Balancing unit-based medium-voltage power system balancing system and optimization method

Abstract

The invention provides a medium-voltage power system balancing system based on a balancing unit and an optimization method, and relates to the technical field of power regulation. The medium-voltage power system balance system based on the balance unit is a technical system for realizing dynamic balance of power, voltage and frequency of a medium-voltage power distribution network through an intelligent control strategy, and comprises a power equipment module, a strategy control module, an energy management module, a service application module and a network communication module, the power equipment module comprises a balance adjusting unit, a modular multilevel converter and an energy router. The balance adjusting unit comprises a static var compensator, a static synchronous compensator and a dynamic voltage restorer. Dynamic balance adjustment of the medium-voltage power system is realized by adopting MPC model prediction and an edge AI algorithm, technologies of edge calculation, intelligent optimization, power electronics and the like are fused, limitation of traditional scheduling is broken through, and flexibility and expandability of the system are improved through modular design and a marketization mechanism.

Description

Technical Field

[0001] This invention relates to the field of power regulation technology, specifically to a balancing system and optimization method for a medium-voltage power system based on a balancing unit. Background Technology

[0002] With the rapid development of renewable energy sources such as photovoltaics and wind power, medium-voltage distribution networks are gradually transforming from unidirectional radial networks to multi-source collaborative and interactive networks. However, traditional medium-voltage power systems have structural defects. Distributed power output fluctuations may cause medium-voltage line power deviations to exceed ±5% of the rated capacity, leading to voltage over-limits. When distributed power output is excessive, it may cause power flow reversal, increase line losses, and threaten equipment safety. Furthermore, with the surge in the number of distributed resources, the communication burden and computational complexity of centralized architectures increase exponentially.

[0003] Therefore, this invention proposes a medium-voltage power system balancing system and optimization method based on balancing units, thereby effectively solving the above-mentioned problems and difficulties. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a medium-voltage power system balancing system and optimization method based on balancing units. It uses MPC model prediction and edge AI algorithms to achieve dynamic balance adjustment of the medium-voltage power system, integrates edge computing, intelligent optimization, power electronics and other technologies, breaks through the limitations of traditional dispatching, and improves the system's flexibility and scalability through modular design and market mechanisms.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A medium-voltage power system balancing system and optimization method based on a balancing unit. The balancing system is a technical system that achieves dynamic balance of power, voltage and frequency in a medium-voltage distribution network through intelligent control strategies. The balancing system includes a power equipment module, a strategy control module, an energy management module, a service application module and a network communication module. The power equipment module includes a balance regulation unit, a modular multilevel converter, and an energy router. The balance regulation unit includes a static var compensator, a static synchronous compensator, and a dynamic voltage restorer. The modular multilevel converter uses a sub-module cascade structure to increase redundancy and achieve stable operation in high-voltage and high-capacity scenarios. The energy router adopts a modular intelligent switch and uses sensors to collect data to achieve temperature monitoring and real-time monitoring of partial discharge. The strategy control module includes a hierarchical collaborative control module, an intelligent algorithm module, and an energy management model. The hierarchical system control module monitors the bus voltage, frequency, and tidal distribution in real time and generates global scheduling instructions through MPC model predictive control. The intelligent algorithm module realizes autonomous coordination between distributed energy and load on the transformer area or user side through edge AI algorithm. The energy management model achieves production and consumption balance through multi-media model, combined with probabilistic power flow calculation and robust optimization algorithm and mixed integer nonlinear programming. The energy management module includes an energy storage system, a distributed energy management module, and a collaborative operation guarantee module. The energy storage system adopts medium-voltage direct-connected energy storage with three-level energy balancing at the cluster, module, and cell levels, and is designed with 1:1 redundancy. It utilizes the rapid charging and discharging capability of medium-voltage direct-connected energy storage to compensate for active power deficit in real time. The distributed energy management module achieves production and consumption balance through mixed integer nonlinear programming. The collaborative operation guarantee module adds hardware redundancy and software fault tolerance functions, adopts dual-machine hot standby in the medium-voltage power system, and adds data backup and fault self-diagnosis technology. The service application module includes grid connection management, grid interaction, and market platform integration functions; The network communication module includes a communication architecture protocol module and a data security protection module. The communication architecture protocol module adopts medium-voltage carrier communication, uses power distribution lines as the transmission medium, and supports GPRS and fiber optic uplink communication. The data security protection module adopts a dedicated power firewall and a one-way physical isolation device to ensure data security between the production control area and the management information area.

[0006] Furthermore, the static var compensator of the power balancing equipment module controls the reactor and capacitor bank through thyristors to quickly adjust the reactive power and stabilize the voltage. The static synchronous compensator continuously adjusts the reactive power within the inductive to capacitive range based on the voltage source inverter. The dynamic voltage restorer is connected in series in the line to compensate for voltage sags and fluctuations in real time, ensuring the power supply quality of sensitive loads.

[0007] Furthermore, the hierarchical system control module of the strategy control module generates global scheduling instructions through MPC model predictive control, establishes an MPC predictive model, describes the system power deviation and energy storage state, and defines it as: ; in, Let k be the system active power deviation. Let the energy storage state of charge at time k be , then the energy storage charging and discharging power is . The uncontrollable load fluctuation prediction error is: : Discrete time, sampling period The state equation is:

[0008] Matrix parameters: ： in The self-discharge coefficient of energy storage; ; The energy storage charge and discharge efficiency is (0~1). This refers to the rated capacity of the energy storage. ; C is the coefficient representing the impact of disturbance on power deviation.

[0009] Furthermore, in the hierarchical system control module of the strategy control module, the MPC model prediction, by clearly defining the energy storage module and power deviation, minimizes the tracking error and control cost in the prediction time domain, thereby achieving power balance control and voltage control. The objective function of the balance control is: ; in To predict the time domain, To control the time domain, The power deviation at time t (with the target being tracked being 0) is the power deviation at time t. This is a reference value for the State of Charge (SOC) of energy storage. Let be the energy storage control quantity at time t. This represents the error weight; a larger weight indicates that the variable is controlled with higher priority. Weights for changes in control quantities; The voltage control objective function is: ; in Let be the voltage deviation at node i at time t. Let j be the control variable for the j-th reactive power device. For node voltage weights, The cost of adjusting reactive power devices.

[0010] Furthermore, the edge AI algorithm of the policy control module achieves fast classification and dynamic balance decision-making at the edge through random forest, and reinforcement learning learns the optimal control strategy for energy storage charging and discharging through interaction with the environment. ; in, For state Next action value, For learning rate, As a discount factor, For instant rewards, For the next state The optimal action value is achieved by using edge AI algorithms to enable the balancing system to make real-time decisions on energy storage charging and discharging and distributed power output adjustment, thereby minimizing system power deviation.

[0011] Furthermore, in the energy management model of the strategy control module, probabilistic power flow calculation is used to quantify the probabilistic impact of input variables on system state variables. Uncertainty is described through a probabilistic model, and the probability distribution of system state variables is derived. In medium-voltage systems, random variables mainly include node loads. Distributed power generation photovoltaic and wind power By establishing its probability density function PDF: ; in, The average load ， If the standard deviation is given, then the work is done. The node power balance equation is: ; No effect The node power balance equations are as follows: ; in, Let i be the voltage magnitude at nodes i and j. Let be the voltage phase angle difference between nodes i and j. Nodal admittance matrix Element.

[0012] Furthermore, the distributed energy management module of the energy management module optimizes decision variables through mixed-integer nonlinear programming to ensure a dynamic balance between energy output and consumption. Its core constraint on capacity is that at any time t, total energy output must equal total consumption. ; in For energy storage charging / discharging efficiency; The energy storage state of charge constraints are: ; in To schedule the time step, This refers to the rated capacity of the energy storage. By planning and optimizing decisions based on capacity and storage constraints, we can ensure the best distributed energy dispatch strategy, achieving dynamic balance between production and consumption while meeting all constraints and minimizing operating costs.

[0013] Furthermore, the service application module achieves real-time monitoring of voltage and power factor through grid connection management, and realizes data sharing with the upper-level power grid dispatch center. It also supports one-click load transfer and aggregator collaborative control, and connects with the market platform to support the business of power ancillary services market and spot trading.

[0014] Furthermore, the optimization method for the balanced system is implemented using the following steps: Step 1: Real-time data acquisition of the balancing unit to obtain the real-time operating status of each component within the balancing unit. The acquisition targets include distributed power sources, energy storage systems, and controllable loads. Data is refreshed at the second or sub-second level through synchronous phasor measurement units, smart meters, and edge terminals to ensure data timeliness. Step 2: Power forecasting, predicting power fluctuations in the near future to provide a forward-looking basis for optimization decisions; Step 3: Balance unit status assessment and optimization target determination. Based on real-time data acquisition and short-term power forecast, determine the priority of the medium-voltage power system and sort the targets for optimization. Step 4: Based on the current state and predicted data, use the MPC model to predict the optimal control commands for each controllable device in the balance unit, and use edge AI algorithms to determine the dynamic balance decision. Step 5: Control command generation and execution, converting the optimization results into executable commands for specific devices to achieve real-time adjustment; Step 6: Real-time anomaly handling: When the power deviation > 10% of the rated capacity, the voltage exceeds the limit by > ±10%, or there is a equipment failure, the anomaly handling mechanism is triggered. The redundancy mechanism is invoked to make up for the power shortfall or interruptible loads are cut off first. An anomaly signal is sent to the dispatch center to request coordination and support from the superior.

[0015] This invention provides a balancing system and optimization method for medium-voltage power systems based on balancing units. It offers the following advantages: 1. This invention provides a medium-voltage power system balancing system and optimization method based on balancing units. By adopting MPC model prediction and edge AI algorithm, the dynamic balance adjustment of the medium-voltage power system is realized. It integrates edge computing, intelligent optimization, power electronics and other technologies to break through the limitations of traditional dispatching. Furthermore, through modular design and market mechanism, the system's flexibility and scalability are improved.

[0016] 2. This invention provides a medium-voltage power system balancing system and optimization method based on a balancing unit. Through hardware redundancy and software fault tolerance, it has the ability to respond quickly to faults and operate in an islanded manner. Through distributed energy storage management and collaborative operation, combined with dynamic reactive power compensation devices and reactive power regulation of energy storage, the power factor of medium-voltage lines is increased to above 0.95, the harmonic distortion rate is reduced to below 5%, and equipment losses and fault risks are reduced. Attached Figure Description

[0017] Figure 1 This invention relates to the system architecture of a medium-voltage power system balancing system based on a balancing unit. Detailed Implementation

[0018] 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, and 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. Example 1:

[0019] like Figure 1 As shown, this embodiment of the invention provides a medium-voltage power system balancing system and optimization method based on a balancing unit. The balancing system is a technical system that achieves dynamic balance of power, voltage and frequency in a medium-voltage power distribution network through intelligent control strategies. The balancing system includes a power equipment module, a strategy control module, an energy management module, a service application module and a network communication module. The power equipment module includes a balance regulation unit, a modular multilevel converter, and an energy router. The balance regulation unit includes a static var compensator, a static synchronous compensator, and a dynamic voltage restorer. The modular multilevel converter uses a sub-module cascade structure to increase redundancy and achieve stable operation in high-voltage and high-capacity scenarios. The energy router adopts a modular intelligent switch and uses sensors to collect data to achieve temperature monitoring and real-time monitoring of partial discharge. The strategy control module includes a hierarchical collaborative control module, an intelligent algorithm module, and an energy management model. The hierarchical system control module monitors the bus voltage, frequency, and tidal distribution in real time and generates global scheduling instructions through MPC model predictive control. The intelligent algorithm module achieves autonomous coordination between distributed energy and load on the transformer area or user side through edge AI algorithm. The energy management model achieves production and consumption balance through multi-media model, combined with probabilistic power flow calculation and robust optimization algorithm and mixed integer nonlinear programming. The energy management module includes an energy storage system, a distributed energy management module, and a collaborative operation guarantee module. The energy storage system adopts medium-voltage direct-connected energy storage with three-level energy balancing at the cluster, module, and cell levels, and is designed with 1:1 redundancy. It utilizes the rapid charging and discharging capability of medium-voltage direct-connected energy storage to compensate for active power deficit in real time. The distributed energy management module achieves production and consumption balance through mixed integer nonlinear programming. The collaborative operation guarantee module adds hardware redundancy and software fault tolerance functions, adopts dual-machine hot standby in the medium-voltage power system, and adds data backup and fault self-diagnosis technology. The service application module includes grid connection management, grid interaction, and market platform integration functions; The network communication module includes a communication architecture protocol module and a data security protection module. The communication architecture protocol module adopts medium-voltage carrier communication, uses power distribution lines as the transmission medium, and supports GPRS and fiber optic uplink communication. The data security protection module adopts a dedicated power firewall and a one-way physical isolation device to ensure data security between the production control area and the management information area. The static var compensator of the power balancing equipment module controls the reactor and capacitor bank through thyristors to quickly adjust reactive power and stabilize voltage. The static synchronous compensator continuously adjusts reactive power within the inductive to capacitive range based on the voltage source inverter. The dynamic voltage restorer is connected in series in the line to compensate for voltage sags and fluctuations in real time and ensure the power supply quality of sensitive loads. The hierarchical system control module of the strategy control module generates global scheduling instructions through MPC model predictive control, establishes an MPC predictive model, describes the system power deviation and energy storage state, and defines it as: ; in, Let k be the system active power deviation. Let the energy storage state of charge at time k be , then the energy storage charging and discharging power is . The uncontrollable load fluctuation prediction error is: : Discrete time, sampling period The state equation is:

[0020] Matrix parameters: ： in The self-discharge coefficient of energy storage; ; The energy storage charge and discharge efficiency is (0~1). This refers to the rated capacity of the energy storage. ; C is the coefficient representing the impact of disturbance on power deviation; In the hierarchical system control module of the strategy control module, the MPC model predicts that by clearly defining the energy storage module and power deviation, the tracking error and control cost are minimized in the prediction time domain, thereby achieving power balance control and voltage control. The objective function of the balance control is: ; in To predict the time domain, To control the time domain, The power deviation at time t (with the target being tracked being 0) is the power deviation at time t. This is a reference value for the State of Charge (SOC) of energy storage. Let be the energy storage control quantity at time t. This represents the error weight; a larger weight indicates that the variable is controlled with higher priority. Weights for changes in control quantities; The voltage control objective function is: ; in Let be the voltage deviation at node i at time t. Let j be the control variable for the j-th reactive power device. For node voltage weights, The cost of adjusting reactive power devices; The edge AI algorithm of the policy control module achieves fast classification and dynamic equilibrium decision-making at the edge through random forest, and reinforcement learning learns the optimal control strategy for energy storage charging and discharging through interaction with the environment. ; in, For state Next action value, For learning rate, As a discount factor, For instant rewards, For the next state The optimal action value is achieved by using edge AI algorithms to enable the balancing system to make real-time decisions on energy storage charging and discharging and distributed power output adjustment, thereby minimizing system power deviation. In the energy management model of the strategy control module, probabilistic power flow calculations are used to quantify the probabilistic impact of input variables on system state variables. The uncertainty is described through a probabilistic model, and the probability distribution of system state variables is derived. In medium-voltage systems, random variables mainly include node loads. Distributed power generation photovoltaic and wind power By establishing its probability density function PDF: ; in, The average load ， If the standard deviation is given, then the work is done. The node power balance equation is: ; No effect The node power balance equations are as follows: ; in, Let i be the voltage magnitude at nodes i and j. Let be the voltage phase angle difference between nodes i and j. Nodal admittance matrix Element; The distributed energy management module of the energy management module optimizes decision variables through mixed-integer nonlinear programming to ensure a dynamic balance between energy production and consumption. Its core constraint on capacity is that at any time t, total energy output must equal total consumption. ; in For energy storage charging / discharging efficiency; The energy storage state of charge constraints are: ; in To schedule the time step, This refers to the rated capacity of the energy storage. By planning and optimizing decisions based on capacity and storage constraints, we can ensure the best distributed energy dispatch strategy, achieving dynamic balance between production and consumption while meeting all constraints and minimizing operating costs. The service application module enables real-time monitoring of voltage and power factor through grid connection management, achieves data sharing with the upper-level power grid dispatch center, supports one-click load transfer and aggregator collaborative control, and connects with the market platform to support the business of power ancillary services market and spot trading. Example 2:

[0021] This invention provides a medium-voltage power system balancing system and optimization method based on a balancing unit. The optimization method includes the following steps: Step 1: Real-time data acquisition of the balancing unit to obtain the real-time operating status of each component within the balancing unit. The acquisition targets include distributed power sources, energy storage systems, and controllable loads. Data is refreshed at the second or sub-second level through synchronous phasor measurement units, smart meters, and edge terminals to ensure data timeliness. Step 2: Power forecasting, predicting power fluctuations in the near future to provide a forward-looking basis for optimization decisions; Step 3: Balance unit status assessment and optimization target determination. Based on real-time data acquisition and short-term power forecast, determine the priority of the medium-voltage power system and sort the targets for optimization. Step 4: Based on the current state and predicted data, use the MPC model to predict the optimal control commands for each controllable device in the balance unit, and use edge AI algorithms to determine the dynamic balance decision. Step 5: Control command generation and execution, converting the optimization results into executable commands for specific devices to achieve real-time adjustment; Step 6: Real-time anomaly handling: When the power deviation > 10% of the rated capacity, the voltage exceeds the limit by > ±10%, or there is a equipment failure, the anomaly handling mechanism is triggered. The redundancy mechanism is invoked to make up for the power shortfall or interruptible loads are cut off first. An anomaly signal is sent to the dispatch center to request coordination and support from the superior.

[0022] The following points should be noted in this article: 1. The accompanying drawings of the embodiments disclosed herein only relate to the structures involved in the embodiments disclosed herein; other structures can be referred to in a general design.

[0023] 2. Where there is no conflict, the embodiments of this disclosure and the features in the embodiments can be combined with each other to obtain new embodiments.

[0024] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

Claims

1. A balancing system for a medium-voltage power system based on a balancing unit, characterized in that: The balancing system is a technical system that achieves dynamic balance of power, voltage and frequency in medium-voltage power distribution networks through intelligent control strategies. The balancing system includes a power equipment module, a strategy control module, an energy management module, a service application module and a network communication module. The power equipment module includes a balance regulation unit, a modular multilevel converter, and an energy router. The balance regulation unit includes a static var compensator, a static synchronous compensator, and a dynamic voltage restorer. The modular multilevel converter uses a sub-module cascade structure to increase redundancy and achieve stable operation in high-voltage and high-capacity scenarios. The energy router adopts a modular intelligent switch and uses sensors to collect data to achieve temperature monitoring and real-time monitoring of partial discharge. The strategy control module includes a hierarchical collaborative control module, an intelligent algorithm module, and an energy management model. The hierarchical system control module monitors the bus voltage, frequency, and tidal distribution in real time and generates global scheduling instructions through MPC model predictive control. The intelligent algorithm module realizes autonomous coordination between distributed energy and load on the transformer area or user side through edge AI algorithm. The energy management model achieves production and consumption balance through multi-media model, combined with probabilistic power flow calculation and robust optimization algorithm and mixed integer nonlinear programming. The energy management module includes an energy storage system, a distributed energy management module, and a collaborative operation guarantee module. The energy storage system adopts medium-voltage direct-connected energy storage with three-level energy balancing at the cluster, module, and cell levels, and is designed with 1:1 redundancy. It utilizes the rapid charging and discharging capability of medium-voltage direct-connected energy storage to compensate for active power deficit in real time. The distributed energy management module achieves production and consumption balance through mixed integer nonlinear programming. The collaborative operation guarantee module adds hardware redundancy and software fault tolerance functions, adopts dual-machine hot standby in the medium-voltage power system, and adds data backup and fault self-diagnosis technology. The service application module includes grid connection management, grid interaction, and market platform integration functions; The network communication module includes a communication architecture protocol module and a data security protection module. The communication architecture protocol module adopts medium-voltage carrier communication, uses power distribution lines as the transmission medium, and supports GPRS and fiber optic uplink communication. The data security protection module adopts a dedicated power firewall and a one-way physical isolation device to ensure data security between the production control area and the management information area.

2. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: The static var compensator of the power balancing equipment module controls the reactor and capacitor bank through thyristors to quickly adjust reactive power and stabilize voltage. The static synchronous compensator continuously adjusts reactive power within the inductive to capacitive range based on the voltage source inverter. The dynamic voltage restorer is connected in series in the line to compensate for voltage sags and fluctuations in real time, ensuring the power supply quality of sensitive loads.

3. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: The hierarchical system control module of the strategy control module generates global scheduling instructions through MPC model predictive control, establishes an MPC prediction model, describes the system power deviation and energy storage state, and defines it as: ； in, Let k be the system active power deviation. Let k be the energy storage state of charge at time k, then the energy storage charging and discharging power is: The uncontrollable load fluctuation prediction error is: ： Discrete time, sampling period The state equation is: ； Matrix parameters: ： in The self-discharge coefficient of energy storage; ； The energy storage charge and discharge efficiency is (0~1). This refers to the rated capacity of the energy storage. ； C is the coefficient representing the impact of disturbance on power deviation.

4. A medium-voltage power system balancing system based on a balancing unit according to claim 3, characterized in that: In the hierarchical system control module of the strategy control module, the MPC model prediction, by clearly defining the energy storage module and power deviation, aims to minimize the tracking error and control cost in the prediction time domain, thereby achieving power balance control and voltage control. The objective function of the balance control is: ； in To predict the time domain, To control the time domain, The power deviation at time t (with the target being tracked being 0) is the power deviation at time t. This is a reference value for the SOC of energy storage. Let be the energy storage control quantity at time t. This represents the error weight; a larger weight indicates that the variable is controlled with higher priority. Weights for changes in control quantities; The voltage control objective function is: ； in Let be the voltage deviation at node i at time t. Let j be the control variable for the j-th reactive power device. For node voltage weights, The cost of adjusting reactive power devices.

5. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: The edge AI algorithm of the policy control module achieves fast classification and dynamic balance decision-making at the edge through random forest, and reinforcement learning learns the optimal control strategy for energy storage charging and discharging through interaction with the environment. ； in, For state Next action value, For learning rate, As a discount factor, For instant rewards, For the next state The optimal action value is achieved by using edge AI algorithms to enable the balancing system to make real-time decisions on energy storage charging and discharging and distributed power output adjustment, thereby minimizing system power deviation.

6. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: In the energy management model of the strategy control module, probabilistic power flow calculations are used to quantify the probabilistic impact of input variables on system state variables. Uncertainty is described through a probabilistic model, and the probability distribution of system state variables is derived. In medium-voltage systems, random variables mainly include node loads. Distributed power generation photovoltaic and wind power By establishing its probability density function PDF: ； in, The average load ， If the standard deviation is given, then the work is done. The node power balance equation is: ； No effect The nodal power balance equations are as follows: ； in, Let i be the voltage magnitude at nodes i and j. Let be the voltage phase angle difference between nodes i and j. Nodal admittance matrix Element.

7. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: The distributed energy management module of the energy management module optimizes decision variables through mixed integer nonlinear programming to ensure a dynamic balance between energy production and consumption. Its core constraint on production capacity is that at any time t, total energy output must equal total consumption. ； in For energy storage charging / discharging efficiency; The energy storage state of charge constraints are: ； in To schedule the time step, This refers to the rated capacity of the energy storage. By planning and optimizing decisions based on capacity and storage constraints, we can ensure the best distributed energy dispatch strategy, achieving dynamic balance between production and consumption while meeting all constraints and minimizing operating costs.

8. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: The service application module achieves real-time monitoring of voltage and power factor through grid connection management, and shares data with the upper-level power grid dispatch center. It also supports one-click load transfer and aggregator collaborative control, and connects with the market platform to support the electricity ancillary services market and spot trading business.

9. A medium-voltage power system balancing system based on a balancing unit according to claim 1, characterized in that: The optimization method for the balanced system is implemented as follows: Step 1: Real-time data acquisition of the balancing unit to obtain the real-time operating status of each component within the balancing unit. The acquisition targets include distributed power sources, energy storage systems, and controllable loads. Data is refreshed at the second or sub-second level through synchronous phasor measurement units, smart meters, and edge terminals to ensure data timeliness. Step 2: Power forecasting, predicting power fluctuations in the near future to provide a forward-looking basis for optimization decisions; Step 3: Balance unit status assessment and optimization target determination. Based on real-time data acquisition and short-term power forecast, determine the priority of the medium-voltage power system and sort the targets for optimization. Step 4: Based on the current state and predicted data, use the MPC model to predict the optimal control commands for each controllable device in the balance unit, and use edge AI algorithms to determine the dynamic balance decision. Step 5: Control command generation and execution, converting the optimization results into executable commands for specific devices to achieve real-time adjustment; Step 6: Real-time anomaly handling: When the power deviation > 10% of the rated capacity, the voltage exceeds the limit by > ±10%, or there is a equipment failure, the anomaly handling mechanism is triggered. The redundancy mechanism is invoked to make up for the power shortfall or interruptible loads are cut off first. An anomaly signal is sent to the dispatch center to request coordination and support from the superior.

Images