Energy storage collaborative control system and method based on multi-time scale hierarchical cascade

Abstract

The application relates to the technical field of energy storage control, and discloses an energy storage collaborative control system and method based on multi-time scale hierarchical cascading. The system adopts a centralized and distributed hybrid energy storage architecture to construct a three-layer cascading control system of millisecond-level decentralized autonomous control, second-level centralized coordination and minute-level energy optimization. The millisecond-level layer responds quickly based on local electrical quantities and reports over-limit power according to a dynamic power threshold; the second-level layer calculates a dynamic threshold, transfers excess power of the distributed unit to the centralized unit for unified adjustment, differentiates power allocation according to state of charge, state of health and system demand, and adaptively switches the control mode according to the power grid frequency and voltage deviation; and the minute-level layer formulates a charging and discharging plan, evaluates capacity configuration based on operation data and outputs optimization suggestions. The application realizes multi-time scale control decoupling and collaboration, improves response speed and system reliability, optimizes energy storage balance and life management, and significantly improves the overall performance of the energy storage system.

Description

Technical Field

[0001] This invention relates to the field of energy storage control technology, and in particular to a multi-timescale hierarchical energy storage collaborative control system and method. Background Technology

[0002] Currently, large-scale energy storage power plants typically employ either a fully distributed control architecture or a fully centralized control architecture. Distributed control, where each energy storage converter responds independently, lacks overall coordination, resulting in low capacity utilization and an inability to achieve energy sharing between subgrids. Centralized control, while capable of global optimization, suffers from system paralysis upon central controller failure, exhibiting poor reliability and extremely high real-time communication requirements. Furthermore, existing control strategies are often designed for single objectives. When the grid experiences multiple disturbances simultaneously, such as frequency deviations, voltage over-limits, and power oscillations, different control modes are prone to conflict or insufficient response. Simultaneously, energy storage capacity planning is often independent of operational control strategies, leading to underutilization of the configured capacity or failure to meet actual control needs. Moreover, in multi-energy storage converter systems, battery state-of-charge management is often rudimentary, lacking refined balancing strategies, which can easily cause some batteries to overcharge and over-discharge, shortening the overall lifespan of the energy storage system. Summary of the Invention

[0003] This invention provides a hierarchical energy storage collaborative control system and method based on multiple time scales to solve the problem of poor overall performance of energy storage systems caused by the inherent contradiction and coordination difficulties between centralized and decentralized control in the prior art.

[0004] The first aspect of this invention provides an energy storage collaborative control system, comprising: an energy storage physical architecture including a centralized energy storage unit and at least two distributed energy storage units, each energy storage unit being connected via a grid connection point bus; a millisecond-level distributed autonomous control layer including energy storage converters corresponding to each energy storage unit, each energy storage converter having a built-in autonomous controller for performing fast power response based on locally acquired electrical quantities and receiving dynamic power thresholds; when real-time power fluctuations exceed the dynamic power thresholds, reporting the excess power demand to a second-level centralized coordination control layer as needed; and a second-level centralized coordination control layer configured with an energy storage coordination controller for performing second-level coordination control, including: acquiring the current state of charge, health status, and historical power fluctuation characteristics of each distributed energy storage unit to calculate the dynamic power threshold of the corresponding distributed energy storage unit. The calculated dynamic power threshold is then sent to the millisecond-level distributed autonomous control layer, where the state of charge, health status, and historical power fluctuations are positively correlated with the dynamic power threshold. The layer receives over-limit power demands reported by the millisecond-level distributed autonomous control layer and transfers the aggregated over-limit power demands to the centralized energy storage unit for unified adjustment. Based on the status information of each energy storage unit and the total system power demand, the layer dynamically allocates the charging and discharging power of each energy storage unit. The status information includes the state of charge, health status, and real-time charging and discharging power. A minute-level energy optimization management layer, equipped with an energy management system, is used to formulate minute-level timescale energy storage charging and discharging plans and send them to the second-level centralized coordination control layer, as well as receive operational data reported by the second-level centralized coordination control layer and dynamically evaluate the rationality of capacity configuration based on the operational data.

[0005] A second aspect of this invention provides an energy storage collaborative control method, the method comprising: a minute-level energy optimization step: the minute-level energy optimization management layer formulates a minute-level timescale energy storage charging and discharging plan and sends it to the second-level centralized coordination control layer to determine the total system power demand; a dynamic threshold calculation and distribution step: the second-level centralized coordination control layer obtains the current state of charge, health status, and historical power fluctuation characteristics of each distributed energy storage unit to calculate the dynamic power threshold of the corresponding distributed energy storage unit, and distributes the calculated dynamic power threshold to the millisecond-level decentralized autonomous control layer, wherein the state of charge, health status, and historical power fluctuation are positively correlated with the dynamic power threshold; and a local monitoring and judgment step: the millisecond-level decentralized autonomous control layer performs a fast power response based on local electrical quantities to monitor real-time power fluctuations; when the real-time power fluctuation does not exceed the distributed threshold, the method determines the dynamic power threshold. When the dynamic power threshold is reached, local autonomous processing continues. When the dynamic power threshold is exceeded, the excess is reported as over-limit power demand to the second-level centralized coordination and control layer. Over-limit power coordination step: The second-level centralized coordination and control layer receives the over-limit power demand reported by all distributed energy storage units and transfers the aggregated over-limit power demand to the centralized energy storage unit for unified adjustment. Global power dynamic allocation step: The second-level centralized coordination and control layer obtains the status information of each energy storage unit and dynamically allocates the charging and discharging power of each energy storage unit in combination with the total power demand of the system. Operation feedback and optimization step: The second-level centralized coordination and control layer records the operation data of each energy storage unit and reports it to the minute-level energy optimization management layer. The minute-level energy optimization management layer dynamically evaluates the rationality of the system capacity configuration based on the operation data and outputs capacity adjustment suggestions.

[0006] In one feasible implementation, the calculation of the dynamic power threshold of the corresponding distributed energy storage unit includes: determining a first coefficient, a second coefficient, and a third coefficient based on the state of charge, the health state, and the historical power fluctuation characteristics, respectively; multiplying the first coefficient by the state of charge, the second coefficient by the health state, and the third coefficient by the historical power fluctuation characteristics, and then summing the three products and multiplying them by the rated power of the distributed energy storage unit to obtain the dynamic power threshold.

[0007] In one feasible implementation, transferring the aggregated excess power demand to the centralized energy storage unit for unified adjustment includes: preferentially allocating the aggregated excess power demand to the portion of the centralized energy storage unit whose state of charge (SBC) is within a preset intermediate range; if the SBC of the centralized energy storage unit is not within the preset intermediate range, then allocating the excess power demand to the portion whose SBC deviates from the intermediate value the least; if the SBC of the centralized energy storage unit exceeds a preset boundary, then triggering an alarm and requesting an adjustment to the charging and discharging plan from the upper layer, while temporarily lowering the dynamic power threshold of each distributed energy storage unit.

[0008] In one feasible implementation, the dynamic allocation of charging and discharging power for each energy storage unit includes: For flywheel or supercapacitor-type energy storage units, their state of charge (SOC) is not included in the calculation of power allocation weight, and the charging and discharging power is determined by the total power demand of the system and its own response speed; for lithium-ion battery-type energy storage units, the charging and discharging power is dynamically allocated according to the SOC, wherein units with a higher SOC bear a larger proportion during discharge, and units with a lower SOC bear a larger proportion during charging; for flow battery-type energy storage units, the charging and discharging power is dynamically allocated according to the SOC, and the SOC balance is stronger than that of lithium-ion battery-type energy storage units.

[0009] In one feasible implementation, the dynamic allocation of charging and discharging power of each energy storage unit further includes: when the absolute value of the total power demand of the system is greater than a preset first threshold, the power allocation weight of each energy storage unit tends to be average, prioritizing the power response speed; when the absolute value of the total power demand of the system is less than a preset second threshold, the difference in power allocation weight of each energy storage unit increases, prioritizing the achievement of state of charge balance; when the absolute value of the total power demand of the system is between the first threshold and the second threshold, the power allocation weight continuously transitions with the change of the total power demand of the system.

[0010] In one feasible implementation, the dynamic assessment of capacity configuration rationality and the output of capacity adjustment suggestions include: when the power threshold of a distributed energy storage unit is in a saturated state for a long period of time and its lifetime consumption rate is greater than a first preset consumption threshold, outputting a suggestion to replace the corresponding distributed energy storage unit with a higher power density energy storage type; when the cycle lifetime consumption rate of a centralized energy storage unit due to bearing excess power demand is greater than a second preset consumption threshold, and its state of charge exceeding risk index is higher than a preset threshold, outputting a suggestion to increase the energy capacity of the centralized energy storage unit; when the power threshold of a distributed energy storage unit is in a low position for a long period of time and the state of charge balancing effect among units is less than a preset standard, outputting a suggestion to reduce the rated power of the corresponding distributed energy storage unit or reduce the number of units.

[0011] In one feasible implementation, the second-level centralized coordination control layer further performs the following steps: acquiring the system operating status, which includes frequency deviation and voltage deviation; dynamically switching control modes according to the system operating status and prioritizing each control mode: when the frequency deviation or the voltage deviation is greater than the corresponding preset threshold, switching to grid support mode, increasing the weight of power response-related control functions, and downgrading or suspending state-of-charge (SOC) related control functions, so that power response takes precedence over SOC; when both the frequency deviation and the voltage deviation are less than the corresponding preset dead zone threshold, switching to economic operation mode, where the economic operation mode dominates power allocation, and maintaining the weight of SOC related control functions at the baseline value; and weighting and fusing the power commands output by each control mode according to their priority weights to obtain the total power command.

[0012] The second-level centralized coordination control layer further performs the following steps: Based on the state of the frequency deviation and voltage deviation, dynamically adjust the power allocation strategy of each energy storage unit: when the frequency deviation or voltage deviation is greater than the corresponding preset threshold, the power allocation weight of each energy storage unit tends to be averaged, prioritizing power response speed; when both the frequency deviation and voltage deviation are less than the corresponding preset dead zone threshold, the difference in power allocation weight among each energy storage unit increases, prioritizing state-of-charge balance. Based on the total power command and the power allocation strategy, determine and issue the charging and discharging power of each energy storage unit.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described energy storage coordinated control method.

[0014] The technical solution provided by this invention achieves natural decoupling and coordination of multi-timescale control tasks by constructing a hybrid energy storage physical architecture of "centralized + distributed" and a three-layer hierarchical control system of "millisecond-second-minute". The millisecond-level distributed autonomous control layer independently executes fast power response based on local electrical quantities without waiting for upper-level instructions, ensuring instantaneous stability during grid disturbances. The second-level centralized coordination control layer calculates dynamic power thresholds based on state of charge, health status, and historical power fluctuation characteristics, absorbing high-frequency small fluctuations locally and only aggregating the excess of low-frequency large fluctuations to the centralized energy storage unit for unified adjustment, significantly reducing communication burden and computational pressure on the centralized layer. At the same time, it dynamically allocates charging and discharging power according to the state of charge of each energy storage unit and the total power demand of the system, avoiding control conflicts between different modes. The minute-level energy optimization management layer feeds operational data back into capacity planning, dynamically evaluates the rationality of configuration, and outputs capacity adjustment suggestions, forming a continuous iterative closed loop of "planning-operation-optimization". Therefore, this invention improves system reliability and response speed while optimizing energy balance and lifespan management among energy storage units, achieving optimization of the entire life cycle cost and significantly improving the reliability and overall performance of the energy storage system. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of an embodiment of the energy storage collaborative control system based on a multi-timescale hierarchical structure in this invention. Figure 2 This is a schematic diagram of an embodiment of the energy storage collaborative control method based on multi-timescale hierarchical structure in this invention. Detailed Implementation

[0016] This invention provides a hierarchical energy storage collaborative control system and method based on multiple time scales, which improves the overall performance of the energy storage system through hierarchical decoupling and dynamic coordination.

[0017] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the energy storage collaborative control system method based on multi-timescale hierarchical structure in this invention includes: It should be noted that, Figure 1 The inclusion of centralized and distributed energy storage units within the millisecond-level decentralized autonomous control layer indicates that each of these energy storage units possesses millisecond-level autonomous control capabilities implemented by a built-in autonomous controller, rather than limiting the energy storage unit itself to a control layer component.

[0019] The energy storage physical architecture includes one centralized energy storage unit and at least two distributed energy storage units, all connected via a grid connection point bus. The centralized energy storage unit can utilize lithium-ion batteries or flow batteries, possessing high capacity and high energy density characteristics, and is used for long-term energy regulation tasks. The distributed energy storage units can utilize supercapacitors or flywheel energy storage devices, possessing fast response and high power density characteristics, and are used for short-term power smoothing tasks. The centralized and distributed energy storage units are connected via the grid connection point bus, forming a hybrid energy storage topology, providing the physical basis for hierarchical control.

[0020] The millisecond-level distributed autonomous control layer includes energy storage converters corresponding to each energy storage unit. Each energy storage converter has a built-in autonomous controller. The millisecond-level distributed autonomous control layer performs fast power response based only on locally acquired electrical quantities. It is used to perform fast power response based on locally acquired electrical quantities and receive dynamic power thresholds. When real-time power fluctuations exceed the dynamic power thresholds, it reports the excess power demand to the second-level centralized coordination control layer as needed.

[0021] The second-level centralized coordination control layer, equipped with an energy storage coordination controller, performs second-level coordination control, including: receiving energy storage charging and discharging plans and determining the total system power demand based on these plans; acquiring the current state of charge, health status, and historical power fluctuation characteristics of each distributed energy storage unit to calculate the dynamic power threshold for the corresponding distributed energy storage unit, and distributing the calculated dynamic power threshold to the millisecond-level distributed autonomous control layer, where the state of charge, health status, and historical power fluctuation are positively correlated with the dynamic power threshold; receiving over-limit power demands reported by the millisecond-level distributed autonomous control layer and transferring the aggregated over-limit power demands to the centralized energy storage unit for unified adjustment; dynamically allocating charging and discharging power based on the status information of each energy storage unit and the total system power demand, where the status information includes the state of charge, health status, and real-time charging and discharging power; and recording and reporting the operating data of each energy storage unit. The energy storage coordination controller, as the core of the system, adopts a multi-core heterogeneous architecture.

[0022] The energy storage coordination controller comprises a communication interface unit, a multi-core heterogeneous data processing unit, and a coordination computing unit. The communication interface unit includes at least one GOOSE fiber optic network interface for high-speed real-time communication with each energy storage converter, and at least one standard Ethernet interface for communication with the upper-level energy management system or dispatch center. The multi-core heterogeneous data processing unit employs a multi-core heterogeneous processor, including a first processor core and a second processor core. The first processor core handles millisecond-level instruction transmission and reception and status monitoring for the distributed autonomous control layer, while the second processor core runs optimization algorithms for the second-level centralized coordination control layer. For example, a high-real-time ARM Cortex-R core is dedicated to handling millisecond-level instruction transmission and reception and status monitoring for the distributed autonomous control layer, ensuring low latency, while a high-performance ARM Cortex-A core is used to run complex optimization algorithms for the second-level centralized coordination control layer. The coordination computing unit is responsible for executing core algorithms such as dynamic power limit determination, SOC coupling allocation, and mode switching. Furthermore, the system periodically (e.g., every 200 milliseconds) collects the status information (including SOC, SOH, and charging / discharging power) of all energy storage units and the global information of the grid connection point (including total power demand, voltage deviation, and frequency deviation) through a high-speed redundant communication network (e.g., GOOSE fiber optic network with a dual-ring network redundancy architecture). The system connects the energy storage coordinating controller with all energy storage converters to ensure the real-time and reliable transmission of control commands and status information.

[0023] The minute-level energy optimization management layer is equipped with an energy management system, which is used to formulate energy storage charging and discharging plans on a minute-level time scale and distribute them to the second-level centralized coordination and control layer; and to receive operating data reported by the second-level centralized coordination and control layer, dynamically evaluate the rationality of capacity configuration and output capacity adjustment suggestions.

[0024] In this embodiment of the invention, by constructing a hybrid energy storage physical architecture of "centralized + distributed" and a three-layer hierarchical control system of "millisecond-second-minute", the natural decoupling and coordination of control tasks at multiple time scales are achieved. The millisecond-level distributed autonomous control layer independently executes rapid power response based on local electrical quantities, without waiting for upper-level instructions, ensuring instantaneous stability during grid disturbances. The second-level centralized coordination control layer calculates dynamic power thresholds based on state of charge, health status, and historical power fluctuation characteristics, absorbing high-frequency small fluctuations locally and only aggregating the excess of low-frequency large fluctuations to the centralized energy storage unit for unified adjustment, significantly reducing communication burden and computational pressure on the centralized layer. At the same time, it dynamically allocates charging and discharging power based on the status information of each energy storage unit and the total power demand of the system, avoiding control conflicts between different modes. The minute-level energy optimization management layer feeds operational data back into capacity planning, dynamically evaluates the rationality of configuration, and outputs capacity adjustment suggestions, forming a continuous iterative closed loop of "planning-operation-optimization". Therefore, this invention improves system reliability and response speed while optimizing energy balance and lifespan management among energy storage units, achieving optimization of the entire life cycle cost and significantly improving the reliability and overall performance of the energy storage system.

[0025] Please see Figure 2 One embodiment of the energy storage collaborative control method based on multi-timescale hierarchical structure in this invention includes: 201. Minute-level energy optimization steps: The minute-level energy optimization management layer formulates a minute-level timescale energy storage charging and discharging plan and sends it to the second-level centralized coordination and control layer to determine the total power demand of the system; The minute-level energy optimization management layer is equipped with an energy management system, which connects to the upper-level dispatch center or market trading platform via a standard Ethernet interface. Based on the received day-ahead or intraday dispatch plans from the upper-level dispatch center, real-time market electricity price signals, and ultra-short-term renewable energy forecast data, the energy management system uses a rolling optimization algorithm to formulate a charging and discharging plan curve for the energy storage system within a preset future time period. The preset time can be 15 minutes, one hour, etc. This energy storage system charging and discharging plan curve includes the total system power command and power adjustment direction at each time step, reflecting the charging and discharging behavior that the energy storage system should follow in the future. After formulation, the energy management system distributes the energy storage charging and discharging plan to the second-level centralized coordination and control layer via Ethernet. The second-level centralized coordination and control layer parses the received energy storage charging and discharging plan and extracts the total system power demand corresponding to the current moment. The total system power demand reflects the magnitude and direction of the power that the energy storage system as a whole needs to inject into or absorb from the grid at the current point in time.

[0026] 202. Dynamic threshold calculation and distribution steps: The second-level centralized coordination control layer obtains the current state of charge, health status and historical power fluctuation characteristics of each distributed energy storage unit to calculate the dynamic power threshold of the corresponding distributed energy storage unit, and distributes the calculated dynamic power threshold to the millisecond-level decentralized autonomous control layer. Among them, the state of charge, health status, historical power fluctuation and dynamic power threshold are positively correlated. The second-level centralized coordination control layer periodically collects the current state of charge, health status, and historical power fluctuation characteristics of each distributed energy storage unit through a high-speed redundant communication network. Based on this information, the energy storage coordination controller calculates a dynamic power threshold for each distributed energy storage unit. This dynamic power threshold reflects the upper limit of the corresponding distributed energy storage unit's ability to autonomously handle local power fluctuations. The "positive correlation" is reflected in the fact that units with higher state of charge, higher health status, and more significant historical power fluctuations have correspondingly larger dynamic power thresholds. After calculation, the second-level centralized coordination control layer periodically sends these dynamic power thresholds to the millisecond-level decentralized autonomous control layers of each distributed energy storage unit through the high-speed communication network, serving as boundary conditions for local autonomous control.

[0027] 203. Local monitoring and judgment steps: The millisecond-level distributed autonomous control layer performs fast power response based on local electrical quantities and monitors real-time power fluctuations; when the real-time power fluctuation does not exceed the issued dynamic power threshold, it continues to be processed locally autonomously; when it exceeds the dynamic power threshold, the excess part is reported as an over-limit power demand to the second-level centralized coordination control layer. The millisecond-level distributed autonomous control layer consists of autonomous controllers built into each energy storage converter. Each autonomous controller first receives the dynamic power threshold issued by the second-level centralized coordination control layer, and simultaneously collects electrical quantities such as voltage, current, and frequency at the grid connection point in real time through local sensors. Based on these locally collected electrical quantities, the autonomous controller independently executes a fast power response according to a preset control strategy, without waiting for instructions from the upper layer. During the execution of the fast response, the autonomous controller also continuously monitors its own real-time power fluctuations and calculates the change in current power relative to the steady state. The autonomous controller compares the absolute value of the real-time monitored power fluctuation with the pre-received dynamic power threshold: if the current power fluctuation does not exceed the threshold, it indicates that the fluctuation is a high-frequency, small-amplitude disturbance, which the autonomous controller can continue to handle independently without reporting to the upper layer; if the current power fluctuation exceeds the threshold, it indicates that the amplitude of the fluctuation is large and exceeds the local autonomous adjustment capability. In this case, the autonomous controller treats the power demand exceeding the threshold as an over-limit power demand and reports it to the second-level centralized coordination control layer on demand through the high-speed communication network, requesting assistance from the upper layer for adjustment.

[0028] 204. Excess Power Coordination Steps: The second-level centralized coordination control layer receives the excess power demands reported by all distributed energy storage units and transfers the aggregated excess power demands to the centralized energy storage unit for unified adjustment. The second-level centralized coordination control layer receives over-limit power demands reported by each distributed energy storage unit through a high-speed communication network and aggregates all over-limit demands. Subsequently, the energy storage coordination controller transfers these aggregated over-limit power demands to the centralized energy storage unit for unified adjustment. During the allocation process, over-limit power demands are preferentially allocated to the portion of the centralized energy storage unit whose state of charge (SOC) is within a preset intermediate range. If the SOC of the centralized energy storage unit is not within the preset intermediate range, the over-limit power demands are allocated to the portion whose SOC deviates from the intermediate value the least. If the SOC of the centralized energy storage unit exceeds a preset boundary, an alarm is triggered, and a request is made to the upper-level minute-level energy optimization management layer to adjust the charging and discharging plan. At the same time, the dynamic power threshold of each distributed energy storage unit is temporarily lowered. Once the SOC of the centralized energy storage unit returns to the normal range, the original dynamic power threshold calculation method is restored.

[0029] 205. Global power dynamic allocation steps: The state information of each energy storage unit is obtained by the second-level centralized coordination control layer, and the charging and discharging power of each energy storage unit is dynamically allocated in combination with the total power demand of the system. The centralized coordination control layer acquires the state of charge (SBC), health status, and real-time charge / discharge power of all energy storage units within seconds. Under the premise of meeting the total system power demand, the energy storage coordination controller dynamically allocates charge / discharge power based on the SBC status information of each unit. Different allocation strategies are adopted for energy storage units of different technologies: for flywheel or supercapacitor energy storage units, their SBC does not participate in the power allocation weight calculation; the charge / discharge power is determined by the total system power demand and its own response speed. For lithium-ion battery energy storage units, charge / discharge power is dynamically allocated according to SBC, with units with higher SBCs bearing a larger proportion during discharge and units with lower SBCs bearing a larger proportion during charging. For flow battery energy storage units, charge / discharge power is dynamically allocated according to SBC, and the intensity of SBC balance is greater for flow batteries than for lithium-ion battery energy storage units. Meanwhile, the allocation strategy is dynamically adjusted according to the size of the total power demand of the system: when the total power demand of the system is large, the power allocation weight of each energy storage unit tends to be average, and the power response speed is given priority; when the total power demand of the system is small, the difference in the power allocation weight of each energy storage unit increases, and the state of charge balance is given priority; when the total power demand of the system is in an intermediate state, the power allocation weight transitions continuously with the change of the total power demand of the system.

[0030] 206. Operation Feedback and Optimization Steps: The second-level centralized coordination control layer records the operation data of each energy storage unit and reports it to the minute-level energy optimization management layer. The minute-level energy optimization management layer dynamically evaluates the rationality of the system capacity configuration based on the operation data and outputs capacity adjustment suggestions.

[0031] The second-level centralized coordination and control layer continuously records key operational data of each energy storage unit throughout the entire operation, including actual power fluctuation data, charging and discharging frequency, state of charge change trajectory, and cycle life consumption. This data reflects the behavioral characteristics and performance of the energy storage system in actual operation. The second-level centralized coordination and control layer organizes and summarizes this recorded data and periodically reports it via Ethernet to the minute-level energy optimization management layer, providing data support for upper-level capacity configuration assessment and optimization decisions. The energy management system, with a minute-level energy optimization management layer, receives operational data reported by the second-level centralized coordination control layer and performs capacity configuration rationality assessments based on this data: when the dynamic power threshold of a distributed energy storage unit is in a state of saturation for a long time and its lifespan is consumed rapidly, it outputs a suggestion to replace the corresponding distributed energy storage unit with a higher power density energy storage type; when the cycle life of a centralized energy storage unit is consumed significantly due to bearing excess power demand and its state of charge exceeds the limit risk is high, it outputs a suggestion to increase the energy capacity of the centralized energy storage unit; when the dynamic power threshold of a distributed energy storage unit is in a low position for a long time and the state of charge balancing effect among the units is poor, it outputs a suggestion to reduce the rated power of the corresponding distributed energy storage unit or reduce the number of units.

[0032] Furthermore, the dynamic power threshold of the corresponding distributed energy storage unit is calculated, including: determining a first coefficient, a second coefficient, and a third coefficient based on the state of charge, the state of health, and historical power fluctuation characteristics; multiplying the first coefficient by the state of charge, the second coefficient by the state of health, and the third coefficient by the historical power fluctuation characteristics; summing these three products and multiplying the sum by the rated power of the distributed energy storage unit to obtain the dynamic power threshold. Specifically, the formula for calculating the dynamic power threshold is:

[0033] Wherein, the function f takes the form of a linear function:

[0034] in, The rated power of distributed energy storage unit i, , and Let be the weighting coefficients, representing the first, second, and third coefficients respectively, and satisfy the following condition: , This represents the current state of charge of distributed energy storage unit i. The health status of distributed energy storage unit i. This represents the normalized standard deviation of power fluctuation for distributed energy storage unit i. Weighting coefficients. , , The minute-level energy optimization management layer updates at preset intervals, such as every 5 minutes. This updates when the absolute value of the local power demand fluctuation of unit i... At that time, the fluctuation is absorbed by the millisecond-level distributed autonomous control layer; when At that time, the excess part It needs to be reported to the centralized coordination and control layer at the second level.

[0035] The state of charge (SOC) coefficient is the dominant energy equilibrium factor. When the SOC is larger, cells with higher SOC gain greater "autonomy" (higher SOC). ( ), discharge more actively; and vice versa. The design should be coupled with the battery's charge and discharge efficiency curve. The State of Health (SOH) coefficient governs health protection; the higher the SOH, the greater the autonomy of the healthy unit; the aging units are handled by the centralized units, and the degradation rate of the distributed aging units is slowed down by the moderate sacrifice of centralized resources, thereby extending the replacement cycle of the entire energy storage cluster and optimizing the total life cycle cost. The fluctuation coefficient governs the power smoothness; the more significant the historical power fluctuation of a unit, the larger its power threshold, allowing it to continue to be autonomous and preventing it from frequently exceeding the limit and increasing the burden on the centralization layer.

[0036] Furthermore, the aggregated excess power demand is transferred to centralized energy storage units for unified adjustment. This includes: prioritizing the allocation of excess power demand to centralized energy storage units whose state of charge (SOC) is within a preset intermediate range; if the SOC of a centralized energy storage unit is not within the preset intermediate range, allocating the excess power demand to the unit whose SOC deviates from the intermediate value the least; and triggering an alarm and requesting adjustments to the charging and discharging plan from the upper layer if the SOC of a centralized energy storage unit exceeds a preset boundary, while temporarily lowering the dynamic power threshold of each distributed energy storage unit. Specifically, the second-level centralized coordination control layer aggregates the excess power demand reported by all distributed energy storage units to obtain the total excess power. The total excess power is transferred to centralized energy storage units according to the following priority rules: First priority, the total excess power is preferentially allocated to centralized energy storage units whose state of charge is in a preset intermediate range. (e.g., [40%, 60%]) partial responsibility; second priority: if no centralized unit has a state of charge in the preset middle range, the excess power will be allocated to the unit with the smallest deviation of its state of charge from the middle value by 50%, achieving "minimum cost equilibrium"; third priority: if the state of charge of all centralized units exceeds the preset boundary (e.g., above 80% or below 20%), an alarm will be triggered and a request will be made to the upper-level minute-level energy optimization management layer to adjust the charging and discharging plan, while the power threshold of all distributed energy storage units will be multiplied by the global attenuation factor. (For example =0.5, or dynamically calculated based on the degree of deviation of the centralized unit's state of charge from 50%; the greater the deviation, the higher the value. (The smaller the value), the more power fluctuations are forced to be submitted. Once the state of charge of the centralized unit returns to the normal range, the original power threshold calculation method will be restored.

[0037] Furthermore, the charging and discharging power of each energy storage unit is dynamically allocated, including: for flywheel or supercapacitor-type energy storage units, their state of charge (SOC) is not included in the power allocation weight calculation; the charging and discharging power is entirely determined by the total system power demand and their own response speed. For lithium-ion battery-type energy storage units, charging and discharging power is dynamically allocated according to SOC, with units with higher SOC bearing a larger proportion during discharge and units with lower SOC bearing a larger proportion during charging. For flow battery-type energy storage units, charging and discharging power is dynamically allocated according to SOC, and the strength of SOC balance is greater than that of lithium-ion battery-type energy storage units. Specifically, the second-level centralized coordination control layer adopts differentiated power allocation strategies based on the technology type of each energy storage unit. For flywheel or supercapacitor-type energy storage units, their SOC is only used for safety protection and does not participate in the power allocation weight calculation; their charging and discharging power is entirely determined by the total system power demand and the response speed of each unit, prioritizing their use to handle high-frequency power fluctuations. For lithium-ion battery-type energy storage units, the discharge weight of unit i during discharge... Calculate using the following formula:

[0038] Charging weight of cell i during charging The calculation is as follows:

[0039] Furthermore, the dynamic allocation of charging and discharging power among each energy storage unit also includes: when the absolute value of the total system power demand is greater than a preset first threshold, the power allocation weight of each energy storage unit tends to be average, prioritizing power response speed; when the absolute value of the total system power demand is less than a preset second threshold, the difference in power allocation weight among each energy storage unit increases, prioritizing the achievement of state-of-charge equilibrium; when the absolute value of the total system power demand is between the first and second thresholds, the power allocation weight continuously transitions with changes in the total system power demand. Specifically, dynamic adjustment is achieved by introducing a dynamic adjustment factor λ for the state-of-charge equilibrium weight, where λ is the absolute value of the total system power demand. The function is calculated using the following formula:

[0040] in, For adjustment coefficients, The rated power of the energy storage system is the absolute value of the total power demand of the system. When the value exceeds a preset first threshold, λ approaches 0, and the power allocation weight of each energy storage unit tends to be evenly distributed, prioritizing rapid power response capability; when... When the value is less than the preset second threshold, λ approaches 1, the power allocation weight difference between units increases, and charge state balance is achieved preferentially; when When the value is between the first and second thresholds, λ changes continuously between 0 and 1, and the power allocation weight transitions smoothly with the change in the total power demand of the system, avoiding control oscillations caused by sudden changes in weight.

[0041] The essence of λ is a "flexible priority scheduler". Traditional priority scheduling (such as frequency modulation mode having the highest priority when the frequency deviation is large) is a "hard switch", which is prone to oscillation. The λ factor realizes continuous, smooth and differentiable priority adjustment of the control target.

[0042] Specifically, λ is assigned a differentiated value for energy storage units of different technology types: For flywheel or supercapacitor-type energy storage units, λ=0 is taken, and their state of charge is only used for safety protection and does not participate in the calculation of power allocation weight. The charging and discharging power is entirely determined by the total system power demand ΔP. total Its own response speed determines this; For lithium-ion battery energy storage units, let λ=λ Li The calculation, based on the above formula, takes into account both power response and state of charge balance.

[0043] For flow battery-type energy storage units, take ,in (For example, Because the lifespan of flow batteries is not sensitive to deep charge-discharge cycles and is more suitable for equalization regulation, a larger λ is assigned to make its state of charge converge more aggressively to the mean.

[0044] Furthermore, the adjustment coefficient Adjustments are made based on the power grid's operating status; the calculation formula is as follows:

[0045] in, Based on the value, and This is the sensitivity coefficient. When the frequency deviation... or voltage deviation When it increases, linearly increasing, It rapidly approaches 0, and SOC equalization gives way to power response. Simultaneously, a dead zone is introduced: when... <0.02Hz and When <0.01 pu, = This allows the system to focus on SOC balancing. This approach prioritizes system stability metrics (…). , ) directly embedded into unit-level control parameters ( The method in the adjustment logic of the system achieves a deep penetration from "system-level goals" to "unit-level algorithms", rather than a simple mode switch.

[0046] Furthermore, the system dynamically assesses the rationality of capacity configuration and outputs capacity adjustment recommendations, including: when the power threshold of a distributed energy storage unit is consistently saturated and its lifetime consumption rate exceeds a first preset consumption threshold, it outputs a recommendation to replace the corresponding distributed energy storage unit with a higher power density energy storage type; when the cycle lifetime consumption rate of a centralized energy storage unit exceeds a second preset consumption threshold due to bearing excess power demand, and its state-of-charge (SOC) exceedance risk index is higher than a preset threshold, it outputs a recommendation to increase the energy capacity of the centralized energy storage unit; when the power threshold of a distributed energy storage unit is consistently low and the SOC balancing effect among units is less than a preset standard, it outputs a recommendation to reduce the rated power of the corresponding distributed energy storage unit or reduce the number of units. Specifically, the minute-level energy optimization management layer outputs targeted capacity adjustment recommendations based on the following diagnostic logic. Diagnosis A (Insufficient Rated Power of Distributed Units): When the power threshold of a certain distributed energy storage unit... If the unit is in a saturated state for a long period of time (e.g., exceeding 90% of the rated power for 24 consecutive hours), and the effective throughput η of the unit is 1 Exceeding If the percentage of total energy / total fluctuating energy) × 100% remains below 80% for an extended period, and the unit's lifespan consumption rate exceeds the first preset consumption threshold (e.g., the equivalent cycle count per unit time exceeds 20% of the design value), a suggestion will be output to replace the corresponding distributed energy storage unit with a higher power density energy storage type (e.g., replacing some lithium batteries with supercapacitors). Diagnosis B (Insufficient Energy Capacity of Centralized Units): Calculate the additional equivalent cycle count that centralized energy storage units incur due to bearing the excess fluctuations of distributed units. If this value exceeds 20% of the design expectation, and the state-of-charge risk index of centralized units (defined as the proportion of sampling points where the state of charge of centralized units enters the danger zone, i.e., below 10% or above 90%, to the total number of sampling points) exceeds the preset threshold (e.g., 15%), a suggestion will be output to increase the energy capacity of centralized energy storage units (e.g., adding battery clusters). Diagnosis C (Oversized Distributed Unit Configuration): If the power threshold of a certain distributed energy storage unit is... If the power level remains low for an extended period (e.g., below 20% of the rated power) and the state of charge (SCC) balance among distributed units is poor (e.g., SCC variance is consistently higher than 0.05), it is recommended to reduce the rated power of the corresponding distributed energy storage unit or reduce the number of units.

[0047] Furthermore, the second-level centralized coordination control layer also performs the following steps: acquiring the system operating status, including frequency deviation and voltage deviation; dynamically switching control modes according to the system operating status and prioritizing each control mode: when the frequency deviation or voltage deviation is greater than the corresponding preset threshold, switching to grid support mode, increasing the weight of power response-related control functions, and downgrading or suspending state-of-load (SOL) related control functions, making power response take priority over SOL; when both frequency deviation and voltage deviation are less than the corresponding preset dead zone threshold, switching to economic operation mode, where the economic operation mode dominates power allocation, and maintaining the weight of SOL related control functions at the baseline value; and weighting and fusing the power commands output by each control mode according to priority to obtain the total power command. Specifically, the second-level centralized coordination control layer integrates multiple control function modules, including grid support mode (primary frequency support, secondary voltage regulation, damped oscillation control), economic operation mode (new energy power smoothing, dispatch plan tracking, demand peak control), and safety protection mode (main transformer overload protection, black start recovery). A second-level centralized coordination control layer constructs a decision engine based on fuzzy logic. The input feature vector of the decision engine includes frequency deviation, voltage deviation, mean state of charge (SCC), total power demand, and market electricity price, and the output is the weight coefficient of each control mode. When the frequency deviation exceeds a threshold (e.g., 0.2Hz), the weight of the frequency support mode increases significantly and has the highest priority. At this time, the adjustment target is active power. Through a single frequency support function, the active power output of the energy storage unit is rapidly increased or decreased within milliseconds to suppress frequency fluctuations. In this mode, SCC is temporarily downgraded or suspended to ensure power response speed is prioritized and to avoid slowing down the adjustment action due to SCC control. When the grid state is stable (both frequency deviation and voltage deviation are less than the corresponding preset dead zone threshold), the system switches to the economic operation mode. At this time, the adjustment target is the charging and discharging power curve and the SCC allocation strategy. Through functions such as power smoothing, scheduling plan tracking, and peak demand control, the power commands of each energy storage unit are actively adjusted to make the SCC of multiple energy storage units more consistent and prevent overcharging and over-discharging. In this mode, the power response speed requirement is relatively low, allowing for fine-grained adjustments on a minute-level timescale. Simultaneously, a safety protection mode continuously monitors the system to prevent sudden limit exceedances. The final total system power command is obtained by multiplying the power commands output from each control mode by their respective weighting coefficients and then summing them, achieving adaptive switching between multiple control modes and dynamic prioritization of targets.

[0048] Furthermore, the second-level centralized coordination control layer also includes: dynamically adjusting the power allocation strategy of each energy storage unit based on the state of frequency and voltage deviations: when the frequency or voltage deviation is greater than the corresponding preset threshold, the power allocation weight of each energy storage unit tends to be averaged, prioritizing power response speed; when both the frequency and voltage deviations are less than the corresponding preset dead zone thresholds, the difference in power allocation weights among the energy storage units increases, prioritizing the achievement of state-of-charge equilibrium. The charging and discharging power of each energy storage unit is determined and issued based on the total power command and power allocation strategy. Specifically, the second-level centralized coordination control layer dynamically adjusts the power allocation strategy of each energy storage unit by introducing a dynamic adjustment factor for state-of-charge equilibrium weights. This dynamic adjustment factor is a monotonically decreasing function of the absolute value of the system's total power demand, and it also adaptively adjusts based on the state of frequency and voltage deviations. When the frequency or voltage deviation exceeds the corresponding preset threshold, the dynamic adjustment factor approaches 0, and the power allocation weights of each energy storage unit tend to be evenly distributed, prioritizing rapid power response. When both the frequency and voltage deviations are less than the corresponding preset dead-zone thresholds, the dynamic adjustment factor approaches 1, and the differences in power allocation weights among the energy storage units increase, prioritizing state-of-charge equilibrium. When the frequency or voltage deviation is in an intermediate state, the dynamic adjustment factor continuously changes between 0 and 1, and the power allocation weights smoothly transition with changes in the system's operating state, avoiding control oscillations caused by sudden weight changes. Simultaneously, when the absolute value of the total system power demand is large, the dynamic adjustment factor also approaches 0, and the power allocation weights of each energy storage unit tend to be evenly distributed, prioritizing power response speed. When the absolute value of the total system power demand is small, the dynamic adjustment factor approaches 1, and the differences in power allocation weights among the energy storage units increase, prioritizing state-of-charge equilibrium. The dynamic adjustment factor adaptively adjusts based on the magnitude of frequency and voltage deviations: when either the frequency or voltage deviation increases, the dynamic adjustment factor rapidly approaches zero, prioritizing power response over state-of-charge equilibrium; when both the frequency and voltage deviations are less than a preset dead-zone threshold, the dynamic adjustment factor remains at its baseline value, allowing the system to focus on state-of-charge equilibrium. This method, which directly embeds system stability indicators into the adjustment logic of unit-level control parameters, achieves a deep integration from system-level objectives to unit-level algorithms, rather than a simple mode switch.

[0049] In this embodiment of the invention, a three-layer hierarchical architecture consisting of a millisecond-level distributed autonomous control layer, a second-level centralized coordination control layer, and a minute-level energy optimization management layer is constructed. This achieves natural decoupling and coordination of control tasks across multiple time scales: the millisecond-level distributed autonomous control layer independently executes rapid power response based on local electrical quantities, without waiting for instructions from the upper layer, ensuring instantaneous stability during grid disturbances; the second-level centralized coordination control layer calculates dynamic power thresholds based on state of charge, health status, and historical power fluctuation characteristics, absorbing high-frequency small-amplitude fluctuations locally and only aggregating the excess portion of low-frequency large-amplitude fluctuations to a centralized energy storage unit. The adjustment significantly reduces communication burden and centralized layer computing pressure. Simultaneously, it introduces a dynamic adjustment factor for state-of-charge (POC) balance weights to achieve flexible priority scheduling of allocation strategies that continuously change with the total power demand of the system. Furthermore, it dynamically switches between grid support mode and economic operation mode based on frequency and voltage deviations, prioritizing power response speed during grid disturbances and achieving POC balance during grid stability, thus avoiding control conflicts between different modes. The minute-level energy optimization management layer feeds operational data back into capacity planning, dynamically assessing the rationality of configuration and outputting capacity adjustment suggestions, forming a continuous iterative closed loop of "planning-operation-optimization." Therefore, this invention improves system reliability and response speed while optimizing energy balance and lifespan management among energy storage units, achieving optimization of the entire lifecycle cost and significantly improving the reliability and overall performance of the energy storage system.

[0050] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the energy storage coordinated control method.

[0051] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0052] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0053] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A hierarchical energy storage collaborative control system based on multiple time scales, characterized in that, The system includes: The energy storage physical architecture includes a centralized energy storage unit and at least two distributed energy storage units, which are connected via a grid connection point bus. The millisecond-level distributed autonomous control layer includes energy storage converters corresponding to each energy storage unit. Each energy storage converter has a built-in autonomous controller for performing fast power response based on locally collected electrical quantities and receiving dynamic power thresholds. When real-time power fluctuations exceed the dynamic power thresholds, the controller reports the excess power demand to the second-level centralized coordination control layer as needed. A second-level centralized coordination control layer, configured with an energy storage coordination controller, is used to perform second-level coordination control, including: acquiring the current state of charge, health status, and historical power fluctuation characteristics of each distributed energy storage unit to calculate the dynamic power threshold of the corresponding distributed energy storage unit, and sending the calculated dynamic power threshold to the millisecond-level decentralized autonomous control layer, wherein the state of charge, health status, and historical power fluctuation are positively correlated with the dynamic power threshold; receiving the over-limit power demand reported by the millisecond-level decentralized autonomous control layer, and transferring the aggregated over-limit power demand to the centralized energy storage unit for unified adjustment; dynamically allocating the charging and discharging power of each energy storage unit according to the status information of each energy storage unit and the total power demand of the system, wherein the status information includes the state of charge, health status, and real-time charging and discharging power; The minute-level energy optimization management layer is equipped with an energy management system, which is used to formulate energy storage charging and discharging plans on a minute-level time scale and send them to the second-level centralized coordination and control layer, as well as receive the operation data reported by the second-level centralized coordination and control layer, and dynamically evaluate the rationality of capacity configuration based on the operation data.

2. A hierarchical energy storage collaborative control method based on multiple time scales, applied to the energy storage collaborative control system as described in claim 1, characterized in that, The method includes: Minute-level energy optimization steps: The minute-level energy optimization management layer formulates a minute-level timescale energy storage charging and discharging plan and sends it to the second-level centralized coordination and control layer to determine the total power demand of the system; Dynamic threshold calculation and distribution steps: The second-level centralized coordination control layer obtains the current state of charge, health status and historical power fluctuation characteristics of each distributed energy storage unit to calculate the dynamic power threshold of the corresponding distributed energy storage unit, and distributes the calculated dynamic power threshold to the millisecond-level decentralized autonomous control layer. The state of charge, health status and historical power fluctuation are positively correlated with the dynamic power threshold. Local monitoring and judgment steps: The millisecond-level distributed autonomous control layer performs a fast power response based on local electrical quantities to monitor real-time power fluctuations; when the real-time power fluctuations do not exceed the issued dynamic power threshold, local autonomous processing continues; when the dynamic power threshold is exceeded, the excess portion is reported as an over-limit power demand to the second-level centralized coordination control layer. Excess power coordination steps: The second-level centralized coordination control layer receives the excess power demand reported by all distributed energy storage units, and transfers the aggregated excess power demand to the centralized energy storage unit for unified adjustment; Global power dynamic allocation steps: The state information of each energy storage unit is obtained by the second-level centralized coordination control layer, and the charging and discharging power of each energy storage unit is dynamically allocated in combination with the total power demand of the system. Operational feedback and optimization steps: The second-level centralized coordination and control layer records the operational data of each energy storage unit and reports it to the minute-level energy optimization management layer. The minute-level energy optimization management layer dynamically evaluates the rationality of the system capacity configuration based on the operational data and outputs capacity adjustment suggestions.

3. The energy storage coordinated control method according to claim 2, characterized in that, The calculation of the dynamic power threshold corresponding to the distributed energy storage unit includes: The first coefficient, the second coefficient, and the third coefficient are determined based on the state of charge, the health state, and the historical power fluctuation characteristics, respectively. The dynamic power threshold is obtained by multiplying the first coefficient by the state of charge, the second coefficient by the health state, and the third coefficient by the historical power fluctuation characteristics, summing the three products, and then multiplying the sum by the rated power of the distributed energy storage unit.

4. The energy storage coordinated control method according to claim 2, characterized in that, The process of transferring the aggregated excess power demand to a centralized energy storage unit for unified regulation includes: The aggregated excess power demand will be preferentially allocated to the portion of the centralized energy storage unit whose state of charge is in the preset middle range. If the state of charge of the centralized energy storage unit is not in the preset intermediate range, the excess power demand will be allocated to the part whose state of charge deviates from the intermediate value the least. If the state of charge of the centralized energy storage unit exceeds the preset boundary, an alarm is triggered and a request is made to the upper layer to adjust the charging and discharging plan. At the same time, the dynamic power threshold of each distributed energy storage unit is temporarily reduced.

5. The energy storage coordinated control method according to claim 2, characterized in that, The dynamic allocation of charging and discharging power for each energy storage unit includes: For flywheel or supercapacitor energy storage units, their state of charge does not participate in the calculation of power allocation weight, and the charging and discharging power is determined by the total power demand of the system and its own response speed. For lithium-ion battery energy storage units, charging and discharging power is dynamically allocated according to the state of charge. Units with a higher state of charge bear a larger proportion during discharging, while units with a lower state of charge bear a larger proportion during charging. For flow battery energy storage units, the charging and discharging power is dynamically allocated according to the state of charge, and the strength of its state of charge balance is greater than that of lithium battery energy storage units.

6. The energy storage coordinated control method according to claim 2, characterized in that, The dynamic allocation of charging and discharging power for each energy storage unit also includes: When the absolute value of the total power demand of the system is greater than the preset first threshold, the power allocation weight of each energy storage unit tends to be average, giving priority to ensuring power response speed; When the absolute value of the total power demand of the system is less than the preset second threshold, the difference in power allocation weight of each energy storage unit increases, and the state of charge balance is achieved first. When the absolute value of the total power demand of the system is between the first threshold and the second threshold, the power allocation weight transitions continuously with the change of the total power demand of the system.

7. The energy storage coordinated control method according to claim 2, characterized in that, The process of dynamically evaluating the rationality of system capacity configuration based on the operational data and outputting capacity adjustment suggestions includes: When the dynamic power threshold of a distributed energy storage unit is saturated for a long period of time and its lifespan consumption rate is greater than the first preset consumption threshold, the output will suggest replacing the corresponding distributed energy storage unit with a higher power density energy storage type. When the cycle life consumption rate of the centralized energy storage unit exceeds the second preset consumption threshold due to the excessive power demand it undertakes, and its state of charge over-limit risk index is higher than the preset threshold, a suggestion to increase the energy capacity of the centralized energy storage unit is output. When the dynamic power threshold of a distributed energy storage unit remains low for an extended period and the state-of-charge balancing effect among units is less than a preset standard, a suggestion is made to reduce the rated power of the corresponding distributed energy storage unit or to reduce the number of units.

8. The energy storage coordinated control method according to claim 2, characterized in that, The second-level centralized coordination and control layer also performs the following steps: The system operating status is obtained, including frequency deviation and voltage deviation; The system dynamically switches control modes based on its operating status and prioritizes each control mode: when the frequency deviation or voltage deviation is greater than the corresponding preset threshold, it switches to grid support mode, increases the weight of power response-related control functions, and degrades or suspends state-of-charge (SOC) balancing-related control functions, prioritizing power response over SOC balancing; when both the frequency deviation and voltage deviation are less than the corresponding preset dead zone threshold, it switches to economic operation mode, where power allocation is dominated by economic operation mode, and the weight of SOC balancing-related control functions is maintained at the baseline value. The power commands output from each control mode are weighted and merged according to their priority to obtain the total power command.

9. The energy storage coordinated control method according to claim 8, characterized in that, The second-level centralized coordination and control layer also performs the following steps: Based on the state of the frequency deviation and voltage deviation, the power allocation strategy of each energy storage unit is dynamically adjusted: when the frequency deviation or the voltage deviation is greater than the corresponding preset threshold, the power allocation weight of each energy storage unit tends to be average, and the power response speed is given priority; when both the frequency deviation and the voltage deviation are less than the corresponding preset dead zone threshold, the difference in power allocation weight of each energy storage unit increases, and the state of charge balance is given priority. Based on the total power command and the power allocation strategy, the charging and discharging power of each energy storage unit is determined and issued.

10. An energy storage coordination controller, characterized in that, The energy storage coordination controller is applied to the energy storage collaborative control system, and the energy storage coordination controller includes: The communication interface unit includes at least one GOOSE fiber optic network interface for high-speed real-time communication with each energy storage converter, and at least one standard Ethernet interface for communication with the upper-level energy management system or dispatch center. The multi-core heterogeneous data processing unit adopts a multi-core heterogeneous processor, including a first processor core and a second processor core. The first processor core is used to process the instruction transmission and reception and status monitoring of the millisecond-level distributed autonomous control layer, and the second processor core is used to run the optimization algorithm of the second-level centralized coordination control layer. The coordination and computing unit is connected to the multi-core heterogeneous data processing unit and is used to perform dynamic power threshold determination, state of charge coupling allocation, and control mode switching algorithms.

Images