Micro-grid energy storage converter control method, device, equipment and medium

Abstract

The application discloses a micro-grid energy storage converter control method, device, equipment and medium, relates to the micro-grid control technical field, and the method comprises the following steps: acquiring state data of any energy storage converter in a micro-grid; obtaining global average state parameters of the energy storage converter by using an optimized distributed average consistency algorithm according to the state data; obtaining droop reference states of the energy storage converter by using a droop control strategy based on quadratic compensation according to a preset state reference value, the global average state parameters and measured values of state parameters of the energy storage converter; obtaining a voltage modulation signal by using a voltage drive modulation method according to the droop reference states, the global average state parameters and the measured values of state parameters of the energy storage converter; and controlling the energy storage converter according to the voltage modulation signal. The application ensures high-precision current sharing of multiple energy storage converters, and comprehensively improves transient robustness and operation resilience of the micro-grid under extreme working conditions.

Description

Technical Field

[0001] This application relates to the field of microgrid control technology, and in particular to a microgrid energy storage converter control method, device, equipment and medium. Background Technology

[0002] With the development of distributed generation and microgrid technologies, the parallel operation of multiple power conversion systems (PCS) has become crucial for improving system capacity and power supply reliability. Traditional multi-PCS parallel systems mainly rely on droop control. Although this method can achieve "plug and play," due to the differences in physical feeder impedance in microgrids, droop control inevitably suffers from low power distribution accuracy and steady-state voltage and frequency drops.

[0003] To address the aforementioned issues, centralized secondary control has been widely adopted, but it suffers from high single-point failure risk and poor scalability. In recent years, distributed control based on Multi-Agent Systems (MAS) has become a research hotspot. However, traditional distributed average consensus algorithms have inherent steady-state calculation errors. After directly superimposing secondary compensation droop control, frequency and voltage steady-state deviations will persist, making it difficult to achieve high-precision power sharing among multiple energy storage converters. Furthermore, conventional pre-synchronization phase error direct difference calculation will produce phase jumps, which can easily cause controller instability, resulting in insufficient anti-disturbance capability of microgrids under transient conditions such as sudden load changes and grid-connected / off-grid switching. Summary of the Invention

[0004] The purpose of this application is to provide a microgrid energy storage converter control method, device, equipment and medium, which can ensure high-precision current sharing of multiple energy storage converters and comprehensively improve the transient robustness and operational resilience of microgrids under extreme operating conditions.

[0005] To achieve the above objectives, this application provides the following solution.

[0006] In a first aspect, this application provides a microgrid energy storage converter control method, comprising: acquiring state data of any energy storage converter in a microgrid; the microgrid includes multiple energy storage converters; the state data includes measured values ​​of state parameters, globally estimated values ​​of state parameters, and globally estimated values ​​of state parameters of interconnected energy storage converters; the state parameters include frequency, voltage, active power, and reactive power; the interconnected energy storage converters are energy storage converters in the microgrid that are communicatively connected to the energy storage converters; and, based on the state data, using an optimized distributed average consensus algorithm to obtain the state data... The global average state parameters of the energy storage converter; the optimized distributed average consensus algorithm is based on an integral compensation mechanism to eliminate steady-state errors in the distributed average consensus algorithm; based on the preset state reference value, the global average state parameters, and the measured state parameters of the energy storage converter, a droop control strategy based on secondary compensation is adopted to obtain the droop reference state of the energy storage converter; based on the droop reference state, the global average state parameters, and the measured state parameters of the energy storage converter, a voltage-driven modulation method is adopted to obtain a voltage modulation signal, and the energy storage converter is controlled according to the voltage modulation signal.

[0007] Secondly, this application provides a microgrid energy storage converter control device, comprising: a state data acquisition module, used to acquire state data of any energy storage converter in a microgrid including multiple energy storage converters; the state data includes measured values ​​of state parameters, globally estimated values ​​of state parameters, and globally estimated values ​​of state parameters of interconnected energy storage converters; the state parameters include frequency, voltage, active power, and reactive power; the interconnected energy storage converters are energy storage converters in the microgrid that are communicatively connected to the energy storage converters; and a global average state parameter calculation module, used to obtain the energy storage converter's state data based on the state data using an optimized distributed average consensus algorithm. The global average state parameters of the converter; the optimized distributed average consensus algorithm is based on an integral compensation mechanism to eliminate steady-state errors in the distributed average consensus algorithm; the droop reference state calculation module is used to obtain the droop reference state of the energy storage converter by adopting a droop control strategy based on secondary compensation according to the preset state reference value, the global average state parameters, and the measured state parameters of the energy storage converter; the energy storage converter control module is used to obtain a voltage modulation signal by adopting a voltage drive modulation method according to the droop reference state, the global average state parameters, and the measured state parameters of the energy storage converter, and control the energy storage converter according to the voltage modulation signal.

[0008] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described microgrid energy storage converter control method.

[0009] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described microgrid energy storage converter control method.

[0010] Based on the specific embodiments provided in this application, the following technical effects are disclosed.

[0011] This application provides a microgrid energy storage converter control method, device, equipment, and medium. By collecting measured values, global estimates, and global estimates of the state parameters of the energy storage converter itself and interconnected energy storage converters, it completes the acquisition and aggregation of multi-dimensional state variables, providing a complete input data source for subsequent distributed consensus calculations and droop control. An optimized distributed average consensus algorithm with integral compensation is used to solve for the global average state parameters. The integral compensation mechanism eliminates the steady-state calculation error of traditional distributed average consensus algorithms, making the solved global average state parameters closer to the true system mean. Simultaneously, it ensures that the global parameters obtained by each energy storage converter node convergence remain consistent, providing a unified global reference benchmark for subsequent coordinated control. By combining preset state reference values, global average state parameters, and local measured values ​​to execute a droop control strategy based on secondary compensation, secondary compensation correction can be superimposed on the primary droop control, offsetting the inherent frequency and voltage steady-state deviations of traditional droop control and generating more accurate droop reference voltage and reference frequency. By using a voltage-driven modulation method to generate a voltage modulation signal and control the operation of the energy storage converter, the drooping reference state can be quickly converted into a modulation signal that can directly drive the energy storage converter, improving the tracking response speed of the reference command, reducing the output deviation during the modulation process, and ensuring that the output voltage of the energy storage converter accurately matches the target reference value, thus guaranteeing the stability of the single-unit output and the basic power quality. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is an application environment diagram of a microgrid energy storage converter control method according to an embodiment of this application.

[0014] Figure 2This is a flowchart illustrating a microgrid energy storage converter control method provided in one embodiment of this application.

[0015] Figure 3 This is a schematic diagram of the overall architecture of hierarchical adaptive cooperative control provided in an embodiment of this application.

[0016] Figure 4 This is a schematic diagram of a cyber-physical convergence architecture and communication topology provided in an embodiment of this application.

[0017] Figure 5 This is a schematic diagram of the functional modules of a microgrid energy storage converter control device provided in an embodiment of this application.

[0018] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] The microgrid energy storage converter control method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on other servers. Terminal 102 can send the status data of any energy storage converter in the microgrid to server 104. After receiving the status data, server 104 uses an optimized distributed average consensus algorithm to obtain the global average state parameters of the energy storage converter, employs a droop control strategy based on secondary compensation to obtain the droop reference state of the energy storage converter, and finally uses a voltage-driven modulation method to obtain a voltage modulation signal to control the energy storage converter. Server 104 can feed back the obtained voltage modulation signal to terminal 102. Furthermore, in some embodiments, the microgrid energy storage converter control method can also be implemented separately by the server 104 or the terminal 102. For example, the terminal 102 can directly use the energy storage converter's state data, employ an optimized distributed average consensus algorithm to obtain the global average state parameters of the energy storage converter, employ a droop control strategy based on secondary compensation to obtain the droop reference state of the energy storage converter, and finally employ a voltage-driven modulation method to obtain a voltage modulation signal to achieve control of the energy storage converter. Alternatively, the server 104 can obtain the energy storage converter's state data from the data storage system, employ an optimized distributed average consensus algorithm to obtain the global average state parameters of the energy storage converter, employ a droop control strategy based on secondary compensation to obtain the droop reference state of the energy storage converter, and finally employ a voltage-driven modulation method to obtain a voltage modulation signal to achieve control of the energy storage converter.

[0022] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0023] In one exemplary embodiment, such as Figure 2 As shown, a microgrid energy storage converter control method is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 204.

[0024] Step 201: For any energy storage converter in the microgrid, acquire the status data of that energy storage converter. The microgrid includes multiple energy storage converters. The status data includes measured values ​​of status parameters, globally estimated values ​​of status parameters, and globally estimated values ​​of status parameters of interconnected energy storage converters. The status parameters include frequency, voltage, active power, and reactive power. The interconnected energy storage converters are those in the microgrid that are communicatively connected to the energy storage converters.

[0025] Step 202: Based on the state data, an optimized distributed average consensus algorithm is used to obtain the global average state parameters of the energy storage converter. The optimized distributed average consensus algorithm is based on an integral compensation mechanism to eliminate steady-state errors in the distributed average consensus algorithm.

[0026] Step 203: Based on the preset state reference value, the global average state parameter, and the measured state parameter value of the energy storage converter, a droop control strategy based on secondary compensation is adopted to obtain the droop reference state of the energy storage converter.

[0027] Step 204: Based on the droop reference state, the global average state parameters, and the measured values ​​of the state parameters of the energy storage converter, a voltage-driven modulation method is used to obtain a voltage modulation signal, and the energy storage converter is controlled according to the voltage modulation signal.

[0028] In one example, each energy storage converter in the microgrid is configured as an independent intelligent agent node, and each intelligent agent node communicates using a time-sensitive network protocol to achieve clock synchronization.

[0029] Distributed consensus algorithms are highly dependent on the performance of the communication network. When faced with sudden heavy loads or extremely short-term processes such as mode switching, the uncertainty of communication delays and data packet jitter of traditional industrial Ethernet often lead to asynchronous interaction of state information among multiple agents. This will seriously reduce the accuracy of secondary compensation proportional-integral adjustment and pre-synchronization tracking, and may even lead to control loop oscillation and system collapse.

[0030] In one example, a Time-Sensitive Network (TSN) and intelligent agent are used to autonomously and collaboratively control complex operating conditions of multiple PCS (Power Control System) units. The microgrid scenario is equipped with multiple PCS units operating in parallel and local sensitive loads. The microgrid system can switch between grid-connected and islanded operation modes based on the external grid status, and faces complex electrical conditions such as nonlinear load switching, grid voltage fluctuations, and frequency deviations during operation. In one example, through methods such as... Figure 3 The multi-agent hierarchical adaptive cooperative control architecture shown enables precise power distribution and rapid stabilization of system voltage and frequency among these parallel PCS devices.

[0031] like Figure 3 As shown, the hierarchical adaptive collaborative control architecture is divided into a three-level vertical architecture: a top-level multi-agent collaborative layer, a middle-level local autonomous control layer, and a bottom-level electrical and physical layer.

[0032] The top layer comprises multiple parallel PCSs, each managed by a local agent. It integrates distributed average consensus interaction, adaptive droop coefficient adjustment, frequency and voltage compensation, fuzzy adaptive proportional-integral (PI) control, and state machine management and bumpless switching. Agents exchange state information such as power, frequency, and voltage, and output control commands such as PI parameters, droop coefficients, and power reference values ​​to the lower layers. Top-level communication uses TSN compliant with the IEEE 802.1 standard.

[0033] The intermediate layer enables local autonomous operation of a single energy storage converter, and includes droop control, adaptive virtual impedance, pre-synchronous phase tracking, voltage outer loop and current inner loop PI dual closed-loop controller. It uploads local electrical status variables upwards and outputs pulse width modulation (PWM) drive signals downwards.

[0034] The bottom layer is an electrical physical power circuit, consisting of a DC source, a three-phase inverter bridge, an inductor-capacitor (LC) filter, a local load, a grid-connected circuit breaker, and a large power grid. It collects three-phase voltage and current sampling signals and transmits them upward to the control layer, fully presenting the vertical control logic of the entire link from the hardware power circuit to the upper-layer distributed collaborative algorithm.

[0035] In another example, for Figure 3 The internal sub-modules have been integrated and simplified, Figure 3 The complete refactoring of the collaborative control architecture is divided into the following: Figure 4 The diagram illustrates a three-layer vertical architecture consisting of an information agent layer, a local control layer, and a physical layer. It focuses on showcasing the horizontally distributed communication topology and inter-layer signal interaction among multiple energy storage converters. The information agent layer comprises multiple independent agents, which communicate with each other via a communication topology adjacency matrix weighting coefficient (a). 12 a 21 a 23 a 32The communication link completes distributed state interaction. Each agent receives local state (active power, reactive power, frequency, and voltage) operating data uploaded from the lower layer, and after collaborative calculation, sends compensation commands (voltage and frequency compensation amounts and adaptive droop coefficients) to the lower layer. The local control layer is simplified to a local droop control module, which realizes basic power regulation of a single converter, outputs PWM drive signals downward, and collects three-phase voltage and current physical quantities to transmit back to the upper layer. The physical layer is a multi-PCS (PCS1, PCS2, PCS3) parallel AC bus topology, with multiple energy storage converters connected in parallel to a common AC bus and uniformly driving local loads, clearly demonstrating the correspondence between the multi-machine chain communication network, the inter-layer uplink and downlink information flow, and the parallel physical topology.

[0036] In one example, a TSN-based multi-agent communication network is first constructed: multiple parallel PCSs in the microgrid are configured as independent agent nodes, and each node is interconnected through a switch that supports the TSN protocol, providing global precise clock synchronization and deterministic low-latency data transmission, and ensuring strict alignment of the collaborative control cycles of multiple PCSs.

[0037] Specifically, each PCS agent node accesses the network through a TSN switch that supports the IEEE 802.1AS protocol and establishes a global master clock using the generalized precise time protocol. Each agent node calculates path delay and frequency drift by exchanging Precision Time Protocol (PTP) messages to achieve strict alignment of sampling time and control cycle at the nanosecond level, ensuring that the phase difference of physical sampling of each PCS in the parallel system approaches zero.

[0038] The interaction information of a multi-agent system (MAS) is divided into the following three types of business flows.

[0039] Critical Traffic (CAC) includes frequency and voltage synchronization vectors and power allocation instructions, mapped to the highest priority queue defined by the IEEE 802.1Qav standard.

[0040] Class B alarm traffic includes fault tripping and mode switching status variables, and has the second highest priority.

[0041] Class C (Best-effort Traffic): Contains non-real-time environmental parameters and background monitoring data, mapped to a low-priority queue.

[0042] Using a time-aware shaper defined by the IEEE 802.1Qbv standard, a cyclic gating list is set for a control step size of 0.1ms. At the beginning of each cycle, the output gating of non-control flow queues is forcibly closed through the gating list, opening a dedicated "green channel" for Class A control flow, and ensuring that the transmission delay jitter of data packets required for collaborative control in the network is controlled within 5μs.

[0043] Based on the IEEE 802.1Qca standard, multiple non-overlapping deterministic paths are preset as communication redundancy for multi-agent agents. When a physical failure occurs in a TSN switching link, the agents achieve seamless switching through the zero-loss redundancy mechanism (IEEE 802.1CB), ensuring the communication continuity of the distributed consensus algorithm when the network topology changes dynamically and preventing control divergence caused by packet loss.

[0044] In another example, the control architecture is divided into the following three layers.

[0045] Bottom Autonomous Layer: Adaptive virtual impedance droop control based on local electrical quantities.

[0046] Upper-layer collaboration layer: Distributed average consistency observation and compensation based on TSN deterministic communication.

[0047] Intelligent decision-making layer: PI parameter dynamic optimization and mode switching state machine based on fuzzy logic.

[0048] In a multi-PCS parallel scenario, the control cycle is set to To eliminate latency jitter in traditional Ethernet under complex operating conditions, the specific implementation steps are as follows.

[0049] Clock synchronization: Each agent node runs the IEEE 802.1AS protocol. By exchanging PTP synchronization messages, the sampling clock deviation of the entire network's PCS is controlled within a specified range. Within this range. This ensures that the PWM waveform sequence of the distributed controller is strictly aligned in time, physically reducing the probability of circulating current generation.

[0050] Time-Aware Shaping (TAS): Configure a Gate Control List (GCL) in the TSN switch. This involves setting each... The cycle is divided into critical time periods and normal time periods. During the critical time period, the switch only allows the highest priority agent state packets (including active power, reactive power, frequency, voltage, etc.) to pass through. This "dedicated lane" mechanism ensures that the end-to-end latency of critical data is fixed. This solves the dependency of distributed algorithms on the determinism of communication.

[0051] In one example, the optimized distributed average consensus algorithm solves for the global average state parameters of the energy storage converter through iterative computation. The global average state parameters include a global average frequency estimate and a global average voltage estimate. The global estimates of the energy storage converter's state parameters include the global average frequency estimate and global average voltage estimate obtained in the previous iteration. The global estimates of the interconnected energy storage converter's state parameters include the global average frequency estimate and global average voltage estimate of the interconnected energy storage converter in the previous iteration.

[0052] The optimized distributed average consensus algorithm uses the following formula to calculate the global average frequency estimate at any iteration time.

[0053] .

[0054] Where i represents the identifier of the energy storage converter in the microgrid. Let be the estimated global average frequency of energy storage converter i at time t. The measured frequency of energy storage converter i at time t. For energy storage converter i, it is an interconnected set of energy storage converters. denoted by , where is the number of interconnected energy storage converters of energy storage converter i, and j is the identifier of the interconnected energy storage converter of energy storage converter i. The weighting coefficients of the adjacency matrix of the communication topology between energy storage converter i and interconnected energy storage converter j are given. For interconnected energy storage converters j in The global average frequency estimate at time t. For energy storage converter i in The estimated global average frequency at time t.

[0055] The optimized distributed average consensus algorithm uses the following formula to calculate the global average voltage estimate at any iteration time.

[0056] .

[0057] in, Let be the estimated global average voltage of energy storage converter i at time t. Let be the measured voltage value of energy storage converter i at time t. For interconnected energy storage converters j in The estimated global average voltage at time t. For energy storage converter i in The estimated global average voltage at time t.

[0058] In one example, after configuring each energy storage converter in the microgrid as an independent intelligent agent node, upper-level state observation and secondary compensation are performed based on an optimized distributed average consensus algorithm. The agent periodically obtains the active power, reactive power, frequency, and voltage estimates of its neighboring nodes through the TSN network, and uses the optimized distributed average consensus algorithm to calculate the global secondary compensation for frequency and voltage, eliminating the inherent steady-state error of droop control.

[0059] Specifically, a distributed average consensus algorithm is used to average the global frequency. and average voltage The mathematical expression for its iterative observer is as follows, which performs real-time observation.

[0060] .

[0061] in, Let be the local state variable of the interconnected energy storage converter j at time t. for The derivative of represents the instantaneous rate of change of the local state variable. Let be the state variable of the interconnected energy storage converter j at time t.

[0062] According to the consistency control protocol, the internal frequency state observation integral term and the internal voltage state observation integral term The dynamic update equations are designed as follows.

[0063] .

[0064] .

[0065] in, for The derivative, For interconnected energy storage converters j in The global average frequency estimate at time t. for The derivative, For interconnected energy storage converters j in The estimated global average voltage at time t.

[0066] Based on the dynamic update equations of the internal frequency state observation integral term and the internal voltage state observation integral term, the calculation formulas for the above-mentioned global average frequency estimate and global average voltage estimate can be further obtained.

[0067] In one example, the preset state reference values ​​include a rated frequency reference value and a rated voltage reference value. The global average state parameters include a global average frequency estimate, a global average voltage estimate, a global average active power estimate, and a global average reactive power estimate.

[0068] The above step 203 can be replaced by the following steps 2031 to 2033.

[0069] Step 2031: Calculate the global average tracking error of the energy storage converter based on the rated frequency reference value, the rated voltage reference value, the global average frequency estimate, and the global average voltage estimate. The global average tracking error includes the global average frequency tracking error and the global average voltage tracking error.

[0070] Step 2032: Based on the global average tracking error, a fuzzy control algorithm is used to obtain the proportional gain and integral gain. The proportional gain includes a frequency-second-compensated proportional gain and a voltage-second-compensated proportional gain. The integral gain includes a frequency-second-compensated integral gain and a voltage-second-compensated integral gain.

[0071] Step 2033: Based on the global average tracking error, the proportional gain, and the integral gain, a proportional-integral control algorithm is used to obtain the secondary state compensation amount of the energy storage converter. The secondary state compensation amount includes the frequency secondary state compensation amount and the voltage secondary state compensation amount.

[0072] Step 2034: Based on the preset state reference value, the measured active power value of the energy storage converter, the measured reactive power value of the energy storage converter, the estimated global average active power value, the estimated global average reactive power value, and the secondary state compensation amount, a droop control method is used to obtain the droop reference state. The droop reference state includes the droop reference voltage and the droop reference frequency.

[0073] In this example, the rated frequency reference value of the microgrid system is set. 50Hz, rated voltage reference value The voltage is 311V. The intelligent agent corresponding to energy storage converter i calculates the voltage internally. Global average frequency tracking error at time 1 and Global average voltage tracking error at time 1 As shown in the following formulas respectively.

[0074] .

[0075] .

[0076] To balance dynamic response speed and steady-state error-free characteristics in error elimination, this example uses a PI controller to handle consistency tracking error. The proportional gain provides rapid compensation and correction capability based on the instantaneous magnitude of the current error, while the integral gain fundamentally overcomes the static deviation of droop control by continuously accumulating historical errors. After determining the proportional and integral gains, the intelligent agent corresponding to energy storage converter i is generated... Frequency quadratic state compensation quantity at time 1 and Voltage secondary state compensation at time t The control law designs are shown in the following equations.

[0077] .

[0078] .

[0079] in, This is a frequency-secondary compensation proportional gain. This is the frequency-secondary compensation integral gain. for Global average frequency tracking error at time t. This is the voltage secondary compensation proportional gain. This is the voltage-secondary compensation integral gain. for The global average voltage tracking error at any given time.

[0080] The intelligent agent corresponding to energy storage converter i sends out the calculated secondary state compensation quantity in real time and superimposes it onto the underlying physical control layer. From the perspective of the droop control mechanism, the following is introduced... and This is equivalent to dynamically shifting the steady-state droop characteristic curve of the energy storage converter up and down on the vertical coordinate axis. In this example, the reconstructed PCS local distributed cooperative droop control equations are as follows.

[0081] .

[0082] .

[0083] in, Let be the droop reference frequency of energy storage converter i at time t. Let i be the frequency droop factor of the energy storage converter. Let be the measured active power of energy storage converter i at time t. This is the estimated global average active power of energy storage converter i. Let be the droop reference voltage of energy storage converter i at time t. Let be the voltage droop factor of energy storage converter i. Let be the measured reactive power of energy storage converter i at time t. This is the estimated global average reactive power of energy storage converter i.

[0084] In one example, step 2032 above can be replaced by steps 0321 to 0325.

[0085] Step 0321: Differentiate the global average frequency tracking error and the global average voltage tracking error respectively to obtain the rate of change of the global average frequency tracking error and the rate of change of the global average voltage tracking error.

[0086] Step 0322: Based on the global average frequency tracking error and the rate of change of the global average frequency tracking error, obtain the frequency tracking error input, and based on the global average voltage tracking error and the rate of change of the global average voltage tracking error, obtain the voltage tracking error input.

[0087] Step 0323: Based on the triangular membership function and the direct universe of discourse, construct the frequency fuzzy optimization rule base and the voltage fuzzy optimization rule base according to the frequency tracking error input and the voltage tracking error input, respectively.

[0088] Step 0324: Using the Midani-type fuzzy inference method, rule matching operations are performed based on the frequency fuzzy optimization rule base and the voltage fuzzy optimization rule base respectively to obtain the frequency output fuzzy interval and the voltage output fuzzy interval.

[0089] Step 0325: The centroid method is used to perform defuzzification operations on the frequency output fuzzy range and the voltage output fuzzy range respectively to obtain the proportional gain and the integral gain.

[0090] A two-dimensional fuzzy logic controller is embedded in the secondary collaborative control layer. The global average tracking error and its rate of change of consistent observation are used as inputs to dynamically tune the proportional and integral gain of the PI controller online, thereby optimizing the dynamic overshoot and recovery delay under load change conditions.

[0091] Taking frequency compensation as an example, the fuzzy controller uses the global average frequency tracking error and its rate of change as dual input variables. It employs a physical direct universe of discourse and a triangular membership function, and defuzzifies the frequency through Mamdani fuzzy inference and the Center of Gravity (COG) method, outputting the frequency secondary compensation proportional gain of the PI controller in real time. Frequency-second compensation integral gain .

[0092] In the initial stage of the transient drop (when the error is extremely large), the fuzzy controller outputs a large value. Increase damping while outputting minimal values To prevent integral saturation; to smoothly reduce [the load] at the end of the steady-state recovery phase. and increase To eliminate minute steady-state errors.

[0093] In one example, the global average state parameters include a global average reactive power estimate. The droop reference state includes a droop reference voltage and a droop reference frequency.

[0094] Step 204 above can be replaced by steps 2041 to 2045.

[0095] Step 2041: Based on the droop reference voltage and droop reference frequency, a coordinate transformation preprocessing method is used to obtain the droop control voltage command. The droop control voltage command includes a d-axis voltage command component and a q-axis voltage command component.

[0096] Step 2042: Based on the estimated global average reactive power and the measured reactive power of the energy storage converter, the virtual impedance corresponding to the energy storage converter is calculated using an impedance correction method.

[0097] The formula for calculating virtual impedance is as follows.

[0098] .

[0099] in, For virtual impedance, For the Laplace operator, The integral coefficient is... This represents the measured reactive power value of the energy storage converter. This is the setpoint for the reactive power output of the energy storage converter.

[0100] An integrator can automatically adjust the virtual impedance according to the reactive power, unaffected by inconsistent line impedance, to adapt to changes in operating conditions and achieve reasonable distribution of reactive power.

[0101] In a centralized control microgrid, The load power and the power of each PCS are collected by the central controller and calculated, and then sent to the local controller of each PCS. The central controller allocates reactive power according to the PCS capacity; if the PCS capacity is the same, the load power is equally distributed among the PCS; if the capacity is different, the load power is allocated proportionally according to the capacity of each PCS.

[0102] Step 2043: Calculate the feedback current of the energy storage converter based on the measured voltage, active power, and reactive power values. The feedback current includes a d-axis feedback current component and a q-axis feedback current component.

[0103] Step 2044: Based on the droop control voltage command, the virtual impedance, and the feedback current, a corrected control voltage command is calculated using a virtual impedance droop correction method. The corrected control voltage command includes a d-axis corrected voltage command component and a q-axis corrected voltage command component.

[0104] Specifically, the calculation formula for the modified control voltage command is as follows.

[0105] .

[0106] in, Correct the voltage command component for the d-axis. Correct the voltage command component for the q-axis. This is the d-axis voltage command component. This is the q-axis voltage command component. The d-axis feedback current component. This represents the q-axis feedback current component.

[0107] Step 2045: According to the modified control voltage command, a vector control modulation method is used to obtain a voltage modulation signal, and the energy storage converter is controlled according to the voltage modulation signal.

[0108] In this example, a bottom-level adaptive virtual impedance autonomous control is adopted. In the bottom-level droop control loop, an adaptive virtual impedance is dynamically generated in real time based on the deviation of the local reactive power of each PCS to compensate for the impedance mismatch of the physical feeder, realize the reasonable allocation of the initial reactive power and suppress the parallel transient circulating current.

[0109] In one example, the droop reference state includes a droop reference frequency.

[0110] Following step 204 above, the microgrid energy storage converter control method further includes steps 205 to 212.

[0111] Step 205: Calculate the frequency change rate of the energy storage converter based on the measured frequency value of the energy storage converter, and perform low-pass filtering on the frequency change rate to obtain the filtered frequency change rate.

[0112] Step 206: Determine the operating condition of the microgrid based on the filter frequency change rate. The operating conditions include normal grid-connected operation and islanded operation. Under normal grid-connected operation, the main grid and the microgrid are normally connected. Under islanded operation, the connection between the main grid and the microgrid is disconnected.

[0113] Step 207: When the operating condition is islanded, the voltage of the main power grid is collected in real time.

[0114] Step 208: When the voltage of the main power grid is greater than a preset threshold, the phase of the main power grid is obtained by using a phase-locked loop phase extraction method.

[0115] Step 209: Based on the measured voltage value of the energy storage converter, the phase of the energy storage converter is obtained using a phase-locked loop phase extraction method.

[0116] Step 210: Based on the sinusoidal mapping phase error extraction method, calculate the pre-synchronization phase error between the phase of the main power grid and the phase of the energy storage converter.

[0117] Step 211: Based on the pre-synchronization phase error, a proportional-integral control algorithm is used to obtain the frequency compensation amount of the energy storage converter.

[0118] Step 212: Compensate the droop reference frequency according to the frequency compensation amount.

[0119] In complex operating conditions such as switching between islanded and grid-connected modes or sudden load changes, traditional control is prone to generating huge transient circulating currents, inrush currents, and long recovery delays. This example implements autonomous sensing and pre-synchronized, disturbance-free switching under extreme conditions. It achieves extremely rapid autonomous sensing of the grid state based on the Rate of Change of Frequency (ROCOF). During the off-grid to grid-connected transition, the agent actively freezes the secondary recovery integrator, enables pre-synchronized phase tracking with frequency and voltage decoupling at the underlying level, and combines a historical state back-calculation mechanism to eliminate state jumps and inrush currents at the moment of circuit breaker closure.

[0120] In this example, ROCOF perception logic with rapid dynamic capture capabilities is introduced into the decision-making algorithm layer of the agent. The physical essence of the ROCOF algorithm originates from the swing equation of a synchronous generator. During normal grid-connected operation, the microgrid's point of common coupling (PCC) frequency is rigidly clamped near its rated value (50Hz) by the massive rotational inertia H of the main grid, with the rate of frequency change approaching zero. Once the main grid experiences a fault disconnection (i.e., the instant of switching from grid connection to islanding), the system loses the inertia support of the main grid. The microgrid is powered only by the power electronic PCS, with almost no rotational inertia, resulting in an imbalance of active power within the microgrid. The effect will be entirely on the low-inertia system of the microgrid itself, causing the local frequency to decrease or increase drastically, and its transient rate of change satisfies the following formula.

[0121] .

[0122] Where f is the real-time frequency of the microgrid system.

[0123] As an independent decision-making part of the distributed control, the agent reads the internally generated discrete frequency state data f(t) from the underlying primary control (such as the vertical control power loop) in real time, and extracts the smoothed ROCOF observation value through discrete difference and first-order low-pass filtering within the agent.

[0124] In the continuous time domain, the rate of change of frequency is the derivative of frequency with respect to time. However, in the discrete digital domain of a microprocessor, the system can only operate in each control cycle. Since a discrete point is acquired, the formula for the discrete difference element is as follows.

[0125] .

[0126] in, This represents the original ROCOF observation value at the k-th sampling time. The real-time frequency of the microgrid is obtained by sampling at the k-th sampling time. The real-time frequency of the microgrid is obtained by sampling at the (k-1)th sampling time.

[0127] Because high-frequency measurement noise or white noise inevitably exists in actual power grid operation, even Extremely minute puncture fluctuations occur due to the denominator. Extremely small, even after division, it will still be in This generates extremely large numerical spikes. If the original value is used directly for island detection, the system will frequently malfunction and will be unable to operate normally. To eliminate the high-frequency glitches amplified by the differential operation, a classic low-pass filter is constructed in this example as shown in the following equation.

[0128] .

[0129] in, The final ROCOF observation value after filtering at the k-th sampling time is... This represents the final ROCOF observation value after filtering at the (k-1)th sampling time. These are the filter weight coefficients.

[0130] At the pre-synchronization control layer, the system first extracts the real-time phase of the power grid at time t through an independent three-phase phase-locked loop (PLL). The output phase of the microgrid's local energy storage converter at time t In traditional phase error calculation, directly subtracting the two phases has a fatal flaw: since the AC phase cycles in a periodic sawtooth pattern between 0 and 2π, when the two phases cross the boundary of 2π, direct subtraction will cause a huge numerical step change, leading to instantaneous instability of the controller.

[0131] To completely eliminate this defect, this example introduces a nonlinear sinusoidal mapping function into the error extraction loop. The microgrid pre-synchronization phase error at time t... The observation equations are reconstructed into the following form.

[0132] .

[0133] In the critical region of grid connection, the phase difference between the large power grid and the microgrid gradually decreases. Based on the small-angle approximation principle of Taylor series, the following condition is met: This mechanism perfectly avoids the abrupt change in the phase reversal point without sacrificing tracking accuracy.

[0134] The frequency compensation command is sent to the discretized PI controller to solve the discretized solution equation as follows.

[0135] .

[0136] in, This represents the original frequency compensation amount output by the PI controller at the k-th sampling time. This is the proportional coefficient of the pre-synchronous PI controller, its function is to quickly respond to the current phase error. Let be the microgrid pre-synchronization phase error at the k-th sampling time. These are the integral coefficients of the pre-synchronous PI converter. Their function is to eliminate steady-state phase error. As long as a phase error exists, the integral term will continue to accumulate until the error is zero. Let be the microgrid pre-synchronization phase error at the j-th sampling time.

[0137] If the phase of the large power grid leads that of the microgrid ( When the frequency deviation is greater than 0, the PI controller will output a positive frequency compensation, prompting the microgrid inverter to accelerate its operation at the base frequency (e.g., 50Hz). Since the microgrid frequency is higher than the main grid frequency at this time, its phase will gradually approach that of the main grid. As the phase deviation continues to decrease, the output compensation of the PI controller will also adaptively decrease. When the two are completely in phase and frequency, the error returns to zero, and the frequency compensation also dynamically converges to zero.

[0138] Based on the same inventive concept, this application also provides a microgrid energy storage converter control device for implementing the microgrid energy storage converter control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the microgrid energy storage converter control device provided below can be found in the limitations of the microgrid energy storage converter control method described above, and will not be repeated here.

[0139] In one exemplary embodiment, such as Figure 5 As shown, a microgrid energy storage converter control device is provided, including a state data acquisition module 001, a global average state parameter calculation module 002, a droop reference state calculation module 003, and an energy storage converter control module 004.

[0140] The status data acquisition module 001 is used to acquire the status data of any energy storage converter in a microgrid including multiple energy storage converters. The status data includes measured values ​​of status parameters, globally estimated values ​​of status parameters, and globally estimated values ​​of status parameters of interconnected energy storage converters. The status parameters include frequency, voltage, active power, and reactive power. The interconnected energy storage converters are those in the microgrid that are communicatively connected to the energy storage converters.

[0141] The global average state parameter calculation module 002 is used to obtain the global average state parameters of the energy storage converter based on the state data using an optimized distributed average consensus algorithm. The optimized distributed average consensus algorithm is based on an integral compensation mechanism to eliminate steady-state errors in the distributed average consensus algorithm.

[0142] The droop reference state calculation module 003 is used to obtain the droop reference state of the energy storage converter by adopting a droop control strategy based on secondary compensation, according to the preset state reference value, the global average state parameter, and the measured value of the state parameter of the energy storage converter.

[0143] The energy storage converter control module 004 is used to obtain a voltage modulation signal by using a voltage drive modulation method based on the droop reference state, the global average state parameters and the measured values ​​of the state parameters of the energy storage converter, and to control the energy storage converter according to the voltage modulation signal.

[0144] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores the status data of any energy storage converter in the microgrid. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a microgrid energy storage converter control method.

[0145] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer equipment to which the present application is applied. Specific computer equipment may include, for example, [the following is a list of possible additional structures]. Figure 6 The diagram shows more or fewer components, or combinations of certain components, or different component arrangements.

[0146] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0147] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0148] To address the problems of low power distribution accuracy, voltage and frequency steady-state error, and mode switching impact in traditional droop control during multi-PCS parallel operation in microgrids, this application proposes a hierarchical control architecture. In the bottom-level electrical control, an adaptive virtual impedance is introduced to suppress transient reactive circulating current caused by line impedance mismatch. In the upper-level coordination layer, the agents corresponding to each PCS engage in deterministic, low-latency state interaction through a TSN network, utilizing a distributed consensus algorithm and a fuzzy adaptive PI controller to dynamically tune the droop coefficient and secondary compensation commands online. Simultaneously, a ROCOF-based autonomous sensing and decoupling pre-synchronization mechanism is combined to achieve seamless switching between grid-connected and off-grid environments. This application ensures high-precision current sharing and zero voltage and frequency steady-state error recovery across multiple PCS while leveraging TSN to solve the latency and jitter problems in multi-agent communication, comprehensively improving the transient robustness and operational resilience of the microgrid under extreme conditions.

[0149] The core idea of ​​this application is to deeply integrate deterministic network communication with intelligent adaptive control algorithms. By introducing a communication network based on a TSN architecture at the physical layer, microsecond-level clock synchronization and bounded low-latency data interaction channels are provided for each PCS agent. At the control logic layer, the lower layer uses droop control with adaptive virtual impedance for basic support; the upper layer uses high-precision state variables based on TSN interaction, and a distributed average consensus algorithm for secondary compensation. Simultaneously, fuzzy logic is introduced to adaptively adjust PI parameters online, fundamentally solving the overshoot and slow recovery problems when the system faces sudden load changes. Finally, by combining ROCOF sensing and decoupled pre-synchronization logic, seamless switching between microgrid grid connection and off-grid operation is achieved.

[0150] The primary objective of this application is to achieve high steady-state accuracy and transient stability of a microgrid parallel PCS system under dynamic processes of grid connection / disconnection and load abrupt changes, minimizing circulating currents between multiple converters and eliminating frequency and voltage steady-state errors generated by traditional control methods. Through a distributed multi-agent control algorithm, based on nanosecond-level clock synchronization and deterministic low-latency communication provided by the TSN network, the virtual impedance parameters and secondary compensation commands of each converter are adjusted in real time, thereby optimizing the overall system's operational robustness and the smoothness of mode switching.

[0151] In one example of this application, a multi-agent cooperative control algorithm is used to dynamically tune the control gain and reference vector of each PCS controller. First, a distributed average consensus frequency and voltage observer is designed to obtain the global operating state of the microgrid, and a rapid sensing mechanism based on the rate of change of frequency (ROCOF) is introduced. By setting fuzzy adaptive control logic, the algorithm aims to automatically adjust the parameters of the PI controller for load disturbances of varying severity. To achieve this goal, the system dynamically distributes secondary frequency recovery and pre-synchronization phase compensation quantities by real-time monitoring of the converter output current, bus voltage, and network synchronization status, utilizing the time-aware shaping mechanism of the TSN network.

[0152] This application utilizes a distributed consensus and collaborative strategy to improve the global synchronization and response speed of control decisions through deterministic information interaction among intelligent agent nodes. Specifically, within each 0.1ms control cycle, the electrical state of each PCS intelligent agent (such as locally measured power, frequency deviation, calculated virtual impedance value, etc.) is used as input and transmitted conflict-free through the TSN network. The intelligent agent selects the optimal parameter compensation operation based on the current consensus convergence status and uses a historical state back-calculation mechanism to evaluate and reset the controller state at the moment of mode switching. As the system operates, each intelligent agent gradually achieves adaptive collaborative control under various extreme transient conditions (such as three-phase load imbalance, circuit breaker operation).

[0153] The beneficial effects of this application are as follows: By introducing TSN technology, the risk of state synchronization loss caused by communication jitter and delay in traditional multi-agent control is eliminated from the network layer, ensuring the safety and stability of the 0.1ms-level high-speed cooperative control loop. The bottom-layer adaptive virtual impedance compensates for the difference in physical impedance voltage drop, and the upper-layer consensus algorithm dynamically fine-tunes the droop coefficient. The two are orthogonally decoupled, achieving high-precision active / reactive current sharing of multiple PCS under arbitrarily complex lines. The fuzzy adaptive PI controller breaks the rigidity of fixed parameters, automatically suppressing the integral term to prevent integral saturation in the initial stage of heavy load surges, and strengthening the integral term to eliminate steady-state error during the steady-state period, significantly reducing system overshoot and recovery time. The ROCOF algorithm achieves highly sensitive autonomous perception of the grid state, combined with a non-disruptive mechanism of "historical state back-calculation" at the moment of switching, fundamentally eliminating the inrush current and power oscillation at the moment of circuit breaker closing or opening, improving system lifespan and resilience. This application can effectively solve the problems of low power distribution accuracy, voltage and frequency steady-state error, and mode switching impact of traditional droop control when multiple PCS are connected in parallel in microgrids, thereby enhancing the safety and stability of the power grid and having broad application prospects and promotion value.

[0154] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0155] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0156] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0157] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A control method for a microgrid energy storage converter, characterized in that, The method includes: For any energy storage converter in a microgrid, acquire the status data of the energy storage converter; the microgrid includes multiple energy storage converters; the status data includes measured values ​​of status parameters, globally estimated values ​​of status parameters, and globally estimated values ​​of status parameters of interconnected energy storage converters; the status parameters include frequency, voltage, active power, and reactive power; the interconnected energy storage converters are energy storage converters in the microgrid that are communicatively connected to the energy storage converters. Based on the state data, an optimized distributed average consensus algorithm is used to obtain the global average state parameters of the energy storage converter; the optimized distributed average consensus algorithm is based on an integral compensation mechanism to eliminate steady-state errors in the distributed average consensus algorithm. Based on the preset state reference value, the global average state parameters, and the measured state parameters of the energy storage converter, a droop control strategy based on secondary compensation is adopted to obtain the droop reference state of the energy storage converter. Based on the droop reference state, the global average state parameters, and the measured values ​​of the state parameters of the energy storage converter, a voltage-driven modulation method is used to obtain a voltage modulation signal, and the energy storage converter is controlled according to the voltage modulation signal.

2. The microgrid energy storage converter control method according to claim 1, characterized in that, The optimized distributed average consensus algorithm solves for the global average state parameters of the energy storage converter through iterative calculations. The global average state parameters include global average frequency estimates and global average voltage estimates. The global estimates of the state parameters of the energy storage converter include the global average frequency estimates and global average voltage estimates obtained in the previous iteration. The global estimates of the state parameters of the interconnected energy storage converter include the global average frequency estimates and global average voltage estimates of the interconnected energy storage converter in the previous iteration. The optimized distributed average consensus algorithm calculates the global average frequency estimate at any iteration time using the following formula: ； Where i represents the identifier of the energy storage converter in the microgrid. Let be the estimated global average frequency of energy storage converter i at time t. The measured frequency of energy storage converter i at time t. For energy storage converter i, it is an interconnected set of energy storage converters. denoted by , where is the number of interconnected energy storage converters of energy storage converter i, and j is the identifier of the interconnected energy storage converter of energy storage converter i. The weighting coefficients of the adjacency matrix of the communication topology between energy storage converter i and interconnected energy storage converter j are given. For interconnected energy storage converters j in The global average frequency estimate at time t. For energy storage converter i in The global average frequency estimate at time t; The optimized distributed average consensus algorithm calculates the global average voltage estimate at any iteration time using the following formula: ； in, Let be the estimated global average voltage of energy storage converter i at time t. Let be the measured voltage value of energy storage converter i at time t. For interconnected energy storage converters j in The estimated global average voltage at time t. For energy storage converter i in The estimated global average voltage at time t.

3. The microgrid energy storage converter control method according to claim 1, characterized in that, The preset state reference values ​​include the rated frequency reference value and the rated voltage reference value; the global average state parameters include the global average frequency estimate, the global average voltage estimate, the global average active power estimate, and the global average reactive power estimate. Based on the preset state reference value, the global average state parameters, and the measured state parameters of the energy storage converter, a droop control strategy based on secondary compensation is adopted to obtain the droop reference state of the energy storage converter, specifically including: The global average tracking error of the energy storage converter is calculated based on the rated frequency reference value, the rated voltage reference value, the global average frequency estimate, and the global average voltage estimate; the global average tracking error includes the global average frequency tracking error and the global average voltage tracking error. Based on the global average tracking error, a fuzzy control algorithm is used to obtain the proportional gain and integral gain; the proportional gain includes frequency-secondary compensation proportional gain and voltage-secondary compensation proportional gain; the integral gain includes frequency-secondary compensation integral gain and voltage-secondary compensation integral gain. Based on the global average tracking error, the proportional gain, and the integral gain, a proportional-integral control algorithm is used to obtain the secondary state compensation amount of the energy storage converter; the secondary state compensation amount includes the frequency secondary state compensation amount and the voltage secondary state compensation amount. Based on the preset state reference value, the measured active power value of the energy storage converter, the measured reactive power value of the energy storage converter, the estimated global average active power value, the estimated global average reactive power value, and the secondary state compensation amount, a droop control method is used to obtain the droop reference state; the droop reference state includes the droop reference voltage and the droop reference frequency.

4. The microgrid energy storage converter control method according to claim 3, characterized in that, Based on the global average tracking error, a fuzzy control algorithm is used to obtain the proportional gain and integral gain, specifically including: By taking the derivatives of the global average frequency tracking error and the global average voltage tracking error respectively, the rate of change of the global average frequency tracking error and the rate of change of the global average voltage tracking error are obtained. The frequency tracking error input is obtained based on the global average frequency tracking error and the rate of change of the global average frequency tracking error, and the voltage tracking error input is obtained based on the global average voltage tracking error and the rate of change of the global average voltage tracking error. Based on the triangular membership function and the direct universe of discourse, a frequency fuzzy optimization rule base and a voltage fuzzy optimization rule base are built according to the frequency tracking error input and the voltage tracking error input, respectively. The Midani-type fuzzy inference method is used to perform rule matching operations based on the frequency fuzzy optimization rule base and the voltage fuzzy optimization rule base respectively, to obtain the frequency output fuzzy interval and the voltage output fuzzy interval. The centroid method is used to perform defuzzification operations on the frequency output fuzzy range and the voltage output fuzzy range respectively to obtain the proportional gain and the integral gain.

5. The microgrid energy storage converter control method according to claim 1, characterized in that, The global average state parameters include the estimated global average reactive power; the droop reference state includes the droop reference voltage and the droop reference frequency. Based on the droop reference state, the global average state parameters, and the measured values ​​of the state parameters of the energy storage converter, a voltage-driven modulation method is used to obtain a voltage modulation signal, and the energy storage converter is controlled according to the voltage modulation signal, specifically including: Based on the droop reference voltage and droop reference frequency, a coordinate transformation preprocessing method is used to obtain the droop control voltage command; the droop control voltage command includes a d-axis voltage command component and a q-axis voltage command component. Based on the estimated global average reactive power and the measured reactive power of the energy storage converter, the virtual impedance of the energy storage converter is calculated using the impedance correction method. The feedback current of the energy storage converter is calculated based on the measured voltage, active power, and reactive power values ​​of the energy storage converter; the feedback current includes a d-axis feedback current component and a q-axis feedback current component. Based on the droop control voltage command, the virtual impedance, and the feedback current, a virtual impedance droop correction method is used to calculate the corrected control voltage command; the corrected control voltage command includes a d-axis corrected voltage command component and a q-axis corrected voltage command component. According to the corrected control voltage command, a vector control modulation method is used to obtain a voltage modulation signal, and the energy storage converter is controlled according to the voltage modulation signal.

6. The microgrid energy storage converter control method according to claim 1, characterized in that, The drooping reference state includes the drooping reference frequency; After obtaining the droop reference state of the energy storage converter based on a droop control strategy with secondary compensation, according to the preset state reference value, the global average state parameter, and the measured state parameter value of the energy storage converter, the microgrid energy storage converter control method further includes: Based on the measured frequency value of the energy storage converter, the frequency change rate of the energy storage converter is calculated, and the frequency change rate is subjected to low-pass filtering to obtain the filtered frequency change rate. The operating conditions of the microgrid are determined based on the rate of change of the filter frequency; the operating conditions include normal grid-connected conditions and islanded conditions; under normal grid-connected conditions, the main grid and the microgrid are normally connected; under islanded conditions, the connection between the main grid and the microgrid is disconnected. When the operating condition is islanded, the voltage of the main power grid is collected in real time; When the voltage of the main power grid is greater than a preset threshold, a phase-locked loop phase extraction method is used to obtain the phase of the main power grid; Based on the measured voltage value of the energy storage converter, the phase of the energy storage converter is obtained using a phase-locked loop phase extraction method. Based on the sinusoidal mapping phase error extraction method, the pre-synchronization phase error between the phase of the main power grid and the phase of the energy storage converter is calculated; Based on the pre-synchronization phase error, the frequency compensation amount of the energy storage converter is obtained by using a proportional-integral control algorithm. The droop reference frequency is compensated according to the frequency compensation amount.

7. The microgrid energy storage converter control method according to claim 1, characterized in that, Each energy storage converter in the microgrid is configured as an independent intelligent agent node, and each intelligent agent node communicates using a time-sensitive network protocol to achieve clock synchronization.

8. A microgrid energy storage converter control device, used to implement the microgrid energy storage converter control method according to any one of claims 1-7, characterized in that, The device includes: The status data acquisition module is used to acquire the status data of any energy storage converter in a microgrid including multiple energy storage converters. The status data includes measured values ​​of status parameters, global estimated values ​​of status parameters, and global estimated values ​​of status parameters of interconnected energy storage converters. The status parameters include frequency, voltage, active power, and reactive power. The interconnected energy storage converters are the energy storage converters in the microgrid that are communicatively connected to the energy storage converters. The global average state parameter calculation module is used to obtain the global average state parameters of the energy storage converter based on the state data and an optimized distributed average consensus algorithm. The optimized distributed average consensus algorithm is based on an integral compensation mechanism to eliminate steady-state errors in the distributed average consensus algorithm. The droop reference state calculation module is used to obtain the droop reference state of the energy storage converter by adopting a droop control strategy based on secondary compensation, according to the preset state reference value, the global average state parameter, and the measured value of the state parameter of the energy storage converter. The energy storage converter control module is used to obtain a voltage modulation signal by using a voltage-driven modulation method based on the droop reference state, the global average state parameters and the measured values ​​of the state parameters of the energy storage converter, and to control the energy storage converter according to the voltage modulation signal.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the microgrid energy storage converter control method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the microgrid energy storage converter control method according to any one of claims 1-7.

Images