Energy storage cluster primary frequency modulation control method and system based on analog quantity control

Abstract

The invention belongs to the technical field of energy storage cluster primary frequency modulation control, and particularly discloses an energy storage cluster primary frequency modulation control method and system based on analog quantity control, and the method comprises the steps: collecting a power grid frequency signal in real time, and calculating a frequency deviation; identifying a frequency disturbance type, dynamically determining a frequency modulation coefficient, and calculating a primary frequency modulation power instruction based on the frequency deviation and the frequency modulation coefficient; performing power amplitude limiting and distribution optimization on the primary frequency modulation power instruction according to the charge state of each battery pack in the energy storage cluster; and the optimized power instruction is converted into a 4-20 mA current signal, the 4-20 mA current signal is broadcasted and sent to all the energy storage converters, each energy storage converter dynamically adjusts a power distribution coefficient according to the difference between the own charge state and the average charge state of the cluster, and primary frequency modulation control of the energy storage cluster is completed. According to the invention, the problems of insufficient response speed, asynchronous cluster operation, unbalanced SOC management, insufficient anti-interference capability and insufficient disturbance adaptive capability in the prior art are solved.

Description

Technical Field

[0001] This invention belongs to the field of primary frequency regulation control technology for energy storage clusters, specifically relating to a primary frequency regulation control method and system for energy storage clusters based on analog quantity control. Background Technology

[0002] Primary frequency regulation (PFRD) is the first stabilization measure initiated by a power system when frequency deviations occur. Its main objective is to suppress further frequency deviations from the rated value (e.g., 50 Hz) by rapidly adjusting the active power on the power source side. Traditional PFRD often relies on the speed control systems of thermal and hydroelectric power units, adjusting mechanical power output to follow frequency changes. However, due to limitations in the combustion process and mechanical inertia, its power ramp-up or ramp-down speed is typically between hundreds of milliseconds and several seconds, which is insufficient to meet the sub-millisecond rapid response requirements of modern power grids.

[0003] With the increasing proportion of renewable energy and the decrease in system inertia, the requirements for the response speed and accuracy of the power grid in primary frequency regulation have significantly increased. Energy storage systems (ESS) have become an important technical means for primary frequency regulation due to their millisecond-level power regulation capabilities. However, in practical applications, energy storage systems are often connected to the grid in a cluster mode, that is, multiple energy storage converters (PCS) and battery packs are connected in parallel to the same AC bus to meet the demand for larger capacity and power.

[0004] In primary frequency regulation control of energy storage, power commands are typically calculated by the Energy Management System (EMS) and distributed to each PCS. Currently, the mainstream approach is to distribute power commands to each PCS via digital communication protocols (such as Modbus TCP and IEC 61850). While this method facilitates information exchange, it suffers from communication delays, long command parsing times, and limited data refresh rates, making it difficult for the cluster response time to exceed 200–500 ms. Furthermore, multiple PCS independently receiving and parsing commands can easily lead to asynchronous responses, uneven power distribution, and overload of some units.

[0005] Analog control interfaces (4–20 mA current signals or 0–10 V voltage signals) are widely used in industrial automation due to their low transmission delay and strong anti-interference capabilities. However, in primary frequency regulation of energy storage, analog interfaces are mostly used for auxiliary control of a single PCS, and there are few integrated analog control schemes for energy storage clusters, as well as a lack of unified SOC equalization management and cluster power consistency optimization strategies.

[0006] In addition, existing primary frequency regulation systems for energy storage lack the ability to identify disturbance types in the signal processing stage. They cannot dynamically adjust the response coefficient based on the duration of frequency changes and waveform characteristics (such as harmonic components), which may lead to over-response to short-term fluctuations, increasing the number of energy storage charge and discharge cycles and the amplitude of power fluctuations, thus affecting equipment lifespan and system stability. Summary of the Invention

[0007] The purpose of this invention is to address the problems of insufficient response speed, asynchronous operation of energy storage clusters, unbalanced SOC management, insufficient anti-interference capability, and lack of disturbance adaptive capability in existing technologies. It proposes a primary frequency regulation control method and system for energy storage clusters based on analog quantity control.

[0008] The technical solution of the present invention is as follows: Firstly, a primary frequency regulation control method for an energy storage cluster based on analog quantity control, comprising the following steps: Real-time acquisition of power grid frequency signals and calculation of frequency deviation; Analyze the power grid frequency signal using Fast Fourier Transform to identify the type of frequency disturbance; Based on the type of frequency disturbance, the frequency modulation coefficient is dynamically determined, and then the primary frequency modulation power command is calculated based on the frequency deviation and the frequency modulation coefficient. Based on the state of charge of each battery pack in the energy storage cluster, the SOC equalization strategy is used to optimize the power limiting and allocation of the primary frequency regulation power command, resulting in the optimized power command. The optimized power command is converted into an analog signal and broadcast to all energy storage converters in the energy storage converter cluster. In response to analog signals, each energy storage converter adjusts its charging and discharging power within milliseconds and dynamically adjusts the power distribution coefficient based on the difference between its own state of charge and the average state of charge of the cluster, thus completing the primary frequency regulation control of the energy storage cluster.

[0009] The beneficial effects of this invention are: When generating power commands, this invention dynamically adjusts the power command based on the SOC status of each battery pack to prevent overcharging or over-discharging of individual energy storage units, thereby improving the overall available capacity and lifespan. It also identifies the type of frequency disturbance and dynamically adjusts the frequency modulation coefficient to achieve both speed and stability in frequency modulation. Simultaneously, this invention broadcasts analog signals to all PCSs to ensure that each unit in the cluster receives the same power command, and within the PCS, it combines SOC and rated power to achieve proportional power sharing, ensuring consistent power response.

[0010] Preferably, after identifying the frequency disturbance type, the frequency modulation coefficient is dynamically determined according to the frequency disturbance type, and before calculating the frequency modulation power command based on the frequency deviation and the frequency modulation coefficient, the method further includes dead zone and amplitude limiting processing, specifically: when |Δf| is less than a preset dead zone threshold, the frequency modulation action is not triggered; when the communication interruption lasts for more than a set time, the last valid power command remains unchanged; Δf represents the frequency deviation.

[0011] The beneficial effects of the above preferred solution are: Enhanced anti-interference capability. This invention incorporates dead-time, filtering, and signal hold functions between the PLC and the analog module to suppress frequent charging and discharging caused by high-frequency noise and short-term disturbances, thereby improving the stability and reliability of system operation.

[0012] As a preferred method, the specific method for dynamically determining the frequency modulation coefficient Kp is as follows: Within the preset sliding time window T w The acquired power grid frequency signal is subjected to a Fast Fourier Transform (FFT) to obtain the frequency domain spectrum. Based on frequency domain spectral statistics, the spectral energy of the first frequency band and the spectral energy of the second frequency band are statistically analyzed, and the proportion of high-frequency energy and the proportion of low-frequency energy are calculated based on the spectral energy of the first frequency band and the spectral energy of the second frequency band. If the proportion of low-frequency energy is not less than the first threshold θ L When the duration reaches the first judgment period, it is judged as a continuous load disturbance, and the frequency modulation coefficient Kp is increased from the reference value Kp0 to Kp_up, and Kp_up = min(Kp_max, γ_up·Kp0); where γ_up is the amplification coefficient and Kp_max is the upper limit of the frequency modulation coefficient; If the proportion of high-frequency energy is not less than the second threshold θ H Furthermore, when the duration reaches the second judgment period, it is judged as transient or harmonic interference, and the frequency modulation coefficient Kp is reduced from the reference value Kp0 to Kp_down, and Kp_down = max(Kp_min, γ_down·Kp0); where Kp_min is the lower limit of the frequency modulation coefficient, and γ_down is the attenuation coefficient; If the proportion of high-frequency energy is not less than the shielding threshold θ block If the high-frequency disturbance continues for longer than the shielding determination time, the frequency modulation response will be shielded and the frequency modulation coefficient Kp will be set to zero to maintain the shielding time.

[0013] The beneficial effects of the above preferred solution are: This invention introduces disturbance feature identification and adaptive adjustment. It combines FFT analysis and other methods to identify frequency disturbance types, improves the response coefficient for persistent frequency deviations, and reduces the response amplitude for transient disturbances, balancing frequency modulation speed and system stability while avoiding over-adjustment.

[0014] As a preferred approach, the SOC equalization strategy employs a power allocation algorithm of "weighted allocation + cut-off protection," specifically: Discharge allocation: If the state of charge (SOC) of any battery pack is lower than the first SOC threshold, the discharge allocation weight of that battery pack is reduced; if the SOC of any battery pack is lower than the second SOC threshold, the discharge allocation weight of that battery pack is reset to zero; where SOC_cut_low... <SOC_low; Charging allocation: If the state of charge of any battery pack is higher than the third state of charge threshold SOC_high, the charging allocation weight of that battery pack is reduced; if the state of charge of any battery pack is higher than the fourth state of charge threshold SOC_cut_high, the charging allocation weight of that battery pack is reset to zero; where SOC_cut_high>SOC_low.

[0015] The beneficial effects of the above preferred solution are: This invention enables balanced SOC management. On the EMS side, it dynamically adjusts the power allocation strategy based on the cluster's SOC status, preventing individual units from prematurely exiting frequency regulation due to SOC anomalies. This achieves balanced utilization and extended lifespan of the cluster's battery capacity, thereby improving overall available capacity.

[0016] Preferably, the analog signal is a 4–20mA current signal. In response to the 4–20mA current signal, each energy storage converter converts the 4–20mA current signal into a digital power command, and superimposes local SOC protection logic and power limiting constraints to dynamically adjust the power allocation coefficient and complete the primary frequency regulation control of the energy storage cluster.

[0017] The beneficial effects of the above preferred solution are: Achieving millisecond-level rapid response. This invention broadcasts power commands in parallel via a 4–20 mA analog interface, combined with the millisecond-level execution capability of the PCS, achieving an end-to-end response time of <50 ms, significantly shortening the frequency modulation intervention time and substantially improving frequency stability.

[0018] Preferably, the 4–20mA current signal is generated by a programmable logic controller (PLC) or a dedicated analog output card and transmitted in parallel to the analog input ports of each energy storage converter via shielded twisted-pair cables.

[0019] The beneficial effects of the above preferred solution are: To ensure consistent cluster operation, this invention broadcasts analog signals to all PCS, synchronously performs power adjustments, and performs proportional corrections based on the unit's SOC, ensuring consistent cluster power allocation and a smooth frequency modulation process.

[0020] Secondly, a primary frequency regulation control system for an energy storage cluster based on analog signal control includes: a frequency acquisition module, an energy management subsystem, a programmable logic controller (PLC) and an FFT analysis module, an analog output module, an energy storage converter cluster, a battery pack, a step-up transformer, and a grid-connected bus subsystem; the energy management subsystem is electrically connected to the frequency acquisition module, the PLC and FFT analysis module, the energy storage converter cluster, and the analog output module; each energy storage converter in the energy storage converter cluster is electrically connected to its corresponding battery pack; and the PLC and FFT analysis module are connected to the step-up transformer and the grid-connected bus subsystem. Frequency acquisition module, used to acquire power grid frequency signals in real time; The energy management subsystem is used to calculate frequency deviation, generate primary frequency modulation power commands, and perform SOC equalization management. The programmable logic controller and FFT analysis module are used to analyze power grid frequency signals through fast Fourier transform and identify the type of frequency disturbance. The analog output module is used to convert power commands into 4–20mA current signals and broadcast them to the energy storage converter cluster. Energy storage converter clusters are used to receive analog commands and perform power regulation; Battery packs are used for energy storage and release. The step-up transformer and grid bus are used to step up the voltage of the energy storage converter to the grid voltage and connect it to the grid.

[0021] The beneficial effects of this invention are: This invention clarifies the responsibilities and connections of each module, constructs a complete and hierarchical control system, and realizes full-chain coordination from frequency acquisition, signal processing, power optimization to command issuance and execution, ensuring the smooth and efficient primary frequency regulation control of the energy storage cluster.

[0022] Preferably, the DC side of each energy storage converter is electrically connected to the corresponding battery pack, and the AC side of each energy storage converter is connected in parallel to the step-up transformer and the grid-connected bus in the grid-connected bus subsystem.

[0023] The beneficial effects of the above preferred solution are: The system adopts a corresponding connection method of one energy storage converter and one battery pack, which facilitates system expansion and maintenance. At the same time, it is connected in parallel to the grid bus, which simplifies the electrical structure and reduces system complexity.

[0024] Preferably, the analog output module is a multi-channel isolated output, using a star or bus signal distribution topology.

[0025] The beneficial effects of the above preferred solution are: Multi-channel isolated outputs avoid signal crosstalk and ground loop interference, improving signal purity and stability; star or bus topology facilitates wiring and expansion, adapting to energy storage clusters of different sizes. Attached Figure Description

[0026] Figure 1 The diagram shows a flowchart of a primary frequency regulation control method for an energy storage cluster based on analog quantity control. Detailed Implementation

[0027] Exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the embodiments shown and described in the drawings are merely exemplary and are intended to illustrate the principles and spirit of the invention, and are not intended to limit the scope of the invention.

[0028] Explanation of key terms: Primary frequency regulation (PFRD) is the process by which a power system, when its frequency deviates from its rated value, rapidly adjusts the active power output of generators or energy storage devices to suppress frequency fluctuations and restore the system frequency. PFRD requires intervention in the initial stages of frequency disturbances, with a response time typically of several seconds or less.

[0029] Energy Storage System (ESS): Composed of battery packs (such as lithium batteries, sodium-ion batteries, etc.), battery management system (BMS), temperature control system, etc. Energy storage systems can absorb or release electrical energy in milliseconds to seconds for grid frequency regulation, peak shaving, backup, and power quality optimization.

[0030] Power Conversion System (PCS): A bidirectional power conversion device that converts the DC power from the energy storage battery to the AC power from the grid. It can both charge (absorb electrical energy) and discharge (release electrical energy), and can adjust the output power in milliseconds according to external control commands.

[0031] Battery Management System (BMS): Used to monitor the voltage, current, temperature, SOC and other states of the battery pack, perform overvoltage, undervoltage, overcurrent and overtemperature protection, and exchange information with PCS and EMS.

[0032] SOC (State of Charge): Represents the percentage of a battery's remaining usable capacity relative to its rated capacity, and is a key parameter for energy storage operation scheduling and protection. In frequency regulation control, SOC is used for power limiting, start-stop strategies, and equalization control.

[0033] Frequency deviation (Δf): The difference between the real-time operating frequency of the power grid and the rated frequency (such as 50 Hz or 60 Hz), in Hz or mHz, used to determine the direction and amplitude of frequency modulation.

[0034] Frequency Regulation Coefficient (Kp): A proportional coefficient that converts frequency deviation into active power command, measured in kW / Hz. The larger the Kp value, the stronger the frequency regulation response.

[0035] Analog Control Interface: An industry-standard physical signal interface that transmits power commands or frequency deviation information via 4–20 mA current signals or 0–10 V voltage signals. It features low transmission delay, strong anti-interference capability, and high real-time performance.

[0036] 4–20 mA current signal: A commonly used industrial analog signal transmission standard, where 4 mA typically represents the lower limit of the range and 20 mA represents the upper limit. Suitable for long-distance, interference-resistant analog signal transmission.

[0037] 690 V Bus: The AC busbar on the low-voltage side of the energy storage converter, which collects the output power of each PCS and steps it up to the high-voltage side and connects it to the power grid via a step-up transformer.

[0038] Step-up transformer: The 690 V bus voltage is increased to the medium or high voltage level required for grid connection (such as 10 kV or 35 kV) in order to supply power to the grid or load.

[0039] Energy Management System (EMS): The core dispatching system of an energy storage power station, responsible for collecting operating data such as frequency, voltage, and power, calculating frequency regulation power commands, and combining SOC and operating strategies to perform power allocation and operation optimization.

[0040] PLC (Programmable Logic Controller Module): A programmable logic controller used for high-speed acquisition of voltage and current signals and performing logic processing, such as dead-time detection, FFT analysis, and signal filtering. The real-time performance and stability of PLCs make them suitable for frequency modulation signal processing.

[0041] FFT (Fast Fourier Transform Analysis): The Fast Fourier Transform algorithm decomposes a time-domain signal into frequency-domain components, making it easier to identify frequency disturbance characteristics (such as persistent deviations, harmonic components, etc.), thereby optimizing the frequency modulation control strategy.

[0042] AVC (Automatic Voltage Control): An automatic control system used in power systems to regulate and stabilize bus voltage. It is usually used in conjunction with frequency regulation control to ensure that frequency and voltage meet requirements simultaneously.

[0043] AGC (Automatic Generation Control): A control system that automatically adjusts the power output of generators or energy storage to maintain grid frequency stability and optimize unit output distribution. Primary frequency regulation signals can be combined with AGC to achieve comprehensive control of frequency and active power.

[0044] Deadband: When the absolute value of the frequency deviation is less than a certain set threshold, no power adjustment is performed to avoid frequent charging and discharging caused by small fluctuations.

[0045] Filter Time Constant: A parameter that controls the response speed of a signal filter, measured in milliseconds (ms). It is used in frequency modulation to suppress high-frequency noise signals.

[0046] Power Limiting: Limits the frequency regulation power command to a range within which the energy storage system can operate safely. It is usually determined by the PCS rated power and the upper and lower limits of SOC.

[0047] Cluster Power Consistency Control: In an energy storage cluster, a unified power command and SOC equalization algorithm are used to ensure that each PCS shares the power load equally in the frequency regulation response, avoiding single-unit overload or insufficient response.

[0048] Example 1: like Figure 1 As shown, a primary frequency regulation control method for an energy storage cluster based on analog quantity control includes the following steps: S1. Real-time acquisition of power grid frequency signals, and calculation of frequency deviation Δf; S2. Analyze the power grid frequency signal using Fast Fourier Transform (FFT) to identify the type of frequency disturbance; S3. Based on the frequency deviation Δf and the type of frequency disturbance, dynamically determine the frequency modulation coefficient Kp and calculate the primary frequency modulation power command P_set; S4. Based on the state of charge (SOC) of each battery pack in the energy storage cluster, the power limit and allocation of the primary frequency regulation power command P_set are optimized to obtain the optimized power command; S5. The optimized power command is converted into an analog signal through the analog output module and broadcast to all energy storage converters in the energy storage converter (PCS) cluster; S6. In response to analog signals, each energy storage converter adjusts its charging and discharging power within milliseconds and dynamically adjusts the power distribution coefficient based on the difference between its own SOC and the average SOC of the cluster, thus completing the primary frequency regulation control of the energy storage cluster.

[0049] In this embodiment, after identifying the frequency disturbance type, the frequency modulation coefficient is dynamically determined according to the frequency disturbance type. Before calculating the frequency modulation power command based on the frequency deviation and the frequency modulation coefficient, the method also includes dead zone and limiting processing. Specifically, when |Δf| is less than a preset dead zone threshold, the frequency modulation action is not triggered; when the communication interruption lasts for more than a set time, the last valid power command remains unchanged; Δf represents the frequency deviation.

[0050] In this embodiment, the method for dynamically determining the frequency modulation coefficient Kp is as follows: Within the preset sliding time window T w The acquired power grid frequency signal is subjected to Fast Fourier Transform (FFT) to obtain the frequency domain spectrum; the spectral energy EL of the first frequency band and the spectral energy EH of the second frequency band are statistically analyzed, and the high-frequency energy ratio RH=EH / (EL+EH) and the low-frequency energy ratio RL=EL / (EL+EH) are calculated; when RL is not less than the first threshold θ L If the load continues to be disturbed for the first determination period TL, it is determined to be a persistent load disturbance, and the frequency regulation coefficient Kp is increased from the baseline value Kp0 to Kp_up, where Kp_up = min(Kp_max, γ_up·Kp0); when RH is not less than the second threshold θ H If the interference persists for the second determination time TH, it is determined to be transient or harmonic interference. The frequency modulation coefficient Kp is reduced from the reference value Kp0 to Kp_down, and Kp_down = max(Kp_min, γ_down·Kp0); when RH is not less than the shielding threshold θ block If the high-frequency disturbance continues for more than the shielding determination time Tblock, the frequency modulation response is shielded and Kp is set to zero to maintain the shielding time Thold; where γ_up is the amplification coefficient, γ_down is the attenuation coefficient, and Kp_max and Kp_min are the upper and lower limits of the frequency modulation coefficient.

[0051] The determination of "predominantly low-frequency components / high-frequency components exceeding the threshold" is implemented as follows: frequency sample values ​​are collected at a sampling frequency fs, and a sliding time window T is used. w Perform spectrum analysis; T wThe time interval can be 1–10 s (preferably 2–5 s), and fs can be 50–200 Hz. After removing the DC component and windowing the frequency sequence within each time window, perform an FFT to obtain the frequency domain amplitude or power spectrum. Divide the frequency domain into a first frequency band and a second frequency band. The first frequency band is used to characterize persistent frequency deviation (low-frequency slow-changing) components, and the second frequency band is used to characterize transient fluctuations, switching noise, or fast-changing components introduced by harmonics. As an example, the first frequency band can be set to 0–fL (fL can be 0.1–0.3 Hz), and the second frequency band can be set to fH–fN (fH can be 0.5–1.0 Hz, fN is the Nyquist frequency or a preset upper limit such as 5–10 Hz). Integrate / accumulate the spectral energy within the two frequency bands to obtain EL and EH, and calculate RL = EL / (EL+EH) and RH = EH / (EL+EH).

[0052] When RL≥θ L If the TL condition is consistently met (e.g., 0.5–3 s, corresponding to several consecutive sliding window judgments), the frequency disturbance is considered to be mainly low-frequency slow-changing and is judged as a continuous load disturbance. In this case, the frequency modulation coefficient is increased from the baseline value Kp0 to Kp_up = min(Kp_max, γ_up·Kp0), where γ_up can be 1.2–2.0, and Kp_max is used to limit excessive response. Correspondingly, when RH ≥ θ... H If the TH condition is consistently met (e.g., 0.2–2 s), the high-frequency component is considered to be significant, and the interference is judged to be transient or harmonic interference. At this point, the frequency modulation coefficient is reduced to Kp_down = max(Kp_min, γ_down·Kp0), where γ_down can be 0.2–0.8, and Kp_min is used to ensure that the necessary frequency modulation capability is still maintained. Furthermore, when RH ≥ θ... block If the high-frequency determination continues for more than Tblock (e.g., 0.2–1 s), the shielding logic is executed: Kp is set to zero and held for Thold (e.g., 0.5–5 s) to avoid frequent charging and discharging caused by short-term noise / harmonics; after the shielding period expires, Kp is restored to Kp0 or Kp is re-determined and updated according to the above rules. To avoid jitter near the threshold, θ can be used. L θ H θ block Introduce hysteresis interval and minimum hold time.

[0053] Regarding the determination of the threshold, in one embodiment, θ L θ H θ block This can be obtained through "steady-state noise calibration + margin": Frequency signals are collected during periods of relatively stable grid operation without significant disturbances, and then calibrated according to the same T... w, the fs calculates the statistical distribution of RH, takes its mean value μ and standard deviation σ or quantile (e.g., 95% quantile), and then adds a preset margin δ (e.g., 0.05 - 0.15) to obtain θ H and θ block , to ensure that reduction or shielding is triggered only when the high-frequency energy is significantly higher than the normal noise level; θ L can take 1 θ H or be determined by calibrating RL in the same way. In another embodiment, the above threshold can also be updated online according to the historical operation data in the station, grid connection specification requirements or operation and maintenance settings, but it needs to be restricted by upper and lower limits to ensure control stability.

[0054] In this embodiment, the SOC balancing strategy is implemented by a power distribution algorithm of "weight reduction allocation + cut-off protection". The EMS obtains the SOC_i of N battery packs in the cluster and calculates SOC_avg. It is processed in two cases according to the direction of P_set, specifically: Discharge allocation (P_set > 0): If the state of charge of any battery pack is lower than the first state of charge threshold SOC_low, reduce the discharge allocation weight of this battery pack; if the state of charge of any battery pack is lower than the second state of charge threshold SOC_cut_low, set the discharge allocation weight of this battery pack to zero; Charge allocation (P_set < 0): If the state of charge of any battery pack is higher than the third state of charge threshold SOC_high, reduce the charge allocation weight of this battery pack; if the state of charge of any battery pack is higher than the fourth state of charge threshold SOC_cut_high, set the charge allocation weight of this battery pack to zero.

[0055] To give an example of "reducing the allocation ratio" that can be implemented, the weight reduction coefficient η_i can be determined in a piecewise linear manner: During discharge: If SOC_i ≥ SOC_low, then η_i = 1; if SOC_cut_low < SOC_i < SOC_low, then η_i linearly decreases with SOC_i in the interval [η_min, 1]; if SOC_i ≤ SOC_cut_low, then η_i = 0; During charge: If SOC_i ≤ SOC_high, then η_i = 1; if SOC_high < SOC_i < SOC_cut_high, then η_i linearly decreases with SOC_i in the interval [η_min, 1]; if SOC_i ≥ SOC_cut_high, then η_i = 0; where η_min is the minimum retained weight, which is used to avoid the sudden drop of the single unit weight causing allocation jumps.

[0056] After obtaining η_i, the EMS normalizes η_i to obtain the allocation coefficient λ_i=η_i / Ση_i, thereby decomposing the cluster primary frequency modulation power command into the power command P_i=λ_i·P_set of each battery pack; and superimposes the available power upper limit (such as the maximum charging power P_ch,max,i and the maximum discharging power P_dis,max,i) given by the BMS / PCS of each battery pack for limiting. If there is residual power after limiting, the battery packs that have not been triggered for cut-off are re-normalized and allocated according to the same rules until the power balance is satisfied.

[0057] Regarding threshold settings: SOC_cut_low and SOC_cut_high are preferably taken from the BMS safe operating range (e.g., SOC_min and SOC_max) to prevent over-discharge / over-charge; SOC_low and SOC_high are used for operation equalization and frequency regulation margin reservation, and can be set according to the target SOC_ref and the allowable SOC difference ΔSOC_allow. For example, SOC_ref can be taken as the target value (which can be set to 50% or the operation and maintenance setting value), and SOC_low = SOC_ref. Δb and SOC_high = SOC_ref + Δb, where Δb is half of the equalization bandwidth. Δb can be taken as 5% to 20% to ensure up / down frequency modulation margin; a smaller Δb is used when stronger equalization is needed, and a larger Δb is used when a larger frequency modulation margin is needed. To avoid frequent switching near the threshold, hysteresis and minimum hold time can be introduced for SOC_low and SOC_high.

[0058] (Numerical Example) For example: SOC_cut_low=10%, SOC_cut_high=90%, SOC_ref=50%, SOC_low=40%, SOC_high=60%, η_min=0.2. During discharge frequency regulation, if a battery pack has SOC_i=35%, then its η_i is reduced within the range [0.2,1] (e.g., linearly reduced to 0.6), and the discharge allocation ratio obtained by the battery pack after normalization is correspondingly reduced; if SOC_i≤10%, then η_i=0, and the battery pack does not participate in discharge frequency regulation.

[0059] In this embodiment, the analog signal is a 4–20 mA current signal. In response to the 4–20 mA current signal, each energy storage converter converts the 4–20 mA current signal into a digital power command, and superimposes local SOC protection logic and power limiting constraints to dynamically adjust the power allocation coefficient and complete the primary frequency regulation control of the energy storage cluster.

[0060] In this embodiment, the 4–20 mA current signal is generated by a programmable logic controller (PLC) or a dedicated analog output card, and is transmitted in parallel to the analog input ports of each energy storage converter via shielded twisted-pair cables.

[0061] In this embodiment, the 4–20 mA current signal can be replaced with 0–10 V, ±10 V, 0–5 V, frequency / duty cycle encoded (PWM / PPM), F / V (frequency-to-voltage) conversion signal, or analog signal with HART / Fieldbus overlay; essentially, it is still a rapid analog signal transmission.

[0062] In this embodiment, the optimized power command broadcast can be a power setting P_set, a power increment ΔP, a charge / discharge current command I_set, or a primary frequency modulation slope (droop) parameter, which is converted locally by the PCS.

[0063] In this embodiment, a single AO channel is distributed in a "star" configuration via a multi-channel isolation distributor / signal replicator; or transmitted in a "bus-type / cascaded" configuration; or each PCS has a built-in A / D converter that is output and connected to the EMS via near-end opto-isolation.

[0064] In this embodiment, redundant channels are set up: A / B dual analog hot standby; when the analog signal is abnormal, it automatically falls back to the local droop mode or switches to the digital backup link (without changing the idea of ​​unified distribution).

[0065] In this embodiment, while maintaining "unified broadcasting + synchronous execution", high-speed protocols such as IEC61850 GOOSE / SV with PTP clock alignment, synchronous Ethernet, and CAN FD broadcast can be used; or broadcast Δf / dfdt, and the PCS locally calculates the power according to unified parameters.

[0066] In this embodiment, the control law can be a proportional P replaced by PI / PID, a P+FF with feedforward (based on Δf and RoCoF=df / dt), or a piecewise linear / nonlinear gain Kp (a function of |Δf|, df / dt, SOC, and temperature).

[0067] In this embodiment, the filtering / dead-time replacement is as follows: the first-order low-pass filter can be replaced with EMA, Kalman filter, Savitzky-Golay filter, or limiting median filter; the dead-time filter can be replaced with hysteresis filter or adaptive dead-time filter.

[0068] In this embodiment, vibration suppression and slope limiting are achieved by adding dP / dt limiting, lead-lag compensation, and virtual inertia / damping injection to improve stability.

[0069] In this embodiment, the hierarchical / consistent allocation is as follows: the EMS only distributes the total power of the cluster, and the underlying PCS self-allocates the power according to weights (SOC / SOH / temperature / internal resistance) through a consensus or averaging algorithm.

[0070] In this embodiment, the equalization strategy is replaced by distributed weighting instead of centralized amplitude limiting; priority is given to using cells with high SOH / low internal resistance; and a "virtual SOC buffer" is set up to avoid boundary oscillations.

[0071] In this embodiment, the hybrid energy storage frequency division is as follows: the high and low frequencies are shared by the battery and the supercapacitor / flywheel (the high frequency is handled by the supercapacitor, and the low frequency is handled by the battery), and they are all subject to the same analog / broadcast command scheduling.

[0072] In this embodiment, the dynamic capability curves, P_max and I_max, are adaptive to temperature, SOH, and terminal voltage; the controller employs inverse integral saturation (anti-windup) and soft limiting.

[0073] In this embodiment, the frequency signal can come from the PMU, relay protection device, PCS local PLL, and multiple sources can be switched via voting / selection.

[0074] In this embodiment, grid connection and island mode are controlled in a unified manner; black start, grid connection soft power boost, and superposition (feedforward) coordination with AGC / AVC are supported.

[0075] In this embodiment, the voltage level / bus structure can be any level such as 400 V / 480 V / 690 V / 1 kV; pre- / post-boost injection and single bus / ring network / double bus structure are all applicable.

[0076] In this embodiment, the FFT / PLC function can be integrated into the EMS or moved down to the edge controller on the PCS side; the analog module can be built into the PCS or an external DIN rail module.

[0077] In this embodiment, long-distance transmission employs fiber-optic current loop isolation, differential transmission, or lightning protection isolation modules; field wiring utilizes terminal cabinets / prefabricated wiring harnesses.

[0078] In this embodiment, new PCS can be added by connecting them in parallel to the analog broadcast bus without configuring communication parameters for each PC; or hot-swappable expansion can be achieved by using a "pluggable signal card".

[0079] Example 2: Based on Embodiment 1, this embodiment of the invention provides a primary frequency regulation control system for an energy storage cluster based on analog quantity control. This system can be used to implement the primary frequency regulation control method for an energy storage cluster based on analog quantity control as described in the previous embodiments. The system includes a frequency acquisition module, an energy management subsystem, a programmable logic controller (PLC) and an FFT analysis module, an analog quantity output module, an energy storage converter cluster, a battery pack, a step-up transformer, and a grid-connected bus subsystem. The energy management subsystem is electrically connected to the frequency acquisition module, the PLC and FFT analysis module, the energy storage converter cluster, and the analog quantity output module. Each energy storage converter in the energy storage converter cluster is electrically connected to its corresponding battery pack. The PLC and FFT analysis module are connected to the step-up transformer and the grid-connected bus subsystem. Frequency acquisition module, used to acquire power grid frequency signals in real time; The energy management subsystem is used to calculate frequency deviation, generate primary frequency modulation power commands, and perform SOC equalization management. The programmable logic controller and FFT analysis module are used to analyze power grid frequency signals through fast Fourier transform and identify the type of frequency disturbance. Analog output module is used to convert power commands into 4–20 mA current signals and broadcast them to the energy storage converter cluster; Energy storage converter clusters are used to receive analog commands and perform power regulation; Battery packs are used for energy storage and release. The step-up transformer and grid bus are used to step up the voltage of the energy storage converter to the grid voltage and connect it to the grid.

[0080] In this embodiment, the DC side of each energy storage converter is electrically connected to the corresponding battery pack, and the AC side of each energy storage converter is connected in parallel to the step-up transformer and the grid-connected bus in the grid-connected bus subsystem.

[0081] In this embodiment, the analog output module is a multi-channel isolated output that supports star or bus signal distribution topologies.

[0082] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A primary frequency regulation control method for an energy storage cluster based on analog quantity control, characterized in that, Includes the following steps: Real-time acquisition of power grid frequency signals and calculation of frequency deviation; Analyze the power grid frequency signal using Fast Fourier Transform to identify the type of frequency disturbance; Based on the type of frequency disturbance, the frequency modulation coefficient is dynamically determined, and then the primary frequency modulation power command is calculated based on the frequency deviation and the frequency modulation coefficient. Based on the state of charge of each battery pack in the energy storage cluster, the SOC equalization strategy is used to optimize the power limiting and allocation of the primary frequency regulation power command, resulting in the optimized power command. The optimized power command is converted into an analog signal and broadcast to all energy storage converters in the energy storage converter cluster. In response to analog signals, each energy storage converter adjusts its charging and discharging power within milliseconds and dynamically adjusts the power distribution coefficient based on the difference between its own state of charge and the average state of charge of the cluster, thus completing the primary frequency regulation control of the energy storage cluster.

2. The primary frequency regulation control method for energy storage clusters based on analog quantity control according to claim 1, characterized in that, After identifying the frequency disturbance type, the frequency modulation coefficient is dynamically determined according to the frequency disturbance type. Before calculating the frequency modulation power command based on the frequency deviation and the frequency modulation coefficient, the method also includes dead zone and amplitude limiting processing. Specifically, when |Δf| is less than the preset dead zone threshold, the frequency modulation action is not triggered; when the communication interruption lasts for more than the set time, the last valid power command remains unchanged; Δf represents the frequency deviation.

3. The primary frequency regulation control method for energy storage clusters based on analog quantity control according to claim 1, characterized in that, The specific method for dynamically determining the frequency modulation coefficient Kp is as follows: Within the preset sliding time window T w The acquired power grid frequency signal is subjected to a Fast Fourier Transform (FFT) to obtain the frequency domain spectrum. Based on frequency domain spectral statistics, the spectral energy of the first frequency band and the spectral energy of the second frequency band are statistically analyzed, and the proportion of high-frequency energy and the proportion of low-frequency energy are calculated based on the spectral energy of the first frequency band and the spectral energy of the second frequency band. If the proportion of low-frequency energy is not less than the first threshold θ L When the duration reaches the first judgment period, it is judged as a continuous load disturbance, and the frequency modulation coefficient Kp is increased from the reference value Kp0 to Kp_up, and Kp_up=min(Kp_max, γ_up·Kp0); where γ_up is the amplification coefficient and Kp_max is the upper limit of the frequency modulation coefficient; If the proportion of high-frequency energy is not less than the second threshold θ H When the duration reaches the second judgment period, it is judged as transient or harmonic interference, and the frequency modulation coefficient Kp is reduced from the reference value Kp0 to Kp_down, and Kp_down=max(Kp_min, γ_down·Kp0); where Kp_min is the lower limit of the frequency modulation coefficient, and γ_down is the attenuation coefficient; If the proportion of high-frequency energy is not less than the shielding threshold θ block If the high-frequency disturbance continues for longer than the shielding determination time, the frequency modulation response will be shielded and the frequency modulation coefficient Kp will be set to zero to maintain the shielding time.

4. The primary frequency regulation control method for energy storage clusters based on analog quantity control according to claim 1, characterized in that, The SOC equalization strategy employs a power allocation algorithm of "weighted allocation + cut-off protection," specifically: Discharge allocation: If the state of charge (SOC) of any battery pack is lower than the first SOC threshold, the discharge allocation weight of that battery pack is reduced; if the SOC of any battery pack is lower than the second SOC threshold, the discharge allocation weight of that battery pack is reset to zero; where SOC_cut_low... <SOC_low; Charging allocation: If the state of charge of any battery pack is higher than the third state of charge threshold SOC_high, the charging allocation weight of that battery pack is reduced; if the state of charge of any battery pack is higher than the fourth state of charge threshold SOC_cut_high, the charging allocation weight of that battery pack is reset to zero; where SOC_cut_high>SOC_low.

5. The primary frequency regulation control method for energy storage clusters based on analog quantity control according to claim 1, characterized in that, The analog signal is a 4–20mA current signal. In response to the 4–20mA current signal, each energy storage converter converts the 4–20mA current signal into a digital power command, and superimposes local SOC protection logic and power limiting constraints to dynamically adjust the power allocation coefficient and complete the primary frequency regulation control of the energy storage cluster.

6. The primary frequency regulation control method for energy storage clusters based on analog quantity control according to claim 5, characterized in that, The 4–20mA current signal is generated by a programmable logic controller (PLC) or a dedicated analog output card and transmitted in parallel to the analog input ports of each energy storage converter via shielded twisted-pair cables.

7. The primary frequency regulation control system for an energy storage cluster based on the analog quantity control method for primary frequency regulation control of an energy storage cluster according to any one of claims 1-6, characterized in that, It includes a frequency acquisition module, an energy management subsystem, a programmable logic controller and FFT analysis module, an analog output module, an energy storage converter cluster, a battery pack, a step-up transformer, and a grid-connected bus subsystem. The energy management subsystem is electrically connected to the frequency acquisition module, the programmable logic controller and FFT analysis module, the energy storage converter cluster, and the analog output module. Each energy storage converter in the energy storage converter cluster is electrically connected to its corresponding battery pack. The programmable logic controller and FFT analysis module are connected to the step-up transformer and the grid-connected bus subsystem. Frequency acquisition module, used to acquire power grid frequency signals in real time; The energy management subsystem is used to calculate frequency deviation, generate primary frequency modulation power commands, and perform SOC equalization management. The programmable logic controller and FFT analysis module are used to analyze power grid frequency signals through fast Fourier transform and identify the type of frequency disturbance. The analog output module is used to convert power commands into 4–20mA current signals and broadcast them to the energy storage converter cluster. Energy storage converter clusters are used to receive analog commands and perform power regulation; Battery packs are used for energy storage and release; The step-up transformer and grid bus are used to step up the voltage of the energy storage converter to the grid voltage and connect it to the grid.

8. The primary frequency regulation control system for energy storage clusters based on analog quantity control according to claim 7, characterized in that, The DC side of each energy storage converter is electrically connected to the corresponding battery pack, and the AC side of each energy storage converter is connected in parallel to the step-up transformer and the grid-connected bus in the grid-connected bus subsystem.

9. The primary frequency regulation control system for energy storage clusters based on analog quantity control according to claim 7, characterized in that, The analog output module is a multi-channel isolated output, using a star or bus signal distribution topology.

Images