Energy storage cross-station mutual assistance control method for centralized multi-traction substation flexible power supply system

Abstract

This application relates to a cross-station mutual control method for energy storage in a centralized multi-traction substation flexible power supply system. The method includes: collecting system status information through an energy storage cluster collaborative controller, including at least the real-time load power and regenerative power of each traction substation and the real-time state of charge of each energy storage unit; based on the collected system status information, the energy storage cluster collaborative controller periodically solves a global economic scheduling model with the goal of minimizing the total system operating cost, generating and broadcasting dynamic virtual electricity price signals related to the node locations of each centralized converter station; the dynamic virtual electricity price signal reflects the marginal cost-effectiveness of energy injection or absorption at the corresponding node; the local controller of each energy storage unit receives the dynamic virtual electricity price signal and, with the goal of maximizing its own virtual revenue, autonomously decides its charging and discharging power based on a built-in battery loss cost model. This method can improve resource utilization.

Description

Technical Field

[0001] This application relates to the field of electrified railway technology, and in particular to a method, controller, computer-readable storage medium, and computer program product for energy storage cross-station mutual assistance control of a centralized multi-traction station flexible power supply system. Background Technology

[0002] To eliminate phase separation and suppress negative sequence current, existing electrified railways generally adopt in-phase power supply technology. Traditional in-phase power supply devices, such as back-to-back converters or railway power regulators, typically employ a distributed deployment scheme, meaning that one device is independently installed in each traction substation. While this distributed approach solves the basic problems, it has significant limitations: each substation requires reserved installation space, making it difficult to modify existing lines; the energy regulation capabilities of each substation are independent, preventing global optimization, resulting in low regenerative braking energy utilization, high overall equipment investment, and complex redundant configurations.

[0003] Currently, to address the bottlenecks of the aforementioned distributed deployment, the industry has proposed a centralized multi-traction substation flexible power supply system. This solution centrally deploys multiple converter stations around substations or along transmission corridors, supplying power to multiple traction substations along the line through a flexible ring network. This overcomes the spatial limitations of existing lines, reduces retrofitting costs, and improves system energy efficiency. Under this new architecture, to cope with the strong fluctuations in regenerative braking energy and improve power supply reliability, large-capacity energy storage systems are typically configured at each centralized converter station, forming a multi-station energy storage cluster.

[0004] However, when the system includes multiple interconnected centralized converter stations, each equipped with energy storage, traditional control strategies will be unable to coordinate and solve cross-station energy scheduling, leading to resource depletion. Summary of the Invention

[0005] Based on this, it is necessary to provide a method, controller, computer-readable storage medium, and computer program product for energy storage cross-station mutual assistance control of a centralized multi-traction station flexible power supply system that can reduce resource consumption, in order to address the above-mentioned technical problems.

[0006] In a first aspect, this application provides a method for energy storage cross-station mutual assistance control in a centralized multi-traction substation flexible power supply system, including:

[0007] The system status information is collected by the energy storage cluster collaborative controller. The system status information includes at least the real-time load power and regenerative power of each traction substation, and the real-time charge status of each energy storage unit. The centralized multi-traction substation flexible power supply system includes multiple centralized converter stations, a flexible ring network connecting each centralized converter station, energy storage units connected to the DC bus of each centralized converter station, and the energy storage cluster collaborative controller.

[0008] Based on the collected system status information, the energy storage cluster collaborative controller periodically solves the global economic scheduling model with the goal of minimizing the total system operating cost, and generates and broadcasts dynamic virtual electricity price signals related to the location of each centralized converter station node; the dynamic virtual electricity price signal is the node location marginal electricity price that reflects the marginal cost-effectiveness of energy injection or absorption at the corresponding node.

[0009] Each energy storage unit receives the dynamic virtual electricity price signal from its local controller and, with the goal of maximizing its own virtual revenue, autonomously decides its charging and discharging power based on the built-in battery loss cost model.

[0010] In one embodiment, the objective function for maximizing its own virtual gains is:

[0011]

[0012] in, The dynamic virtual electricity price signal, This refers to the charging and discharging power of the energy storage unit; the power is positive during discharging and negative during charging. This is the battery loss cost model; This represents the state of charge of the i-th energy storage unit.

[0013] In one embodiment, the battery loss cost model is a function of the charging and discharging power of the energy storage unit and the state of charge of the energy storage unit, expressed as:

[0014]

[0015] in, This is a cost factor related to the battery's rated capacity and cycle life. Represents the absolute value of charging and discharging power; For time step; This is a stress factor function used to quantify the impact of different SOC levels on battery life;

[0016] The stress factor function is:

[0017]

[0018] in, The SOC operating point is the optimal point for battery life; This is the preset penalty coefficient.

[0019] In one embodiment, the method further includes: in the event of a grid fault at the centralized converter station, controlling the energy storage unit of the faulted station to switch to DC voltage control mode;

[0020] Based on the estimated load of the faulty station area, the transmission limit of the flexible ring network line, and the SOC protection constraint of the faulty station energy storage unit, the optimal support power from the normal station to the faulty station area through the flexible ring network is calculated.

[0021] The control station adjusts its power according to the optimal support power.

[0022] In one embodiment, the formula for calculating the optimal support power is:

[0023]

[0024] in, Estimate the load for the faulty station area; K1 is the preset load factor; The thermal stability limit of the flexible ring network line; The current state of charge of the energy storage unit at the faulty station; Preset protection thresholds for energy storage at fault stations; Its capacity; This represents the expected duration of the failure.

[0025] In one embodiment, controlling the normal station to adjust according to the optimal support power includes:

[0026] The converter of the normal station is controlled to increase the power drawn from the grid to the optimal support power, and the energy storage unit of the normal station is controlled to switch to power follower mode.

[0027] In one embodiment, the method further includes:

[0028] In the event of a failure in the flexible ring network that causes the system to be split into multiple electrical islands, the energy storage units in each electrical island are controlled to switch to DC voltage control mode in order to stabilize the DC bus voltage of the station.

[0029] After the DC bus voltage stabilizes, the single-phase MMC inverters in each electrical island are switched to V / f control mode to establish the AC voltage and frequency reference for that island.

[0030] In one embodiment, the method further includes:

[0031] When the fault in the flexible ring network is eliminated, quasi-synchronous grid connection between islands can be achieved by adjusting the frequency setting value of the single-phase MMC in V / f control mode within the island to be connected.

[0032] In a second aspect, this application also provides a controller, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in the first aspect.

[0033] Thirdly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0034] Fourthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0035] The aforementioned centralized multi-traction substation flexible power supply system's energy storage cross-station mutual assistance control method, controller, computer-readable storage medium, and computer program product collect system status information through an energy storage cluster collaborative controller. This system status information includes at least the real-time load power and regenerative power of each traction substation, and the real-time state of charge of each energy storage unit. The centralized multi-traction substation flexible power supply system includes multiple centralized converter stations, a flexible ring network connecting each centralized converter station, energy storage units connected to the DC buses of each centralized converter station, and an energy storage cluster collaborative controller. Based on the collected system status information, the energy storage cluster collaborative controller periodically solves the global economic scheduling model with the goal of minimizing the total system operating cost, generating and broadcasting dynamic virtual electricity price signals related to the node locations of each centralized converter station. The dynamic virtual electricity price signal reflects the marginal cost-effectiveness of energy injection or absorption at the corresponding node. Each energy storage unit's local controller receives the dynamic virtual electricity price signal and, with the goal of maximizing its own virtual revenue, autonomously decides its charging and discharging power based on a built-in battery loss cost model. This application constructs a "centralized optimization-distributed execution" control system. Global economic scheduling is achieved through a collaborative controller of the energy storage cluster, while local energy storage controllers make autonomous decisions. This ensures the overall economic efficiency of the system operation while reducing the communication pressure and control complexity of the upper-level controllers, thus improving the system's fault tolerance and scalability. Furthermore, a dynamic virtual electricity price (marginal electricity price at node location) is introduced as an economic guidance signal for cross-station energy mutual assistance. This transforms the complex optimization problem involving multiple stations, multiple energy sources, and multiple loads into autonomous optimization decisions by each energy storage unit, achieving a unity of decentralized decision-making and global optimization. This effectively improves the utilization rate of energy storage resources and reduces the total operating cost of the system. Real-time load / regenerative power and energy storage SOC, among other core status information, are collected from traction substations, providing accurate data support for virtual electricity price generation and autonomous energy storage decision-making. This ensures the real-time performance and effectiveness of the control strategy and enables rapid response to changes in the system's power supply and demand. Energy storage units autonomously decide on charging and discharging power without requiring high-frequency power commands from the upper level, significantly improving the system's response speed and adapting to the rapidly fluctuating operating characteristics of electrified railway traction loads and regenerative power. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is an application environment diagram of the energy storage cross-station mutual assistance control method for a centralized multi-traction station flexible power supply system in one embodiment;

[0038] Figure 2 This is a flowchart illustrating the energy storage cross-station mutual assistance control method for a centralized multi-traction station flexible power supply system in one embodiment.

[0039] Figure 3 This is a flowchart illustrating the power adjustment steps in one embodiment;

[0040] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0042] Currently, in-phase power supply devices (such as back-to-back converters and railway power regulators) used to eliminate phase separation and suppress negative sequence commonly employ a distributed deployment scheme, meaning they are independently installed in each traction substation. This model suffers from problems such as space constraints and difficulties in modification, poor economic efficiency and resource waste, fragmented energy pools and low energy efficiency, and complex system redundancy configuration. Figure 1 The image shows a centralized multi-traction station flexible power supply system proposed by the applicant prior to this application. This system addresses the aforementioned bottleneck issues because converter stations are primarily deployed near substations or along transmission corridors, offering advantages such as: adaptability to existing infrastructure upgrades, overcoming spatial limitations, significantly reducing upgrade costs, improving regenerative energy utilization and overall energy efficiency, reducing investment in backup facilities, and providing active grid-side support. In flexible power supply systems based on centralized converter stations, configuring energy storage systems has become a key means to address regenerative braking energy fluctuations and improve power quality and reliability. However, when the system comprises multiple independent or interconnected centralized converter stations, each equipped with energy storage, simple local control strategies become clearly insufficient.

[0043] 1. Lack of an efficient and economical guidance mechanism for cross-site energy dispatch: When station A has excess regenerative energy and station B has high load demand, energy flows naturally through the ring network. However, existing strategies cannot intelligently determine the complex choice of "whether to let station B's energy storage discharge to meet local load, or let station A's regenerative energy be directly sent to station B, or let station A's energy storage charge while station B draws power from the grid." There is a lack of a quantitative economic signal and optimization model to guide the energy flow of the entire network, resulting in operating costs that are not the lowest globally.

[0044] 2. Conflict exists between power support and energy storage protection during faults: When a converter station loses power, its energy storage switches to voltage support mode. At this time, there is a lack of refined power allocation strategies regarding how to obtain power support from normal stations through the ring network. Simple power transmission may lead to overload of the ring network lines or excessive consumption of the faulty station's energy storage, causing it to be depleted before the grid is restored, thus losing its support capacity.

[0045] 3. Lack of Autonomous Control in Subnetworks After Ring Network Split: After a flexible ring network fails and splits, the resulting electrical islands need to quickly establish autonomous operation capabilities. Current technologies do not clearly address: which energy storage unit should act as the "main power source" (i.e., the voltage and frequency establisher), how the other units should coordinate, and how to provide early warning and prevention of potential asynchrony issues between subnetworks.

[0046] 4. The impact of energy storage cluster status on system control strategy is not dynamically quantified: Energy storage SOC and SOH not only affect their own capabilities, but should also dynamically influence the operating boundaries and control strategies of the entire system (such as maximum supportable load and peak regenerative power to be tolerated). Existing methods do not incorporate this dynamic coupling relationship into real-time control decisions.

[0047] Therefore, for multi-station centralized power supply architectures, there is an urgent need for an advanced control method that can achieve collaborative optimization, adaptive reconfiguration, and economical operation of energy storage clusters.

[0048] To address the aforementioned issues, in one exemplary embodiment, such as Figure 2 As shown, this application provides a method for cross-station mutual assistance control of energy storage in a centralized multi-traction substation flexible power supply system. Taking the application of this method to the controller of a centralized multi-traction substation flexible power supply system as an example, the method includes the following steps S202 to S206. Wherein:

[0049] Step S202: Collect system status information through the energy storage cluster collaborative controller. The system status information includes at least the real-time load power and regenerative power of each traction substation, and the real-time charge status of each energy storage unit. The centralized multi-traction substation flexible power supply system includes multiple centralized converter stations, a flexible ring network connecting each centralized converter station, energy storage units connected to the DC bus of each centralized converter station, and an energy storage cluster collaborative controller.

[0050] Among them, the energy storage cluster collaborative controller can refer to the upper control unit of the centralized multi-traction station flexible power supply system. It can be integrated into the system's global energy manager or set up independently. It is the core device for realizing the centralized optimization decision-making of energy storage cross-station mutual assistance. It is responsible for collecting the status information of the entire system, solving the optimization model, issuing control commands, and connecting with the local controllers of each converter station through a high-speed communication network.

[0051] System status information can refer to a collection of various data that characterize the operating status of the entire centralized multi-traction substation flexible power supply system. The core data includes the real-time load power (electrical power required by the traction load) and regenerative power (feedback energy generated by train braking) of each traction substation, as well as the real-time state of charge (SOC, which characterizes the ratio of the remaining power of the energy storage unit to its rated capacity) of each energy storage unit. It can also be extended to include the grid operating status of each converter station, the topology and power flow of the flexible ring network, and the state of health (SOH) of the energy storage unit.

[0052] The centralized multi-traction station flexible power supply system refers to a new type of electrified railway traction power supply architecture that is different from the traditional decentralized in-phase power supply system. Its core consists of multiple centralized converter stations, a flexible ring network connecting each converter station, energy storage units corresponding to the DC bus of each converter station, and an energy storage cluster collaborative controller. The converter stations are deployed around the substation or in the transmission corridor, and have advantages such as high resource utilization and low transformation cost.

[0053] A centralized converter station can refer to a centrally deployed core power conversion device that undertakes AC-DC and DC-AC power conversion functions, provides stable power supply to traction substations, and realizes power interaction with flexible ring networks and energy storage units. It is the core node of the centralized power supply architecture.

[0054] A flexible ring network can refer to a flexible power network that connects various centralized converter stations. It has flexible power transmission and topology adjustment capabilities and serves as the physical carrier for realizing power exchange and cross-station power support among converter stations.

[0055] An energy storage unit can refer to an electrical energy storage device connected to the DC bus of a centralized converter station. It can realize the charging, storage and discharging of electrical energy, and is used to absorb the regenerative braking energy of trains, smooth load fluctuations, and provide voltage support and power supplementation during faults.

[0056] For example, the energy storage cluster collaborative controller collects the status information of the entire centralized multi-traction substation flexible power supply system in real time through a high-speed communication network. It collects at least the real-time load power and regenerative power of each traction substation, as well as the real-time charge status of each energy storage unit. For example, the collection cycle can be matched with the scheduling cycle (e.g., 5 minutes). At the same time, it can also collect information such as the SOH of energy storage units, the power flow of the flexible ring network, and the grid status of the converter station, providing comprehensive data support for subsequent optimization decisions.

[0057] Step S204: Based on the collected system status information, the energy storage cluster collaborative controller periodically solves the global economic scheduling model with the goal of minimizing the total system operating cost, and generates and broadcasts dynamic virtual electricity price signals related to the node locations of each centralized converter station; wherein, the dynamic virtual electricity price signal is the node location marginal electricity price that reflects the marginal cost-effectiveness of energy injection or absorption at the corresponding node.

[0058] Among them, the global economic dispatch model can refer to a mathematical optimization model constructed with the goal of minimizing the total operating cost of the entire centralized multi-traction station flexible power supply system. The constraints include flexible ring network power flow constraints, energy storage unit charging and discharging power constraints, converter station power transmission constraints, etc., and it is the core model for realizing the global economic operation of the system.

[0059] The dynamic virtual electricity price signal can refer to the electricity price reference signal that is generated by the energy storage cluster collaborative controller periodically solving the global economic dispatch model and changes dynamically with the system operating status. Specifically, it is the marginal electricity price at the node location and is the core economic signal that guides each energy storage unit to make autonomous decisions on charging and discharging behavior.

[0060] Marginal Price at Node Location (LMP): This is a price signal that reflects the marginal cost-effectiveness of electricity injected into or absorbed at a certain centralized converter station node. When there is excess regenerative energy at the node, the LMP is low (or even negative), and when the load demand at the node is high, the LMP is high. It can accurately characterize the power supply and demand status of each node.

[0061] For example, the energy storage cluster collaborative controller aims to minimize the total system operating cost. Combining the collected system status information, it periodically solves the global economic scheduling model (e.g., the scheduling period is set to 5 minutes, and the model constraints include flexible ring network power flow constraints and energy storage charging and discharging power limit constraints). After the solution is completed, it generates a dynamic virtual electricity price signal (node ​​location marginal electricity price) that corresponds one-to-one with the location of each centralized converter station node, and broadcasts the signal to the local controllers of all energy storage units through a high-speed communication network. For example, converter station nodes with excess regenerative energy generate a low LMP signal, while nodes with tight loads generate a high LMP signal.

[0062] In step S206, the local controller of each energy storage unit receives the dynamic virtual electricity price signal and, with the goal of maximizing its own virtual revenue, autonomously decides its charging and discharging power based on the built-in battery loss cost model.

[0063] Among them, the local controller of the energy storage unit can refer to the underlying control device deployed on the side of each energy storage unit, which communicates with the energy storage cluster co-controller, is responsible for receiving dynamic virtual electricity price signals, autonomously deciding the charging and discharging power of the energy storage unit, and executing various mode switching and power adjustment commands issued by the upper layer.

[0064] Virtual revenue refers to the economic revenue of charging and discharging of an energy storage unit based on the marginal electricity price at the node location, after deducting the cost of battery loss. It is the core optimization objective for the autonomous decision-making of the local energy storage controller.

[0065] A battery loss cost model can be a mathematical model that quantifies the impact of energy storage unit charging and discharging power and state of charge on battery life loss. It transforms battery life loss into economic cost, achieving a synergistic trade-off between economic benefits and battery life.

[0066] For example, the local controller of each energy storage unit receives the marginal electricity price signal of the corresponding node location broadcast by the energy storage cluster coordinating controller. With the core objective of maximizing its own virtual revenue within the rolling time window, and combined with the built-in battery loss cost model, it autonomously calculates and decides the charging and discharging power of its own energy storage unit without the need for the upper-level controller to issue specific power commands. For example, when the marginal electricity price of the node location is negative, the local controller decides to charge the energy storage unit to absorb regenerative energy; when the marginal electricity price of the node location is high, the local controller decides to discharge the energy storage unit to supplement the local load.

[0067] In the aforementioned energy storage cross-station mutual assistance control method for a centralized multi-traction substation flexible power supply system, system status information is collected through an energy storage cluster collaborative controller. This system status information includes at least the real-time load power and regenerative power of each traction substation, and the real-time charge status of each energy storage unit. The centralized multi-traction substation flexible power supply system includes multiple centralized converter stations, a flexible ring network connecting each centralized converter station, energy storage units connected to the DC buses of each centralized converter station, and an energy storage cluster collaborative controller. Based on the collected system status information, the energy storage cluster collaborative controller periodically solves the global economic scheduling model with the goal of minimizing the total system operating cost, generating and broadcasting dynamic virtual electricity price signals related to the node locations of each centralized converter station. The dynamic virtual electricity price signal reflects the marginal cost-benefit of energy injection or absorption at the corresponding node. Each energy storage unit's local controller receives the dynamic virtual electricity price signal and, with the goal of maximizing its own virtual revenue, autonomously decides its charging and discharging power based on a built-in battery loss cost model. This application constructs a "centralized optimization-distributed execution" control system. Global economic scheduling is achieved through a collaborative controller of the energy storage cluster, while local energy storage controllers make autonomous decisions. This ensures the overall economic efficiency of the system operation while reducing the communication pressure and control complexity of the upper-level controllers, thus improving the system's fault tolerance and scalability. Furthermore, a dynamic virtual electricity price (marginal electricity price at node location) is introduced as an economic guidance signal for cross-station energy mutual assistance. This transforms the complex optimization problem involving multiple stations, multiple energy sources, and multiple loads into autonomous optimization decisions by each energy storage unit, achieving a unity of decentralized decision-making and global optimization. This effectively improves the utilization rate of energy storage resources and reduces the total operating cost of the system. Real-time load / regenerative power and energy storage SOC, among other core status information, are collected from traction substations, providing accurate data support for virtual electricity price generation and autonomous energy storage decision-making. This ensures the real-time performance and effectiveness of the control strategy and enables rapid response to changes in the system's power supply and demand. Energy storage units autonomously decide on charging and discharging power without requiring high-frequency power commands from the upper level, significantly improving the system's response speed and adapting to the rapidly fluctuating operating characteristics of electrified railway traction loads and regenerative power.

[0068] In an exemplary embodiment, the objective function for maximizing one's own virtual gains is:

[0069]

[0070] in, For dynamic virtual electricity price signals, This refers to the charging and discharging power of the energy storage unit; the power is positive during discharging and negative during charging. For battery loss cost model; This represents the state of charge of the i-th energy storage unit.

[0071] The objective function can be used as a mathematical expression to characterize the optimization objective of the local controller of the energy storage unit.

[0072] For example, the local controller of each energy storage unit transforms maximizing its own virtual revenue into the aforementioned specific mathematical objective function. Using this objective function as the core optimization criterion, and combining it with the battery loss cost model, its own state of charge constraints, and charge / discharge power limits, the optimal charge / discharge power of the energy storage unit at each moment within the rolling time window is obtained by solving this objective function. The rolling time window can be the time range within which the local controller of the energy storage unit makes charge / discharge power decisions. Charge / discharge power refers to the charging or discharging electrical energy of the energy storage unit per unit time, defined as positive for discharging and negative for charging, quantitatively representing the direction and magnitude of electrical energy interaction between the energy storage unit and the system.

[0073] In this embodiment, the virtual revenue optimization objective of the energy storage unit is transformed into a specific mathematical objective function, realizing the quantitative calculation of the optimization objective. This provides a clear mathematical basis for the autonomous decision-making of the local energy storage controller, improving the accuracy and scientific nature of the decision-making. The objective function simultaneously incorporates the economic benefits of the marginal electricity price at the node location and the battery depreciation cost, achieving a quantitative trade-off between economic benefits and battery life loss. This avoids the energy storage unit from overcharging and discharging in pursuit of short-term economic benefits, thus balancing system economy and energy storage asset lifespan.

[0074] In one exemplary embodiment, the battery loss cost model is a function of the charging and discharging power of the energy storage unit and the state of charge of the energy storage unit, expressed as:

[0075]

[0076] in, This is a cost factor related to the battery's rated capacity and cycle life. Represents the absolute value of charging and discharging power; For time step; This is a stress factor function used to quantify the impact of different SOC levels on battery life;

[0077] The stress factor function is:

[0078]

[0079] in, The SOC operating point is the optimal point for battery life; This is the preset penalty coefficient.

[0080] For example, the local controller of each energy storage unit concretizes the built-in battery loss cost model into the mathematical expression above, and also concretizes the stress factor function into the expression above. When calculating the battery loss cost, the cost coefficient is first calculated based on the battery's rated capacity and cycle life. Determine the preset penalty coefficient and optimal SOC operating point Then, the charging and discharging power of this energy storage unit is obtained in real time. Time step , (Real-time SOC) The stress factor and battery loss cost are calculated sequentially, and finally substituted into the aforementioned objective function to solve for the optimal charge and discharge power.

[0081] In this embodiment, the battery loss cost model is concretized into a quantitative mathematical expression, enabling precise quantitative calculation of battery life loss. This allows the energy storage unit to accurately balance economic benefits and battery loss during autonomous decision-making, improving the intelligence level of battery life management. A stress factor function is introduced to quantify the impact of different SOC levels on battery life. Cost penalties are imposed on charging and discharging behaviors that deviate from the optimal SOC operating point, guiding the energy storage unit to operate near the optimal SOC, effectively delaying battery performance degradation and extending the overall service life of the energy storage unit.

[0082] In one exemplary embodiment, such as Figure 3 As shown, the energy storage cross-station mutual assistance control method for a centralized multi-traction substation flexible power supply system may further include the following steps:

[0083] Step S302: In the event of a grid fault at a centralized converter station, the energy storage unit of the faulted station is switched to DC voltage control mode.

[0084] Step S304: Based on the estimated load of the faulty station area, the transmission limit of the flexible ring network line, and the SOC protection constraint of the faulty station energy storage unit, calculate the optimal support power from the normal station to the faulty station area through the flexible ring network.

[0085] Step S306: Control the normal station to adjust according to the optimal support power.

[0086] Among these, grid faults can refer to voltage drops, power outages, or other faults occurring in the AC grid connected to a centralized converter station, preventing the converter station from obtaining normal power from the grid. DC voltage control mode can refer to an operating control mode of the energy storage unit. In this mode, the energy storage unit acts as a DC voltage source, using its rated DC voltage as a setpoint to stabilize the DC bus voltage of the converter station it is connected to, providing voltage support for the system. This is the core support mode for the energy storage unit during faults. The faulty station area can refer to the centralized converter station where the grid fault occurred and all traction substations and loads within its power supply range. Estimated load can refer to the estimated power consumption required in the faulty station area in real time, based on pre-fault load data and load characteristics. The transmission limit of a flexible ring network line can refer to the maximum electrical power that each line in the flexible ring network can transmit per unit time under constraints such as thermal stability and voltage stability; this is the thermal stability limit and a physical constraint for cross-station power support. State of Charge (SOC) protection constraints refer to the state-of-charge limits set to protect energy storage units from over-discharge, i.e., the SOC protection threshold. During a fault, the SOC of the energy storage unit must not fall below this threshold to prevent it from losing its voltage support capability due to over-discharge. Optimal support power refers to the optimal power value that a normal substation can transmit to the faulty substation area through a flexible ring network, balancing load demand, ring network transmission capacity, and energy storage protection. It is the core control parameter for cross-substation power support. A normal substation refers to a centralized converter station in a centralized multi-traction substation flexible power supply system where the connected AC grid is in normal operating condition.

[0087] For example, fault detection and energy storage mode switching: The energy storage cluster collaborative controller monitors the grid operation status of each centralized converter station in real time. When a grid fault is detected at a converter station, it immediately issues a control command, ordering the energy storage units at the faulted station to switch from the normal charging and discharging decision mode to the DC voltage control mode, stabilizing the DC bus voltage of the faulted station and providing basic voltage support for the faulted station area. Optimal support power calculation: Based on the real-time collected system status information, the energy storage cluster collaborative controller obtains parameters such as the estimated load of the faulted station area, the transmission limit of the flexible ring network line, the SOC protection constraints of the faulted station's energy storage units (current SOC value, SOC protection threshold, energy storage capacity), and the expected fault duration. It then constructs an optimization model constrained by the faulted station's energy storage protection, ring network line safety, and meeting basic load requirements to calculate the optimal support power that normal stations can deliver to the faulted station area through the flexible ring network. Normal station power adjustment control: The energy storage cluster collaborative controller sends control commands to normal stations, instructing them to adjust their power according to the calculated optimal support power, thereby achieving cross-station power support, while ensuring that the SOC of the faulty station's energy storage is not lower than the protection threshold and that the flexible ring network line does not exceed the transmission limit.

[0088] In this embodiment, a dedicated cross-site power support strategy is designed for grid faults, filling the gap in fault power support under traditional local control strategies. This strategy enables energy storage coordination between normal and faulty stations during a fault, significantly improving the system's fault ride-through resilience and power supply reliability. Upon fault occurrence, the energy storage at the faulty station is immediately switched to DC voltage control mode to quickly stabilize the DC bus voltage at the faulty station, preventing DC bus voltage instability from causing system collapse. This buys time for cross-site power support and improves system stability in the initial stage of a fault. The calculation of the optimal support power takes into account the load demand of the faulty station, the transmission limits of the flexible ring network, and the SOC protection constraints of the energy storage. This ensures basic load power supply to the faulty station area while avoiding ring network line overload and excessive discharge of the faulty station's energy storage, achieving coordinated unification of power support and equipment protection. The energy storage cluster collaborative controller uniformly completes fault detection, mode switching, power calculation, and command issuance, realizing centralized collaborative control under fault conditions and improving the efficiency and accuracy of fault handling.

[0089] In one exemplary embodiment, the formula for calculating the optimal support power is:

[0090]

[0091] in, Estimate the load for the faulty station area; K1 is the preset load factor; The thermal stability limit of the flexible ring network line; The current state of charge of the energy storage unit at the faulty station; Preset protection thresholds for energy storage at fault stations; Its capacity; This represents the expected duration of the failure.

[0092] The load-bearing factor characterizes the maximum proportion of the estimated load that the optimal support power can bear in the faulty substation area. It can be used to limit the maximum amplitude of power support, avoiding over-support that could lead to overload of the ring network or abnormal operation of normal substations. The thermal stability limit characterizes the maximum electrical power that a flexible ring network line can transmit under thermal stability constraints; exceeding this limit will cause the line to overheat and be damaged. The preset protection threshold for energy storage at the faulty substation refers to the state-of-charge limit set to protect the energy storage units at the faulty substation during a fault, preventing the energy storage units from over-discharging and losing voltage support and recovery capabilities. The expected fault duration is an estimate of the fault duration based on the type of grid fault and fault handling characteristics.

[0093] In this embodiment, the calculation of the optimal support power is transformed into a specific quantitative mathematical formula, which realizes the accurate and rapid calculation of cross-site support power under fault conditions. The formula adopts the constraint method of taking the minimum value, while taking into account the three core constraints of load sharing, ring network thermal stability, and energy storage SOC protection. From a mathematical perspective, it ensures the safety and rationality of the optimal support power and completely solves the contradiction between power support and equipment protection in the traditional strategy.

[0094] In one exemplary embodiment, controlling a normal station to adjust according to the optimal support power includes: controlling the converter of the normal station to increase the power drawn from the grid to the optimal support power, and controlling the energy storage unit of the normal station to switch to power follower mode.

[0095] Here, "power extraction" can refer to the electrical power that the converter of a normal station obtains from the connected AC power grid. "Power following mode" can refer to an operating control mode of the energy storage unit. In this mode, the energy storage unit no longer makes its own decisions on charging and discharging power, but adjusts its own charging and discharging power in real time according to the power fluctuations of the system to smooth out system power fluctuations and ensure the stability of system power.

[0096] For example, after calculating the optimal support power, the energy storage cluster coordinating controller issues specific power adjustment commands to normal stations, which may include: Normal station converter power intake adjustment: instructing the converter of the normal station to increase the power intake from the AC grid, with the increase equal to the optimal support power, so that the normal station has the electrical energy basis to deliver support power to the faulty station area. Normal station energy storage mode switching: instructing the energy storage units of the normal station to switch from the normal autonomous charging and discharging decision-making mode to the power following mode, to smooth out system power fluctuations caused by increased power intake from the converter and load fluctuations in the faulty station area in real time, ensuring stable power operation of the normal station itself and the flexible ring network.

[0097] In this embodiment, the converter at the normal station precisely increases the power extraction capacity to an amount equivalent to the optimal support power, ensuring the support power supply to the faulty station area and meeting its basic load requirements. The energy storage at the normal station switches to power-following mode, effectively smoothing system power fluctuations during power support and preventing power fluctuations from causing abnormal operation of the normal station's power grid and flexible ring network, thus improving system stability during fault support. The coordinated adjustment of the converter and energy storage at the normal station achieves a balance between power support and power stability, ensuring power supply to the faulty station while also ensuring the safe operation of the normal station and the flexible ring network.

[0098] In an exemplary embodiment, the energy storage cross-station mutual assistance control method of a centralized multi-traction substation flexible power supply system further includes: in the event of a fault in the flexible ring network that causes the system to be split into multiple electrical islands, controlling the energy storage units in each electrical island to switch to DC voltage control mode to stabilize the DC bus voltage of the substation; after the DC bus voltage stabilizes, controlling the single-phase MMC inverters in each electrical island to switch to V / f control mode to establish the AC voltage and frequency reference of the island.

[0099] In this context, a flexible ring network failure can refer to a short circuit, open circuit, or other fault in the backbone lines of the flexible ring network, leading to a disruption of the network's topology and the system being isolated into multiple independent electrical zones by protection devices. An electrical island refers to an independent power supply zone formed after a flexible ring network failure, where the systems are isolated and have no electrical connection to each other. Each island contains at least one centralized converter station, energy storage unit, and traction load, and must possess the capability for autonomous and stable operation. The DC bus voltage can be the bus voltage on the DC side of the centralized converter station, serving as the core voltage reference for energy exchange between the converter station, energy storage unit, and flexible ring network; its stability directly determines the normal operation of the system. A single-phase MMC inverter can be a single-phase form of a modular multilevel converter (MMC), the core power conversion device of the centralized converter station, undertaking DC-AC power conversion functions and enabling precise voltage and frequency control. V / f control mode refers to voltage / frequency control mode, a core operating control mode of the inverter. In this mode, the inverter uses rated voltage and rated frequency as setpoints to provide a stable voltage and frequency reference for the AC side, which is the core control mode for autonomous operation of the electrical island. The AC voltage and frequency references refer to the core reference parameters for the operation of the AC power grid within the electrical island. These parameters are constructed by the single-phase MMC inverter in V / f control mode and are the basis for the normal operation of all loads and equipment within the island.

[0100] For example, this embodiment is a ring network fault propagation strategy under normal control, including ring network fault detection and energy storage DC voltage control: The energy storage cluster collaborative controller monitors the topology and operating status of the flexible ring network in real time. When a fault is detected in the flexible ring network, causing the system to be split into multiple electrical islands, control commands are immediately issued to the energy storage units in each electrical island, ordering all energy storage units in the islands to switch to DC voltage control mode. Using the rated DC voltage as the set value, the DC bus voltage of each station is quickly stabilized, providing a stable DC voltage reference for the autonomous operation of the islands. It also includes island AC voltage and frequency reconstruction: The energy storage cluster collaborative controller monitors the DC bus voltage status in each electrical island in real time. When it is confirmed that the DC bus voltage of a certain island is stable, control commands are immediately issued to the single-phase MMC inverter in that island, ordering it to switch to V / f control mode. Using the system's rated AC voltage and rated frequency as set values, the AC voltage and frequency reference of the electrical island is constructed, realizing the stable operation of the island's AC side.

[0101] In this embodiment, a specialized self-organizing strategy for energy storage sub-clusters is designed for flexible ring network fault isolation scenarios. This fills the gap in autonomous sub-network control after ring network splitting in traditional technologies, enabling rapid, autonomous, and stable operation of electrical islands and significantly improving the system's power supply resilience. Following the "DC voltage priority" control principle, the DC bus voltage is first stabilized through energy storage units, and then AC voltage and frequency references are constructed. This clarifies the control timing and responsibilities of equipment under fault conditions, ensuring the orderliness and stability of islanded autonomy.

[0102] In an exemplary embodiment, the energy storage cross-station mutual assistance control method of the centralized multi-traction station flexible power supply system further includes: in the case of elimination of flexible ring network faults, quasi-synchronous grid connection between islands is achieved by adjusting the frequency setting value of the single-phase MMC in V / f control mode within the island to be connected.

[0103] Fault elimination can refer to the repair of faulty lines in a flexible ring network and the reset of protection devices, creating conditions for restoring the ring network topology and enabling parallel operation of individual electrical islands. Islanding systems awaiting parallel operation can refer to the need to reconnect the flexible ring network into a single, independent electrical island after fault elimination. Quasi-synchronous grid connection refers to a grid connection method where, before two AC power grids are put into parallel operation, the voltage and frequency of one grid are adjusted to match (or deviate within allowable limits) the voltage, frequency, and phase of the other grid, thus completing the parallel operation. This is the core safe grid connection method for parallel operation of power system islands.

[0104] For example, the energy storage cluster collaborative controller monitors the fault status of the flexible ring network in real time. Once the ring network fault is detected and cleared, it continuously monitors the AC voltage, frequency, phase, and other electrical quantities of each island to be paralleled, obtaining the electrical quantity deviations between the islands. Based on the monitored electrical quantity deviations of each island to be paralleled, it sends frequency adjustment commands to the single-phase MMC inverters in V / f control mode within one (or more) of the islands to be paralleled, fine-tuning their frequency setpoints and gradually reducing the frequency and phase deviations between the islands. When the AC voltage, frequency, and phase deviations between the islands are all within the preset allowable range, it controls the switching equipment of the flexible ring network to close, achieving quasi-synchronous grid connection between the islands and restoring the normal ring network operation state of the entire centralized multi-traction substation flexible power supply system.

[0105] In this embodiment, a quasi-synchronous grid connection strategy for islanded systems after ring network fault clearance is designed. This achieves a smooth transition from autonomous operation to normal ring network operation for electrical islanded systems, completely solving the problem of sub-network restoration after ring network splitting in traditional technologies and improving the overall system recovery capability. Quasi-synchronous grid connection is achieved by fine-tuning the frequency setpoint of the single-phase MMC inverter in the island to be paralleled. The adjustment method is precise and stable, avoiding inrush current and power fluctuations during the grid connection process, and ensuring the safety and stability of the grid connection process.

[0106] In one exemplary embodiment, this application also provides a method for cross-station mutual assistance control of energy storage in a centralized multi-traction substation flexible power supply system. This embodiment aims to provide an innovative collaborative control scheme for energy storage clusters to address the unique control challenges of centralized architectures. Specific objectives include:

[0107] A cross-site energy mutual assistance guidance mechanism based on dynamic virtual electricity price is proposed, which transforms the complex optimization problem of multiple stations, multiple energy sources and multiple loads into autonomous optimization decision-making by each energy storage unit based on the local "virtual electricity price" signal, thereby achieving the unity of decentralized decision-making and global optimization.

[0108] A hierarchical and collaborative fault ride-through strategy is designed to clarify the collaborative logic of "power support" and "capacity protection" between normal and faulty station energy storage under grid faults, and to establish a dynamic power allocation model based on ring network transmission capacity constraints and faulty station energy storage SOC state.

[0109] A "DC voltage priority" collaborative self-organizing method is established under flexible ring network faults. It is determined that after the ring network is disconnected, the energy storage immediately switches to DC voltage control to stabilize the operating base point, and the single-phase MMC then switches to V / f control to construct the timing and logic of AC voltage, so as to achieve islanded stability, autonomy and recoverability.

[0110] A decision-making framework is constructed that dynamically couples the state of the energy storage cluster with the system operation mode, making state variables such as energy storage SOC and SOH key parameters for real-time adjustment of system power limits and control mode switching thresholds.

[0111] The core of this embodiment lies in a hybrid control system of "centralized optimization and distributed execution," and its innovation is reflected in the following specific strategy details:

[0112] 1. System Overall Architecture and Energy Storage Cluster Co-controller (ESCC). The Energy Storage Cluster Co-controller (ESCC), as the upper-level control unit, can be physically integrated into the system's Global Energy Manager (GEM) or set up independently. It is connected to the local controllers of each converter station (including the local control of the energy storage converter) via a high-speed communication network.

[0113] ESCC input information includes:

[0114] System status: AC grid status of each converter station (normal / fault), flexible ring network topology and power flow, and current system operating mode (normal / islanding / fault reconfiguration).

[0115] Load and power generation information: real-time traction / regenerative power and forecast data for each traction station.

[0116] Energy storage unit status: Real-time SOC, SOH (health status), rated power and capacity, current operating mode and adjustable range of each energy storage unit.

[0117] ESCC output commands are high-level settings issued to the local controller of each energy storage converter, including: active power commands ( DC voltage command ( (or the command to switch running modes.)

[0118] 2. Routine optimization scheduling strategy based on dynamic virtual electricity price.

[0119] This strategy aims to achieve global economic guidance of energy storage charging and discharging behavior without relying on high-frequency commands from a central controller.

[0120] Virtual electricity price generation: The Central Optimizer (ESCC) solves the global economic scheduling model once every scheduling cycle (e.g., every 5 minutes). The model aims to minimize total cost and includes network power flow constraints. After solving, the ESCC does not directly issue power commands, but instead calculates and broadcasts a unified "basic virtual electricity price" across the entire network. And the "location-based marginal virtual electricity price" for each converter station node. . This reflects the marginal cost-effectiveness of energy injection or absorption at this node.

[0121] Energy storage autonomous decision-making model: Each energy storage local controller receives... Its autonomous decision-making goal is to maximize its "virtual gains" within a rolling time window:

[0122]

[0123] in, This represents the charging and discharging power (discharging is positive). The battery loss cost function calculated locally is specifically expressed as follows:

[0124]

[0125] in, It represents the absolute value of the charging and discharging power. The time step (in hours) for optimization or control. For example, 0.25h in a 15-minute rolling optimization.

[0126]

[0127] in, The total lifespan of a battery under standard reference conditions (typically a specific SOC window, charge / discharge rate, and temperature) is equivalent to the number of full cycles. Rated energy capacity of energy storage (kWh). The SOC stress factor function is used to quantify the accelerating effect of charging and discharging on battery life at different SOC levels (representing the penalty imposed by the control system on deviations from 50% SOC). It is an innovative manifestation of the intelligent battery life management strategy in this application.

[0128]

[0129] in, The optimal SOC (State of Charge) for battery life is typically set at 0.5% (50%). This represents the ideal state with minimal electrochemical stress on the battery.

[0130] Strategic Innovation: This method decomposes the complex central optimization problem into parallel local optimization problems for each energy storage unit. The LMP signal acts as a "conductor": when a station has abundant regenerative energy resulting in a negative LMP, the energy storage at that station tends to charge; when a station has a heavy load and a high LMP, the energy storage at that station tends to discharge. Simultaneously, the loss cost function... This avoids unnecessary actions by the energy storage system when electricity price differences are minor, thus protecting battery life. The central controller only needs to periodically calculate and broadcast price signals, resulting in low communication pressure and high system reliability.

[0131] 3. Hierarchical and coordinated support strategy under power grid faults.

[0132] refer to Figure 1When a grid fault (voltage below 0.2 pu) is detected at converter station #i, ESCC immediately activates this strategy.

[0133] Phase 1: Rapid Isolation and Mode Switching (0-100ms). The ESCC commands the faulty station's energy storage (ES_i) to switch to DC voltage source mode, setting the voltage to the rated value. Simultaneously, it commands the faulty station's three-phase MMC to be locked, and the single-phase MMC to switch to constant power (P / Q) mode, with the power reference value set to 80% of the instantaneous active power before the fault, in order to maintain the basic stability of the ring network power.

[0134] Phase Two: Cross-Site Power Support and Capacity Protection (after 100ms). ESCC initiates a rapid optimization model to calculate the optimal support power from a normal station (e.g., station #j) through the ring network to the faulty station's area. .

[0135]

[0136] in, Estimate the load for the faulty station area; K1 is the load factor (initially 0.8, adjustable as the fault duration increases); This represents the thermal stability limit of the relevant ring network lines; The protection threshold for energy storage at the faulty station (e.g., 30%). Its capacity; This represents the expected duration of the failure.

[0137] ESCC issues a command to normal station #j: its converter should increase power draw from the grid. Its energy storage (ES_j) enters power follower mode to smooth out power changes caused by increased power consumption and load fluctuations.

[0138] Phase 3: Long-term islanding operation and load regulation. If the fault persists, when the SOC of ES_i approaches... At that time, ESCC has the authority to command station #j through the scheduling system to provide more support (closer to) via the ring network. Alternatively, it can send tiered unloading commands to the faulty area to protect the power supply to the core load.

[0139] 4. Self-organizing strategy of energy storage sub-cluster under N-1 fault in flexible ring network.

[0140] When a fault in the ring network backbone causes the network to be isolated into multiple electrical islands, the ESCC executes a collaborative self-organizing strategy based on the principle of "prioritizing DC voltage stability".

[0141] Phase 1: Fault Identification and DC Anchoring (0-200ms). ESCC identifies network disconnection and immediately broadcasts the "Islanding Operation - DC Voltage Priority" command to each isolated energy storage unit. Each isolated energy storage unit (e.g., ES_i) unconditionally switches to DC voltage control mode, using the rated value (e.g., ...). With ±25kV as the setpoint, it quickly stabilizes the local DC bus voltage, providing a stable operating platform for all converters.

[0142] Phase 2: AC side reconfiguration (200-500ms). After confirming that the DC bus voltage is stable, the ESCC commands the single-phase MMC inverters in each island to switch to V / f control mode, setting their output voltage and frequency to the rated values ​​(e.g., 110kV / 50Hz), thereby establishing a stable 110kV AC voltage within the island.

[0143] Phase 3: Coordinated Operation and Recovery Preparation. At this stage, a stable operating paradigm is established within the island: Energy Storage (DC voltage control) → Single-phase MMC (V / f control) → Load. If the island's power grid is normal, its three-phase MMC can operate in PQ mode to assist in power supply. The ESCC monitors the electrical quantities at each island boundary. When the fault is cleared and the ring network needs to be restored, quasi-synchronous grid connection is achieved by fine-tuning the V / f controller frequency setting of the single-phase MMC within the island to be connected.

[0144] Compared with traditional technologies and simple upper-level command issuance schemes, the technical solution in this embodiment has the following outstanding innovations and advantages:

[0145] It achieves true distributed optimization and economic improvement: the "virtual electricity price" mechanism enables each energy storage unit to have "economic intelligence", and to optimize autonomously under the guidance of the central government. The system has strong fault tolerance and scalability, which can reduce the overall system operating cost.

[0146] The fault ride-through strategy combines robustness and precision, greatly enhancing resilience: the hierarchical strategy clarifies the complete chain from millisecond-level mode switching and second-level power optimization to minute-level load management, and explicitly constrains the support power through mathematical formulas. It utilizes mutual assistance capabilities while strictly protecting critical energy storage capacity, thus extending the power supply guarantee time for critical loads.

[0147] The system has overcome the challenges of stable autonomy and safe recovery after ring network disconnection: the "DC voltage priority" self-organizing strategy clarifies the equipment control sequence and responsibility under fault conditions. Through the synergy of energy storage to stabilize DC and single-phase MMC to construct AC, the system can quickly form multiple stable autonomous power supply islands under the guidance of the central controller, and can achieve smooth reconstruction by adjusting the AC side V / f source, which greatly improves the resilience of power supply.

[0148] By deeply embedding the energy storage status into the control closed loop, asset lifespan is extended: SOC and SOH are transformed from passive monitoring parameters into key decision variables that actively influence system operating boundaries and control commands. Through intelligent SOC balancing and charging / discharging strategy optimization, the overall performance degradation of the cluster is slowed down.

[0149] In some embodiments, the following example further illustrates the system configuration, which includes two centralized converter stations (C1, C2) and associated energy storage (ES1, ES2) to power four traction substations (TS1-TS4). System parameters: C1 / C2 capacity 30MVA, DC bus voltage ±25kV; ES1 / ES2 capacity 15MW / 30MWh.

[0150] Scenario 1: Routine optimization operation.

[0151] ESCC calculates LMP every 5 minutes. Assume that at a certain moment, due to TS1 braking, LMP_C1 of node C1 is -0.05 yuan / kWh (encouraging charging); and due to heavy load on node C2, LMP_C2 is +0.10 yuan / kWh (encouraging discharging).

[0152] Based on LMP_C1 and its own SOC=70%, the ES1 local controller calculates that the charging benefits outweigh the loss costs and decides to charge at 5MW power.

[0153] Based on LMP_C2 and SOC=40%, ES2 calculated that the discharge benefit was considerable and decided to discharge at 8MW power to support the ring network.

[0154] Result: Renewable energy is partially absorbed, and high-priced node loads are supported by energy storage discharge, reducing the grid's high electricity purchase cost.

[0155] Scenario 2: C1 station power grid failure (lasting 10 seconds).

[0156] (1) t<100ms: ESCC detects a fault, instructs ES1 to switch to DC voltage control (±25kV), and instructs C1 single-phase MMC to switch to constant power mode (output -20MW).

[0157] (2) t=100ms: ESCC calculates the support power. Given: =25MW, =30MW, =30%, ES1 current SOC=65%, capacity C1=30MWh, assuming =10s. Then we have:

[0158]

[0159] (3) t=100-200ms: ESCC instructs station C2 to draw 20MW more power from the grid and instructs ES2 to enter power smoothing mode.

[0160] Results: The load of station C1 was mainly supplied by the 20MW power of station C2 through the ring network. ES1 only bore a small amount of the difference and loss, and its SOC was effectively maintained, which prepared it for grid restoration.

[0161] Scenario 3: Fault disconnection of the backbone line of the ring network between C1 and C2.

[0162] (1) t<50ms: Protection trips, and the ring network is split into island A (C1, TS1, TS2) and island B (C2, TS3, TS4). ESCC identifies topology changes.

[0163] (2) t=50-200ms: ESCC broadcast command. ES1 and ES2 immediately switch to DC voltage control mode to stabilize the DC bus of C1 and C2 stations respectively.

[0164] (3) t=200-500ms: ESCC command C1 single-phase MMC in island A switches to V / f control (110kV / 50Hz), and C2 single-phase MMC in island B also switches to V / f control (110kV / 50Hz).

[0165] (4) t>500ms: The two islands enter a stable autonomous state. ESCC monitors the frequency difference at the boundary between the two islands to prepare for grid restoration.

[0166] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0167] In one exemplary embodiment, a controller is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4As shown, the controller includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a cross-station mutual assistance control method for energy storage in a centralized multi-traction station flexible power supply system.

[0168] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0169] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0170] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0171] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0172] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0173] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0174] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for energy storage cross-station mutual aid control of a centralized multi-drag station flexible power supply system, characterized in that, The method includes: The system status information is collected by the energy storage cluster collaborative controller. The system status information includes at least the real-time load power and regenerative power of each traction substation, and the real-time charge status of each energy storage unit. The centralized multi-traction substation flexible power supply system includes multiple centralized converter stations, a flexible ring network connecting each centralized converter station, energy storage units connected to the DC bus of each centralized converter station, and the energy storage cluster collaborative controller. Based on the collected system status information, the energy storage cluster collaborative controller periodically solves the global economic scheduling model with the goal of minimizing the total system operating cost, and generates and broadcasts dynamic virtual electricity price signals related to the location of each centralized converter station node; the dynamic virtual electricity price signal is the node location marginal electricity price that reflects the marginal cost-effectiveness of energy injection or absorption at the corresponding node. Each energy storage unit receives the dynamic virtual electricity price signal from its local controller and, with the goal of maximizing its own virtual revenue, autonomously decides its charging and discharging power based on the built-in battery loss cost model.

2. The method of claim 1, wherein, The objective function for maximizing one's own virtual gains is: wherein, is the dynamic virtual price signal, is the charge-discharge power of the energy storage unit, positive for discharging and negative for charging; is the battery wear cost model; is the state of charge of the i-th energy storage unit.

3. The method of claim 2, wherein, The battery loss cost model is a function of the charging and discharging power of the energy storage unit and the state of charge of the energy storage unit, and its expression is: wherein, is a cost coefficient related to the battery rated capacity and cycle life; denotes the absolute value of the charge and discharge power; is the time step; is a stress factor function used to quantify the impact of different SOC levels on the battery life; The stress factor function is: wherein, the SOC working point for the optimal battery life; is a preset penalty coefficient.

4. The method of claim 1, wherein, The method further includes: In the event of a grid fault at the centralized converter station, the energy storage unit of the faulted station is switched to DC voltage control mode. Based on the estimated load of the faulty station area, the transmission limit of the flexible ring network line, and the SOC protection constraint of the faulty station energy storage unit, the optimal support power from the normal station to the faulty station area through the flexible ring network is calculated. The control station adjusts its power according to the optimal support power.

5. The method of claim 4, wherein, The formula for calculating the optimal support power is: wherein, is the load of the fault station area; K1 is a preset load bearing coefficient; is the thermal stability limit of the flexible looped network line; is the current state of charge of the energy storage unit of the fault station; is the preset protection threshold of the energy storage of the fault station; is the capacity thereof; is the predicted fault duration.

6. The method of claim 4, wherein, The control of the normal station to adjust according to the optimal support power includes: The converter of the normal station is controlled to increase the power drawn from the grid to the optimal support power, and the energy storage unit of the normal station is controlled to switch to power follower mode.

7. The method of claim 1, wherein, The method further includes: In the event of a failure in the flexible ring network that causes the system to be split into multiple electrical islands, the energy storage units in each electrical island are controlled to switch to DC voltage control mode in order to stabilize the DC bus voltage of the station. After the DC bus voltage stabilizes, the single-phase MMC inverters in each electrical island are switched to V / f control mode to establish the AC voltage and frequency reference for that island.

8. The method of claim 7, wherein, The method further includes: When the fault in the flexible ring network is eliminated, quasi-synchronous grid connection between islands can be achieved by adjusting the frequency setting value of the single-phase MMC in V / f control mode within the island to be connected.

9. A controller comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images