Source network load storage coordinated control method for multi-configuration network type energy storage coordinated operation

Abstract

The invention relates to the technical field of energy storage systems, in particular to a source network load storage coordination control method for multi-configuration network type energy storage coordination operation. According to the technical scheme, the source-network-load-storage coordinated control method for multi-construction-network-type energy storage coordinated operation comprises the following steps: generating a new energy output and load change trend through a dynamic prediction model based on historical power data, real-time load requirements and meteorological information; optimization is carried out to obtain a power generation plan containing a clean energy consumption target and economical efficiency constraints; decomposing the power generation plan into a photovoltaic power adjusting instruction, an energy storage charging and discharging instruction and a load control instruction according to the power grid safety priority, wherein the emergency control instruction is executed prior to the economic dispatching instruction; and controlling parallel operation of a plurality of pieces of networking type energy storage equipment, and realizing multi-machine power distribution and circulating current suppression through a virtual synchronous control technology. And through layered optimization scheduling and multi-machine cooperative control, the new energy consumption capability is remarkably improved, and the phenomenon of abandoning light and wind is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy storage systems, in particular to a source-grid-load-storage coordinated control method for multi-structure grid type energy storage collaborative operation. BACKGROUND

[0002] In the current new energy high penetration rate power grid, the traditional centralized photovoltaic power station is connected to the grid through a large-capacity inverter and relies on high-voltage transmission lines for long-distance power transmission, which has significant defects: on the one hand, the fluctuation of new energy output leads to instability of power grid voltage and frequency, and centralized access aggravates the risk of sudden change of power flow; on the other hand, the capacity of single energy storage device is limited and the control means is single, and it is difficult to realize multi-machine collaborative operation and dynamic power distribution. The existing coordinated control system mostly adopts hierarchical independent scheduling, and there is a lack of closed-loop feedback between the master station, the substation and the terminal, which cannot adapt to the randomness of new energy and the multi-time scale regulation and control demand of the power grid. In addition, the on-grid and off-grid switching relies on manual intervention, and the response is slow, which is difficult to guarantee the continuity of power supply.

[0003] The centralized access of large-capacity new energy leads to severe fluctuation of power grid voltage and frequency, limited consumption capacity, and lack of multi-machine collaborative control mechanism of energy storage devices; the traditional hierarchical scheduling system has insufficient dynamic response capability and cannot realize global optimization of source-grid-load-storage; the on-grid and off-grid switching relies on manual decision, and the switching time is long and easy to cause secondary failure. SUMMARY

[0004] The present application proposes a source-grid-load-storage coordinated control method for multi-structure grid type energy storage collaborative operation, which solves the problem of existing technology that large-capacity new energy centralized access causes severe fluctuation of power grid voltage and frequency and limited consumption, and the lack of multi-machine collaborative control of energy storage devices leads to circulating current and oscillation, and the dynamic response of the traditional hierarchical scheduling system is insufficient and the on-grid and off-grid switching efficiency is low.

[0005] To solve the above technical problems, the technical solution adopted by the present application is:

[0006] A source-grid-load-storage coordinated control method for multi-structure grid type energy storage collaborative operation, comprising the following steps:

[0007] Based on historical power data, real-time load demand and weather information, generate new energy output and load change trend through dynamic prediction model, and optimize to obtain power generation plan containing clean energy consumption target and economic constraint;

[0008] The power generation plan is divided into photovoltaic power regulation instruction, energy storage charging and discharging instruction and load control instruction according to the priority of power grid safety, and the emergency control instruction is prior to the economic dispatching instruction;

[0009] The plurality of grid-connected energy storage devices are controlled to operate in parallel, power distribution and circulating current suppression of multiple machines are realized through virtual synchronous control technology, the instructions are executed and adjustment results are fed back to the master station, and the prediction model and the scheduling strategy are dynamically corrected according to the feedback data.

[0010] Further, the dynamic prediction model comprises:

[0011] The meteorological satellite data, the ground monitoring station data and the real-time operation parameters of the inverter are fused to construct a photovoltaic power prediction model.

[0012] The sliding window method is used to dynamically correct the ultra-short-term prediction result, when the prediction deviation exceeds the set threshold, the energy storage pre-charge and discharge strategy is triggered and the maximum photovoltaic output is limited.

[0013] Based on the historical load data and user power consumption behavior analysis, a time-of-use load prediction curve is generated, and the time and space matching is performed with the new energy output prediction result.

[0014] Further, the power generation plan optimization comprises:

[0015] A multi-objective optimization model is constructed to minimize the operation cost and maximize the clean energy consumption, and the grid N-1 safety check constraint and the energy storage dynamic charge and discharge limit are embedded.

[0016] An improved genetic algorithm is used to solve the Pareto optimal solution set, and the entropy weight method is used to select the final scheduling scheme.

[0017] In real-time operation, the optimization model parameters are updated according to the power deviation, and the corrected scheduling instruction is generated every 5 minutes.

[0018] Further, the instruction decomposition comprises:

[0019] When the grid frequency limit or the main transformer overload is detected, the machine trip sequence is generated according to the photovoltaic power station conversion efficiency, and the low-efficiency power station is preferentially removed.

[0020] The frequency and voltage regulation weight is dynamically allocated according to the state of charge of the energy storage, the fast power regulation demand is responded, and the frequency regulation weight of the energy storage is set to be negatively correlated with the SOC state.

[0021] Based on the time-of-use price curve, the energy storage charge and discharge timing is optimized, the discharge power in the peak period is set to be 1.5-2 times of that in the valley period, and the charge and discharge threshold is dynamically adjusted combined with the load prediction data.

[0022] Further, the cooperative control of the plurality of grid-connected energy storage devices comprises:

[0023] A master-slave control and peer-to-peer control hybrid architecture is used, the master controller is dynamically elected based on the device health state and communication delay, and the power distribution of multiple machines is coordinated.

[0024] Parallel circulating current is inhibited by harmonic current detection and adaptive adjustment of virtual impedance, and circulating current rate is monitored in real time to ensure that it is below a safety threshold;

[0025] Multi-machine phase synchronization is achieved based on a high-precision clock synchronization protocol, and the phase of the energy storage inverter output voltage is dynamically adjusted to match the grid demand.

[0026] Further, it also includes on-off grid seamless switching control:

[0027] When voltage sag or frequency mutation at the grid-connected point is detected, the grid-connected switch is disconnected within 100ms, the energy storage is switched to voltage source mode, off-grid microgrid voltage frequency support is established, and photovoltaic is controlled to operate according to microgrid energy management instructions;

[0028] When external grid power is restored, the output voltage amplitude, frequency and phase of the energy storage are dynamically adjusted, the electrical parameters on both sides of the grid-connected point are monitored in real time, and the grid-connected switch is closed after the synchronization condition is met, with a switching time of less than 200ms.

[0029] Further, it also includes a dynamic priority control strategy:

[0030] The control instructions are divided into three levels of emergency control, fast regulation and economic dispatch, among which the response time of the emergency control instruction is less than 500ms, the response time of the fast regulation instruction is less than 2s, and the economic dispatch instruction is executed according to a minute-level period;

[0031] When the energy storage power exceeds the limit or the equipment fails, the load is cut off according to the load elasticity coefficient and user contract priority classification, and high-priority loads such as medical treatment and communication are preferentially retained.

[0032] Further, it also includes digital twin assisted decision-making:

[0033] A source-grid-load-storage joint simulation model is constructed to simulate extreme weather, equipment failure and load mutation scenarios, and standby control strategies are pre-generated and stored in a strategy library;

[0034] The health status of the energy storage and the output fluctuation of the photovoltaic are monitored in real time, and when an abnormality is detected, the standby strategy is called to execute load classification cutting or energy storage power emergency support, while updating the digital twin model parameters.

[0035] Further, the virtual synchronous control includes:

[0036] The inertia and damping characteristics of a synchronous generator are simulated, and the virtual inertia parameters of the energy storage are dynamically adjusted to respond to grid frequency changes, with a frequency deviation recovery time of less than 300ms;

[0037] Local voltage stability is achieved through reactive power-voltage droop control, and the droop coefficient is dynamically adjusted according to the grid strength, with a voltage deviation recovery time of less than 300ms.

[0038] Further, it further includes multi-time scale coordinated control:

[0039] Emergency control and fast frequency regulation and voltage regulation are performed at a second-level time scale, economic dispatch instructions are optimized at a minute-level time scale, and a prediction model and a long-term dispatch plan are corrected at a hour-level time scale;

[0040] The control instructions of different time scales are integrated through a data bus, and it is ensured that the instructions are executed in coordination and without conflict.

[0041] The positive effects of the application are as follows: through hierarchical optimization dispatching and multi-machine coordinated control, the new energy consumption capacity is significantly improved, and the phenomena of light and wind curtailment are reduced; the adaptability of the power grid to fluctuating power sources is enhanced, and the voltage and frequency instability problems are effectively alleviated; a closed-loop feedback mechanism dynamically corrects the prediction and dispatching strategy, and the overall response accuracy and reliability of the system are improved.

[0042] The positive effects of the application are as follows: through hierarchical optimization dispatching and multi-machine coordinated control, the new energy consumption capacity is significantly improved, and the phenomena of light and wind curtailment are reduced; the adaptability of the power grid to fluctuating power sources is enhanced, and the voltage and frequency instability problems are effectively alleviated; a closed-loop feedback mechanism dynamically corrects the prediction and dispatching strategy, and the overall response accuracy and reliability of the system are improved. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor also fall within the scope of protection of the application.

[0044] EMBODIMENT

[0045] A source-grid-load-storage coordinated control method for multi-structure network type energy storage collaborative operation, comprising the following steps:

[0046] Based on historical power data, real-time load demand and weather information, a new energy output and load change trend are generated through a dynamic prediction model, and a generation plan containing a clean energy consumption target and an economic constraint is obtained through optimization;

[0047] The generation plan is divided into photovoltaic power regulation instructions, energy storage charging and discharging instructions and load control instructions according to the safety priority of the power grid, wherein the emergency control instructions are executed in priority to the economic dispatch instructions;

[0048] A plurality of network type energy storage devices are controlled to operate in parallel, multi-machine power distribution and circulating current suppression are realized through virtual synchronous control technology, the instructions are executed and the adjustment results are fed back to the master station, and the prediction model and the dispatching strategy are dynamically corrected according to the feedback data.

[0049] The method adopts hierarchical optimization and multi-machine cooperative control architecture, including master station optimization scheduling, substation instruction decomposition and terminal execution feedback three levels:

[0050] Master station optimization scheduling:

[0051] The master station predicts the future new energy output and load change trend based on historical power data, real-time load demand and weather information through a dynamic prediction model (such as time series analysis combined with machine learning algorithm);

[0052] Generate power generation plan using multi-objective optimization algorithm, goals include maximizing clean energy consumption rate, minimizing operating cost, and embedding grid N-1 safety constraints (such as main transformer capacity limit, line transmission capacity);

[0053] According to real-time operation data and prediction deviation, dynamically correct the power generation plan and generate scheduling instructions.

[0054] Substation instruction decomposition:

[0055] The substation receives the master station instructions, decomposes the instructions into photovoltaic power regulation, energy storage charging and discharging, and load control instructions according to the grid safety priority, where emergency control instructions (such as machine tripping, load shedding) are executed before economic dispatch instructions;

[0056] Use priority queue management mechanism to ensure real-time response of high priority instructions.

[0057] Terminal cooperative execution:

[0058] Control multiple grid-connected energy storage devices to run in parallel, simulate the external characteristics of synchronous generators through virtual synchronous control technology, realize multi-machine power distribution and circulating current suppression;

[0059] After executing the instructions, the adjustment results are fed back to the master station to form a "prediction-optimization-execution-feedback" closed-loop control.

[0060] Through hierarchical optimization scheduling and multi-machine cooperative control, the new energy consumption capacity is significantly improved, and the phenomenon of abandoned light and wind is reduced; The adaptability of the grid to fluctuating power sources is enhanced, effectively alleviating the voltage and frequency instability problem; The closed-loop feedback mechanism dynamically corrects the prediction and scheduling strategy, improving the overall response accuracy and reliability of the system.

[0061] Dynamic prediction model includes:

[0062] Integrate meteorological satellite data, ground monitoring station data and real-time operation parameters of inverters to build a photovoltaic power prediction model;

[0063] Use sliding window method to dynamically correct short-term prediction results, when the prediction deviation exceeds the set threshold, trigger the energy storage pre-charging and discharging strategy and limit the maximum photovoltaic output;

[0064] Based on load historical data and user electricity behavior analysis, generate time-of-use load prediction curve, and match with new energy output prediction results in time and space.

[0065] Photovoltaic power prediction realizes dynamic correction by fusing multi-source data:

[0066] Data fusion: Integrate irradiance data from meteorological satellites, cloud movement trends from ground monitoring stations, and real-time operating parameters of inverters (such as MPPT voltage and current) to build a photovoltaic output prediction model;

[0067] Dynamic correction: Use sliding window method to roll over the ultra-short-term prediction results, when the prediction deviation exceeds the preset threshold (such as sudden rain causing output drop), trigger energy storage pre-charge and discharge strategy, and limit photovoltaic maximum output to avoid overload;

[0068] Load matching: Based on user electricity behavior analysis, generate time-of-use load curve, and match with new energy output prediction results in time and space, optimize energy storage charging and discharging timing.

[0069] Dynamic correction of photovoltaic prediction model improves output prediction accuracy and reduces the impact of prediction deviation on the power grid; through energy storage pre-regulation and photovoltaic power limiting strategy, reduce the frequency of emergency control, prolong the service life of equipment, and improve the economic efficiency of system operation.

[0070] Generation plan optimization includes:

[0071] Build a multi-objective optimization model with the goal of minimizing operating cost and maximizing clean energy consumption, embed grid N-1 safety check constraints and energy storage dynamic charging and discharging limits;

[0072] Use improved genetic algorithm to solve the Pareto optimal solution set, and select the final dispatching scheme through entropy weight method;

[0073] In real-time operation, update the optimization model parameters according to the power deviation, and generate corrected dispatching instructions every 5 minutes.

[0074] Multi-objective optimization scheduling is achieved through the following steps:

[0075] Modeling and solving:

[0076] Build a multi-objective optimization model with the goal of economic efficiency (fuel cost, operation and maintenance cost) and clean energy consumption rate, constraint conditions include power balance equation, energy storage SOC dynamic limit value and main transformer N-1 overload threshold;

[0077] Use improved genetic algorithm (such as introduction of elite preservation strategy) to solve the Pareto optimal solution set, and select the comprehensive optimal dispatching scheme through entropy weight method.

[0078] Real-time correction:

[0079] During operation, the optimization model parameters are updated according to the real-time power deviation (such as the difference between the actual output of photovoltaic and the predicted value) and the corrected scheduling instructions are generated every 5 minutes;

[0080] The corrected instructions are synchronized to the substation and terminal through the data bus to ensure consistency of the instructions.

[0081] The multi-objective optimization model takes into account the economy and safety, balances clean energy consumption and grid operation cost; the real-time rolling correction strategy improves the adaptability and real-time of the scheduling scheme, ensuring the power supply reliability in complex scenarios.

[0082] The instruction decomposition includes:

[0083] When the grid frequency limit or the main transformer overload is detected, the generator trip sequence is generated according to the photovoltaic power station conversion efficiency, and the low-efficiency power station is preferentially removed;

[0084] The frequency regulation and voltage regulation weight is dynamically allocated according to the state of charge of the energy storage, and the fast power regulation demand is responded, and the energy storage frequency regulation weight and SOC state are set to be negatively correlated;

[0085] The energy storage charging and discharging time sequence is optimized based on the time-of-use price curve, the discharging power in peak period is set to be 1.5-2 times of that in valley period, and the charging and discharging threshold is dynamically adjusted combined with the load forecast data.

[0086] The instruction decomposition and execution includes a hierarchical control strategy:

[0087] Emergency control:

[0088] When the grid frequency limit or the main transformer overload is detected, the generator trip sequence is generated according to the photovoltaic power station conversion efficiency, and the low-efficiency power station (such as the inverter efficiency lower than the industry average) is preferentially removed;

[0089] Combined with the load elasticity coefficient (such as the interruptibility of industrial users is higher than that of residential users), the low-priority load is removed.

[0090] Fast regulation:

[0091] The frequency regulation and voltage regulation weight is dynamically allocated according to the state of charge of the energy storage, and the frequency regulation weight is increased when the SOC is high to prolong the discharging time;

[0092] The feedforward-feedback composite control strategy is adopted to quickly respond to the grid frequency fluctuation.

[0093] Economic dispatch:

[0094] The energy storage charging and discharging threshold is set based on the time-of-use price curve, and the discharging power in peak period is 1.5-2 times of that in valley period;

[0095] Dynamic adjustment of charging and discharging plan combined with load prediction data, priority to meet the power supply demand in high price period.

[0096] Hierarchical instruction decomposition and dynamic weight allocation optimize resource utilization efficiency, reduce economic loss caused by machine tripping; combined with time-of-use pricing strategy to improve energy storage economic benefits, while reducing the impact of user power outage through load elasticity hierarchical tripping.

[0097] The coordinated control of multiple network-type energy storage includes:

[0098] Adopting master-slave control and peer-to-peer control hybrid architecture, dynamically electing master controller based on device health status and communication delay, coordinating multi-machine power distribution;

[0099] Suppressing parallel circulating current through harmonic current detection and adaptive adjustment of virtual impedance, and monitoring circulating current rate in real time to ensure it is below the safety threshold;

[0100] Based on high-precision clock synchronization protocol, realize multi-machine phase synchronization, and dynamically adjust the phase of energy storage inverter output voltage to match the grid demand.

[0101] The coordinated control of multiple network-type energy storage is achieved through a hybrid architecture:

[0102] Master-slave-peer control:

[0103] Dynamically electing master controller (based on device health status, communication delay and capacity proportion), master controller is responsible for global power distribution, and the rest of the devices follow in peer-to-peer mode;

[0104] When the master controller fails, automatically switch to the backup controller to ensure system redundancy.

[0105] Circulating current suppression:

[0106] Identify circulating current components through harmonic current detection module, use adaptive adjustment strategy of virtual impedance to dynamically increase circulating current path impedance to suppress circulating current;

[0107] Monitor circulating current rate in real time to ensure it is below the safety threshold.

[0108] Phase synchronization:

[0109] Based on high-precision clock synchronization protocol (such as IEEE 1588) to realize multi-machine phase alignment;

[0110] Dynamically adjust the phase of energy storage inverter output voltage to match the grid demand.

[0111] Master-slave-peer hybrid control solves the problem of multi-machine parallel circulating current, improves system stability; dynamic election and redundancy switching mechanism enhances the flexibility of control architecture, ensures continuous and reliable operation of devices in case of device failure.

[0112] Also includes and off-grid seamless switching control:

[0113] When detecting voltage sag or frequency mutation at the grid-connected point, the grid-connected switch is disconnected within 100 ms, the energy storage is switched to voltage source mode, the off-grid microgrid voltage frequency support is established, and the photovoltaic is controlled to operate according to the microgrid energy management instruction;

[0114] When the external power grid is restored, the output voltage amplitude, frequency and phase of the energy storage are dynamically adjusted, the electrical parameters on both sides of the grid-connected point are monitored in real time, and the grid-connected switch is closed after the synchronization condition is met, and the switching time is less than 200 ms.

[0115] The on-grid and off-grid seamless switching control includes the following steps:

[0116] Grid-connected to off-grid:

[0117] The voltage amplitude and frequency of the grid-connected point are monitored in real time, and when the voltage sag exceeds 20% or the frequency deviation exceeds ±1 Hz, the protection device is triggered to disconnect the grid-connected switch;

[0118] Switch the energy storage to voltage source mode to simulate a synchronous generator to establish microgrid voltage frequency support;

[0119] Control the photovoltaic to operate according to the microgrid energy management instruction to limit output fluctuation.

[0120] Off-grid to grid:

[0121] Monitor the restoration of the external power grid, and dynamically adjust the output voltage amplitude, frequency and phase of the energy storage;

[0122] When the voltage amplitude difference on both sides of the grid-connected point is less than 5%, the frequency difference is less than 0.1 Hz, and the phase difference is less than 5°, the grid-connected switch is closed to complete synchronization and parallel operation.

[0123] The seamless switching control ensures the continuity of on-grid and off-grid mode conversion, reduces the power outage time; the synchronization and parallel operation strategy avoids equipment impact, prolongs the service life of switches and energy storage devices, and improves the reliability of microgrid power supply.

[0124] Also includes a dynamic priority control strategy:

[0125] The control instruction is divided into three levels of emergency control, fast regulation and economic dispatch, among which the emergency control instruction response time is less than 500 ms, the fast regulation instruction response time is less than 2 s, and the economic dispatch instruction is executed according to the minute level period;

[0126] When the energy storage power exceeds the limit or the equipment fails, the load is cut off according to the load elasticity coefficient and user contract priority classification, and high-priority loads such as medical treatment and communication are preferentially retained.

[0127] The dynamic priority control strategy is realized through a hierarchical response mechanism:

[0128] Instruction hierarchy:

[0129] Emergency control instructions (such as machine tripping and load shedding) have a response time of less than 500ms and are directly issued to the terminal using the GOOSE protocol;

[0130] Fast regulation instructions (such as frequency and voltage regulation) have a response time of less than 2s and are transmitted through a fiber channel;

[0131] Economic dispatch instructions are executed on a minute-by-minute basis and are communicated through Ethernet.

[0132] Load hierarchical removal:

[0133] When the energy storage power exceeds the limit, the hierarchical load shedding is executed according to the user contract flexibility coefficient (such as medical load priority 1 and commercial load 2);

[0134] Low flexibility load is preferentially removed to preserve high priority load for power supply.

[0135] Multi-level response mechanism optimizes control efficiency, emergency instruction priority execution ensures power grid safety; user hierarchical removal strategy reduces social impact, preferentially guarantees key load power supply, and improves user satisfaction.

[0136] It also includes digital twin assisted decision-making:

[0137] A source-grid-load-storage joint simulation model is constructed to simulate extreme weather, equipment failure and load mutation scenarios, pre-generate backup control strategies and store them in the strategy library;

[0138] Real-time monitoring of energy storage health status and photovoltaic output fluctuation, when detecting abnormalities, call backup strategy, execute load hierarchical removal or energy storage power emergency support, and update digital twin model parameters.

[0139] Digital twin assisted decision-making includes the following steps:

[0140] Model construction:

[0141] Establish a source-grid-load-storage joint simulation model, integrating power grid topology, device parameters and meteorological data;

[0142] Simulate extreme scenarios (such as hurricane-induced photovoltaic shutdown, energy storage fire) to generate a backup control strategy library.

[0143] Strategy invocation:

[0144] Real-time monitoring of energy storage health status (such as sudden increase in battery internal resistance) and photovoltaic output fluctuation (such as single-day fluctuation exceeding 30%);

[0145] Call backup strategy to execute load hierarchical removal or enable redundant energy storage devices, while updating digital twin model parameters.

[0146] Digital twin pre-plays extreme scenarios, enhances the robustness of the system to respond to sudden failures; backup strategy quickly calls to shorten the fault recovery time, reduce the operation and maintenance risk, and improve the intelligent level of the system.

[0147] The virtual synchronous control includes:

[0148] The inertia and damping characteristics of the synchronous generator are simulated, and the virtual inertia parameter of the energy storage is dynamically adjusted to respond to the change of the grid frequency, and the frequency deviation recovery time is less than 300 ms.

[0149] Local voltage stability is achieved through reactive power-voltage droop control, and the droop coefficient is dynamically adjusted according to the grid strength, and the voltage deviation is restored to within ±2% of the rated value.

[0150] The virtual synchronous control technology realizes grid support:

[0151] Inertia simulation:

[0152] The inertia and damping characteristics of the synchronous generator are simulated through the virtual synchronous generator (VSG) algorithm, and the virtual inertia parameter is dynamically adjusted to respond to the change of the grid frequency;

[0153] When the frequency deviation is large, the virtual inertia is increased to slow down the rate of frequency change.

[0154] Voltage recovery:

[0155] The reactive power-voltage droop control strategy is adopted, and the droop coefficient is dynamically adjusted according to the grid strength (such as short-circuit capacity);

[0156] Under weak grid conditions, the droop coefficient is reduced to avoid voltage oscillation.

[0157] The virtual synchronous technology enhances the inertia support capability of the grid, and alleviates the frequency instability problem caused by high proportion of new energy access; the dynamic voltage recovery strategy improves the voltage quality of the local grid and guarantees the stability of the user side power supply.

[0158] It also includes multi-time scale coordinated control:

[0159] Emergency control and fast frequency and voltage regulation are performed on the second time scale, economic dispatch instructions are optimized on the minute time scale, and prediction models and long-term dispatch plans are corrected on the hour time scale;

[0160] The control instructions of different time scales are integrated through the data bus to ensure that the instructions are executed in coordination and without conflict.

[0161] The multi-time scale coordinated control is realized through the following mechanisms:

[0162] Time scale division:

[0163] Second-level control: perform emergency shutdown, frequency and voltage regulation, etc.

[0164] Minute-level control: optimize economic dispatch instructions and revise prediction models;

[0165] Hour-level control: update long-term generation plan and operation strategy.

[0166] Instruction coordination:

[0167] Integrate control instructions of different time scales through data bus (such as message middleware);

[0168] Use conflict detection algorithm (such as rule engine) to ensure that instructions are conflict-free, for example, economic dispatch does not interfere with emergency control.

[0169] Multi-time scale coordination realizes global optimization of short, medium and long-term control, and improves system operation efficiency; conflict detection mechanism ensures that instruction coordination has no contradiction, and enhances the ability of coordinated control in complex scenarios.

[0170] The above-described embodiments are more detailed and specific, express the preferred embodiments of the present application, and are only used to illustrate the technical ideas and characteristics of the present application. The purpose is to enable those skilled in the art to understand the content of the present application and implement it, but it is not limited to the present application, and the patent range of the present application cannot be limited to the present embodiment, that is, any equivalent changes or modifications made according to the spirit disclosed by the present application, within the structure of the present application, local improvement and change between subsystems, transformation, etc., are still within the patent range of the present application.

Claims

1. A source-grid-load-storage coordinated control method for the coordinated operation of multi-configuration grid-type energy storage, characterized in that: The following steps are involved: Based on historical power data, real-time load demand, and meteorological information, a dynamic forecasting model is used to generate renewable energy output and load change trends, and an optimized power generation plan is obtained that incorporates clean energy consumption targets and economic constraints. The power generation plan is decomposed into photovoltaic power regulation instructions, energy storage charging and discharging instructions, and load control instructions according to the grid security priority, among which emergency control instructions take precedence over economic dispatch instructions; Control multiple grid-connected energy storage devices to operate in parallel, realize multi-machine power distribution and circulation suppression through virtual synchronous control technology, execute the instructions and feedback the adjustment results to the master station, and dynamically modify the prediction model and scheduling strategy based on the feedback data.

2. A source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1, characterized in that: The dynamic prediction model includes: Integrate meteorological satellite data, ground monitoring station data, and inverter real-time operating parameters to build a photovoltaic power prediction model; The sliding window method is used to dynamically correct the ultra-short-term forecast results. When the forecast deviation exceeds the set threshold, the energy storage pre-charge and discharge strategy is triggered and the maximum photovoltaic output is limited. Based on the analysis of historical load data and user electricity consumption behavior, a time-of-use load forecast curve is generated and matched with the new energy output forecast results in time and space.

3. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: The power generation plan optimization includes: Construct a multi-objective optimization model with the goals of minimizing operating costs and maximizing clean energy consumption, embedding the grid's N-1 safety verification constraints and energy storage dynamic charging and discharging restrictions; An improved genetic algorithm is used to solve the Pareto optimal solution set, and the final scheduling solution is selected through the entropy weight method; In real-time operation, the optimization model parameters are updated based on the power deviation, and the revised scheduling instructions are generated every 5 minutes.

4. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: The instruction decomposition includes: When the grid frequency exceeds the limit or the main transformer is overloaded, a shutdown sequence is generated based on the conversion efficiency of the photovoltaic power stations, with inefficient power stations being prioritized. Dynamically allocate frequency and voltage regulation weights based on the energy storage state of charge to respond to rapid power regulation needs, and set the energy storage frequency regulation weight to be negatively correlated with the SOC state; The energy storage charging and discharging sequence is optimized based on the time-of-use electricity price curve, the discharge power during peak hours is set to 1.5-2 times that of the valley hours, and the charging and discharging thresholds are dynamically adjusted in combination with load forecast data.

5. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: The coordinated control of multiple grid-type energy storage units includes: A hybrid architecture of master-slave and peer-to-peer control is adopted, dynamically electing a master controller based on device health status and communication delay to coordinate power distribution among multiple machines. Parallel circulating current is suppressed through harmonic current detection and adaptive adjustment of virtual impedance, and the circulating current rate is monitored in real time to ensure it is below the safety threshold; Multi-machine phase synchronization is achieved based on a high-precision clock synchronization protocol, and the output voltage phase of the energy storage inverter is dynamically adjusted to match the grid demand.

6. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: Also includes on-grid and off-grid seamless switching control: When a sudden voltage drop or frequency change is detected at the grid connection point, the grid-connected switch is disconnected within 100ms, the energy storage is switched to voltage source mode, the voltage and frequency support of the off-grid microgrid is established, and the photovoltaic operation is controlled according to the microgrid energy management instructions; When the external power grid resumes power supply, the energy storage output voltage amplitude, frequency and phase are dynamically adjusted. By real-time monitoring of the electrical parameters on both sides of the grid connection point, the grid connection switch is closed when the synchronization conditions are met, and the switching time is less than 200ms.

7. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: Also includes dynamic priority control strategies: Control instructions are divided into three levels: emergency control, rapid adjustment, and economic dispatch. The response time of emergency control instructions is less than 500ms, the response time of rapid adjustment instructions is less than 2s, and the economic dispatch instructions are executed in a minute-level cycle. When the energy storage power exceeds the limit or the equipment fails, the load is cut off in a graded manner according to the load elasticity coefficient and the user contract priority, with priority given to high-priority loads such as medical and communication loads.

8. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: It also includes digital twin-assisted decision-making: Build a joint simulation model of power generation, grid, load and storage to simulate extreme weather, equipment failure and load mutation scenarios, pre-generate backup control strategies and store them in the strategy library; Real-time monitoring of energy storage health status and photovoltaic output fluctuations. When an anomaly is detected, the backup strategy is invoked to perform load shedding or energy storage power emergency support, while updating the digital twin model parameters.

9. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1 is characterized in that: The virtual synchronization control includes: Simulate the inertia and damping characteristics of synchronous generators, dynamically adjust the virtual inertia parameters of energy storage to respond to grid frequency changes, and achieve frequency deviation recovery time of less than 300ms; Local voltage stabilization is achieved through reactive power-voltage droop control, and the droop coefficient is dynamically adjusted according to the grid strength, so that the voltage deviation is restored to within ±2% of the rated value.

10. The source-grid-load-storage coordinated control method for multi-configuration grid-type energy storage coordinated operation according to claim 1, characterized in that: Also includes multi-timescale coordinated control: Execute emergency control and rapid frequency and voltage regulation on a time scale of seconds, optimize economic dispatch instructions on a time scale of minutes, and revise forecast models and long-term dispatch plans on a time scale of hours; Control instructions of different time scales are integrated through the data bus to ensure that the instructions are executed in a coordinated and conflict-free manner.