Network construction type energy storage semi-physical simulation test platform and method

Abstract

The invention discloses a networking type energy storage semi-physical simulation test platform and method, and particularly relates to the field of simulation testing, and the platform comprises a digital twin collaborative modeling module, an impedance adaptive hardware coupling module, a networking control depth verification module, a multi-dimensional performance evaluation module, and an iterative optimization module. According to the method, a dynamic impedance network model and a virtual energy storage unit are integrated by constructing a power grid-energy storage digital twin; the dynamic impedance network is generated by adopting a frequency-dependent parameter transmission line equation and combining actual measurement impedance scanning data in a fitting manner; in the loop test, a full-band disturbance injection method is adopted to monitor impedance characteristics, and virtual inertia and physical droop control parameters are coordinated under a preset weak power grid working condition; the multi-dimensional performance evaluation system quantifies the system performance from three dimensions of dynamic response, stability margin and transient energy function, and when the phase margin deviation is greater than 15 degrees, triggers an iterative optimization mechanism, reconstructs a virtual inertia coefficient and repeatedly verifies the virtual inertia coefficient until a stability threshold is met.

Description

Technical Field

[0001] The present invention relates to the technical field of simulation testing, and more specifically, to a grid-forming energy storage semi-physical simulation testing platform and method. Background Art

[0002] With the rapid development of renewable energy and the construction of smart grids, energy storage systems are increasingly widely used in power systems. As a new type of energy storage technology, the grid-forming energy storage system (Grid-Forming Energy Storage System) can actively support grid voltage and frequency, improve the stability and reliability of the grid, and has become a research hotspot in the energy storage field; however, the complexity and dynamic characteristics of the grid-forming energy storage system pose many challenges to its testing and verification.

[0003] However, in actual use, there are still some disadvantages. For example, traditional physical testing usually needs to be carried out in an actual power grid environment, with high testing costs, high risks, and it is difficult to simulate various extreme working conditions and fault scenarios; in addition, the repeatability and controllability of physical testing are poor, and it is difficult to comprehensively evaluate the dynamic characteristics of the system. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a grid-forming energy storage semi-physical simulation testing platform and method to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] Digital twin collaborative modeling module: Construct a power grid-energy storage digital twin in the real-time simulation unit, and the digital twin includes a dynamic impedance network model and a virtual grid-forming energy storage unit; the virtual grid-forming energy storage unit synchronizes real-time data with the physical energy storage device controller through an error compensation algorithm;

[0007] Impedance adaptive hardware coupling module: Connect the physical energy storage device to the simulation loop through a power amplifier, and collect the impedance characteristics of the device output end in real time; dynamically adjust the damping compensation parameters of the power amplifier based on the impedance scanning results;

[0008] Grid-forming control depth verification module: Inject a preset weak grid condition into the digital twin to activate the grid-forming control strategy of the physical device; synchronously adjust the inertia simulation coefficient of the virtual energy storage unit and the droop control parameters of the physical device to generate a two-way parameter interaction instruction;

[0009] Multi-dimensional performance evaluation module: Calculate the active power response rate of the physical device under frequency disturbance; analyze the phase margin deviation of the virtual-physical dual-end data at the impedance mutation point to determine the control stability boundary;

[0010] Iterative optimization module: When the phase margin deviation is greater than 15°, the virtual inertia coefficient is reconstructed based on the Lyapunov function, and the network construction control depth verification module to the multi-dimensional performance evaluation module is repeated until the stability threshold is met.

[0011] Preferably, in the digital twin collaborative modeling module, a power grid - energy storage digital twin is constructed in the real-time simulation unit. The digital twin includes a dynamic impedance network model and a virtual network-forming energy storage unit, and real-time data synchronization with the physical energy storage device controller is achieved through an error compensation algorithm. The dynamic impedance network model uses a frequency-variable parameter transmission line equation;

[0012] The virtual network-forming energy storage unit is constructed based on a multi-time scale state space model. At the millisecond level, the IGBT loss - thermal coupling equation is used to simulate the switching dynamics of power devices, and at the second level, the virtual synchronous machine rotor motion equation is executed;

[0013] At the same time, an integrated control instruction pre-compiler is used to convert the PWM signal output by the physical controller into an equivalent voltage source that can be parsed by the simulation;

[0014] The error compensation real-time synchronization mechanism adopts a two-way delay observation - compensation channel design. The forward channel marks the time stamp of the physical device control signal and then inputs it into the digital twin, and the transmission delay is estimated through a sliding window filter.

[0015] Preferably, in the impedance adaptive hardware coupling module, the physical energy storage device is connected to the real-time simulation loop through an interface via a power amplifier to construct a hardware-in-the-loop test system, and the impedance characteristics of the device output end are collected in real time and the system parameters are dynamically adjusted. Specifically: the full-frequency band perturbation injection method is adopted, and white noise test signals with an amplitude not exceeding 2% of the rated voltage are injected in the wide frequency range of 0.1 - 2000Hz, and the voltage and current responses are monitored in real time through a high-sampling rate data acquisition card;

[0016] Among them, the full-frequency band perturbation injection method is used to perform wide-band excitation testing on the device under test; the test signal is generated by a high-performance arbitrary waveform generator, and white noise signals with an amplitude strictly controlled within 2% of the rated voltage are injected in the frequency range of 0.1 - 2000Hz. After being shaped by a 6th-order Butterworth band-pass filter, the signal is coupled to the port of the device under test through an isolation amplifier; a 24-bit high-precision data acquisition card with a synchronous sampling rate of 500kS / s is used to synchronously collect three-phase voltage and current signals in a hardware-triggered manner, the sampling time window is set to 10 power frequency cycles, and the collected time-domain data is converted into frequency-domain impedance characteristics through real-time FFT processing;

[0017] Based on the frequency-domain analysis method, the ratio curve of the output impedance of the physical device to the equivalent impedance of the power grid is calculated. The Nyquist stability criterion is used to perform a closed-loop analysis on the trajectory of the impedance ratio. When the trajectory approaches the point (-1, j0) and the phase margin is less than 30 degrees, it is determined that there is a risk of interactive resonance;

[0018] For the identified resonance points, the module automatically activates a multi-level compensation mechanism. In the low-frequency band below 500 Hz, a passive damping strategy is adopted by adjusting the cut-off frequency of the output filter of the power amplifier; in the medium-frequency band of 500 - 1500 Hz, an active damper is enabled to dynamically generate a reverse compensation current according to the resonance point frequency.

[0019] Preferably, in the grid-forming control depth verification module, a closed-loop test environment with virtual-real interaction is established to verify the control performance of the grid-forming energy storage system under extremely weak grid conditions; this module first presets two types of typical weak grid test scenarios in the digital twin: one is the impedance phase jump condition, where the equivalent impedance angle of the power grid is quickly switched from 30° to 75° within 100 ms to simulate the drastic change of the power grid characteristics caused by the sudden switching of transmission lines; the other is the SCR gradual perturbation condition, where the system short-circuit capacity ratio linearly decreases from SCR = 3 to SCR = 1.5 at a rate not lower than 1 SCR / second to simulate the continuous deterioration process of the grid strength caused by the large-scale disconnection of new energy.

[0020] During the test process, the module synchronously coordinates the interaction of control parameters between the virtual and physical systems: the virtual energy storage unit adjusts the inertia simulation coefficient in real time, and its adjustment algorithm is designed based on the Lyapunov stability theory;

[0021] In the steady-state characteristic analysis, by calculating the THD change rate of the voltage at the point of common coupling before and after the impedance phase jump, the improvement effect of the control strategy on the power quality is evaluated;

[0022] Among them, in the steady-state characteristic analysis, the voltage waveform at the point of common coupling is monitored in real time by a high-precision power quality analyzer, and the sampling rate is set to 256 points / cycle; when a preset impedance phase jump (such as a step change from 30° to 75°) is detected, the THD dynamic analysis process is automatically triggered: first, the voltage waveforms of the previous 5 power frequency cycles before the jump are collected as reference data, and then the dynamic process of 20 cycles after the jump is continuously recorded; the improved IEC 61000-4-7 standard algorithm is used to calculate the harmonic contents of each order, especially focusing on the characteristic harmonics of 2 - 25 times; when the THD change rate is detected to be greater than 15% and remains out of limit, the control parameter optimization is automatically triggered to adjust the bandwidth of the phase-locked loop. If it does not converge within 3 cycles, it switches to the harmonic compensation mode and injects a reverse harmonic current component into the d-q axis current loop.

[0023] Preferably, in the multi-dimensional performance evaluation module, a three-dimensional evaluation system including dynamic response, stability margin, and energy characteristics is constructed to comprehensively quantify the performance of the network-forming energy storage system; in the dimension of dynamic response evaluation, the module monitors the active power change rate of physical devices under frequency perturbation in real time. The calculation uses the least squares method to fit the slope of the power-time curve, and the dynamic response benchmark that satisfies the active power change rate greater than 10% of the rated power of the device per second is met; at the same time, the response time difference between the virtual model and the physical device is collected to establish a dynamic consistency index;

[0024] The specific calculation method is as follows:

[0025]

[0026] Among them, is expressed as the dynamic consistency index, is expressed as the actual response time, is expressed as the reference time constant, and its value is 20 ms; when DCI < 0.9, the control parameter re-tuning is triggered, that is, when the dynamic consistency index is lower than the threshold of 0.9, the control parameters are automatically or manually adjusted;

[0027] In the stability margin analysis, for the impedance mutation condition, based on the dual-end data synchronous acquisition unit, a closed-loop transfer function matrix including virtual and physical systems is constructed , by solving the root locus of the characteristic equation , the phase margin deviation Δφ of the key frequency point is calculated; when it is detected that Δφ > 15°, it is automatically marked as a stability risk point.

[0028] Preferably, in the iterative optimization module, in the iterative optimization stage, when it is detected that the phase margin deviation exceeds 15°, the system automatically triggers a parameter adjustment mechanism based on the Lyapunov stability theory; first, a Lyapunov function including the phase margin error term and the virtual inertia offset term is constructed;

[0029] The updated virtual inertia will be sent to the virtual energy storage unit in the digital twin in real time, and at the same time, the droop coefficient of the physical device is synchronously adjusted through the bidirectional parameter interaction interface; then the dynamic test process of the network-forming control depth verification module is re-executed, and new phase margin data is collected under the same perturbation conditions. If the convergence condition of | | ≤ 5° is not met, the iteration continues until the maximum iteration number is reached; during each iteration process, the evolution trajectories of , g, and key state variables are recorded to form a parameter optimization path report.

[0030] The technical effects and advantages of the present invention:

[0031] The present invention constructs a power grid - energy storage digital twin containing a dynamic impedance network model and a virtual energy storage unit through a digital twin collaborative modeling module, and realizes real - time data synchronization with the physical device controller through an error compensation algorithm; the impedance adaptive hardware coupling module connects the physical energy storage device to the simulation loop, collects impedance characteristics in real - time, and dynamically adjusts the parameters of the power amplifier to suppress high - frequency oscillations; the grid - forming control depth verification module presets weak grid conditions and synchronously adjusts the parameters of the virtual and physical systems; the multi - dimensional performance evaluation module constructs a three - dimensional evaluation system to quantitatively evaluate the system performance; when the phase margin deviation > 15°, the iterative optimization module reconstructs the virtual inertia coefficient based on the Lyapunov function, repeats verification and evaluation until the stability threshold is met, and records the parameter evolution trajectory to form a report.

[0032] The present invention breaks through the limitations of traditional methods in dynamic performance verification through a closed - loop coupling architecture of "digital simulation - power interface - physical device", provides a highly reliable test method for the grid - connection stability of grid - forming energy storage, and improves the control performance and stability evaluation accuracy of the grid - forming energy storage system. Brief Description of the Drawings

[0033] Figure 1 It is a schematic diagram of the connection of the platform modules of the present invention.

[0034] Figure 2 It is a schematic diagram of the method flow of the present invention. Detailed Embodiments

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0036] Please refer to Figure 1 As shown, the present invention provides a grid - forming energy storage hardware - in - the - loop simulation test platform, including a digital twin collaborative modeling module, an impedance adaptive hardware coupling module, a grid - forming control depth verification module, a multi - dimensional performance evaluation module, and an iterative optimization module.

[0037] Digital twin collaborative modeling module: Construct a power grid - energy storage digital twin in the real - time simulation unit. The digital twin includes a dynamic impedance network model and a virtual grid - forming energy storage unit; the virtual grid - forming energy storage unit realizes real - time data synchronization with the physical energy storage device controller through an error compensation algorithm.

[0038] In the digital twin collaborative modeling module, a power grid - energy storage digital twin is constructed in the real-time simulation unit. The digital twin includes a dynamic impedance network model and a virtual grid-forming energy storage unit, and real-time data synchronization with the physical energy storage device controller is achieved through an error compensation algorithm. The dynamic impedance network model uses a frequency-variable parameter transmission line equation:

[0039]

[0040] Among them, represents the power grid impedance, represents the real part resistance of the impedance, represents the imaginary part reactance of the impedance, represents the amplitude of the k-th frequency-variable component, represents the time constant of the k-th frequency-variable component, f represents the frequency, k represents the index number of the frequency-variable component, and j represents the imaginary unit;

[0041] and are generated by fitting the measured impedance scan data; j is used to distinguish the real part and the imaginary part of the impedance. The real part represents the resistance, and the imaginary part represents the reactance;

[0042] The virtual grid-forming energy storage unit is constructed based on a multi-time scale state space model. At the millisecond level, the IGBT loss - thermal coupling equation is used to simulate the switching dynamics of power devices, and at the second level, the virtual synchronous machine rotor motion equation is executed:

[0043]

[0044] Among them, represents the rotor angular acceleration, represents the rotor angular velocity, H represents the inertia constant, represents the input mechanical torque, represents the output electromagnetic torque, represents the damping coefficient, represents the angular velocity deviation;

[0045] At the same time, an integrated control instruction pre-compiler is used to convert the PWM signal output by the physical controller into an equivalent voltage source that can be parsed by the simulation;

[0046] The error compensation real-time synchronization mechanism adopts a two-way delay observation - compensation channel design. The forward channel marks the time stamp of the physical device control signal and then inputs it into the digital twin, and estimates the transmission delay through a sliding window filter;

[0047] Among them, the forward channel adopts a hardware timestamp marking unit based on FPGA, inserts 16-bit precision synchronous timestamps at both the rising edge and the falling edge of the PWM drive signal output by the physical device controller, and transmits them to the digital twin through a dedicated communication protocol with CRC check; the digital twin side deploys an adaptive sliding window delay estimation algorithm, which adopts a variable-weight exponential forgetting strategy;

[0048] The feedback channel injects the output of the virtual unit into the physical controller after dynamic gain compensation, and at the same time establishes a data consistency check protocol, compares the phase difference of the terminal voltage between the physical device and the virtual unit every 10 ms, and triggers online correction of the model parameters when the phase difference is greater than 2°.

[0049] Impedance adaptive hardware coupling module: Connect the physical energy storage device to the simulation loop through a power amplifier, and collect the impedance characteristics of the device output end in real time; Dynamically adjust the damping compensation parameters of the power amplifier based on the impedance scanning results to suppress high-frequency oscillation;

[0050] In the impedance adaptive hardware coupling module, the physical energy storage device is connected to the real-time simulation loop through a power amplifier through an interface, a hardware-in-the-loop test system is constructed, and the impedance characteristics of the device output end are collected in real time and the system parameters are dynamically adjusted. Specifically: The full-band perturbation injection method is adopted to inject a white noise test signal with an amplitude not exceeding 2% of the rated voltage in the wide frequency range of 0.1~2000Hz, and the voltage and current responses are monitored in real time through a high-sampling-rate data acquisition card;

[0051] Among them, the full-band perturbation injection method is used to perform broadband excitation tests on the device under test; the test signal is generated by a high-performance arbitrary waveform generator, and a white noise signal with an amplitude strictly controlled within 2% of the rated voltage is injected in the frequency range of 0.1~2000Hz. After being shaped by a 6th-order Butterworth band-pass filter, the signal is coupled to the port of the device under test through an isolation amplifier; A 24-bit high-precision data acquisition card with a synchronous sampling rate of 500kS / s is used to synchronously collect three-phase voltage and current signals in a hardware-triggered manner. The sampling time window is set to 10 power frequency cycles, and the collected time-domain data is converted into frequency-domain impedance characteristics through real-time FFT processing;

[0052] Based on the frequency-domain analysis method, the ratio curve of the output impedance of the physical device to the equivalent impedance of the power grid is calculated, and the Nyquist stability criterion is used to perform a closed-loop analysis on the trajectory of the impedance ratio. When the trajectory approaches the point (-1,j0) and the phase margin is less than 30 degrees, it is determined that there is a risk of interactive resonance;

[0053] For the identified resonance points, the module automatically activates a multi-level compensation mechanism. In the low-frequency band below 500 Hz, a passive damping strategy is adopted by adjusting the cut-off frequency of the output filter of the power amplifier; in the medium-frequency band of 500 - 1500 Hz, an active damper is enabled to dynamically generate a reverse compensation current according to the resonance point frequency. The specific calculation method is as follows:

[0054]

[0055] Among them, represents the reverse compensation current generated by the active damper, represents the damping gain coefficient, represents the PCC point voltage, represents the rate of change of the PCC point voltage;

[0056] The specific calculation method of the gain coefficient is as follows:

[0057]

[0058] Among them, represents the damping gain coefficient, represents the frequency of the resonance point, represents the equivalent capacitance parameter;

[0059] For high-frequency resonances above 1500 Hz, it switches to the digital notch filter mode, and a second-order IIR filter is implanted in the control loop to attenuate the gain of a specific frequency band.

[0060] Grid-forming control depth verification module: Inject a preset weak grid condition into the digital twin, and activate the grid-forming control strategy of the physical device; synchronously adjust the inertia simulation coefficient of the virtual energy storage unit and the droop control parameter of the physical device to generate a two-way parameter interaction command;

[0061] In the grid-forming control depth verification module, a closed-loop test environment for virtual-real interaction is established to verify the control performance of the grid-forming energy storage system under extremely weak grid conditions; this module first presets two types of typical weak grid test scenarios in the digital twin: one is the impedance phase jump condition, where the equivalent impedance angle of the grid is quickly switched from 30° to 75° within 100 ms to simulate the sudden change of grid characteristics caused by the sudden switching of transmission lines; the other is the SCR gradual perturbation condition, where the system short-circuit capacity ratio decreases linearly from SCR = 3 to SCR = 1.5, and the change rate is not less than 1 SCR / second, to simulate the continuous deterioration process of grid strength caused by the large-scale disconnection of new energy.

[0062] During the test, the module synchronously coordinates the control parameter interaction between the virtual and physical systems: The virtual energy storage unit adjusts the inertia simulation coefficient in real time, and its adjustment algorithm is designed based on the Lyapunov stability theory. The specific calculation method is as follows:

[0063]

[0064] Among them, represents the real-time inertia simulation coefficient of the virtual energy storage unit, represents the initial inertia value, represents the adaptive gain coefficient, frequency deviation;

[0065] Meanwhile, the calculation method of the droop control parameter on the physical device side is specifically as follows:

[0066]

[0067] Among them, represents the real-time droop control parameter on the physical device side, represents the reference droop reference value, represents the change sensitivity factor, represents the change amount of the short-circuit capacity ratio; Two-way parameter interaction is realized through a high-speed communication bus to ensure that the control characteristics of the virtual model and the physical device evolve synchronously;

[0068] In the transient process evaluation, a comprehensive performance index is used to quantify the system dynamic response:

[0069]

[0070] Among them, CPI represents the comprehensive performance index, represents the maximum frequency deviation in the dynamic process, represents the rated frequency, represents the recovery time, represents the reference value;

[0071] In the steady-state characteristic analysis, by calculating the change rate of the THD of the voltage at the common connection point before and after the impedance phase jump, the improvement effect of the control strategy on the power quality is evaluated;

[0072] Among them, in the steady-state characteristic analysis, the voltage waveform at the point of common coupling is monitored in real time by a high-precision power quality analyzer, and the sampling rate is set to 256 points per cycle; when a preset jump in impedance phase (such as a step change from 30° to 75°) is detected, the THD dynamic analysis process is automatically triggered: First, the voltage waveforms of the first 5 power frequency cycles before the jump are collected as reference data, and then the dynamic process of the 20 cycles after the jump is continuously recorded; the improved IEC 61000-4-7 standard algorithm is used to calculate the harmonic contents of each order, and special attention is paid to the characteristic harmonics from the 2nd to the 25th order; when the change rate of THD is detected to be greater than 15% and remains out of limit, the control parameter optimization is automatically triggered, the PLL bandwidth is adjusted, and if it does not converge within 3 cycles, it switches to the harmonic compensation mode, and a reverse harmonic current component is injected into the d-q axis current loop.

[0073] Multi-dimensional performance evaluation module: Calculate the active power response rate of physical devices under frequency disturbances; Analyze the phase margin deviation of virtual-physical dual-end data at the impedance mutation point to determine the control stability boundary.

[0074] In the multi-dimensional performance evaluation module, the multi-dimensional performance evaluation module constructs a three-dimensional evaluation system including dynamic response, stability margin and energy characteristics for a comprehensive quantitative evaluation of the performance of the network-forming energy storage system; In the dynamic response evaluation dimension, the module monitors in real time the change rate of the active power of physical devices under frequency disturbances, and its calculation uses the least squares method to fit the slope of the power-time curve, meeting the dynamic response benchmark of the rated power of the device with an active power change rate greater than 10% per second; At the same time, the response time difference between the virtual model and the physical device is collected to establish a dynamic consistency index, and the calculation method is specifically:

[0075]

[0076] Among them, is expressed as the dynamic consistency index, is expressed as the actual response time, is expressed as the reference time constant, takes a value of 20 ms; when DCI < 0.9, the control parameter re-tuning is triggered, that is, when the dynamic consistency index is lower than the threshold of 0.9, the control parameters are automatically or manually adjusted again.

[0077] In the stability margin analysis, for the impedance mutation condition, based on the dual-end data synchronous acquisition unit, a closed-loop transfer function matrix including virtual and physical systems is constructed , by solving the root locus of the characteristic equation , the phase margin deviation Δφ at the key frequency point is calculated; when Δφ > 15° is detected, it is automatically marked as a stability risk point.

[0078] The small-signal stability assessment uses the singular value decomposition method in the dq coordinate system. First, establish the state-space equation of the control loop:

[0079]

[0080] where x represents the state vector, u represents the input vector, M represents the state matrix, and N represents the input matrix;

[0081] u

[0082] y represents the output vector, P represents the output matrix, and Q represents the direct transfer matrix.

[0083] Among them, the state matrix M includes the virtual synchronous machine rotor dynamics, the phase-locked loop dynamics, and the current control loop dynamics; perform singular value decomposition on the system matrix A to obtain , set the stability threshold , and generate an analysis report containing the dominant oscillation mode when the limit is exceeded;

[0084] The transient energy function evaluation dimension uses an improved Lyapunov function, and the calculation method is specifically as follows:

[0085]

[0086] where represents the transient energy function, represents the equivalent inertia, represents the voltage sensitivity coefficient, represents the power angle deviation, represents the power angular velocity deviation, represents the mechanical torque;

[0087] The module calculates the rate of change of the energy function in real time. When it is detected that the rate of change of the energy function is continuously greater than 0 and the duration exceeds 5 cycles, it is determined that the platform has a risk of transient instability.

[0088] Iterative optimization module: When the phase margin deviation > 15°, reconstruct the virtual inertia coefficient based on the Lyapunov function, and repeat from the network construction control depth verification module to the multi-dimensional performance evaluation module until the stability threshold is met;

[0089] In the iterative optimization module, during the iterative optimization stage, when it is detected that the phase margin deviation exceeds 15°, the system automatically triggers a parameter adjustment mechanism based on the Lyapunov stability theory; first, construct a Lyapunov function containing the phase margin error term and the virtual inertia offset term, and the calculation method is specifically as follows:

[0090]

[0091] where Denoted as the Lyapunov function, The phase margin tracking error, where g represents the actual phase, Denoted as the weight factor, Denoted as the virtual inertia update amount;

[0092]

[0093] Wherein, The phase margin tracking error is denoted, and g represents the actual phase, Denoted as the reference phase;

[0094] By solving the negative definite condition of the derivative of the Lyapunov function The virtual inertia update law is derived, and the calculation method is specifically as follows:

[0095]

[0096] Wherein, Denoted as the virtual inertia update amount, m represents the convergence factor, Denoted as the partial derivative of the phase with respect to the virtual inertia, Denoted as the phase margin tracking error;

[0097] Obtained by online looking up the table in the pre-established frequency domain sensitivity model;

[0098] The updated virtual inertia will be sent to the virtual energy storage unit in the digital twin in real time, and at the same time, the droop coefficient of the physical device will be synchronously adjusted through the bidirectional parameter interaction interface; then the dynamic test process of the network-forming control depth verification module will be re-executed, and new phase margin data will be collected under the same disturbance conditions. If the convergence condition of | | ≤ 5° is not satisfied, continue to iterate until the maximum number of iterations is reached; during each iteration, record , g and the evolution trajectories of the key state variables to form a parameter optimization path report.

[0099] Please refer to Figure 2 As shown, in this embodiment, it should be specifically noted that the present invention provides a network-forming energy storage semi-physical simulation test method, including:

[0100] Step A1: Construct a power grid - energy storage digital twin in the real-time simulation unit, and the digital twin includes a dynamic impedance network model and a virtual network-forming energy storage unit; the virtual network-forming energy storage unit synchronizes real-time data with the physical energy storage device controller through an error compensation algorithm;

[0101] Step A2: Connect the physical energy storage device to the simulation loop through a power amplifier, and collect the impedance characteristics at the output end of the device in real time; dynamically adjust the damping compensation parameters of the power amplifier based on the impedance scanning results;

[0102] Step A3: Inject a preset weak grid condition into the digital twin to activate the grid-forming control strategy of the physical device; synchronously adjust the inertia simulation coefficient of the virtual energy storage unit and the droop control parameters of the physical device to generate a two-way parameter interaction command;

[0103] Step A4: Calculate the active power response rate of the physical device under frequency disturbance; analyze the phase margin deviation of the virtual-physical dual-end data at the impedance mutation point to determine the control stability boundary;

[0104] Step A5: When the phase margin deviation is greater than 15°, reconstruct the virtual inertia coefficient based on the Lyapunov function, and repeat from the grid-forming control depth verification module to the multi-dimensional performance evaluation module until the stability threshold is met.

[0105] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A grid-forming energy storage hardware-in-the-loop simulation test platform, characterized in that, Including: Digital twin collaborative modeling module: Build a power grid - energy storage digital twin in the real-time simulation unit. The digital twin includes a dynamic impedance network model and a virtual network-forming energy storage unit. The virtual network-forming energy storage unit synchronizes real-time data with the physical energy storage device controller through an error compensation algorithm. Impedance adaptive hardware coupling module: Connect the physical energy storage device to the simulation loop through a power amplifier, and collect the impedance characteristics of the device output end in real time. Dynamically adjust the damping compensation parameters of the power amplifier based on the impedance scanning results. Network-forming control depth verification module: Inject a preset weak power grid condition into the digital twin to activate the network-forming control strategy of the physical device. Synchronously adjust the inertia simulation coefficient of the virtual energy storage unit and the droop control parameter of the physical device to generate a two-way parameter interaction instruction. Multi-dimensional performance evaluation module: Calculate the active power response rate of the physical device under frequency disturbance. Analyze the phase margin deviation of the virtual - physical dual-end data at the impedance mutation point to determine the control stability boundary. Iterative optimization module: When the phase margin deviation is greater than 15°, reconstruct the virtual inertia coefficient based on the Lyapunov function, and repeat from the network-forming control depth verification module to the multi-dimensional performance evaluation module until the stability threshold is met.

2. The network-constructing energy storage semi-physical simulation test platform according to claim 1, characterized in that: In the digital twin collaborative modeling module, a power grid - energy storage digital twin is built in the real-time simulation unit. The twin includes a dynamic impedance network model and a virtual network-forming energy storage unit, and realizes real-time data synchronization with the physical energy storage device controller through an error compensation algorithm. The dynamic impedance network model adopts a frequency-variable parameter transmission line equation: ； Among them, is represented as the grid impedance, is represented as the real part resistance of the impedance, is represented as the imaginary part reactance of the impedance, is represented as the amplitude of the k-th frequency-varying component, is represented as the time constant of the k-th frequency-varying component, f is represented as the frequency, k is represented as the index number of the frequency-varying component, and j is represented as the imaginary unit.

3. The network-constructing energy storage semi-physical simulation test platform according to claim 2, characterized in that: The virtual network-forming energy storage unit is built based on a multi-time scale state space model. At the millisecond level, the IGBT loss - thermal coupling equation is used to simulate the switching dynamics of power devices. At the second level, the virtual synchronous machine rotor motion equation is executed: ； Wherein, is represented as the rotor angular acceleration, is represented as the rotor angular velocity, and H is represented as the inertia constant, is represented as the input mechanical torque, is represented as the output electromagnetic torque, is represented as the damping coefficient, is represented as the angular velocity deviation; At the same time, an integrated control instruction pre-compiler is used to convert the PWM signal output by the physical controller into an equivalent voltage source that can be parsed by the simulation.

4. A semi-physical simulation test platform for grid-forming energy storage according to claim 1, characterized in that: In the impedance adaptive hardware coupling module, the physical energy storage device is connected to the real-time simulation loop through a power amplifier through an interface to build a hardware-in-the-loop test system, and collect the impedance characteristics of the device output end in real time and dynamically adjust the system parameters. Calculate the ratio curve of the output impedance of the physical device to the equivalent impedance of the power grid based on the frequency domain analysis method, and perform a closed-loop analysis on the trajectory of the impedance ratio using the Nyquist stability criterion. When the trajectory approaches the (-1,j0) point and the phase margin is less than 30 degrees, it is determined that there is a risk of interactive resonance. For the identified resonance point, the module automatically starts a multi-level compensation mechanism. In the low-frequency band below 500Hz, a passive damping strategy is adopted by adjusting the cut-off frequency of the output filter of the power amplifier. In the middle-frequency band of 500 - 1500Hz, an active damper is enabled to dynamically generate a reverse compensation current according to the resonance point frequency. The calculation method is as follows: ； Among them, represents the reverse compensation current generated for the active damper, represents the damping gain coefficient, represents the PCC point voltage, represents the rate of change of the PCC point voltage.

5. A network-constructing energy storage semi-physical simulation test platform according to claim 1, characterized in that: In the network-forming control depth verification module, a closed-loop test environment for virtual-real interaction is established to verify the control performance of the network-forming energy storage system under extremely weak power grid conditions. This module first presets two types of typical weak power grid test scenarios in the digital twin. During the test, the module synchronously coordinates the interaction of control parameters between the virtual and physical systems: the virtual energy storage unit adjusts the inertia simulation coefficient in real time, and its adjustment algorithm is designed based on the Lyapunov stability theory. The specific calculation method is as follows: ； Among them, is expressed as the real-time inertia simulation coefficient of the virtual energy storage unit, is expressed as the initial inertia value, is expressed as the adaptive gain coefficient, Frequency deviation; At the same time, the specific calculation method of the droop control parameters on the physical device side is as follows: ； Among them, is expressed as the real-time droop control parameter on the physical device side, is expressed as the reference droop reference value, is expressed as the change sensitivity factor, is expressed as the change amount of the short-circuit capacity ratio.

6. The semi-physical simulation test platform for grid-forming energy storage according to claim 1, characterized in that: In the multi-dimensional performance evaluation module, the multi-dimensional performance evaluation module constructs a three-dimensional evaluation system including dynamic response, stability margin, and energy characteristics to comprehensively quantify the performance of the grid-forming energy storage system; in the dimension of dynamic response evaluation, the module monitors the active power change rate of physical devices under frequency disturbances in real time. The calculation uses the least squares method to fit the slope of the power-time curve, and meets the dynamic response benchmark of the rated power of the device with an active power change rate greater than 10% per second; at the same time, the response time difference between the virtual model and the physical device is collected to establish a dynamic consistency index. The calculation method is as follows: ； Among them, is expressed as the dynamic consistency index, is expressed as the actual response time, is expressed as the reference time constant, and its value is 20 ms.

7. A network-constructing energy storage semi-physical simulation test platform according to claim 1, characterized in that: In the iterative optimization module, in the iterative optimization stage, when it is detected that the phase margin deviation exceeds 15°, the system automatically triggers a parameter adjustment mechanism based on the Lyapunov stability theory; first, a Lyapunov function including the phase margin error term and the virtual inertia offset term is constructed. The calculation method is as follows: ； Among them, is expressed as a Lyapunov function, the phase margin tracking error, where g is expressed as the actual phase, is expressed as a weight factor, is expressed as the virtual inertia update amount; By solving the negative definite condition of the derivative of the Lyapunov function the virtual inertia update law is derived, and the specific calculation method is as follows: ； Among them, is expressed as the virtual inertia update amount, and m is expressed as the convergence factor, is expressed as the partial derivative of the phase with respect to the virtual inertia, is expressed as the phase margin tracking error.

8. A grid-forming energy storage hardware-in-the-loop simulation test method, using a grid-forming energy storage hardware-in-the-loop simulation test platform according to any one of claims 1-7, characterized in that: Step A1: Build a power grid-energy storage digital twin in the real-time simulation unit. The digital twin includes a dynamic impedance network model and a virtual grid-forming energy storage unit; the virtual grid-forming energy storage unit synchronizes real-time data with the physical energy storage device controller through an error compensation algorithm. Step A2: Connect the physical energy storage device to the simulation loop through a power amplifier, and collect the impedance characteristics of the device output terminal in real time; dynamically adjust the damping compensation parameters of the power amplifier based on the impedance scan results. Step A3: Inject a preset weak grid condition into the digital twin to activate the grid-forming control strategy of the physical device. Synchronously adjust the inertia simulation coefficient of the virtual energy storage unit and the droop control parameters of the physical device to generate a two-way parameter interaction command. Step A4: Calculate the active power response rate of the physical device under frequency disturbances; analyze the phase margin deviation of the virtual-physical dual-end data at the impedance mutation point to determine the control stability boundary. Step A5: When the phase margin deviation is greater than 15°, reconstruct the virtual inertia coefficient based on the Lyapunov function, and repeat from the grid-forming control depth verification module to the multi-dimensional performance evaluation module until the stability threshold is met.

Images