Control method and device of energy storage converter multi-machine parallel system and storage medium

Abstract

The invention discloses a control method and device of an energy storage converter multi-machine parallel system and a storage medium, and belongs to the technical field of energy storage converter control. According to the control method, high-frequency circulation is suppressed in real time through dynamic virtual impedance compensation; dividing a multi-mode response strategy according to the voltage drop depth of the power grid, and triggering the characteristics of the virtual synchronous machine during deep drop; and a distributed consistency algorithm is adopted to realize balanced power distribution. The problems of circulating current suppression, rapid voltage recovery and weak communication fault tolerance of a multi-machine parallel system are solved, and THDlt is realized; and the voltage recovery time is less than or equal to 80ms, so that the stability and the reliability of the energy storage system under a power grid fault are remarkably improved.

Description

Technical Field

[0001] This invention relates to a control method, device, and storage medium for a multi-unit parallel energy storage converter system, belonging to the field of energy storage converter control technology. Background Technology

[0002] Existing parallel multi-unit energy storage converter systems have three major technical drawbacks under grid voltage dip conditions: 1. Insufficient adaptability of circulating current suppression; Existing technologies use virtual impedance control methods with fixed parameters, which are difficult to track dynamic changes in harmonic frequency bands, resulting in the total harmonic distortion (THD) of circulating current generally exceeding 10%; 2. Poor voltage recovery timeliness; Existing technology relies on a centralized control architecture, which is highly sensitive to communication latency. The voltage recovery time often exceeds 150 milliseconds, making it difficult to meet the requirements for rapid support. 3. Weak system fault tolerance: The master-slave control architecture used in existing technologies is prone to system crashes when communication is interrupted.

[0003] These shortcomings limit the reliability of energy storage converters in applications such as grid frequency regulation. Summary of the Invention

[0004] The purpose of this invention is to provide a control method, device, and storage medium for a multi-machine parallel energy storage converter system, which solves the following defects of the existing multi-machine parallel energy storage converter system under grid voltage drop conditions: insufficient circulating current suppression adaptability, poor voltage recovery timeliness, and weak system fault tolerance.

[0005] To achieve the above objectives, the present invention employs the following technical solution: In a first aspect, the present invention provides a control method for a multi-unit parallel system of energy storage converters, comprising: The output current of each energy storage converter is acquired in real time. The dominant harmonic frequency within a set frequency range is extracted from the output current through FFT analysis. The virtual impedance parameters in the set virtual impedance transfer function are dynamically updated according to the dominant harmonic frequency. The adaptive virtual impedance control of each energy storage converter is performed according to the dynamically updated virtual impedance parameters to complete the dynamic virtual impedance compensation of each energy storage converter. In response to a voltage dip in the power grid where the system is located, the voltage dip depth is obtained, and multi-mode hierarchical response control is performed on each energy storage converter after dynamic virtual impedance compensation based on the voltage dip depth to complete the grid response; After obtaining the impedance of each energy storage converter, the electrical distance weight of each energy storage converter is calculated. Based on the electrical distance weight of each energy storage converter, a consensus algorithm is used to allocate power to each energy storage converter after completing the grid response, thereby completing the distributed power balance of each energy storage converter.

[0006] Furthermore, the set frequency range is 0.1~2kHz.

[0007] Furthermore, the virtual impedance parameters include proportional gain parameters, integral gain parameters, and differential gain parameters; The virtual impedance parameters in the virtual impedance transfer function, which are dynamically updated based on the dominant harmonic frequency, are determined by the following formula: ; ; ; in, This represents the proportional gain parameter. This represents the integral gain parameter. Represents the differential gain parameter. Indicates the dominant harmonic frequency. Indicates the reference frequency; After dynamically updating the virtual impedance parameters in the predefined virtual impedance transfer function, the process also includes a step of verifying the virtual impedance parameters. If the proportional gain parameter is greater than the preset first upper limit value, the proportional gain parameter will be adjusted to the first upper limit value; if the proportional gain parameter is less than the preset first lower limit value, the proportional gain parameter will be adjusted to the first lower limit value. If the integral gain parameter is greater than the preset second upper limit value, the integral gain parameter will be adjusted to the second upper limit value; if the integral gain parameter is less than the preset second lower limit value, the integral gain parameter will be adjusted to the second lower limit value. If the differential gain parameter is greater than the preset third upper limit value, the differential gain parameter will be adjusted to the third upper limit value; if the differential gain parameter is less than the preset third lower limit value, the differential gain parameter will be adjusted to the third lower limit value.

[0008] Furthermore, the adaptive virtual impedance control of each energy storage converter based on the dynamically updated virtual impedance parameters includes: The virtual impedance parameters for control are calculated based on the dynamically updated virtual impedance parameters and the virtual impedance parameters before the update. The virtual impedance parameter is injected into the control loop of the energy storage converter in real time to dynamically adjust the virtual impedance of the energy storage converter, and the output voltage of the energy storage converter is controlled according to the virtual impedance.

[0009] Furthermore, the control loop that injects virtual impedance parameters into the energy storage converter in real time to dynamically adjust the virtual impedance of the energy storage converter is achieved through the following formula: ; in, Represents virtual impedance. This represents the proportional gain parameter. This represents the integral gain parameter. The differential gain parameter, proportional gain parameter, integral gain parameter, and differential gain parameter are all types of virtual impedance parameters. It is a natural constant. Represents the complex frequency variable of the Laplace transform. Indicates the time constant of the integral / differential filter; The method of controlling the output voltage of the energy storage converter based on the virtual impedance includes: calculating the output voltage of the energy storage converter based on the virtual impedance, thereby controlling the output voltage of the energy storage converter; The calculation of the output voltage of the energy storage converter based on the virtual impedance is performed using the following formula: ; in, This indicates the output voltage of the energy storage converter. This indicates the original control voltage of the energy storage converter. This indicates the amount of circulating flow detected.

[0010] Furthermore, the multi-mode hierarchical response control of each energy storage converter after dynamic virtual impedance compensation based on the voltage sag depth includes: like Then, reactive power incremental compensation is performed on the energy storage converter after dynamic virtual impedance compensation, increasing the reactive power of the energy storage converter to [amount missing]. ; in, , This indicates the initial reactive power of the energy storage converter. Indicates the voltage drop depth. This indicates the reactive power of the energy storage converter after reactive power increment compensation. like Then, active power linear unloading is performed on the energy storage converter after dynamic virtual impedance compensation, reducing the active power of the energy storage converter to [a certain value]. Furthermore, the reactive power of the energy storage converter after dynamic virtual impedance compensation is increased to a value greater than the preset value of rated reactive power output; in, , This indicates the initial active power of the energy storage converter. Indicates the nominal voltage drop depth. This represents the active power of the energy storage converter after active power linear unloading; like Then the virtual synchronous machine rotor equation of the energy storage converter after dynamic virtual impedance compensation is activated, and part of the active power of the energy storage converter after dynamic virtual impedance compensation is unloaded. in, This indicates the per-unit value.

[0011] Furthermore, before activating the virtual synchronous machine rotor equation of the energy storage converter after activating the dynamic virtual impedance compensation, the following steps are also included: Obtain the grid voltage waveform and extract the phase and amplitude of the grid voltage from the waveform; The phase of the output voltage of the energy storage converter is obtained, and the phase of the output voltage of the energy storage converter is dynamically adjusted by the phase regulator inside the energy storage converter, so that the deviation between the phase of the output voltage of the energy storage converter and the phase of the grid voltage is reduced to within 0.5°. The amplitude of the output voltage of the energy storage converter is obtained, and the amplitude of the output voltage of the energy storage converter is dynamically adjusted by the amplitude regulator inside the energy storage converter, so that the deviation between the amplitude of the output voltage of the energy storage converter and the amplitude of the grid voltage is reduced to within 1%.

[0012] Furthermore, the electrical distance weight of each energy storage converter is calculated after obtaining its impedance using the following formula: ; in, Indicates the first The first energy storage converter is affected by the Electrical distance weighting of each energy storage converter Indicates the first The impedance of the energy storage converter Indicates the first The impedance of the energy storage converter.

[0013] Secondly, the present invention provides a control device for a multi-unit parallel system of energy storage converters, comprising: The dynamic virtual impedance compensation module is configured to: acquire the output current of each energy storage converter in real time, extract the dominant harmonic frequency within a set frequency range from the output current through FFT analysis, dynamically update the virtual impedance parameters in the set virtual impedance transfer function according to the dominant harmonic frequency, and perform adaptive virtual impedance control on each energy storage converter according to the dynamically updated virtual impedance parameters, thereby completing the dynamic virtual impedance compensation of each energy storage converter. The grid response module is configured to: respond to a voltage drop in the grid where the system is located, obtain the voltage drop depth, and perform multi-mode hierarchical response control on each energy storage converter after dynamic virtual impedance compensation based on the voltage drop depth to complete the grid response; The distributed power balancing module is configured to: obtain the impedance of each energy storage converter, calculate the electrical distance weight of each energy storage converter, and, based on the electrical distance weight of each energy storage converter, use a consensus algorithm to perform power allocation on each energy storage converter after completing the grid response, thereby completing the distributed power balancing of each energy storage converter.

[0014] Thirdly, the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of the control method for a multi-machine parallel system of energy storage converters as described in the first aspect.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention are: This invention provides a control method, device, and storage medium for a multi-unit parallel energy storage converter system. It addresses the high-frequency circulating current within the energy storage converter through dynamic virtual impedance compensation, providing a stable internal control environment for all subsequent control operations. When the grid voltage drops, the energy storage converter needs to accurately execute unloading / compensation actions. Therefore, this invention suppresses the interference of high-frequency circulating current on command tracking by performing multi-mode hierarchical response control within a stable internal control environment, ensuring accurate unloading / compensation actions and thus completing grid response. This enables the system to handle grid anomalies. Without prior dynamic virtual impedance compensation, high-frequency circulating current can cause yaw or even trigger protection mechanisms. However, this invention enables rapid voltage recovery when grid anomalies occur. When the energy storage converter itself is stable... After achieving stability (through dynamic virtual impedance compensation) and the ability to cope with grid anomalies (through grid response), the power distribution problem of multiple parallel energy storage converters needs to be addressed. At this point, this invention uses electrical distance weighting to allow each energy storage converter to coordinate according to electrical distance, achieving distributed power balance. In summary, this invention focuses on the stable operation and low-voltage ride-through capability of multi-parallel energy storage converters. Through dynamic virtual impedance compensation, grid response, and distributed power balance, it forms a closed-loop logic of micro-level stability, macro-level external response, and multi-converter coordination. It constructs a hierarchical and progressively driven collaborative framework, solving circulating current suppression, achieving rapid voltage recovery, and addressing weak communication fault tolerance issues, significantly improving the stability and reliability of multi-parallel energy storage converter systems under grid fault conditions. Attached Figure Description

[0016] Figure 1 This is a flowchart of a control method for a multi-machine parallel system of energy storage converters provided in an embodiment of the present invention; Figure 2This is a schematic diagram of the connection relationship between the energy storage converter multi-machine parallel system and the power grid provided in the embodiment of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solution of the present invention more clearly, and should not be used to limit the scope of protection of the present invention.

[0018] Example 1

[0019] like Figure 1 As shown, this embodiment provides a control method for a multi-unit parallel system of energy storage converters, including: The output current of each energy storage converter is acquired in real time. The dominant harmonic frequency within a set frequency range is extracted from the output current through FFT analysis. The virtual impedance parameters in the set virtual impedance transfer function are dynamically updated according to the dominant harmonic frequency. The adaptive virtual impedance control of each energy storage converter is performed according to the dynamically updated virtual impedance parameters to complete the dynamic virtual impedance compensation of each energy storage converter. In response to a voltage dip in the power grid where the system is located, the voltage dip depth is obtained, and multi-mode hierarchical response control is performed on each energy storage converter after dynamic virtual impedance compensation based on the voltage dip depth to complete the grid response; After obtaining the impedance of each energy storage converter, the electrical distance weight of each energy storage converter is calculated. Based on the electrical distance weight of each energy storage converter, a consensus algorithm is used to allocate power to each energy storage converter after completing the grid response, thereby completing the distributed power balance of each energy storage converter.

[0020] This invention addresses the high-frequency circulating current within the energy storage converter through dynamic virtual impedance compensation, providing a stable internal control environment for all subsequent control operations. When the grid voltage drops, the energy storage converter needs to precisely execute unloading / compensation actions. Therefore, this invention suppresses the interference of high-frequency circulating current on command tracking by implementing multi-mode hierarchical response control within a stable internal control environment, ensuring accurate unloading / compensation actions and thus completing grid response. This enables the converter to handle grid anomalies. Without prior dynamic virtual impedance compensation, high-frequency circulating currents can cause yaw or even trigger protection mechanisms. However, this invention enables rapid voltage recovery when grid anomalies occur. Furthermore, when the energy storage converter itself stabilizes (achieved through dynamic virtual impedance compensation),... Furthermore, it can handle the power distribution problem of multiple parallel energy storage converters after grid anomalies (achieved through grid response). In this case, the present invention uses electrical distance weights to enable each energy storage converter to coordinate according to electrical distance, thereby achieving distributed power balance. In summary, the present invention focuses on the stable operation and low-voltage ride-through capability of multi-parallel energy storage converters. Through dynamic virtual impedance compensation, grid response, and distributed power balance, it forms a closed-loop logic of micro-level stability, macro-level external response, and multi-machine coordination. It constructs a hierarchical and progressive collaborative framework with mutually driving results, solves circulating current suppression, achieves rapid voltage recovery, and solves the problem of weak communication fault tolerance, significantly improving the stability and reliability of multi-parallel energy storage converter systems under grid faults.

[0021] Example 2

[0022] This embodiment provides a control method for a multi-unit parallel system of energy storage converters, specifically for a medium-power energy storage microgrid system in an industrial or commercial setting. Its core objective is to ensure a rapid and stable system response when the grid experiences a short-term voltage dip (a common power quality issue), preventing equipment downtime and providing continuous, high-quality power to critical local loads. In this application scenario, the connection relationships of the units are as follows: Figure 2 As shown, this control method is executed by a central controller (master node), using fiber optic communication, with the PCS cluster acting as a slave node, and includes a battery pack.

[0023] The system is configured as follows: five 200kW power storage converters (PCS) are connected in parallel. This means the total power handling capacity of the system is 1 megawatt (1MW). This parallel architecture is a standard method for building medium-to-high power systems, providing good redundancy (the system can still operate at derating in the event of a single unit failure) and scalability.

[0024] The system's technical specifications are as follows: The heat dissipation method employs liquid cooling, which offers extremely high heat dissipation efficiency compared to common air cooling, making it particularly suitable for scenarios involving high power density and long-term high-current operation. It ensures that core power devices (such as IGBTs, Insulated-Gate Bipolar Transistors) maintain a low temperature rise even in high-temperature environments, thereby improving the overall system reliability and lifespan. Critical temperature limits explicitly require IGBT junction temperatures to be below 125°C. IGBTs are the core switching devices in a PCS (Power Supply System), and their junction temperature is the most critical parameter affecting lifespan and reliability. Strict control of junction temperature is fundamental to ensuring long-term stable system operation. Communication performance: The internal communication latency of the system is no higher than 5 milliseconds. This is crucial for achieving precise synchronization and rapid collaboration among multiple PCS units. Low-latency communication ensures that all energy storage converters can respond to grid disturbances simultaneously like a single machine, avoiding internal circulating currents or oscillations caused by inconsistent responses.

[0025] In this embodiment, the core problem to be solved is dealing with the severe operating condition where the grid voltage drops to 60% (i.e., the voltage at the grid connection point drops instantaneously to 0.6 pu), where pu represents the per-unit value, meaning that the grid voltage drops to 60% of the rated voltage under this condition. This is usually caused by faults such as grid-side short circuits or the starting of large motors.

[0026] This embodiment provides a control method for a multi-machine parallel system of energy storage converters, based on... Figure 1 The flowchart shown includes steps S1, S2, and S3.

[0027] Step S1: Dynamic virtual impedance compensation.

[0028] Step S1 must be executed first because, in multi-unit parallel operation, the high-frequency circulating current inside the energy storage converter is a significant internal problem. It can interfere with current sampling accuracy, cause control oscillations, and even lead to overcurrent in the device. If the circulating current is not suppressed first, the subsequent grid response (step S2) and power balancing (step S3) will fail due to internal losses. Therefore, dynamic virtual impedance compensation is fundamental, providing a stable internal control environment for all subsequent controls.

[0029] Step S1 is specifically divided into: current acquisition and preprocessing, signal conditioning and AD sampling, FFT harmonic frequency extraction, dynamic updating of virtual impedance parameters and adaptive virtual impedance control.

[0030] The core objective of current acquisition and preprocessing is to accurately capture the high-frequency current signal from the PCS output side, providing reliable data for subsequent harmonic analysis.

[0031] The current sensor selected is a Hall current sensor with a bandwidth of not less than 5kHz (covering the harmonic range of 0.1~2kHz), high accuracy (specifically ±0.5% in this embodiment), and fast response speed. It is installed at the output end of the PCS (one output for each of the three phases) to collect the three-phase current in real time.

[0032] Signal conditioning and AD sampling: The three-phase current output from the current sensor passes through a filter circuit (preserving the 0.1~2kHz frequency band and suppressing out-of-band noise) and an amplifier circuit (matching the ADC input range), and is then fed into a high-speed ADC (specifically, the TI ADS8329 in this embodiment, with 16-bit resolution and a sampling rate ≥8kHz) for AD sampling to obtain the output current of the energy storage converter. ADC stands for Analog-to-Digital Converter.

[0033] Sampling frequency The Nyquist theorem must be satisfied. , Therefore In this embodiment, 9kHz is actually used to improve frequency resolution.

[0034] FFT harmonic frequency extraction is used to locate the dominant harmonic frequency. The core is to separate the harmonic frequency with the largest amplitude (dominant harmonic frequency) from the time-domain current (output current of the energy storage converter) within the range of 0.1~2kHz through digital signal processing. The specific steps are as follows: Hanning windowing is applied to the continuously acquired output current to balance frequency resolution and spectral leakage, avoiding distortion of the FFT results due to signal truncation; FFT is performed on the windowed output current using a hardware acceleration module of DSP or FPGA to obtain the amplitude and phase spectra, decomposing the time-domain current into the amplitude and phase information of each harmonic; Dominant harmonic determination involves traversing the FFT results within the 0.1~2kHz frequency band and selecting the harmonic frequency with the largest amplitude as the dominant harmonic frequency. To avoid misjudgment due to transient interference, a multi-frame consistency check needs to be added. Only when the amplitude of the frequency is consistently among the top three in three consecutive FFT results can it be determined as the dominant harmonic frequency.

[0035] The virtual impedance parameter is dynamically updated to dynamically update the virtual impedance parameters in the predefined virtual impedance transfer function based on the dominant harmonic frequency, thus generating the control gain. This is specifically achieved through the following formula: ; ; ; in, This represents the proportional gain parameter. This represents the integral gain parameter. Represents the differential gain parameter. Indicates the dominant harmonic frequency. Indicates the reference frequency.

[0036] and It is directly proportional to the frequency, and enhances the proportional regulation effect at high frequencies, enabling rapid response to harmonic disturbances.

[0037] Proportional to frequency, the integral element dominates the elimination of steady-state error, and converges to the error-free state faster at high frequencies.

[0038] Inversely proportional to frequency, the differential element suppresses abrupt changes in harmonic amplitude while avoiding high-frequency noise amplification (the differential is sensitive to noise, and low gain can suppress noise).

[0039] When multiple energy storage converters are connected in parallel, slight differences in device parameters and line impedance can generate internal circulating currents (circulating currents) that do not pass through the power grid. These circulating currents do no work but increase device stress, cause heating, and can even trigger system oscillations. This problem is particularly pronounced during grid faults. This embodiment uses the suppression of 500Hz high-frequency circulating currents as an example. This embodiment sets a set of PID parameters, namely the proportional gain parameter, integral gain parameter, and derivative gain parameter mentioned above. , , This set of precisely calculated parameters can achieve a fast, stable, and zero-steady-state-error suppression effect for disturbance frequencies up to 500Hz.

[0040] Adaptive virtual impedance control is performed on each energy storage converter based on the dynamically updated virtual impedance parameters to achieve dynamic virtual impedance compensation for each converter. The core steps are: injecting the dynamically updated virtual impedance parameters into the control loop of the energy storage converter in real time, dynamically adjusting the virtual impedance of the energy storage converter, and controlling the output voltage of the energy storage converter based on the virtual impedance.

[0041] The virtual impedance model is designed in the inner current loop of the energy storage converter, and the virtual impedance is... Designed in PID form: ; in, Represents virtual impedance. This represents the proportional gain parameter. This represents the integral gain parameter. The differential gain parameter, proportional gain parameter, integral gain parameter, and differential gain parameter are all types of virtual impedance parameters. It represents the complex frequency variable of the Laplace transform. It is a key bridge connecting the dynamic behavior in the time domain and the analysis of the characteristics in the frequency domain, supporting the frequency domain design of virtual impedance, system stability verification and control parameter optimization.

[0042] Controlling the output voltage of the energy storage converter based on virtual impedance includes: calculating the output voltage of the energy storage converter based on virtual impedance, thereby controlling the output voltage of the energy storage converter; The output voltage of the energy storage converter is calculated based on the virtual impedance using the following formula: ; in, This indicates the output voltage of the energy storage converter. This indicates the original control voltage of the energy storage converter. This indicates the amount of circulating flow detected.

[0043] The virtual impedance parameter update step is triggered immediately after each FFT completes the identification of the dominant harmonic frequency.

[0044] Dynamic adjustments are prone to causing system oscillations, and optimization is needed in the following aspects: Gain limiting: Upper and lower limits are set for the three virtual impedance parameters. In this embodiment, specifically: , , To prevent sudden changes in gain; Gradual update: New parameters are mixed proportionally with old parameters to smoothly transition the control loop and avoid transient shocks; the specific mixing formula is as follows: ,in, This indicates the virtual impedance parameter used for control (i.e., the new parameter). This represents the virtual impedance parameter before dynamic update. This represents the dynamically updated virtual impedance parameter.

[0045] Step S2, Power Grid Response.

[0046] Step S2 must be executed after step S1. When the grid voltage drops, the energy storage converter needs to accurately execute active / reactive power commands (such as active power unloading). At this time, the optimized parameters output from step S1 (dynamically updated virtual impedance parameters) are embedded in the control loop, which can suppress the interference of circulating current on command tracking and ensure accurate unloading / compensation actions. If step S1 is skipped, the circulating current will cause the power response to yaw, or even trigger protection.

[0047] In response to a voltage dip in the power grid where the system is located, the voltage dip depth is obtained, and multi-mode hierarchical response control is performed on each energy storage converter after dynamic virtual impedance compensation based on the voltage dip depth.

[0048] In this embodiment, the core issue is handling a severe voltage drop to 60%, responding to a momentary voltage drop of 0.6 at the grid connection point. pu This is 60% of the rated voltage, which is usually caused by faults such as short circuits on the power grid side or starting of large motors.

[0049] The corresponding control shall be executed according to the following rules: Mode 1: If Then, reactive power incremental compensation is performed on the energy storage converter after dynamic virtual impedance compensation, increasing the reactive power of the energy storage converter to [amount missing]. ; in, , This indicates the initial reactive power of the energy storage converter. Indicates the voltage drop depth. This indicates the reactive power of the energy storage converter after reactive power increment compensation. Mode 2: If Then, active power linear unloading is performed on the energy storage converter after dynamic virtual impedance compensation, reducing the active power of the energy storage converter to [a certain value]. Furthermore, the reactive power of the energy storage converter after dynamic virtual impedance compensation is increased to a preset value greater than the rated reactive power output (specifically 120% of the rated reactive power output in this embodiment). in, , This indicates the initial active power of the energy storage converter. Indicates the nominal voltage drop depth. This represents the active power of the energy storage converter after active power linear unloading; Mode 3: If Then, the virtual synchronous machine rotor equation of the energy storage converter after dynamic virtual impedance compensation is activated, and part of the active power of the energy storage converter after dynamic virtual impedance compensation is unloaded (specifically, 80% of the active power is unloaded in this embodiment). in, This indicates the per-unit value.

[0050] In this embodiment, It is 0.6 pu The control corresponding to Mode 2 is executed. This is typically a set of specific control algorithms for severe voltage dips, which may include subroutines such as active / reactive power regulation, islanding detection, and circulating current suppression.

[0051] The purpose of dynamic active power offloading in Mode 2 is to address the limited power capacity of the grid during severe voltage drops. If the energy storage converter continues to inject full power into the grid, it will cause a surge in DC-side voltage, jeopardizing equipment safety. Therefore, it is necessary to rapidly reduce the output of active power.

[0052] In this embodiment, This means that the output active power of each energy storage converter will rapidly decrease from 200kW to approximately 92kW. The exponent 1.5 in the formula indicates that the power decrease rate is faster than the voltage decrease rate, which is a proactive protective strategy to ensure the system's safety margin under extreme operating conditions.

[0053] Step S3: After obtaining the impedance of each energy storage converter, calculate the electrical distance weight of each energy storage converter. Based on the electrical distance weight of each energy storage converter, use the consensus algorithm to allocate power to each energy storage converter after completing the grid response, and complete the distributed power balance of each energy storage converter.

[0054] Electrical distance weights are calculated using the following formula: ; in, Indicates the first The first energy storage converter is affected by the Electrical distance weighting of each energy storage converter Indicates the first The impedance of the energy storage converter Indicates the first The impedance of the energy storage converter.

[0055] Through step S3, within 5 seconds after the fault occurs, ensure that the output power deviation among the 5 parallel energy storage converters is less than 1.8%.

[0056] If the output power of each energy storage converter is uneven, with some under light load and others under heavy load, the heavily loaded energy storage converter will overheat and shorten its lifespan, failing to fully utilize the total system capacity. The extremely high balance requirement of 1.8% reflects the system's precise coordinated control capability, maximizing the overall system performance and lifespan.

[0057] When the system experiences a communication interruption with a single energy storage converter—that is, upon detecting the loss of connection of a particular energy storage converter—the main controller immediately issues a command to rapidly switch the remaining energy storage converters from the original communication topology (which may be bus or ring network) to a star network centered on the main controller within 12 milliseconds. This completes the automatic reconfiguration of the star topology. The core purpose of this operation is fault isolation; the system removes the faulty unit from the control network to prevent it from sending erroneous signals or failing to respond to commands, thus disrupting overall stability. This extremely rapid reconfiguration within 12 milliseconds demonstrates the system's powerful self-diagnosis and self-healing capabilities, which is crucial for realizing unattended intelligent power plants.

[0058] After experiencing severe disturbances such as communication outages and network reconfiguration, the output power deviation among the remaining healthy energy storage converters was strictly controlled within 2.3%. This demonstrates the extremely powerful power coordination control algorithm of the system. Even under partial damage, the remaining units can still work together precisely, avoiding overloading some devices (affecting their lifespan) and underloading others (wasting capacity). This is the cornerstone of the system's stability and efficiency.

[0059] Through the control described in this embodiment, the total harmonic distortion (THD) can be controlled at 2.5%, and the voltage recovery time is 76 milliseconds.

[0060] Example 3 This embodiment provides a control method for a multi-unit parallel energy storage converter system, specifically for a wind-solar-storage integrated power station. The specific implementation steps include steps S1, S2, and S3.

[0061] Step S1: Dynamic virtual impedance compensation.

[0062] Step S1 must be executed first because, in multi-unit parallel operation, the high-frequency circulating current inside the energy storage converter is a significant internal problem. It can interfere with current sampling accuracy, cause control oscillations, and even lead to overcurrent in the device. If the circulating current is not suppressed first, the subsequent grid response (step S2) and power balancing (step S3) will fail due to internal losses. Therefore, dynamic virtual impedance compensation is fundamental, providing a stable internal control environment for all subsequent controls.

[0063] The specific steps of step S1 are consistent with those of embodiment 2.

[0064] Step S2, Power Grid Response.

[0065] Step S2 must be executed after step S1. When the grid voltage drops, the energy storage converter needs to accurately execute active / reactive power commands (such as active power unloading). At this time, the optimized parameters output from step S1 (dynamically updated virtual impedance parameters) are embedded in the control loop, which can suppress the interference of circulating current on command tracking and ensure accurate unloading / compensation actions. If step S1 is skipped, the circulating current will cause the power response to yaw, or even trigger protection.

[0066] In this embodiment, the core issue is handling a severe voltage drop to 40%, responding to a momentary voltage drop of 0.4 at the grid connection point. pu That is, 40% of the rated voltage.

[0067] The corresponding control shall be executed according to the following rules: Mode 1: If Then, reactive power incremental compensation is performed on the energy storage converter after dynamic virtual impedance compensation, increasing the reactive power of the energy storage converter to [amount missing]. ; in, , This indicates the initial reactive power of the energy storage converter. Indicates the voltage drop depth. This indicates the reactive power of the energy storage converter after reactive power increment compensation. Mode 2: If Then, active power linear unloading is performed on the energy storage converter after dynamic virtual impedance compensation, reducing the active power of the energy storage converter to [a certain value]. Furthermore, the reactive power of the energy storage converter after dynamic virtual impedance compensation is increased to a preset value greater than the rated reactive power output (specifically 120% of the rated reactive power output in this embodiment). in, , This indicates the initial active power of the energy storage converter. Indicates the nominal voltage drop depth. This represents the active power of the energy storage converter after active power linear unloading; Mode 3: If Then, the virtual synchronous machine rotor equation of the energy storage converter after dynamic virtual impedance compensation is activated, and part of the active power of the energy storage converter after dynamic virtual impedance compensation is unloaded (specifically, 80% of the active power is unloaded in this embodiment). in, This indicates the per-unit value.

[0068] In this embodiment, It is 0.4 pu The control corresponding to mode three is executed. This is usually a set of specific control algorithms for severe voltage dips, which may include subroutines such as active / reactive power regulation, islanding detection, and circulating current suppression.

[0069] Step S3: After obtaining the impedance of each energy storage converter, calculate the electrical distance weight of each energy storage converter. Based on the electrical distance weight of each energy storage converter, use the consensus algorithm to allocate power to each energy storage converter after completing the grid response, and complete the distributed power balance of each energy storage converter.

[0070] The specific steps of step S1 are consistent with those of embodiment 2.

[0071] When the system experiences a communication interruption with a single energy storage converter—that is, upon detecting the loss of connection of a particular energy storage converter—the main controller immediately issues a command to rapidly switch the remaining energy storage converters from the original communication topology (which may be bus or ring network) to a star network centered on the main controller within 12 milliseconds. This completes the automatic reconfiguration of the star topology. The core purpose of this operation is fault isolation; the system removes the faulty unit from the control network to prevent it from sending erroneous signals or failing to respond to commands, thus disrupting overall stability. This extremely rapid reconfiguration within 12 milliseconds demonstrates the system's powerful self-diagnosis and self-healing capabilities, which is crucial for realizing unattended intelligent power plants.

[0072] After experiencing severe disturbances such as communication outages and network reconfiguration, the output power deviation among the remaining healthy energy storage converters was strictly controlled within 2.5%. This demonstrates the extreme power coordination control algorithm of the system. Even under partial damage, the remaining units can still work together precisely, avoiding overloading some devices (affecting their lifespan) and underloading others (wasting capacity). This is the cornerstone of the system's stability and efficiency.

[0073] In addition to detecting a disconnection of a certain energy storage converter, a voltage drop of 0.4 was also detected. pu This is a typical multi-fault stress test, and the fact that this invention can recover within 76 milliseconds demonstrates the system's strong fault tolerance and redundant design. Even when not in optimal condition, it can still prioritize support for the power grid, quickly restoring the voltage to normal levels and preventing grid collapse. This is crucial for maintaining the stability of the regional power grid.

[0074] Through the control described in this embodiment, the total harmonic distortion (THD) can be controlled at 2.9%, and the voltage recovery time is 78 milliseconds.

[0075] Example 4 This embodiment provides a control device for a multi-unit parallel system of energy storage converters, including: The dynamic virtual impedance compensation module is configured to: acquire the output current of each energy storage converter in real time, extract the dominant harmonic frequency within a set frequency range from the output current through FFT analysis, dynamically update the virtual impedance parameters in the set virtual impedance transfer function according to the dominant harmonic frequency, and perform adaptive virtual impedance control on each energy storage converter according to the dynamically updated virtual impedance parameters, thereby completing the dynamic virtual impedance compensation of each energy storage converter. The grid response module is configured to: respond to a voltage drop in the grid where the system is located, obtain the voltage drop depth, and perform multi-mode hierarchical response control on each energy storage converter after dynamic virtual impedance compensation based on the voltage drop depth to complete the grid response; The distributed power balancing module is configured to: obtain the impedance of each energy storage converter, calculate the electrical distance weight of each energy storage converter, and, based on the electrical distance weight of each energy storage converter, use a consensus algorithm to perform power allocation on each energy storage converter after completing the grid response, thereby completing the distributed power balancing of each energy storage converter.

[0076] Example 5 This embodiment provides a computer-readable storage medium storing a computer program / instructions. When executed by a processor, the computer program / instructions implement the steps of the control method for a multi-machine parallel energy storage converter system provided in Embodiment 1: The output current of each energy storage converter is acquired in real time. The dominant harmonic frequency within a set frequency range is extracted from the output current through FFT analysis. The virtual impedance parameters in the set virtual impedance transfer function are dynamically updated according to the dominant harmonic frequency. The adaptive virtual impedance control of each energy storage converter is performed according to the dynamically updated virtual impedance parameters to complete the dynamic virtual impedance compensation of each energy storage converter. In response to a voltage dip in the power grid where the system is located, the voltage dip depth is obtained, and multi-mode hierarchical response control is performed on each energy storage converter after dynamic virtual impedance compensation based on the voltage dip depth to complete the grid response; After obtaining the impedance of each energy storage converter, the electrical distance weight of each energy storage converter is calculated. Based on the electrical distance weight of each energy storage converter, a consensus algorithm is used to allocate power to each energy storage converter after completing the grid response, thereby completing the distributed power balance of each energy storage converter.

[0077] Based on the above embodiments, the present invention can achieve a THD of less than 3%, while existing technologies all exceed 10%, achieving a 70% improvement; the present invention achieves a voltage drop of 0.5 pu In the case of a single device, the recovery time can be no more than 80 milliseconds, while existing technologies are no less than 150 milliseconds, achieving a 47% improvement. With 20 energy storage converters connected in parallel, the present invention can achieve a power distribution deviation of less than 2.5%, while existing technologies are typically 5% to 8%, achieving a 60% improvement. Existing technologies will experience system crashes when communication is interrupted (energy storage converter failure), while the present invention can achieve a response within 10 milliseconds.

[0078] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0080] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0082] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A control method for a multi-unit parallel system of energy storage converters, characterized in that, include: The output current of each energy storage converter is acquired in real time. The dominant harmonic frequency within a set frequency range is extracted from the output current through FFT analysis. The virtual impedance parameters in the set virtual impedance transfer function are dynamically updated according to the dominant harmonic frequency. The adaptive virtual impedance control of each energy storage converter is performed according to the dynamically updated virtual impedance parameters to complete the dynamic virtual impedance compensation of each energy storage converter. In response to a voltage dip in the power grid where the system is located, the voltage dip depth is obtained, and multi-mode hierarchical response control is performed on each energy storage converter after dynamic virtual impedance compensation based on the voltage dip depth to complete the grid response; After obtaining the impedance of each energy storage converter, the electrical distance weight of each energy storage converter is calculated. Based on the electrical distance weight of each energy storage converter, a consensus algorithm is used to allocate power to each energy storage converter after completing the grid response, thereby completing the distributed power balance of each energy storage converter.

2. The control method for a multi-unit parallel system of energy storage converters according to claim 1, characterized in that, The set frequency range is 0.1~2kHz.

3. The control method for a multi-machine parallel system of energy storage converters according to claim 1, characterized in that, The virtual impedance parameters include proportional gain parameters, integral gain parameters, and differential gain parameters; The virtual impedance parameters in the virtual impedance transfer function, which are dynamically updated based on the dominant harmonic frequency, are determined by the following formula: ； ； ； in, This represents the proportional gain parameter. This represents the integral gain parameter. Represents the differential gain parameter. Indicates the dominant harmonic frequency. Indicates the reference frequency; After dynamically updating the virtual impedance parameters in the predefined virtual impedance transfer function, the process also includes a step of verifying the virtual impedance parameters. If the proportional gain parameter is greater than the preset first upper limit value, the proportional gain parameter will be adjusted to the first upper limit value; if the proportional gain parameter is less than the preset first lower limit value, the proportional gain parameter will be adjusted to the first lower limit value. If the integral gain parameter is greater than the preset second upper limit value, the integral gain parameter will be adjusted to the second upper limit value; if the integral gain parameter is less than the preset second lower limit value, the integral gain parameter will be adjusted to the second lower limit value. If the differential gain parameter is greater than the preset third upper limit value, the differential gain parameter will be adjusted to the third upper limit value; if the differential gain parameter is less than the preset third lower limit value, the differential gain parameter will be adjusted to the third lower limit value.

4. The control method for a multi-machine parallel system of energy storage converters according to claim 1, characterized in that, The adaptive virtual impedance control of each energy storage converter based on the dynamically updated virtual impedance parameters includes: The virtual impedance parameters for control are calculated based on the dynamically updated virtual impedance parameters and the virtual impedance parameters before the update. The virtual impedance parameter is injected into the control loop of the energy storage converter in real time to dynamically adjust the virtual impedance of the energy storage converter, and the output voltage of the energy storage converter is controlled according to the virtual impedance.

5. The control method for a multi-machine parallel system of energy storage converters according to claim 4, characterized in that, The control loop that injects virtual impedance parameters for control into the energy storage converter in real time, and dynamically adjusts the virtual impedance of the energy storage converter, is achieved through the following formula: ； in, Represents virtual impedance. This represents the proportional gain parameter. This represents the integral gain parameter. The differential gain parameter, proportional gain parameter, integral gain parameter, and differential gain parameter are all types of virtual impedance parameters. It is a natural constant. Represents the complex frequency variable of the Laplace transform. Indicates the time constant of the integral / differential filter; The method of controlling the output voltage of the energy storage converter based on the virtual impedance includes: calculating the output voltage of the energy storage converter based on the virtual impedance, thereby controlling the output voltage of the energy storage converter; The calculation of the output voltage of the energy storage converter based on the virtual impedance is performed using the following formula: ； in, This indicates the output voltage of the energy storage converter. This indicates the original control voltage of the energy storage converter. This indicates the amount of circulating flow detected.

6. The control method for a multi-machine parallel system of energy storage converters according to claim 1, characterized in that, The multi-mode hierarchical response control of each energy storage converter after dynamic virtual impedance compensation based on voltage sag depth includes: like Then, reactive power incremental compensation is performed on the energy storage converter after dynamic virtual impedance compensation, increasing the reactive power of the energy storage converter to [amount missing]. ; in, , This indicates the initial reactive power of the energy storage converter. Indicates the voltage drop depth. This indicates the reactive power of the energy storage converter after reactive power increment compensation. like Then, active power linear unloading is performed on the energy storage converter after dynamic virtual impedance compensation, reducing the active power of the energy storage converter to [a certain value]. Furthermore, the reactive power of the energy storage converter after dynamic virtual impedance compensation is increased to a value greater than the preset value of rated reactive power output; in, , This indicates the initial active power of the energy storage converter. Indicates the nominal voltage drop depth. This represents the active power of the energy storage converter after active power linear unloading; like Then the virtual synchronous machine rotor equation of the energy storage converter after dynamic virtual impedance compensation is activated, and part of the active power of the energy storage converter after dynamic virtual impedance compensation is unloaded. in, This indicates the per-unit value.

7. The control method for a multi-machine parallel system of energy storage converters according to claim 6, characterized in that, Before activating the virtual synchronous machine rotor equation of the energy storage converter after dynamic virtual impedance compensation, the following steps are also included: Obtain the grid voltage waveform and extract the phase and amplitude of the grid voltage from the waveform; The phase of the output voltage of the energy storage converter is obtained, and the phase of the output voltage of the energy storage converter is dynamically adjusted by the phase regulator inside the energy storage converter, so that the deviation between the phase of the output voltage of the energy storage converter and the phase of the grid voltage is reduced to within 0.5°. The amplitude of the output voltage of the energy storage converter is obtained, and the amplitude of the output voltage of the energy storage converter is dynamically adjusted by the amplitude regulator inside the energy storage converter, so that the deviation between the amplitude of the output voltage of the energy storage converter and the amplitude of the grid voltage is reduced to within 1%.

8. The control method for a multi-machine parallel system of energy storage converters according to claim 1, characterized in that, The electrical distance weight of each energy storage converter is calculated after obtaining its impedance using the following formula: ； in, Indicates the first The first energy storage converter is affected by the Electrical distance weighting of each energy storage converter Indicates the first The impedance of the energy storage converter Indicates the first The impedance of the energy storage converter.

9. A control device for a multi-unit parallel system of energy storage converters, characterized in that, include: The dynamic virtual impedance compensation module is configured to: acquire the output current of each energy storage converter in real time, extract the dominant harmonic frequency within a set frequency range from the output current through FFT analysis, dynamically update the virtual impedance parameters in the set virtual impedance transfer function according to the dominant harmonic frequency, and perform adaptive virtual impedance control on each energy storage converter according to the dynamically updated virtual impedance parameters, thereby completing the dynamic virtual impedance compensation of each energy storage converter. The grid response module is configured to: respond to a voltage drop in the grid where the system is located, obtain the voltage drop depth, and perform multi-mode hierarchical response control on each energy storage converter after dynamic virtual impedance compensation based on the voltage drop depth to complete the grid response; The distributed power balancing module is configured to: obtain the impedance of each energy storage converter, calculate the electrical distance weight of each energy storage converter, and, based on the electrical distance weight of each energy storage converter, use a consensus algorithm to perform power allocation on each energy storage converter after completing the grid response, thereby completing the distributed power balancing of each energy storage converter.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the control method for a multi-machine parallel system of energy storage converters as described in any one of claims 1 to 8.

Images