Charging and discharging cooperative control method of green power direct connection energy storage system

Abstract

The invention discloses a charging and discharging cooperative control method for a green power direct connection energy storage system, and aims to solve the problems of low efficiency, short service life and weak power grid support caused by renewable energy source volatility and energy storage response lag. The method comprises the following steps: collecting green power output, energy storage state and load demand data in real time; based on a three-layer collaborative architecture, an upper-layer energy scheduling layer determines target power according to an operation mode (such as green power consumption, peak clipping and valley filling), a middle-layer power distribution layer dynamically distributes power to each sub-cluster according to a health state and temperature and voltage consistency, and a lower-layer equipment execution layer drives a converter to perform precise execution through double-closed-loop control; and a closed-loop feedback correction and hierarchical protection mechanism is introduced. The green power consumption rate is effectively improved, the energy storage life is prolonged, and the system safety and the power grid supporting capacity are enhanced.

Description

Technical Field

[0001] This invention belongs to the field of power system and energy storage control technology, specifically, it relates to a charging and discharging coordinated control method for a green electricity direct-connected energy storage system. Background Technology

[0002] Currently, the proportion of green electricity, represented by wind power and photovoltaics, in the energy structure continues to increase. Green electricity has inherent characteristics such as intermittency, volatility, and uncontrollability; its direct grid connection or power supply can easily cause grid frequency instability, voltage flicker, and power backflow. To improve the absorption capacity of green electricity and the stability of system operation, direct-connection green electricity storage systems have been widely studied as a novel architecture. By directly coupling renewable energy generation units and energy storage devices on the DC side, this avoids the multiple energy conversion stages in traditional AC grid connection, thereby reducing losses and improving efficiency. However, this architecture places higher demands on charging and discharging control strategies, requiring dynamic power matching and energy synergy between the source and storage systems under grid-free conditions.

[0003] Among these, the coordinated charging and discharging control of the green electricity direct-connected energy storage system becomes the core element determining system performance. This control method aims to dynamically adjust the charging and discharging power of the energy storage units based on real-time green electricity output, load demand, and energy storage status, in order to maintain stable DC bus voltage, maximize green electricity utilization, and extend energy storage life. An ideal control strategy should possess multi-objective optimization capabilities, balancing economy, safety, and response speed, and be able to achieve millisecond-level dynamic adjustment under complex operating conditions.

[0004] In existing technologies, most solutions still employ discrete control logic: during the charging phase, constant current or constant voltage modes are triggered based on fixed thresholds, while the discharging phase relies on preset discharge curves or simple power allocation rules. These methods do not fully consider the coupling effects of sudden changes in green energy output and rapid load fluctuations, making it difficult to achieve real-time coordination between the source, storage, and load. Some studies have introduced fuzzy control or model predictive control, but their parameter tuning is complex, computationally burdensome, and they lack a unified optimization objective function for the charging and discharging process. This leads to problems such as overcharging / over-discharging of energy storage, exceeding bus voltage limits, or increased green energy abandonment rates in extreme scenarios such as sudden drops in sunlight or sudden increases in load. Especially under multi-timescale operational requirements (such as second-level fluctuation smoothing and hourly energy dispatch), existing algorithms lack a hierarchical coordination mechanism for control commands, failing to balance dynamic response and long-term economic efficiency. Therefore, there is an urgent need for a charging and discharging coordinated control method driven by optimization algorithms, capable of embedding physical constraints, adaptively adjusting control weights, and achieving efficient and robust real-time decision-making through analytical formulas. Summary of the Invention

[0005] The purpose of this invention is to provide a charging and discharging coordinated control method for a green electricity direct-connected energy storage system. This method mainly addresses the technical problems in the prior art, such as low system operating efficiency, shortened equipment lifespan, and insufficient grid support capacity, caused by the lack of a dynamic matching mechanism between the volatility of renewable energy generation and the lag in the response of the energy storage system.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for coordinated charging and discharging control of a green electricity direct-connected energy storage system includes the following steps:

[0008] S1, acquire real-time power output data of distributed photovoltaic arrays and wind turbine generators;

[0009] S2, obtain the state of charge, terminal voltage, charge and discharge current, temperature distribution and health status assessment indicators of the energy storage battery pack;

[0010] S3, obtain the local load power demand at the current moment and the load forecast curve within the future preset time window;

[0011] S4, perform multi-source sensing fusion processing on the real-time power output data, energy storage battery pack status parameters and local load power demand data to obtain a fusion characterization of the system operating status;

[0012] S5. Based on the fusion characterization of the system's operating status, execute a hierarchical collaborative control strategy to generate charging and discharging power commands for the energy storage converter.

[0013] Furthermore, in this invention, the hierarchical collaborative control strategy includes an upper energy scheduling layer, a middle power allocation layer, and a lower device execution layer. The upper energy scheduling layer determines the system operation mode and the target charging and discharging power of the energy storage system based on the net load power, energy storage state of charge boundary constraints, and grid dispatch instructions. The middle power allocation layer dynamically allocates the target charging and discharging power to each energy storage sub-cluster based on the health status assessment indicators of the energy storage battery pack, temperature distribution uniformity, and individual cell voltage consistency. The lower device execution layer receives the allocated power instructions and generates a pulse width modulation signal to drive the energy storage converter to perform precise power output.

[0014] Furthermore, in this invention, the operating modes of the upper energy dispatch layer include a green electricity priority consumption mode, a peak shaving and valley filling mode, a grid support mode, and an emergency supply guarantee mode; wherein,

[0015] In the green electricity priority consumption mode, the energy storage system absorbs excess green electricity at the maximum allowable charging power;

[0016] In peak shaving and valley filling mode, the energy storage system supplies power to the load at a set discharge power.

[0017] In the grid support mode, the energy storage system provides active or reactive power support according to the instructions.

[0018] In emergency power supply mode, the energy storage system switches to off-grid operation to maintain continuous power supply to critical loads.

[0019] Furthermore, in this invention, under the green electricity priority consumption mode, the method for determining the target charging and discharging power is as follows: calculate the net load power, that is, the total green electricity output minus the local load power; if the net load power is greater than zero, the target charging power is the smaller value between the net load power and the maximum allowable charging power of the energy storage system; if the net load power is less than or equal to zero, the target charging power is zero.

[0020] Furthermore, in this invention, under peak shaving and valley filling mode, the method for determining the target charging and discharging power is as follows: Based on the time-of-use electricity price table, it is determined whether the current period is a peak period; if it is a peak period and the energy storage state of charge is higher than a preset lower threshold, the target discharge power is the smaller value between the local load power and the maximum allowable discharge power of the energy storage system; if it is a peak period and the energy storage state of charge is lower than a preset lower threshold, the target discharge power is zero; if it is a valley period and the energy storage state of charge is lower than a preset upper threshold, the target charging power is the maximum allowable charging power of the energy storage system; if it is a valley period and the energy storage state of charge is higher than a preset upper threshold, the target charging power is zero.

[0021] Furthermore, in this invention, the specific method by which the middle power allocation layer dynamically allocates the target charging and discharging power to each energy storage sub-cluster is as follows: Calculate the comprehensive weighting coefficient of each sub-cluster, which is jointly determined by the sub-cluster's health status assessment index, average temperature deviation, and individual cell voltage standard deviation; the higher the health status assessment index, the closer the average temperature is to the rated operating temperature, and the smaller the individual cell voltage standard deviation, the larger the comprehensive weighting coefficient of the sub-cluster; and allocate the target charging and discharging power proportionally according to the comprehensive weighting coefficient of each sub-cluster.

[0022] Furthermore, in this invention, the formula for calculating the comprehensive weighting coefficient is as follows:

[0023]

[0024] in, For the first The overall weighting coefficient of each sub-cluster Assess their health status indicators. Its average temperature, For the rated operating temperature, Its standard deviation of individual unit voltage, The rated voltage of a single battery cell. , , The normalization correction factor satisfies , This is the temperature sensitivity coefficient.

[0025] Furthermore, in this invention, the specific process by which the lower-level device executes the generation of the pulse width modulation signal includes:

[0026] The power commands allocated to each sub-cluster are converted into corresponding DC-side current reference values;

[0027] It adopts a dual closed-loop control structure, with the outer loop being the power loop and the inner loop being the current loop;

[0028] The output of the power loop serves as the input of the current loop, which employs a proportional-integral regulator. Its output is used to generate six complementary pulse width modulation signals via a space vector modulation algorithm, which drive the insulated-gate bipolar transistor bridge arms.

[0029] Furthermore, the method of the present invention also includes a closed-loop feedback correction mechanism:

[0030] At the end of each control cycle, the deviation between the actual charging / discharging power and the commanded power is collected;

[0031] The deviation value is input to the feedforward compensation module, which generates a compensation amount that is added to the power command of the next cycle.

[0032] The feedforward compensation module adopts a first-order inertial element transfer function, and the time constant is tuned online according to the response characteristics of the energy storage system.

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

[0034] (1) This invention solves the problem that the discrete control in the prior art cannot cope with the coupled effects of sudden changes in green electricity and sudden load changes through a three-layer collaborative control architecture. The upper-layer scheduling layer dynamically switches the operating mode (such as green electricity consumption and peak shaving and valley filling) according to the net load, energy storage status and grid commands; the middle-layer allocation layer allocates power based on a comprehensive weighting coefficient of health status, temperature uniformity and voltage consistency; the lower-layer execution layer ensures accurate output through dual closed-loop control and space vector modulation. This architecture effectively reduces the green electricity abandonment rate and the risk of overcharging and over-discharging of energy storage, and takes into account both short-term dynamic response and long-term economic efficiency.

[0035] (2) This invention introduces a power allocation strategy based on a comprehensive weighting coefficient, which solves the problem of uneven lifespan of energy storage units caused by simple allocation rules in the prior art. The strategy prioritizes allocating power to subclusters with good health, temperatures close to the rated value, and high voltage consistency, avoiding overcharging and discharging of some subclusters, significantly extending the overall lifespan of the system, and reducing operation and maintenance costs.

[0036] (3) This invention solves the problems of insufficient control accuracy and poor safety under extreme conditions in the prior art through closed-loop feedback correction and hierarchical protection strategies. The closed-loop feedback adopts a first-order inertial link to compensate for power commands, ensuring that the tracking error is minimized; the hierarchical protection starts power reduction, fault isolation or shutdown according to the degree of current, voltage and temperature exceeding the limit (response time <10ms). This mechanism improves the robustness of the system, effectively prevents equipment damage and enhances the operational safety under extreme scenarios.

[0037] (4) This invention supports adaptive switching of green electricity consumption, peak shaving and valley filling, grid support, and emergency supply modes, solving the problem of the lack of multi-scenario adaptability in existing technologies. For example, when there is a surplus of green electricity, it automatically absorbs the excess electricity, supplies power to the load to reduce costs during peak electricity prices, and maintains power supply to critical loads during grid failures. This capability maximizes the utilization rate of green electricity, while improving the grid's frequency and voltage regulation support and emergency supply capabilities. Attached Figure Description

[0038] Figure 1 This is a schematic diagram illustrating the core principle framework of the hierarchical collaborative control strategy in this invention;

[0039] Figure 2 This is a diagram illustrating the overall architecture of the green electricity direct-connected energy storage system in an embodiment of the present invention. Detailed Implementation

[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments. The embodiments of the present invention include, but are not limited to, the following embodiments.

[0041] like Figure 1 As shown, this invention discloses a charging and discharging coordinated control method for a green electricity direct-connected energy storage system. Its core lies in constructing a coordinated control architecture integrating multi-source sensing, hierarchical decision-making, and closed-loop feedback to achieve dynamic linkage control between distributed photovoltaic arrays, wind turbine generators, energy storage battery packs, and local loads. This method strictly and synchronously collects green electricity output, energy storage status parameters, and load demand data, and refines the instruction generation logic level by level in a three-layer control structure, ultimately driving the energy storage converter to execute precise power output. This improves system operating efficiency, extends equipment lifespan, and enhances grid support capabilities.

[0042] This method first requires acquiring real-time power output data from distributed photovoltaic arrays and wind turbine generators. Specifically, it uses a high-precision Hall current sensor and voltage divider network installed on the DC side of the photovoltaic inverter to collect the DC current and voltage signals output by the photovoltaic modules. After analog-to-digital conversion, the instantaneous active power is calculated. The filtered active power output value of the wind turbine is read through the communication interface of the wind turbine's main control system, with a sampling frequency of at least 100Hz. The Hall current sensor's range covers from zero to the maximum design short-circuit current, with an accuracy class of at least 0.5. The voltage divider network uses high-resistance precision resistors with a fixed voltage division ratio and a temperature drift compensation mechanism. The analog-to-digital conversion module has a resolution of at least 16 bits, and its sampling period is strictly synchronized with the system's main control clock to ensure time consistency in power calculation. The wind turbine's active power data is transmitted via the controller's local area network bus or industrial Ethernet protocol. The receiving end is configured with data verification and retransmission mechanisms to prevent power misjudgment due to communication packet loss.

[0043] Then, the state of charge (SOC), terminal voltage, charge / discharge current, temperature distribution, and health status assessment indicators of the energy storage battery pack are obtained. Specifically, the terminal voltage of each individual cell in the battery management system is obtained through a polling method using a single-cell voltage acquisition chip; the total charge / discharge current of the battery cluster is obtained through a shunt or Hall effect current sensor integrated on the bus; real-time temperature values ​​at no fewer than eight temperature measurement points are obtained through a digital temperature sensor array arranged within each module; the current SOC is calculated based on the open-circuit voltage method combined with the ampere-hour integral method, and a temperature compensation coefficient is introduced; and the battery health status assessment indicators are calculated based on the internal resistance incremental analysis method combined with historical cycle data. The single-cell voltage acquisition chip supports multi-channel synchronous sampling, with an inter-channel isolation voltage of no less than 1500V and a sampling error of less than ±5mV; the rated current of the shunt matches the maximum continuous charge / discharge current of the system, and the temperature drift coefficient is less than 10ppm / ℃; the digital temperature sensor uses a one-wire or two-wire digital interface, with a temperature measurement range covering -40℃ to +125℃ and an accuracy of ±0.5℃. During the state of charge calculation, the initial value of the ampere-hour integral is determined by looking up the open-circuit voltage under static conditions. The coulomb efficiency correction factor is added in real time during the integration process. This factor is dynamically adjusted according to the charge / discharge rate and temperature. The health status assessment index is obtained by monitoring the rate of change of the voltage slope at the end of each constant current charging stage. Its benchmark value is established from the factory calibration data and is updated every 50 complete charge / discharge cycles.

[0044] Simultaneously, the system acquires the current local load power demand and the load forecast curve for the future within a preset time window, i.e., by reading the total active power on the low-voltage side in real time through smart meters or power monitoring terminals. Based on the time-series characteristics of historical load data, a long short-term memory neural network model is used to perform rolling predictions of load power for the next 15 minutes to 2 hours, with a prediction step size of 5 minutes. The communication interface of the smart meter supports power line carrier or wireless radio frequency transmission, with a data reporting cycle of 1 second. The input features of the long short-term memory neural network include the load power every 5 minutes in the past 24 hours, the day of the week, the season identifier, and the daily maximum temperature and sunshine duration from the weather forecast data. The network structure contains two hidden layers, with 128 neurons in each layer, and the activation function is the hyperbolic tangent function. The model training adopts a sliding window mechanism, triggering online fine-tuning every 24 hours of newly added measured data, with a fixed learning rate of 0.001 and the loss function being the mean squared error.

[0045] This embodiment, based on the three types of data mentioned above, executes a hierarchical collaborative control strategy to generate charging and discharging power commands for the energy storage converter. The hierarchical collaborative control strategy includes an upper-level energy dispatch layer, a middle-level power allocation layer, and a lower-level device execution layer. The upper-level energy dispatch layer determines the system operating mode and the target charging and discharging power of the energy storage system based on the net load power, energy storage state-of-charge boundary constraints, and grid dispatch commands. The middle-level power allocation layer dynamically allocates the target charging and discharging power to each energy storage sub-cluster based on the energy storage battery pack's health status assessment indicators, temperature distribution uniformity, and individual cell voltage consistency. The lower-level device execution layer receives the allocated power commands and generates pulse-width modulation signals to drive the energy storage converter to execute precise power output.

[0046] The upper-level energy dispatch layer operates in four modes: green electricity priority consumption mode, peak shaving and valley filling mode, grid support mode, and emergency supply guarantee mode. When the system enters green electricity priority consumption mode, the energy storage system absorbs excess green electricity at its maximum allowable charging power. When the system enters peak shaving and valley filling mode, the energy storage system supplies power to the load at a set discharge power. When the system receives frequency or voltage regulation commands from the grid dispatch center, it enters grid support mode, and the energy storage system provides active or reactive power support as required by the commands. When an external grid power outage is detected and the energy storage's state of charge is higher than a safe threshold, the system enters emergency supply guarantee mode, and the energy storage system switches to off-grid operation to maintain continuous power supply to critical loads. The switching logic between modes is implemented using a state machine, and hysteresis intervals are set for the switching conditions to avoid frequent oscillations. For example, the exit condition for green electricity priority consumption mode is that the net load power is less than 0 for 10 consecutive seconds or the state of charge reaches 95%.

[0047] Under the green energy priority consumption mode, the target charging and discharging power is determined as follows: calculate the net load power, which is the total green energy output minus the local load power; if the net load power is greater than zero, the target charging power is the smaller value between the net load power and the maximum allowable charging power of the energy storage system; if the net load power is less than or equal to zero, the target charging power is zero. The maximum allowable charging power is calculated in real time by the battery management system based on the current state of charge, temperature, and health status, and is uploaded to the upper-level scheduling module via the communication bus.

[0048] In peak shaving and valley filling mode, the target charging and discharging power is determined as follows: The time-of-use (TOU) electricity price meter indicates whether the current period is peak. If it is peak and the energy storage state of charge (SBC) is above a preset lower threshold, the target discharge power is the smaller of the local load power and the maximum allowable discharge power of the energy storage system. If it is peak and the SBC is below the preset lower threshold, the target discharge power is zero. If it is valley and the SBC is below a preset upper threshold, the target charging power is the maximum allowable charging power of the energy storage system. If it is valley and the SBC is above the preset upper threshold, the target charging power is zero. The TOU electricity price meter is stored in local non-volatile memory and supports remote updates. Peak hours are defined as from 10:00 AM to 10:00 PM daily, and valley hours are the remaining times.

[0049] The specific method for the middle-layer power allocation layer to dynamically allocate the target charge / discharge power to each energy storage sub-cluster is as follows: The comprehensive weighting coefficient of each sub-cluster is calculated. This coefficient is jointly determined by the sub-cluster's health status assessment index, average temperature deviation, and individual cell voltage standard deviation. Sub-clusters with higher health status assessment indexes, average temperatures closer to the rated operating temperature, and smaller individual cell voltage standard deviations have larger comprehensive weighting coefficients. The target charge / discharge power is then allocated proportionally according to the comprehensive weighting coefficients of each sub-cluster. The formula for calculating the comprehensive weighting coefficient is:

[0050]

[0051] in, For the first The overall weighting coefficient of each sub-cluster Use health status assessment indicators (values ​​ranging from 0 to 100). Its average temperature, The rated operating temperature (usually 25℃). Its standard deviation of individual unit voltage, This refers to the rated voltage of a single battery cell (e.g., 3.2V). , , The normalization correction factor satisfies , This is the temperature sensitivity coefficient, with a value of 0.1.

[0052] Power allocation for each sub-cluster Calculate using the following formula:

[0053] in, The total number of subclusters, The target charge / discharge power.

[0054] The specific process of generating pulse width modulation (PWM) signals at the lower-level device execution layer includes: converting the power commands allocated to each subcluster into corresponding DC-side current reference values; employing a dual-loop control structure, with the outer loop being the power loop and the inner loop being the current loop; the output of the power loop serves as the input to the current loop, which uses a proportional-integral (PI) regulator, and its output is processed by a space vector modulation algorithm to generate six complementary PWM signals to drive the insulated-gate bipolar transistor (IGBT) bridge arms. The input to the power loop is the difference between the actual DC power and the commanded power, and the controller transfer function is... , where proportional gain With integration time constant Based on the online tuning of the converter's rated power, s is a complex variable of the Laplace Transform; the current loop sampling frequency is 20kHz, and the current sensor bandwidth is not less than 5kHz; the space vector modulation algorithm adopts a seven-segment switching sequence, with a fixed dead time of 2μs to suppress the risk of shoot-through between the upper and lower bridge arms.

[0055] During charging and discharging, the AC three-phase current, DC bus voltage, and radiator temperature of the energy storage converter are monitored in real time. If any monitored parameter exceeds a safety threshold, a tiered power reduction protection strategy is immediately activated. Level 1 protection reduces the power command by 10% for 5 seconds; Level 2 protection disconnects the contactor of the corresponding sub-cluster and isolates the faulty unit; Level 3 protection shuts down the entire system and reports a fault code. The AC current threshold is set to 1.2 times the rated current, the DC bus voltage threshold is ±10% of the rated voltage, and the radiator temperature threshold is 85℃. Protection actions are triggered by both hardware comparators and software logic, ensuring a response time of less than 10ms.

[0056] The method of this invention also includes a closed-loop feedback correction mechanism; at the end of each control cycle, the deviation value between the actual charging / discharging power and the commanded power is collected; this deviation value is input to the feedforward compensation module to generate a compensation amount that is superimposed on the power command of the next cycle; the feedforward compensation module adopts a first-order inertial element transfer function, and the time constant is tuned online according to the response characteristics of the energy storage system. The control cycle is fixed at 100ms, and the deviation value is sent to the compensation module after low-pass filtering; the transfer function expression is as follows: Where K is a fixed gain of 0.8, T is a time constant with an initial value of 0.5s, and s is a complex variable of the Laplace Transform, which is dynamically adjusted every 100 control cycles based on the root mean square value of the power tracking error, with an adjustment step size of 0.05s.

[0057] like Figure 2 As shown, the green electricity direct-connected energy storage system upon which the above method relies includes a distributed photovoltaic array, wind turbine generators, energy storage battery packs, energy storage converters, a battery management system, a load monitoring terminal, a grid interface unit, and a central coordinating controller. The distributed photovoltaic array is connected to the DC bus of the energy storage converter via a DC combiner box; the wind turbine generators are connected to the same AC bus via an AC converter; the energy storage battery pack consists of multiple parallel battery sub-clusters, each containing several series-connected battery modules; the energy storage converter adopts a three-phase two-level topology, with its rated power matched to the system's maximum green electricity output; the battery management system is independently deployed in each battery cluster and communicates with the central coordinating controller via a controller area network bus; the load monitoring terminal is installed in the user-side distribution cabinet, uploading the total active power in real time; the grid interface unit includes intelligent circuit breakers and power quality monitoring modules, supporting remote opening and closing and receiving dispatch commands; the central coordinating controller is an industrial-grade embedded computer running a real-time operating system and incorporating the aforementioned three-layer control algorithm modules.

[0058] The hardware platform of the central coordinating controller includes a multi-core processor, a high-speed Ethernet interface, analog input / output modules, and digital input / output modules. In the multi-core processor, one core is dedicated to upper-level energy dispatch calculations, another core handles mid-level power allocation, and a third core performs lower-level pulse-width modulation signal generation. The high-speed Ethernet interface connects to a remote monitoring center and supports the IEC61850 communication protocol. The analog input modules have a 16-bit sampling accuracy and a sampling rate of at least 10kHz. The digital output modules drive contactor coils and fault indicator lights, with an output contact capacity of at least 5A.

[0059] During system initialization, the central coordinating controller first loads the initial health status, rated parameters, and communication address list of the battery sub-clusters; then, it establishes heartbeat connections with each subsystem to verify the integrity of the data link; next, it reads the current time-of-use electricity price table, grid dispatch rules, and load forecasting model parameters; finally, it enters the main control loop, executing data acquisition, mode judgment, power calculation, allocation, and instruction issuance at 100ms intervals. Within the main control loop, each step is strictly scheduled according to time slices, ensuring that upper-level decisions are completed within 10ms, middle-level allocation within 20ms, and lower-level execution instructions are generated and output within the remaining 70ms.

[0060] During long-term operation, the system periodically performs self-test procedures, including battery cell voltage equalization testing, temperature sensor calibration, communication delay measurement, and control parameter drift detection. The self-test results are stored in non-volatile memory and used to update the health status assessment model and control gain parameters. When the health status of a subcluster continuously declines beyond a preset rate, the system automatically reduces its overall weighting coefficient limit, restricting its participation in high-power charging and discharging operations until manual intervention or replacement is required.

[0061] In summary, this embodiment achieves efficient, safe, and long-life operation of the green electricity direct-connected energy storage system under complex operating conditions through a rigorous three-layer control architecture, multi-source data fusion, dynamic weight allocation, and closed-loop feedback mechanism. All technical details are implemented with deterministic logic, without any optional or exemplary descriptions, ensuring a clear and unambiguous scope of patent protection.

[0062] The above embodiments are merely one of the preferred embodiments of the present invention and should not be used to limit the scope of protection of the present invention. Any modifications or refinements made to the main design concept and spirit of the present invention that are not of substantial significance, but solve the same technical problem as the present invention, should be included within the scope of protection of the present invention.

Claims

1. A method for coordinated charge and discharge control of a green electricity direct-connected energy storage system, characterized in that, Includes the following steps: S1, acquire real-time power output data of distributed photovoltaic arrays and wind turbine generators; S2, obtain the state of charge, terminal voltage, charge and discharge current, temperature distribution and health status assessment indicators of the energy storage battery pack; S3, obtain the local load power demand at the current moment and the load forecast curve within the future preset time window; S4, perform multi-source sensing fusion processing on the real-time power output data, energy storage battery pack status parameters and local load power demand data to obtain a fusion characterization of the system operating status; S5. Based on the fusion characterization of the system's operating status, execute a hierarchical collaborative control strategy to generate charging and discharging power commands for the energy storage converter.

2. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 1, characterized in that, The hierarchical collaborative control strategy includes an upper-level energy dispatch layer, a middle-level power allocation layer, and a lower-level device execution layer. The upper-level energy dispatch layer determines the system operation mode and the target charging and discharging power of the energy storage system based on the net load power, energy storage state of charge boundary constraints, and grid dispatch instructions. The middle-level power allocation layer dynamically allocates the target charging and discharging power to each energy storage sub-cluster based on the health status assessment indicators of the energy storage battery pack, temperature distribution uniformity, and individual cell voltage consistency. The lower-level device execution layer receives the allocated power instructions and generates pulse width modulation signals to drive the energy storage converter to perform precise power output.

3. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 2, characterized in that, The operation modes of the upper-level energy dispatch layer include green electricity priority consumption mode, peak shaving and valley filling mode, grid support mode, and emergency supply guarantee mode; among them; In the green electricity priority consumption mode, the energy storage system absorbs excess green electricity at the maximum allowable charging power; In peak shaving and valley filling mode, the energy storage system supplies power to the load at a set discharge power. In the grid support mode, the energy storage system provides active or reactive power support according to the instructions. In emergency power supply mode, the energy storage system switches to off-grid operation to maintain continuous power supply to critical loads.

4. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 3, characterized in that, Under the green electricity priority consumption mode, the target charging and discharging power is determined as follows: calculate the net load power, that is, the total green electricity output minus the local load power; if the net load power is greater than zero, the target charging power is the smaller value between the net load power and the maximum allowable charging power of the energy storage system; if the net load power is less than or equal to zero, the target charging power is zero.

5. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 3, characterized in that, In peak shaving and valley filling mode, the target charging and discharging power is determined as follows: the current peak period is identified according to the time-of-use electricity price table; if the peak period is in effect and the energy storage state of charge is higher than the preset lower limit threshold, then the target discharge power is the smaller value between the local load power and the maximum allowable discharge power of the energy storage system. If it is during peak hours and the energy storage state of charge is below the preset lower threshold, the target discharge power is zero. If it is during a low-temperature period and the energy storage state of charge is below the preset upper limit threshold, the target charging power is the maximum allowable charging power of the energy storage system. If it is during a low-temperature period and the energy storage state of charge is higher than the preset upper limit threshold, the target charging power will be zero.

6. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 3, characterized in that, The specific method by which the middle power distribution layer dynamically allocates the target charging and discharging power to each energy storage sub-cluster is as follows: calculate the comprehensive weighting coefficient of each sub-cluster, which is jointly determined by the sub-cluster's health status assessment index, average temperature deviation, and individual cell voltage standard deviation; the higher the health status assessment index, the closer the average temperature is to the rated operating temperature, and the smaller the individual cell voltage standard deviation, the larger the comprehensive weighting coefficient of the sub-cluster; and allocate the target charging and discharging power proportionally according to the comprehensive weighting coefficient of each sub-cluster.

7. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 6, characterized in that, The formula for calculating the comprehensive weighting coefficient is as follows: in, For the first The overall weighting coefficient of each sub-cluster Assess their health status indicators. Its average temperature, For the rated operating temperature, Its standard deviation of individual unit voltage, The rated voltage of a single battery cell. , , The normalization correction factor satisfies , This is the temperature sensitivity coefficient.

8. The charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 2, characterized in that, The specific process by which the lower-level device executes the generation of pulse width modulation signals includes: The power commands allocated to each sub-cluster are converted into corresponding DC-side current reference values; It adopts a dual closed-loop control structure, with the outer loop being the power loop and the inner loop being the current loop; The output of the power loop serves as the input of the current loop, which employs a proportional-integral regulator. Its output is used to generate six complementary pulse width modulation signals via a space vector modulation algorithm, which drive the insulated-gate bipolar transistor bridge arms.

9. A charging and discharging coordinated control method for a green electricity direct-connected energy storage system according to claim 7 or 8, characterized in that, It also includes a closed-loop feedback correction mechanism: At the end of each control cycle, the deviation between the actual charging / discharging power and the commanded power is collected; The deviation value is input to the feedforward compensation module, which generates a compensation amount that is added to the power command of the next cycle. The feedforward compensation module adopts a first-order inertial element transfer function, and the time constant is tuned online according to the response characteristics of the energy storage system.

Images