Flexible mobile energy storage power grid strength rapid identification and adaptive grid connection method

Abstract

The invention discloses a flexible mobile energy storage power grid strength rapid identification and adaptive grid connection method, and relates to the technical field of electrochemical energy storage. The method sequentially comprises the eight steps of system pre-grid-connection detection, broadband rapid power grid strength identification, active grid-connection mode selection based on the short-circuit ratio, grid-connection parameter optimization and dynamic safety margin setting, high-voltage cascade topology three-layer coordination control, state pre-synchronization seamless switching, grid-connection starting and safety monitoring and periodic retest optimization. And rapid and stable grid connection of mobile energy storage is realized. According to the method, power grid strength identification can be rapidly completed, the optimal control mode is actively selected before grid connection based on a quantitative short-circuit ratio index, passive response delay is avoided, seamless and smooth switching of the control mode is achieved, a complete and rapid deployment scheme from access to stable grid connection is established, and the method is suitable for large-scale popularization and application. The problem of rapid grid connection of mobile energy storage in emergency power supply and other scenes is solved.

Description

Technical Field

[0001] This invention relates to the field of electrochemical energy storage technology, and in particular to a method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength. Background Technology

[0002] With the rapid development of new energy power generation, flexible mobile energy storage has demonstrated unique advantages in emergency power supply, grid peak shaving, and new energy consumption due to its rapid deployment and flexible allocation characteristics. Typical application scenarios for mobile energy storage require that the equipment be connected to the grid and put into operation within 5 minutes of arriving on site; however, existing technologies mainly suffer from the following problems: 1. The speed of power grid strength identification cannot meet the needs of rapid deployment: Traditional power grid impedance identification methods are slow and have a significant impact on power quality. The single-frequency harmonic injection method requires continuous injection for 3-10 seconds to establish a steady-state response, and the injection amplitude is typically 5%-10% of the rated current, which may affect power quality. The disturbance observation method requires 2-5 seconds, and there is a trade-off between the disturbance amplitude and the identification accuracy. The Kalman filter method is computationally complex, with uncertain convergence time, which may exceed 10 seconds under certain operating conditions.

[0003] Mobile energy storage emergency power supply scenarios require a total deployment time of no more than 5 minutes, of which grid strength identification should be controlled within 1 second. Existing methods have too high time costs.

[0004] 2. Lack of a proactive, rapid mode selection mechanism: Existing technologies mostly employ passive detection methods: the system initially operates in a default mode (usually grid-connected), and when abnormalities such as power oscillations, voltage or frequency exceeding limits are detected, the system assesses changes in grid strength and then switches to other modes. This reactive approach has a long response time (typically 3-10 seconds) and may have already impacted the system, even triggering protection actions.

[0005] Mobile energy storage needs to quickly identify grid strength before grid connection and proactively select the optimal control mode to avoid trial and error and switching shocks.

[0006] 3. Power and voltage surges occur during control mode switching: Existing grid-connected / network-connected switching methods have long switching times (3-5 seconds), large power fluctuations (10%-20%), and voltage fluctuations of 5%-10% during the switching process, which may trigger protection devices and affect power supply continuity. They also lack a smooth and seamless switching mechanism.

[0007] 4. Lack of a systematic approach for the rapid deployment of mobile energy storage: Fixed energy storage power stations are designed for specific grid connection points, and their control parameters can be manually tested and tuned offline (usually taking 5-15 minutes). Mobile energy storage needs to complete the entire process from connection to stable grid connection within minutes, including grid strength identification, control mode selection, grid connection parameter optimization, soft start grid connection, and mode switching.

[0008] Existing technologies are mostly designed for the long-term optimized operation of stationary energy storage, lacking an integrated rapid deployment solution that includes "rapid identification, proactive selection, parameter optimization, and seamless switching." This results in long deployment times and low reliability for mobile energy storage in scenarios such as emergency power supply and grid support.

[0009] Therefore, there is an urgent need for an adaptive grid connection method that can quickly identify grid strength, actively select the optimal control mode, smoothly switch control strategies, and adapt to the rapid deployment requirements of mobile energy storage. Summary of the Invention

[0010] The purpose of this invention is to provide a method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength. This method can quickly identify grid strength, actively select the optimal control mode before grid connection based on a quantitative short-circuit ratio index, avoid passive response delay, achieve seamless and smooth switching of control modes, and establish a complete and rapid deployment scheme from access to stable grid connection. This solves the problem of rapid grid connection of mobile energy storage in scenarios such as emergency power supply.

[0011] To achieve the above objectives, the present invention provides the following solution: A method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength includes the following steps: S1, System pre-grid connection test: Before the energy storage system is connected to the grid, electrical quantity detection at the grid connection point, system self-test and pre-charging control are performed; S2, Wideband Fast Grid Strength Identification: Employs a multi-frequency signal injection method to simultaneously collect grid connection point voltage and injection current. After signal processing, impedance identification calculation, and short-circuit ratio calculation, the grid strength is identified, and the identification results are verified. S3, Active grid connection mode selection based on short-circuit ratio: Quantitatively classify the grid strength according to the short-circuit ratio (SCR), and actively select the corresponding control mode based on the classification results; S4, Rapid optimization of grid connection parameters and dynamic safety margin setting: Calculates the optimal reactive power and power limit for different grid strengths, and sets the dynamic safety margin; S5, High Voltage Cascaded Topology Three-Layer Coordinated Control: Establishes a three-layer coordinated control architecture of system layer, bridge arm layer, and module layer to realize command issuance, power distribution, and voltage distribution; S6, Seamless switching of state pre-synchronization: When the change in grid intensity triggers the control mode switching, the new and old controller outputs are made to be consistent through the state pre-synchronization mechanism before a fast switching is performed; S7, Grid-connected startup and safety monitoring: Performs soft-start grid connection, monitors the status of the energy storage system in real time, and performs protection functions and fault isolation; S8, Periodic Retesting and Parameter Optimization: After grid connection, the grid strength is periodically retested, and the control mode is reselected and the control parameters are optimized based on the retest results.

[0012] Furthermore, in S1, the system pre-grid connection detection specifically includes: Grid connection point electrical quantity detection: The three-phase voltage at the grid connection point is collected in real time through voltage sensors and phase-locked loops. The sampling frequency is not less than 10kHz. The detection measures whether the effective voltage value is within 90%-110% of the rated voltage, whether the frequency is within the allowable range, whether the phase sequence is correct, and whether the voltage imbalance is less than 2%. System self-test: Check whether the SOC of each energy storage module is within the safe range of 20%-80%, whether the temperature is within the normal range of 5-45°C, whether the voltage difference between modules is <5%, whether the communication response time is <100ms, and whether the protection functions are intact; Pre-charge control: The DC bus voltage is gradually charged to near the peak voltage of the grid connection point through the pre-charge resistor and contactor, with a voltage difference of <5% and a pre-charge time of 2-5 seconds. After completion, the pre-charge resistor is bypassed and the main contactor is closed.

[0013] Furthermore, in S2, the broadband fast power grid strength identification specifically includes: Multi-frequency injection signal design: Three odd harmonic frequencies of 150Hz, 250Hz, and 350Hz are selected. The amplitude of each frequency signal is set to 2% of the rated current. The sinusoidal signals of the three frequencies are superimposed as the current injection command. The current injection command formula is as follows:

[0014] in, , , , For the injected amplitude; Impedance identification calculation: During the injection period, the grid connection point voltage and injection current are synchronously collected. The collected data are subjected to Fast Fourier Transform (FFT) to extract the voltage amplitude, current amplitude and phase difference at three frequency points. At each frequency point, the complex number of the grid impedance is calculated according to Ohm's law. The grid impedance model adopts the series RL model, and the grid resistance R and inductance L are solved by weighted least squares fitting. Short-circuit ratio calculation: Based on the grid resistance R and inductance L, calculate the impedance modulus at 50Hz power frequency, and calculate the three-phase short-circuit capacity at the grid connection point. With the rated capacity of the energy storage system The ratio of the two values ​​is used to obtain the short-circuit ratio (SCR). Verification of identification results: The first verification is frequency consistency. Check whether the standard deviation of the R and L values ​​obtained by fitting the three frequency points is less than 15% of the average value. The second verification is physical rationality. Check whether the grid resistance R is within the set reasonable range, whether the inductance L is within the set reasonable range, and whether the short-circuit ratio SCR is greater than 1. If the verification fails, repeat step S2.

[0015] Furthermore, in S2, the formula for calculating the complex number of the power grid impedance in the impedance identification calculation is as follows:

[0016] In the formula, Voltage amplitude, The current amplitude, For phase difference, For frequency The current grid impedance angle is as follows; The power grid impedance model adopts a series RL model at frequency. f When the impedance is For the three frequency points, the real and imaginary parts of the impedance are as follows:

[0017]

[0018] The grid resistance R and inductance L are solved by weighted least squares fitting; a system of linear equations is established and solved using matrix methods.

[0019]

[0020] in, Number of frequency points; In the short-circuit ratio calculation, the impedance magnitude at 50Hz power frequency is calculated based on the identified R and L:

[0021] The formula for calculating the three-phase short-circuit capacity is:

[0022] in, Line voltage rating; short-circuit ratio .

[0023] Furthermore, in S3, the active grid connection mode selection based on the short-circuit ratio specifically includes: Grid strength classification: Quantitative classification based on short-circuit ratio (SCR): SCR≥15, strong grid; 5≤SCR<15, medium-strength grid; 2≤SCR<5, weak grid; SCR<2, extremely weak grid / isolated grid; Control mode selection: Strong power grid: Select grid-following control mode, with the energy storage system as the controlled current source. The control targets are active power P and reactive power Q. The reactive power is set to Q=0, and the power limit is the rated power. Medium-intensity power grids: An enhanced grid-following control mode is selected, adding voltage outer loop auxiliary support to the grid-following control. Reactive power is dynamically adjusted based on voltage deviation, calculated using the following formula: ,in This is the reactive power regulation coefficient. This is the reference value for the grid connection point voltage. This is the measured voltage value at the grid connection point; For weak grids, extremely weak grids / islanded grids: select the grid construction control mode, with the energy storage system acting as the controlled voltage source, and the control targets being voltage amplitude U and frequency. f It has active-frequency droop and reactive-voltage droop characteristics; Furthermore, in S4, the rapid optimization of grid connection parameters and the dynamic safety margin setting specifically include: Reactive power setting: Set reactive power Q=0 under strong grid conditions; under medium grid conditions, set an appropriate amount of reactive power according to the voltage deviation; under weak grid conditions, set capacitive reactive power to 10%-20% of the rated capacity. Power limiting calculation: Based on voltage and current constraints, the maximum transmittable power is calculated. The constraints are that the grid connection point voltage is within the range of 0.9-1.1 times the rated voltage, and the system current does not exceed 1.2 times the rated current. Through iterative solutions or numerical optimization, the maximum active power that satisfies all constraints is found. ; Dynamic safety margin setting: Considering the uncertainties of the mobile energy storage site environment, a safety margin is reserved based on the calculated power limit. The final power limit calculation is as follows:

[0024] in, The safety margin factor is dynamically adjusted based on the grid strength: strong grid Medium-intensity power grid Weak power grid Voltage control target: In grid construction mode, the voltage for weak grids is set at 1.02-1.05 times the rated voltage, and the voltage for medium-strength grids is set at 1.00-1.02 times the rated voltage.

[0025] Furthermore, in S5, the three-layer coordinated control of the high-voltage cascaded topology specifically includes: System-level control: Based on the selected control mode, system-level commands are generated. In grid-connected mode, total active and reactive power commands are generated, and in grid-connected mode, voltage amplitude and frequency commands are generated. Power safety limiting is also executed. Bridge arm layer power allocation: An iterative power allocation algorithm is used to allocate system power to each bridge arm. Under discharge conditions, bridge arms with high SOC receive more power, and the allocation ratio is proportional to SOC. Under charging conditions, bridge arms with low SOC receive more power, thus achieving balanced SOC allocation. Module-level voltage distribution: Modules within the bridge arm are connected in series. SOC balance is achieved by adjusting the voltage of each module. During discharge, the voltage of high SOC modules increases, and during charging, the voltage of low SOC modules increases. The voltage adjustment range is proportional to the SOC deviation. The voltage of a single module is within the range of 85%-115% of the rated value, and the total voltage of the bridge arm is equal to the reference value.

[0026] Furthermore, in S6, the seamless state pre-synchronization switching specifically includes: Switching trigger conditions: Continuously monitor the power grid strength. If the SCR crosses the grade boundary and the duration is >3 seconds, or if the upper-level dispatching system issues a switching command, the switching process will be triggered. State pre-synchronization mechanism: The new controller corresponding to the new control mode enters the pre-running state and runs in parallel with the old controller corresponding to the currently running old control mode. The power inner loop of the new controller tracks the current actual output power, the voltage inner loop tracks the current actual output voltage, and the current inner loop tracks the current actual output current. The internal state variables of the new controller are adjusted so that the output command of the new controller is the same as that of the old controller. Quick switchover execution: When the output errors of the new and old controllers meet the switching conditions, the handover of control is completed within one control cycle; after the switchover, the new controller works normally and the old controller exits operation.

[0027] Furthermore, in S7, grid connection startup and security monitoring specifically include: Soft start strategy: Power gradually increases from 0 to the target value, with the increase rate limited to 20% / second of rated power. In network mode, the rated voltage and frequency are established first, and the power output is gradually increased after stabilizing for 100ms. Safety monitoring: Real-time monitoring of grid-side voltage, frequency, power factor, system-side output power, current of each bridge arm, voltage / SOC / temperature of each module, with a monitoring cycle of 1-10ms; Protection functions: Grid voltage over-limit protection: Disconnects from the grid if the voltage exceeds 90%-110% for 2 seconds; Frequency over-limit protection: Disconnects from the grid if the frequency exceeds 49.5-50.5Hz for 2 seconds; System overcurrent protection: Limits power if the current exceeds 1.2 times the rated current for 5 seconds, and disconnects from the grid if the current exceeds 1.5 times the rated current for 0.5 seconds; Module temperature protection: Derating operation if the temperature exceeds 55°C, and isolating the module if the temperature exceeds 60°C; Module SOC protection: Stops discharging if the SOC level is below 10%, and stops charging if the SOC level is above 90%. Fault isolation: When a module fault is detected, the bypass switch is closed within 50ms to isolate the faulty module, update the available power, and re-execute the power allocation.

[0028] Furthermore, in S8, the periodic retesting and parameter optimization specifically include: Periodic retesting: Repeat step S2 for grid strength identification every 10-30 minutes. If the identification result differs from the current value by more than 20%, trigger step S3 for mode reselection and step S4 for parameter re-optimization. Seamlessly switch to the new mode through step S6. Parameter optimization: Based on the performance evaluation results, adjust the control parameters periodically. If the power factor is <0.95, adjust the reactive power setting; if the voltage deviation is >5%, adjust the voltage control target; if the SOC dispersion is >5%, enhance the equalization control strength. The optimization cycle is 10-30 minutes, and the adjustment range is ≤10% of the current value.

[0029] According to specific embodiments provided by the present invention, the method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength disclosed in the present invention has the following technical effects: Fast identification speed: The fast identification method using wideband multi-frequency point small amplitude signal injection can complete the grid strength identification within 300ms, which is 10-30 times faster than the traditional method (3-10 seconds). Moreover, the amplitude of the injected signal is only 2% of the rated current, and the impact on power quality is negligible (THD increase <0.5%), meeting the time requirements for rapid deployment of mobile energy storage.

[0030] Active mode selection: Based on the quantitative short-circuit ratio (SCR) index, the optimal control mode is actively selected before grid connection. The response time is less than 1 second, which is 5-10 times faster than the traditional passive detection method (3-10 seconds). This avoids the trial-and-error process and problems such as system impact and protection actions caused by mode mismatch, thus improving grid connection safety and reliability.

[0031] Seamless and smooth switching: The control mode is switched quickly through a state pre-synchronization mechanism. The switching time is <100ms, the power fluctuation is <3%, and the voltage fluctuation is <1%. Compared with the traditional switching method (3-5 seconds, 10%-20% fluctuation), the impact during the switching process is significantly reduced, ensuring the continuity of power supply.

[0032] Multi-scenario Adaptability: Based on the grid strength classification, different control modes and parameter optimization strategies are designed to adapt to various scenarios such as strong grid, medium-strength grid, weak grid, and extremely weak grid / isolated grid. It can operate stably in grid connection under a wide range of SCR changes (1.63-22.3), which improves the environmental adaptability of mobile energy storage.

[0033] High deployment efficiency: It provides a complete and rapid deployment solution from access to grid connection, with a total deployment time of less than 3 minutes, which is far less than traditional methods (5-15 minutes of manual testing and parameter tuning). It also has a periodic retesting and parameter optimization mechanism, which can continuously adapt to changes in grid intensity and ensure long-term stable operation of the system.

[0034] Dynamically adjustable safety margin: The safety margin is dynamically set according to the grid strength. In a weak grid, sufficient margin is reserved to avoid voltage over-limit and overcurrent. In a strong grid, the capacity is fully utilized to increase power output, thus balancing the safety and economy of the system. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart of the method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to the present invention; Figure 2 This is a hardware topology diagram of the high-voltage cascaded flexible mobile energy storage system of the present invention; Figure 3 This is a schematic diagram of the broadband fast power grid strength identification principle of the present invention, wherein A is the identification time flow, B is the injected current waveform, C is the FFT extraction result, and D is an example of the identification result; Figure 4 This is a schematic diagram comparing the active selection and passive detection methods of the present invention; Figure 5 This is a timing diagram for the seamless pre-synchronization switching of the state in this invention; Figure 6 This is a diagram of the three-layer coordination control architecture of the present invention; Figure 7 This is a schematic diagram illustrating the power limiting calculation principle and dynamic safety margin. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] The relevant technical terms of this invention are explained as follows: Grid strength: Characterizes the ability of a power grid to maintain voltage stability at the grid connection point, usually quantified by the short-circuit ratio (SCR). SCR is defined as the ratio of the three-phase short-circuit capacity of the grid at the grid connection point to the rated capacity of the energy storage system. An SCR ≥ 20 indicates a strong grid, 5 ≤ SCR < 20 indicates a medium-strength grid, and SCR < 5 indicates a weak grid.

[0039] Short Circuit Ratio (SCR): Three-phase short-circuit capacity at the grid connection point. With the rated capacity of the energy storage system The ratio, i.e. It is a core quantitative indicator for measuring power grid strength.

[0040] Flexible mobile energy storage: Energy storage systems that employ containerized or skid-mounted designs, enabling rapid transportation, deployment, and grid connection. Compared to stationary energy storage, they need to adapt to different locations and grid environments of varying intensities within minutes, placing extremely high demands on rapid identification and adaptive grid connection.

[0041] High-Voltage Cascaded Topology: This topology consists of multiple energy storage modules connected in series via power electronic converters to form bridge arms. These bridge arms are then connected in parallel and directly connected to the medium- and high-voltage power grid. It can be directly integrated into 10kV and 35kV power grids without the need for step-up transformers, and offers advantages such as modularity, high efficiency, and high reliability.

[0042] Active selection and passive detection: Active selection refers to actively selecting the optimal control mode by quickly identifying abnormalities before grid connection; passive detection refers to passively adjusting the mode after grid connection by detecting abnormalities (oscillation, exceeding limits, etc.).

[0043] This invention provides a method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength, aiming to achieve: grid strength identification within 300ms; proactive selection of the optimal control mode before grid connection based on a quantitative short-circuit ratio index, avoiding passive response delay; seamless and smooth switching of control modes with a switching time of <100ms and power fluctuation of <3%; and a complete solution from access to grid connection with a total deployment time of <3 minutes.

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] like Figures 1 to 7 As shown, this invention provides a method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength, comprising the following steps: Step 1: System pre-grid connection test Before the energy storage system is connected to the grid, safety testing and initial parameter configuration are carried out.

[0046] Collect the three-phase voltage at the grid connection point and measure it in real time using voltage sensors and phase-locked loops, with a sampling frequency of no less than 10kHz. Check whether the effective voltage value is within 90%-110% of the rated voltage, whether the frequency is within the allowable range (49.5-50.5Hz for a 50Hz system), whether the phase sequence is correct, and whether the voltage imbalance is less than 2%.

[0047] System self-test: Check whether the SOC of each energy storage module is within the safe range of 20%-80%, whether the temperature is within the normal range of 5-45°C, whether the voltage difference between modules is <5%, whether the communication response time is <100ms, and whether the protection functions are intact.

[0048] Pre-charge control: The DC bus voltage is gradually charged to near the peak voltage of the grid connection point (voltage difference <5%) through the pre-charge resistor and contactor. The pre-charge time is 2-5 seconds. After completion, the pre-charge resistor is bypassed and the main contactor is closed.

[0049] Step 2: Broadband Fast Power Grid Strength Identification This step is one of the core innovations of this invention, proposing a rapid identification method based on wideband multi-frequency point small-amplitude signal injection, such as... Figure 3 As shown.

[0050] Multi-frequency injection signal design: Three odd harmonic frequencies are selected: 150Hz (3rd harmonic), 250Hz (5th harmonic), and 350Hz (7th harmonic). The frequency selection principle is as follows: Choose odd harmonics and avoid even harmonics (even harmonics are easily affected by nonlinear loads and have large identification errors). It covers a wider frequency range (150-350Hz), improving identification robustness and anti-interference capability; Avoiding the 75Hz (1.5th harmonic) commonly used in traditional single-frequency injection methods reduces coupling with background harmonics of the power grid; The amplitude of each frequency signal is set to 2% of the rated current (e.g., if the rated current is 500A, then the injected current is 10A), which is much smaller than the 5%-10% of the traditional method, and the impact on power quality is negligible (THD increase <0.5%).

[0051] The three sinusoidal signals of different frequencies are superimposed and used as the current injection command:

[0052] in , , , This is the injection amplitude. The injection time is set to 250ms, including 100ms signal setup time and 150ms steady-state sampling time.

[0053] Impedance identification calculation: During the injection period, the grid connection point voltage and injection current are simultaneously acquired at a sampling frequency of 10kHz. A Fast Fourier Transform (FFT) is performed on the acquired data to extract the voltage amplitude at three frequency points: 150Hz, 250Hz, and 350Hz. Current amplitude and phase difference .

[0054] At each frequency point, calculate the complex impedance of the power grid according to Ohm's law:

[0055] in, For frequency The grid impedance angle is given below. The grid impedance model uses a series RL model; at frequency f, the impedance is... For the three frequency points, the real and imaginary parts of the impedance are respectively:

[0056]

[0057] The grid resistance R and inductance L are solved by least squares fitting. A system of linear equations is established and solved using matrix methods.

[0058] in Here are the number of frequency points. To improve accuracy, the actual calculation uses the weighted least squares method, with slightly lower weights for high-frequency points (due to higher noise in high-frequency impedance measurements). The weighting coefficients are 1.0, 0.9, and 0.8, respectively.

[0059] Short-circuit ratio calculation: Based on the identified R and L, calculate the impedance magnitude at 50Hz power frequency.

[0060] The formula for calculating the three-phase short-circuit capacity is:

[0061] in Line voltage rating (e.g., 10kV). Short-circuit ratio. .

[0062] Identification result verification: Perform double verification to ensure reliability.

[0063] First verification (frequency consistency): Check the consistency of the R and L values ​​obtained from fitting the three frequency points, and calculate whether the standard deviation is less than 15% of the mean. If the standard deviation is too large, it indicates that the measurement data is interfered with or the model is not applicable, and it needs to be re-identified.

[0064] The second verification (physical rationality): Check whether the mains resistance R is within the reasonable range of 0.01-2Ω, whether the inductance L is within the reasonable range of 0.1-100mH, and whether the short-circuit ratio (SCR) is greater than 1. If the parameters are abnormal, the identification is deemed to have failed and the process should be repeated.

[0065] The entire identification process takes about 300ms (including 250ms for signal injection, 35ms for FFT calculation, and 15ms for fitting and solving), which is 10-30 times faster than traditional methods (3-10 seconds).

[0066] Step 3: Active grid connection mode selection based on short-circuit ratio This step is the second core innovation of the present invention. Based on the quantitative short-circuit ratio index, the optimal control mode is actively selected before grid connection, which is different from the traditional passive detection method.

[0067] Power grid strength classification: Quantitative classification is carried out based on the short-circuit ratio (SCR), and mode selection rules are established.

[0068] Short-circuit ratio range: SCR≥15, grid strength level: strong grid, typical scenario: urban main grid, large substation, control mode selection: grid-following control mode; Short-circuit ratio range: 5≤SCR<15, grid strength level: medium, typical scenario: general distribution network, industrial park, control mode selection: enhanced grid-following control mode; Short-circuit ratio range: 2≤SCR<5, grid strength level: weak grid, typical scenario: remote area, small microgrid, control mode selection: grid control mode; Short-circuit ratio range: SCR<2, grid strength level: extremely weak grid / islanded grid, typical scenario: islanded grid operation, control mode selection: fully grid-connected control mode.

[0069] Control mode configuration: Grid-connected control mode: The energy storage system acts as a controlled current source, with the control targets being active power P and reactive power Q. Reactive power is set to Q=0 to achieve unity power factor, and the upper limit of power is the rated power. This mode is suitable for strong power grids, offering simple control and good stability.

[0070] Enhanced grid-following control mode: This mode adds voltage outer loop auxiliary support to the grid-following control. Reactive power is dynamically adjusted based on voltage deviation, calculated using the following formula: ,in It is the reactive power regulation coefficient, which is usually taken as 2-3 times the system capacity. This is the reference value for the grid connection point voltage. This is the measured voltage at the grid connection point. Power needs to be limited according to the grid strength. Applicable to medium-strength grids.

[0071] Grid-based control mode: The energy storage system acts as a controlled voltage source, with the control targets being voltage amplitude U and frequency. f It possesses both active power-frequency droop and reactive power-voltage droop characteristics. The droop factor is configured based on the grid strength and load capacity. The active power-frequency droop factor m is typically set as the upper limit of power divided by the allowable frequency deviation (0.5Hz for weak grids, 0.2Hz for main grids), and the reactive power-voltage droop factor n is set as the reactive power capacity divided by the allowable voltage deviation (typically 0.5kV). Power must be strictly limited within a safe range. It is suitable for weak grids and islanded grids.

[0072] like Figure 4 As shown, a comparison between active selection and passive detection: Traditional passive detection methods: The system first connects to the grid in the default mode (usually grid-connected), and during operation, it detects abnormalities such as voltage / frequency / power oscillations. When a problem is detected, it is determined that the grid is weak, and then it switches to grid-connected mode. The response time is usually 3-10 seconds, and the system may have already been impacted (voltage exceeding limits, overcurrent protection, etc.).

[0073] The active selection method of this invention: grid strength identification is completed before grid connection (300ms), and the optimal mode is selected immediately according to the quantitative SCR index. The response time is <1 second, avoiding trial and error and anomalies caused by mode mismatch.

[0074] Time comparison: Active response time is 5-10 times shorter than passive response time. Safety comparison: Active response avoids problems such as voltage overruns and power oscillations that may occur during trial and error.

[0075] Step 4: Rapid optimization of grid connection parameters and dynamic safety margin setting For different grid strengths, the optimal grid connection parameters can be quickly calculated and dynamic safety margins can be set.

[0076] Reactive power setting: Under strong grid conditions, set Q=0 (power factor 1.0); under medium-strength grid conditions, set an appropriate amount of reactive power based on voltage deviation. In weak power grids, capacitive reactive power is set to support voltage, typically 10%-20% of the rated capacity.

[0077] Power limiting calculation: Based on voltage and current constraints, the maximum transmittable power is calculated. The constraints are that the grid connection point voltage is within the range of 0.9-1.1 times the rated voltage, and the system current does not exceed 1.2 times the rated current.

[0078] By solving the power flow equation, the voltage drop caused by grid impedance is considered. The formula for calculating the voltage drop is:

[0079] Where P is active power, Q is reactive power, and R is grid resistance. U is the grid reactance, and U is the grid voltage. Grid connection point voltage. , must meet .

[0080] The formula for calculating system current is:

[0081] Must meet .

[0082] The maximum active power that satisfies all constraints can be found through iterative solutions or numerical optimization. .

[0083] Dynamic safety margin setting: Considering the uncertainties of the mobile energy storage site environment (may fluctuate grid parameters, lack of professional operation and maintenance personnel, and priority to ensure the stability of emergency power supply), a safety margin is reserved on the basis of the calculated power limit.

[0084] The final power limiting calculation is as follows:

[0085] in The safety margin factor is dynamically adjusted based on the grid strength. High-voltage power grid (SCR≥15): (No restrictions); Medium strength (5≤SCR<15): (Reserve a 10% margin); Weak grid (SCR<5): (Reserve a 15% margin).

[0086] Voltage control target: Under grid construction mode, the voltage control target is adjusted according to the grid strength. For weak grids, it is set to 1.02-1.05 times the rated voltage (to compensate for voltage drops on the load side), and for medium-strength grids, it is set to 1.00-1.02 times the rated voltage.

[0087] The parameter optimization process is completed quickly using a lookup table method and simplified calculations, taking less than 50ms.

[0088] Power limiting calculation principle diagram (see appendix) Figure 7 ): Figure 7This demonstrates the multi-constraint optimization principle for power limiting calculation. The horizontal axis represents the short-circuit ratio (SCR) (2-30), and the vertical axis represents the output power (0-10MW). The figure contains five curves: ① Theoretical maximum power curve (black dashed line): The theoretical upper limit calculated based on grid impedance identification, which increases with the increase of SCR. The formula is as follows: .

[0089] ② Voltage constraint limit curve (dark blue dotted line): The upper limit of power to ensure the grid connection point voltage is within the range of 0.9-1.1 pu. The calculation formula is as follows: When excessive power causes the voltage drop to exceed the allowable range, this curve becomes the main limiting factor.

[0090] ③ Current constraint limit curve (dark cyan dotted line): Ensures the system current does not exceed 1.2 times the rated power limit. The calculation formula is as follows: In weak power grids, current constraints are often more stringent than voltage constraints.

[0091] ④ Actual available power curve (dark orange solid line): Take the minimum value envelope of the above three curves, i.e. , which represents the maximum available power that satisfies all constraints.

[0092] ⑤ Power curve with safety margin (thick red line): A dynamic safety margin is reserved based on the actual available power. The calculation formula is as follows: ,in The system power limit setting is ultimately determined by dynamically adjusting the power grid intensity (0% for strong grids, 10% for medium grids, and 15% for weak grids).

[0093] Figure 7 The three power grid strength zones are clearly shown: In weak grid areas (SCR<5, red background): power is strictly constrained by both current and voltage, and the usable power is only 40%-50% of the rated value, with a safety margin of 15%.

[0094] Medium intensity zone (5≤SCR<15, yellow background): Power is mainly constrained by voltage, and the available power gradually increases to 70%-95%, with a safety margin of 10%.

[0095] Strong power grid area (SCR≥15, green background): The constraints are basically lifted, the available power reaches the rated value, and no margin is required.

[0096] This diagram visually illustrates why mobile energy storage requires strict power limits in weak grid environments: without these limits, excessive power will cause a significant voltage drop at the grid connection point (below 0.9 pu) or system overcurrent (exceeding 1.2 times the rated current), triggering protection actions. This invention, by quickly identifying the SCR (Special Control Ring), proactively calculates the power limit and reserves a safety margin, avoiding the frequent protection tripping problems caused by improper power settings in traditional methods.

[0097] Step 5: Three-layer coordinated control of high-voltage cascaded topology For the high-voltage cascaded topology used in flexible mobile energy storage, a three-layer coordinated control architecture consisting of the system layer, bridge arm layer, and module layer is established, such as... Figure 2 As shown, each layer is controlled by a corresponding controller.

[0098] System-level control: Based on the selected control mode, system-level commands are generated. In grid-connected mode, total active and reactive power commands are generated; in network-connected mode, voltage amplitude and frequency commands are generated. Simultaneously, power safety limiting is implemented to ensure that the commands do not exceed the upper limit calculated in step 4.

[0099] Bridge arm power allocation: System power is allocated to each bridge arm using a SOC-based equalization strategy. During discharge, higher SOC arms receive more power, with the allocation ratio proportional to the SOC. During charging, lower SOC arms receive more power.

[0100] Iterative power allocation algorithm: Input: System target power Each bridge arm Maximum power of each bridge arm ; Output: Power distribution of each bridge arm ; Step (1): Initial allocation for i = 1 to

[0101]

[0102] Step (2): Iterative correction (maximum 10 times) repeat

[0103] for i = 1 to

[0104] if

[0105]

[0106]

[0107] if

[0108] break convergence Calculate the remaining margin of each bridge arm

[0109] for i = 1 to

[0110] if

[0111]

[0112] until iteration count > 10 Module-level voltage distribution: Modules within the bridge arm are connected in series, and SOC balance is achieved by adjusting the voltage of each module. During discharge, the voltage of high SOC modules increases, and during charging, the voltage of low SOC modules increases. The voltage adjustment range is proportional to the SOC deviation. The constraints are that the voltage of a single module is within 85%-115% of its rated value, and the total voltage of the bridge arm is equal to the reference value.

[0113] Step 6: Seamless handover of state pre-synchronization like Figure 5 As shown, this step is the third core innovation of the present invention, achieving seamless and smooth switching of control modes.

[0114] Switching trigger conditions: Continuously monitor grid strength. If the SCR crosses the grade boundary (e.g., from 18 to 12, crossing the boundary of 15) and the duration is >3 seconds, the switching process is triggered. Alternatively, the upper-level dispatching system issues a switching command.

[0115] State pre-synchronization mechanism: Before the switchover, the new mode controller enters a pre-running state and runs in parallel with the currently running old mode controller. The output of the new controller does not directly affect the system, but tracks the output state of the old controller, making the outputs of the two tend to be consistent.

[0116] Specifically, the new controller's power inner loop tracks the current actual output power, voltage inner loop tracks the current actual output voltage, and current inner loop tracks the current actual output current. By adjusting the new controller's internal state variables (integrator initial value, filter state, etc.), the output command of the new controller is made the same as that of the old controller.

[0117] When the output errors of the new and old controllers meet the switching conditions (power error <5%, voltage error <2%, current error <5%, and duration >3 control cycles), fast switching is performed.

[0118] Fast switchover execution: Control handover is completed within one control cycle (2ms), switching the output command from the old controller to the new controller. Since the output state has been pre-synchronized, the output command changes almost instantly (error <5%), resulting in a smooth system transition.

[0119] After the switchover, the new controller operates normally, and the old controller is deactivated. The entire switchover process takes approximately 80-100ms (including pre-synchronization time), with power fluctuations of <3% and voltage fluctuations of <1%.

[0120] Step 7: Grid Connection Start-up and Security Monitoring Perform a soft start and connect to the network, and monitor the system status in real time.

[0121] Soft-start strategy: Power gradually increases from 0 to the target value, with the rate of increase limited to 20% of rated power per second to avoid grid connection impact. In grid-connected mode, rated voltage and frequency are established first, and power output is gradually increased after stabilizing for 100ms.

[0122] Safety monitoring: Real-time monitoring of grid-side voltage, frequency, power factor, system-side output power, current of each bridge arm, and voltage / SOC / temperature of each module. Monitoring cycle: 1-10ms.

[0123] Protection functions: Over-limit protection of grid voltage (disconnects from the grid if the voltage exceeds 90%-110% for 2 seconds), over-limit protection of frequency (disconnects from the grid if the frequency exceeds 49.5-50.5Hz for 2 seconds), overcurrent protection of the system (limits power if the current exceeds 1.2 times the rated current for 5 seconds, disconnects from the grid if the current exceeds 1.5 times the rated current for 0.5 seconds), module temperature protection (derating operation if the current exceeds 55°C, isolating the module if the current exceeds 60°C), and SOC protection of the module (stops discharging if the current exceeds 10%, stops charging if the current exceeds 90%).

[0124] Fault isolation: When a module fault is detected (temperature, voltage, communication abnormalities, etc.), the bypass switch is closed within 50ms to isolate the faulty module, update the available power, and re-execute the power allocation.

[0125] Step 8: Periodic retesting and parameter optimization After grid connection, the grid strength is periodically retested to optimize control parameters.

[0126] Periodic retesting: Repeat step 2, grid strength identification, every 10-30 minutes. If the identification result differs from the current value by more than 20%, trigger step 3, mode reselection, and step 4, parameter re-optimization, and seamlessly switch to the new mode through step 6.

[0127] Parameter optimization: Based on the performance evaluation results, adjust the control parameters periodically. If the power factor is <0.95, adjust the reactive power setting; if the voltage deviation is >5%, adjust the voltage control target; if the SOC dispersion is >5%, strengthen the equalization control. The optimization cycle is 10-30 minutes, and the adjustment range is ≤10% of the current value. Specific Implementation System Configuration: A flexible mobile energy storage system adopts a 10kV high-voltage cascaded topology and a containerized design, with a total capacity of 10MW / 20MWh.

[0129] Topology: Star topology, three-phase, six-arm. Two arms per phase are connected in parallel, and 12 energy storage modules are connected in series per arm, for a total of 72 modules in the system.

[0130] Electrical parameters: System rated capacity 10MVA, rated line voltage 10kV, rated line current 577A. Single bridge arm rated power 1.67MW, rated voltage 5.77kV, rated current 289A. Single module rated voltage 481V, rated power 139kW, rated current 289A.

[0131] Module Composition: Each module consists of a battery stack and an H-bridge power converter. The battery stack comprises 150 series of lithium iron phosphate cells (nominal voltage of 3.2V per cell), with a module rated voltage of 480V, an operating voltage range of 420-540V (corresponding to SOC 10%-90%), a capacity of 580Ah, and a single module energy capacity of 278kWh. The H-bridge converter uses IGBTs (switching frequency 2kHz, DC side voltage 550V).

[0132] Control parameters: control cycle 2ms, current inner loop bandwidth 500Hz, power outer loop bandwidth 20Hz, voltage outer loop bandwidth 5Hz.

[0133] Scenario 1: Rapid Deployment of Urban Main Grid Power Grid T=0: The mobile energy storage device arrives at a 10kV substation in a city and is connected to the 10kV busbar.

[0134] Step 1: Pre-grid connection test (T=0-8s): The grid connection point voltage was measured at 10.12kV (1.012pu), frequency at 50.01Hz, phase sequence correct, and voltage imbalance at 0.4%, meeting grid connection requirements. System self-test showed that the SOC of 72 modules ranged from 50% to 60% (average 54.5%), temperature was 25-30°C, voltage was 452-468V, communication was normal, and protection functions were intact.

[0135] Pre-charging: The DC bus is charged from 0V to 8.16kV (corresponding to the peak phase voltage of 10kV) in 5 seconds, and the voltage difference drops to 0.25kV (<5%), at which point the main contactor is closed. The total time is 8 seconds.

[0136] Step 2: Rapid identification (T=8-8.3s): Inject signals at three frequencies: 150Hz, 250Hz, and 350Hz, each with an amplitude of 12A (approximately 2% of the rated current) and an injection time of 250ms.

[0137] FFT extraction results: 150Hz: U=15.8V, I=11.9A ; 250Hz: U=25.3V, I=12.0A ; 350Hz: U=34.6V, I=11.8A ; Least squares fitting (weighting coefficients 1.0, 0.9, 0.8): R = (1.328×1.0 + 2.108×0.9 + 2.932×0.8) / (1.0+0.9+0.8)×cos(mean phase) ≈ 0.088Ω; L = (1.328×1.0 + 2.108×0.9 + 2.932×0.8) / (1.0+0.9+0.8)×sin(average phase) / (2πf average) ≈ 1.42mH; Power frequency impedance: ; Short-circuit capacity: ; Short-circuit ratio: SCR = 223 / 10 = 22.3; Verification: The standard deviation of the R value at the three frequency points is 0.009Ω (<15%), and the standard deviation of the L value is 0.13mH (<15%). Verification passed. Identification time: 295ms.

[0138] Step 3: Mode Selection (T=8.3s) SCR=22.3 belongs to a strong power grid (≥15), so select the grid-following control mode.

[0139] Step 4: Parameter optimization (T=8.3s) Under a strong power grid, Q=0. Power limit calculation: 10MW active power, 0 reactive power. The grid connection point voltage is 10.03kV (1.003pu), and the current is 574A (99.5% of rated current), which is within the limits. The maximum power is 10MW. The time taken is 35ms.

[0140] Step 5: Three-layer coordination (T=8.3-8.5s) System Layer: Initial power command 2MW. Bridge Arm Layer: Power allocated according to SOC ratio, with six bridge arms having power ratings of 0.342MW, 0.328MW, 0.336MW, 0.340MW, 0.320MW, and 0.335MW respectively, totaling 2.001MW. Module Layer: High SOC modules have approximately 2% higher voltage, while low SOC modules have approximately 1.5% lower voltage.

[0141] Step 7: Grid-connected startup (T=8.5s) Soft start, power increased from 0 to 2MW at a rate of 2MW / second, taking 1 second. Grid connection point voltage: 10.13kV (1.013pu), frequency: 50.01Hz, power factor: 0.998, harmonic THD: 1.4%. Grid connection successful. Total deployment time: 9.5 seconds.

[0142] Stable operation: Power was gradually increased to 10MW rated power in 40 seconds. During rated power operation, the grid connection point voltage was 10.04kV (1.004pu), with a voltage drop of only 0.8%, current of 574A, power factor of 0.997, and module temperature of 37-42°C. After 2 hours of operation, the SOC decreased from 54.5% to 41.2%, releasing 13.3MWh of electricity, with an energy efficiency of 94.3%.

[0143] Scenario 2: Adaptation and Mode Switching of Rural Distribution Networks with Weak Grids That same afternoon, the mobile energy storage was transferred to a 10kV distribution network in a remote township.

[0144] Initial identification (T=0-8.4s) Pre-grid connection test: voltage 9.82kV (0.982pu), frequency 49.95Hz, meeting the requirements. Pre-charging time: 5 seconds.

[0145] Rapid identification: Injected signal 250ms, FFT extraction: 150Hz: U=62.5V, I=12.0A ; 250Hz: U=102.8V, I=11.9A ; 350Hz: U=142.6V, I=12.0A ; Fitting calculations: R = 0.46Ω, L = 7.65mH; Power frequency impedance: ; Short-circuit capacity: ; Short-circuit ratio: SCR = 41.0 / 10 = 4.10; Identification took 298ms. Total time: 8.4 seconds.

[0146] Mode selection (T=8.4s): SCR=4.10 belongs to the critical weak grid (2≤SCR<5), so select the grid construction control mode.

[0147] Parameter optimization (T=8.4s): Network configuration mode, set Q=+1.5MVar (15% of rated capacity). Calculate the power limit (corresponding to the weak grid area with SCR=4.10 in Appendix A): ① Theoretical maximum power: According to curve A in the attached figure, the theoretical upper limit is about 7.5MW when SCR=4.10.

[0148] ② Voltage constraint: Assume 10MW + 1.5MVar, The grid connection point voltage is 9.0kV (0.90pu, close to the lower limit), and the current is about 1050A (1.82 times the rated current), which exceeds the limit.

[0149] ③ Current constraint: The system rated current is 577A, and the upper limit is 692A, which is 1.2 times the upper limit. The power limit is approximately 6.2MW.

[0150] ④ Take the minimum value: When the iterative calculation yields 6.0MW + 1.5MVar, The grid connection point voltage is 9.17kV (0.917pu) and the current is 688A (1.19 times the rated current), which meets all constraints.

[0151] ⑤ Reserve a safety margin: weak current grid upper limit of power .

[0152] The calculation process corresponds exactly to Figure A: SCR=4.10 is in a weak grid area, and the actual available power is about 6.0MW (60% of the rated value). After reserving a 15% margin, it is reduced to 5.1MW (red curve) to ensure that the voltage and current do not exceed the limits.

[0153] Voltage control target 10.5kV (1.05pu), droop factor , .

[0154] The above steps took 48ms.

[0155] Three-layer coordination and grid connection (T=8.4-8.6s): System layer: Grid control, set voltage 10.5kV, frequency 50.00Hz, power 2MW+1.5MVar. Arm layer: Power distribution, circulating current suppression enabled. Module layer: Voltage regulation coefficient reduced to 0.06, fine-tuning range ±1.5%.

[0156] Soft start: First, establish voltage and frequency, stabilize for 100ms, then increase power from 0 to 2MW + 1.5MVar at a rate of 2MW / second. The grid connection voltage increases from 9.82kV to 10.08kV (with a reactive power support increase of 0.26kV), frequency reaches 50.00Hz, and grid connection is successful. Total deployment time: 9.6 seconds.

[0157] Peak load response (T=1h): The grid load surged, causing the voltage to drop to 9.45kV, and the dispatching system required full load. The system output was 5.1MW (maximum power) + 1.5MVar, the grid connection point voltage was 9.15kV (0.915pu, close to the lower limit but not exceeding the limit), the current was 685A (1.18 times the rated current, close to the upper limit), and the module temperature was 47-53°C (close to the alarm). The system was operating safely.

[0158] Without power limiting and safety margin settings, outputting at 10MW would cause the grid connection point voltage to drop to 8.2kV (0.82pu, below the lower limit of 0.90) and the current to 1050A (1.82 times the rated current), triggering protection to disconnect from the grid. The method of this invention successfully avoids this.

[0159] After 6 hours of stable operation: power factor 0.96, voltage deviation <8% (large load fluctuation), harmonic THD 2.7%, SOC standard deviation 3.6%, temperature standard deviation 6.5°C, average output power 4.5MW (limited but optimal under weak grid conditions), energy efficiency 92.5%.

[0160] Scenario 3: Seamless switching during sudden changes in power grid strength The following day, the mobile energy storage was connected to the 10kV microgrid in the industrial park. The initial SCR was 11.2 (medium intensity). The enhanced grid-connected mode was selected, and the operating power was 5MW+1MVar.

[0161] T=2h: The main grid connection line of the park tripped due to a fault, and the microgrid is operating in island mode.

[0162] Periodic retesting (T=2h+10min): Identification results: R increased from 0.15Ω to 0.72Ω, L increased from 2.2mH to 19.5mH, and the power frequency impedance increased from 0.70Ω to 6.15Ω. The short-circuit capacity decreased from 143MVA to 16.3MVA, and the SCR decreased from 11.2 to 1.63 (in extremely weak power grids), a change of 86%, triggering switching.

[0163] Mode reselection (T=2h+10min+0.3s): SCR=1.63 (<2), select full network control mode.

[0164] State pre-synchronization switch: Current enhanced grid output: 5MW+1MVar, voltage 10.15kV, frequency 49.88Hz (frequency drops after islanding).

[0165] Grid controller pre-run: Voltage loop target 10.15kV, frequency loop target 49.88Hz, power loop tracking 5MW+1MVar, adjust the integrator initial value to make the output current command the same as the grid controller.

[0166] Pre-synchronization process (0-85ms): The error decreases from 18% to 8% (0-50ms), and then to 4% (50-85ms), meeting the switching conditions (<5% and lasting for 6ms).

[0167] Fast switching (T=86ms): Control handover is completed within 2ms. At the moment of switching, the power jumps from 5.00MW to 4.96MW (fluctuation of 0.8%), the voltage jumps from 10.15kV to 10.13kV (fluctuation of 0.2%), and the frequency has no significant fluctuation.

[0168] After switching and grid operation: The voltage reference was gradually adjusted to 10.5kV (takes 5 seconds), and the frequency reference was adjusted to 50.00Hz (takes 10 seconds). The energy storage, as the main power source, established the microgrid voltage and frequency. According to load changes (4-7MW), the output was automatically adjusted, with the frequency fluctuating between 49.86-49.94Hz (RoCoF<0.22Hz / s) and the voltage within the range of 10.2-10.6kV (deviation <4.5%), and stable operation for 2.5 hours.

[0169] Main network recovery (T=2h+2.5h): The tie line is reclosed, the SCR is restored to 11.2, the system detects the strength rebound, and performs a reverse switch from network construction to enhanced follow-up network, which takes 92ms and is a smooth transition.

[0170] Implementation effect Rapid identification: The identification times for the three scenes are 295ms, 298ms, and 301ms, all <300ms. Compared with the traditional method (3-10 seconds), the speed is improved by 10-30 times, and the identification accuracy error is 6%-9% (verified by comparison with an offline impedance tester).

[0171] Proactive selection: Based on quantitative SCR indicators, mode selection is completed before grid connection, with a response time of less than 1 second, which is 5-10 times faster than passive detection (3-10 seconds). It automatically selects grid-following / grid-building modes within a wide SCR range of 22.3 to 1.63, ensuring stable grid connection in both modes.

[0172] Seamless switching: In scenarios where SCR drops from 11.2 to 1.63, the switching time is 86ms, with power fluctuation of 0.8% and voltage fluctuation of 0.2%, which is a significant improvement compared to traditional switching (3-5 seconds, 10%-20% fluctuation).

[0173] Rapid deployment: From connection to successful grid connection, it takes 9.5 seconds for a strong grid and 9.6 seconds for a weak grid, both less than 1 minute, far exceeding traditional methods (5-15 minutes of manual testing and parameter tuning). The total deployment process takes less than 3 minutes.

[0174] Multi-scenario adaptability: The power utilization rate of the strong power grid is 88%, while the weak power grid is limited to 51% but no protection is triggered (avoiding multiple overvoltage and overcurrent protections of traditional methods).

[0175] System health: SOC standard deviation of 72 modules <4%, temperature standard deviation <7°C, no module failures, compared with traditional methods, the peak temperature is reduced by about 6°C, and the SOC dispersion is reduced by about 30%.

[0176] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, 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 the present invention.

Claims

1. A method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength, characterized in that, Includes the following steps: S1, System pre-grid connection test: Before the energy storage system is connected to the grid, electrical quantity detection at the grid connection point, system self-test and pre-charging control are performed; S2, Wideband Fast Grid Strength Identification: Employs a multi-frequency signal injection method to simultaneously collect grid connection point voltage and injection current. After signal processing, impedance identification calculation, and short-circuit ratio calculation, the grid strength is identified, and the identification results are verified. S3, Active grid connection mode selection based on short-circuit ratio: Quantitatively classify the grid strength according to the short-circuit ratio (SCR), and actively select the corresponding control mode based on the classification results; S4, Rapid optimization of grid connection parameters and dynamic safety margin setting: Calculates the optimal reactive power and power limit for different grid strengths, and sets the dynamic safety margin; S5, High Voltage Cascaded Topology Three-Layer Coordinated Control: Establishes a three-layer coordinated control architecture of system layer, bridge arm layer, and module layer to realize command issuance, power distribution and voltage distribution; S6, Seamless switching of state pre-synchronization: When the change in power grid intensity triggers the switching of control mode, the outputs of the new and old controllers are made consistent through the state pre-synchronization mechanism before a fast switch is performed; S7, Grid-connected startup and safety monitoring: Performs soft-start grid connection, monitors the status of the energy storage system in real time, and performs protection functions and fault isolation; S8, Periodic Retesting and Parameter Optimization: After grid connection, the grid strength is periodically retested, and the control mode is reselected and the control parameters are optimized based on the retest results.

2. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S1, the system pre-grid connection detection specifically includes: Grid connection point electrical quantity detection: The three-phase voltage at the grid connection point is collected in real time through voltage sensors and phase-locked loops, with a sampling frequency of not less than 10kHz; the effective voltage value is detected to be within 90%-110% of the rated voltage, the frequency is within the allowable range, the phase sequence is correct, and the voltage imbalance is less than 2%; System self-test: Check whether the SOC of each energy storage module is within the safe range of 20%-80%, whether the temperature is within the normal range of 5-45°C, whether the voltage difference between modules is <5%, whether the communication response time is <100ms, and whether the protection functions are intact; Pre-charge control: The DC bus voltage is gradually charged to near the peak voltage of the grid connection point through the pre-charge resistor and contactor, with a voltage difference of < 5% and a pre-charge time of 2-5 seconds. After completion, the pre-charge resistor is bypassed and the main contactor is closed.

3. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S2, the broadband fast power grid strength identification specifically includes: Multi-frequency injection signal design: Three odd harmonic frequencies of 150Hz, 250Hz, and 350Hz are selected. The amplitude of each frequency signal is set to 2% of the rated current. The sinusoidal signals of the three frequencies are superimposed as the current injection command. The current injection command formula is as follows: in, , , , For the injection amplitude; Impedance identification calculation: During the injection period, the grid connection point voltage and injection current are synchronously collected. The collected data are subjected to Fast Fourier Transform (FFT) to extract the voltage amplitude, current amplitude and phase difference at three frequency points. At each frequency point, the complex number of the grid impedance is calculated according to Ohm's law. The grid impedance model adopts the series RL model, and the grid resistance R and inductance L are solved by weighted least squares fitting. Short-circuit ratio calculation: Based on the grid resistance R and inductance L, calculate the impedance modulus at 50Hz power frequency, and calculate the three-phase short-circuit capacity at the grid connection point. With the rated capacity of the energy storage system The ratio of the two values ​​is used to obtain the short-circuit ratio (SCR). Verification of identification results: The first verification is frequency consistency. Check whether the standard deviation of the R and L values ​​obtained by fitting the three frequency points is less than 15% of the average value. The second verification is physical rationality. Check whether the grid resistance R is within the set reasonable range, whether the inductance L is within the set reasonable range, and whether the short-circuit ratio SCR is greater than 1. If the verification fails, repeat step S2.

4. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 3, characterized in that, In S2, the formula for calculating the complex number of the power grid impedance in the impedance identification calculation is as follows: In the formula, Voltage amplitude, The current amplitude, For phase difference, For frequency The current grid impedance angle is as follows; The power grid impedance model adopts a series RL model at frequency. f When the impedance is For the three frequency points, the real and imaginary parts of the impedance are as follows: The grid resistance R and inductance L are solved by weighted least squares fitting; a system of linear equations is established and solved using matrix methods. in, Number of frequency points; In the short-circuit ratio calculation, the impedance magnitude at 50Hz power frequency is calculated based on the identified R and L: The formula for calculating the three-phase short-circuit capacity is: in, Line voltage rating; short-circuit ratio .

5. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S3, the active grid connection mode selection based on the short-circuit ratio specifically includes: Grid strength classification: Quantitative classification based on short-circuit ratio (SCR): SCR≥15, strong grid; 5≤SCR<15, medium-strength grid; 2≤SCR<5, weak grid; SCR<2, extremely weak grid / isolated grid; Control mode selection: Strong power grid: Select grid-following control mode, with the energy storage system as the controlled current source. The control targets are active power P and reactive power Q. The reactive power is set to Q=0, and the power limit is the rated power. Medium-intensity power grids: An enhanced grid-following control mode is selected, adding voltage outer loop auxiliary support to the grid-following control. Reactive power is dynamically adjusted based on voltage deviation, calculated using the following formula: ,in This is the reactive power regulation coefficient. This is the reference value for the grid connection point voltage. This is the measured voltage value at the grid connection point; For weak grids, extremely weak grids / islanded grids: select the grid construction control mode, with the energy storage system acting as the controlled voltage source, and the control targets being voltage amplitude U and frequency. f It possesses active-frequency droop and reactive-voltage droop characteristics.

6. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S4, the rapid optimization of grid connection parameters and the dynamic safety margin setting specifically include: Reactive power setting: Set reactive power Q=0 under strong grid conditions; under medium grid conditions, set an appropriate amount of reactive power according to the voltage deviation; under weak grid conditions, set capacitive reactive power to 10%-20% of the rated capacity. Power limiting calculation: Based on voltage and current constraints, the maximum transmittable power is calculated. The constraints are that the grid connection point voltage is within the range of 0.9-1.1 times the rated voltage, and the system current does not exceed 1.2 times the rated current. Through iterative solutions or numerical optimization, the maximum active power that satisfies all constraints is found. ; Dynamic safety margin setting: Considering the uncertainties of the mobile energy storage site environment, a safety margin is reserved based on the calculated power limit. The final power limit calculation is as follows: in, The safety margin factor is dynamically adjusted based on the grid strength: strong grid Medium-intensity power grid Weak power grid Voltage control target: In grid construction mode, the voltage for weak grids is set at 1.02-1.05 times the rated voltage, and the voltage for medium-strength grids is set at 1.00-1.02 times the rated voltage.

7. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S5, the three-layer coordinated control of the high-voltage cascaded topology specifically includes: System-level control: Based on the selected control mode, system-level commands are generated. In grid-connected mode, total active and reactive power commands are generated, and in grid-connected mode, voltage amplitude and frequency commands are generated. Power safety limiting is also executed. Bridge arm layer power allocation: An iterative power allocation algorithm is used to allocate system power to each bridge arm. Under discharge conditions, bridge arms with high SOC receive more power, and the allocation ratio is proportional to SOC. Under charging conditions, bridge arms with low SOC receive more power, thus achieving balanced SOC allocation. Module-level voltage distribution: Modules within the bridge arm are connected in series. SOC balance is achieved by adjusting the voltage of each module. During discharge, the voltage of high SOC modules increases, and during charging, the voltage of low SOC modules increases. The voltage adjustment range is proportional to the SOC deviation. The voltage of a single module is within the range of 85%-115% of the rated value, and the total voltage of the bridge arm is equal to the reference value.

8. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S6, the seamless switching of state pre-synchronization specifically includes: Switching trigger conditions: Continuously monitor the power grid strength. If the SCR crosses the grade boundary and the duration is >3 seconds, or if the upper-level dispatching system issues a switching command, the switching process will be triggered. State pre-synchronization mechanism: The new controller corresponding to the new control mode enters the pre-running state and runs in parallel with the old controller corresponding to the currently running old control mode. The power inner loop of the new controller tracks the current actual output power, the voltage inner loop tracks the current actual output voltage, and the current inner loop tracks the current actual output current. The internal state variables of the new controller are adjusted so that the output command of the new controller is the same as that of the old controller. Quick switchover execution: When the output errors of the new and old controllers meet the switching conditions, the handover of control is completed within one control cycle; after the switchover, the new controller works normally and the old controller exits operation.

9. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, The grid connection startup and security monitoring in S7 specifically includes: Soft start strategy: Power gradually increases from 0 to the target value, with the increase rate limited to 20% / second of rated power. In network mode, the rated voltage and frequency are established first, and the power output is gradually increased after stabilizing for 100ms. Safety monitoring: Real-time monitoring of grid-side voltage, frequency, power factor, system-side output power, current of each bridge arm, voltage / SOC / temperature of each module, with a monitoring cycle of 1-10ms; Protection functions: Grid voltage over-limit protection: Disconnects from the grid if the voltage exceeds 90%-110% for 2 seconds; Frequency over-limit protection: Disconnects from the grid if the frequency exceeds 49.5-50.5Hz for 2 seconds; System overcurrent protection: Limits power if the current exceeds 1.2 times the rated current for 5 seconds, and disconnects from the grid if the current exceeds 1.5 times the rated current for 0.5 seconds; Module temperature protection: Derating operation if the temperature exceeds 55°C, and isolating the module if the temperature exceeds 60°C; Module SOC protection: Stops discharging if the SOC level is below 10%, and stops charging if the SOC level is above 90%. Fault isolation: When a module fault is detected, the bypass switch is closed within 50ms to isolate the faulty module, update the available power, and re-execute the power allocation.

10. The method for rapid identification and adaptive grid connection of flexible mobile energy storage grid strength according to claim 1, characterized in that, In S8, the periodic retesting and parameter optimization specifically include: Periodic retesting: Repeat step S2 for grid strength identification every 10-30 minutes. If the identification result differs from the current value by more than 20%, trigger step S3 for mode reselection and step S4 for parameter re-optimization. Seamlessly switch to the new mode through step S6. Parameter optimization: Based on the performance evaluation results, adjust the control parameters periodically. If the power factor is <0.95, adjust the reactive power setting; if the voltage deviation is >5%, adjust the voltage control target; if the SOC dispersion is >5%, enhance the equalization control strength. The optimization cycle is 10-30 minutes, and the adjustment range is ≤10% of the current value.

Images