strategy is generated through the energy management system; the global scheduling strategy includes at least power mutual assistance parameters and multi-energy coordinated scheduling instructions. The power router dynamically adjusts the power balance between the AC bus and the DC bus based on power balance parameters. Based on multi-energy collaborative scheduling instructions, energy cross-time shift scheduling is performed through energy storage units and DC loads.

7. The control method for a campus-level flexible microgrid system according to claim 6, characterized in that: The multi-energy coordinated scheduling instructions include the following instructions: When the local area is in a period of low electricity demand and the photovoltaic output exceeds the local load, the surplus electricity is transferred to other distribution areas and / or energy storage units through flexible interconnection devices. When the local area is in peak electricity consumption period and the photovoltaic output is lower than the local load, the energy storage unit and the adjacent DC bus are triggered to provide reverse power to the local area.

8. The control method for a campus-level flexible microgrid system according to claim 6, characterized in that: The energy storage unit performs energy storage during off-peak electricity periods; The energy storage unit is equipped with a phase change energy storage device for converting electrical energy into cold or hot energy for storage; The phase change energy storage device releases energy during peak power periods to supply energy to hot and cold terminals.

9. The control method for a campus-level flexible microgrid system according to claim 6, characterized in that: The generation of the global scheduling policy includes the following steps: The energy management system uses meteorological data and historical load curves to make predictions through artificial intelligence models to obtain power output and load forecast results. Based on the output and load prediction results, a global scheduling strategy is generated in rolling increments from minutes to hours using a multi-timescale optimization algorithm.

10. The control method for a campus-level flexible microgrid system according to claim 6, characterized in that: The energy management system is embedded with a carbon flow tracking algorithm to measure the carbon footprint in real time and generate emission reduction paths.

Images