Method for suppressing broadband oscillation in data center based on mobile energy storage vehicle with plug and play impedance reshaping

Abstract

The application discloses a mobile energy storage vehicle based on plug and play impedance remodeling to suppress wide frequency oscillation in a data center, comprising the following steps: after the mobile energy storage vehicle is connected to a preset energy storage interface of the data center, the access port, UPS operation mode, key bus and self-constraint information are obtained, the voltage and current signals of the grid-connected point, UPS output side, key distribution bus and access port are collected, the wide frequency oscillation mode is identified and the equivalent impedance of the access port is identified online; the target impedance is generated according to the dominant oscillation frequency band, port impedance and data center power supply safety constraint, the impedance remodeling control parameters of the mobile energy storage vehicle PCS are solved, and the frequency band selective damping current is output; after impedance remodeling, the oscillation energy is continuously monitored and the parameters are updated online. The method is helpful to adaptively provide wide frequency damping support for the mobile energy storage vehicle when temporarily connected to the data center, reduces the dependence on manual setting, and takes into account the continuous power supply constraint of key IT loads.

Description

Technical Field

[0001] This invention relates to the fields of power supply and distribution for data centers, power electronic stability control, and mobile energy storage applications, specifically to a method for suppressing broadband oscillations in data centers using a mobile energy storage vehicle based on plug-and-play impedance reshaping. Background Technology

[0002] Data centers typically include grid-connected transformers, UPS, PDUs, server clusters, cooling inverters, and numerous power electronic interfaces. With the increasing demand for AI computing loads, GPU servers, and high-power-density cooling equipment, data center loads exhibit rapid fluctuations, strong nonlinearity, and multi-time-scale coupling characteristics. When there is weak grid access, UPS mode switching, mismatched rectifier / inverter control parameters, or frequent adjustments to the cooling load, data centers may experience low-frequency oscillations, subsynchronous or hypersynchronous oscillations, and mid-to-high-frequency control-coupled oscillations.

[0003] In existing technologies, fixed energy storage, UPS control, power quality management devices, and reactive power compensation equipment can be used to improve power supply stability, but they typically assume that the equipment access location is fixed, system parameters are relatively stable, or control parameters have been set offline. Mobile energy storage vehicles have the advantages of rapid deployment and temporary support, but when connected to data centers, they face problems such as non-fixed access ports, unknown port equivalent impedance, strong constraints on UPS operating modes and continuous power supply to critical IT loads, and the inability to manually and repeatedly adjust control parameters.

[0004] Therefore, a technical solution is needed that can automatically identify the broadband oscillation mode of the data center, identify the impedance of the access port online, and reshape the external equivalent impedance of the PCS under power supply safety constraints after the mobile energy storage vehicle is plug-and-play connected. Summary of the Invention

[0005] Technical Problem: This invention provides a method for suppressing broadband oscillations in data centers using a mobile energy storage vehicle based on plug-and-play impedance reshaping, in order to solve the problems of unknown port impedance, uncertain broadband oscillation frequency band, difficulty in online tuning of damping control parameters, and the potential impact of vibration suppression control on the power supply continuity of critical loads when the mobile energy storage vehicle is temporarily connected to a data center.

[0006] Technical Solution: A method for suppressing broadband oscillations in data centers using a mobile energy storage vehicle based on plug-and-play impedance reshaping, comprising the following steps:

[0007] S1. After the mobile energy storage vehicle connects to the preset energy storage interface of the data center, it obtains the access port information, the data center operation status information and the mobile energy storage vehicle's own operation constraint information. The data center operation status information includes at least the grid connection point status, UPS operation mode, key power distribution bus voltage level and key load power supply constraints.

[0008] S2. Collect voltage, current and power signals at the data center grid connection point, UPS output side, key power distribution bus and mobile energy storage vehicle access port, perform time synchronization, outlier removal, filtering and broadband decomposition to form a multi-point synchronous measurement matrix;

[0009] S3. Identify the broadband oscillation modes of the data center based on the measurement signals after broadband decomposition, obtain the dominant oscillation frequency, oscillation amplitude, damping ratio, node phase vector and node participation vector, and select the set of dominant modes to be suppressed based on the oscillation risk index.

[0010] S4. Based on the voltage disturbance frequency domain vector and current disturbance frequency domain vector at the access port of the mobile energy storage vehicle, identify online the equivalent admittance matrix and equivalent impedance matrix observed by the data center from the access port;

[0011] S5. Based on the set of dominant modes to be suppressed, the equivalent impedance matrix, and the data center security constraints, generate the controllable impedance and target impedance presented by the mobile energy storage vehicle PCS, and solve the impedance reshaping control parameters.

[0012] S6. Under the conditions of satisfying UPS operation authority, critical load power supply continuity, mobile energy storage vehicle state of charge, PCS current limit, output power limit and bus voltage deviation constraint, generate damping current command and control the output of mobile energy storage vehicle PCS according to the impedance reshaping control parameters.

[0013] S7. Continuously calculate the broadband oscillation energy index after impedance reshaping. When the broadband oscillation energy index is not lower than the oscillation energy threshold or a new dominant oscillation mode is identified, update the equivalent impedance identification result and impedance reshaping control parameters to form a closed-loop suppression.

[0014] As a preferred embodiment, in step S2, the multi-point synchronous measurement matrix is ​​constructed as follows:

[0015] ;

[0016] Where k represents the sampling time sequence number; pcc represents the data center grid connection point; ups represents the UPS output side; bus represents the key power distribution bus; and port represents the mobile energy storage vehicle access port. This represents a multi-point synchronous measurement matrix. Represents the three-phase voltage vector. This represents the three-phase current vector, and the superscript T indicates transpose.

[0017] As a preferred embodiment, the broadband decomposition is expressed as:

[0018] ;

[0019] Where q represents the broadband decomposition frequency band number, and Q represents the total number of frequency bands. This represents the measured component of the q-th frequency band; This represents the q-th wideband pass decomposition operator, whose frequency band covers low-frequency oscillations, subsynchronous or supersynchronous oscillations, and mid-to-high frequency controlled coupled oscillations.

[0020] As a preferred embodiment, in step S3, the set of broadband oscillation modes is represented as follows:

[0021] ;

[0022] in, This represents the set of oscillation modes identified at time k. This represents the frequency of the m-th mode. Indicates modal amplitude, Indicates the modal damping ratio. Represents the node phase vector. This represents the node participation vector, where M represents the number of modes;

[0023] The oscillation risk indicator is calculated as follows:

[0024] ;

[0025] in, Indicator of oscillation risk This indicates the avoidance of small positive quantities with a denominator of zero. This indicates the security weights related to UPS mode, IT load level, and access port security margin. This represents the L1 norm.

[0026] Modes whose oscillation risk index exceeds a preset risk threshold will be incorporated into the set of dominant modes to be suppressed.

[0027] As a preferred embodiment, in step S4, the equivalent admittance matrix is ​​identified according to the following formula:

[0028] ;

[0029] in, This represents the equivalent admittance matrix observed by the data center from the access port of the mobile energy storage vehicle. This represents the frequency domain vector of the voltage disturbance at the access port at the angular frequency. This represents the frequency domain vector of the current disturbance at the angular frequency of the access port. This represents the Moore-Penrose generalized inverse.

[0030] As a preferred embodiment, the equivalent impedance matrix is ​​obtained according to the following formula:

[0031] ;

[0032] in, This represents the equivalent impedance matrix observed by the data center from the access port of the mobile energy storage vehicle. Represents the equivalent admittance matrix;

[0033] The voltage and current disturbances are generated by natural load disturbances or small-amplitude multisine detection signals injected by the mobile energy storage vehicle PCS.

[0034] As a preferred embodiment, in step S5, the controllable impedance presented by the mobile energy storage vehicle PCS is expressed as follows:

[0035] ;

[0036] in, This indicates the controllable impedance presented to the outside world by the PCS in the mobile energy storage vehicle. Let s denote the identity matrix, and s denote the Laplace operator. Indicates virtual resistance. Represents virtual inductance. This represents the damping gain of the m-th mode. Indicates the damping filter factor. This represents the angular frequency of the m-th mode. This represents the modal orientation selection matrix. This represents the vector of PCS impedance reshaping control parameters;

[0037] The damping effect is concentrated on the node direction corresponding to the dominant mode to be suppressed by using the mode direction selection matrix.

[0038] As a preferred embodiment, in step S5, the equivalent impedance of the port after connecting to the mobile energy storage vehicle is expressed by the following formula:

[0039] ;

[0040] in, This represents the equivalent impedance of the port after being connected to the mobile energy storage vehicle. This represents the equivalent impedance of a data center port. This indicates the controllable impedance of the PCS in a mobile energy storage vehicle at the corresponding frequency.

[0041] Furthermore, by making the port equivalent impedance close to the target impedance, the positive damping in the dominant oscillation frequency band is improved.

[0042] As a preferred embodiment, in step S5, the impedance reshaping control parameters are solved as follows:

[0043] ;

[0044] in, This represents the optimized impedance reshaping control parameters. Indicates the safe and feasible domain. This represents the set of dominant oscillation frequency bands to be suppressed. Indicates frequency point weight, Indicates the target impedance. This indicates the parameters of the previous control cycle. This represents the penalty coefficient for parameter changes. Denotes the Frobenius norm. Represents the L2 norm;

[0045] Furthermore, the safe and feasible domain is limited by the following formula:

[0046] ;

[0047] in, Indicates the output current of the PCS. Indicates the upper limit of PCS current. Indicates the output power of the mobile energy storage vehicle. This indicates the upper limit of the power of the mobile energy storage vehicle, and SOC indicates the state of charge of the mobile energy storage vehicle. and This indicates the lower and upper limits of the permissible state of charge. Indicates the bus voltage deviation. This indicates the upper limit of the bus voltage deviation. This variable represents the permission variable for the UPS to perform damping actions. It represents the infinite norm.

[0048] As a preferred embodiment, in step S6, the damping current command is generated by the following formula and superimposed on the current control loop or power control loop of the mobile energy storage vehicle PCS:

[0049] ;

[0050] in, This indicates a damping current command. This represents the set of dominant modes selected to participate in control. and These represent the damping coefficients for voltage feedback and current feedback, respectively. This represents the broadband pass decomposition operator. and These represent the port measurement voltage vector and the port measurement current vector, respectively. and These represent the port reference voltage vector and the reference current vector, respectively.

[0051] As a preferred embodiment, in step S7, the broadband oscillation energy index is calculated as follows:

[0052] ;

[0053] in, This represents a broadband oscillation energy index. This represents the energy weight of the m-th dominant mode. This represents the amplitude of the m-th mode at time k;

[0054] When the broadband oscillation energy index is higher than the oscillation energy threshold, the impedance reshaping control parameters are updated according to the following formula:

[0055] ;

[0056] in, This indicates the impedance reshaping control parameters for the next control cycle. This indicates projection onto the safe and feasible region. Indicates the parameters of the current control cycle. This indicates that the step size is updated online. This represents the parameter update direction matrix obtained from sensitivity identification. This represents the oscillation energy threshold.

[0057] As a preferred option, when the UPS is in bypass mode, the PCS experiences current overrun, bus voltage deviation overrun, access port communication interruption, or the oscillation energy continuously increases after impedance reshaping, the mobile energy storage vehicle PCS is switched from impedance reshaping control mode to power-limited safety support mode.

[0058] As an alternative, when multiple mobile energy storage vehicles are connected to a data center or data center campus at the same time, the damping task is allocated according to the node participation vector, the available power of each mobile energy storage vehicle, and the impedance sensitivity of each access port.

[0059] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0060] 1. Compared with conventional fixed energy storage damping control, this invention unifies the plug-and-play access identification of mobile energy storage vehicles, data center UPS safety constraints, online identification of access port impedance, and target impedance optimization into the same closed-loop control chain, which helps to improve the broadband oscillation suppression adaptability in temporary access scenarios.

[0061] 2. This invention helps mobile energy storage vehicles to adaptively provide broadband damping support when temporarily accessing data centers, reducing reliance on manual tuning, while also taking into account the continuous power supply constraints of critical IT loads. Attached Figure Description

[0062] Figure 1 This is a flowchart of the method for suppressing broadband oscillations in data centers using a mobile energy storage vehicle according to the present invention;

[0063] Figure 2 This is a schematic diagram illustrating the plug-and-play access, measurement points, and control links between the data center and the mobile energy storage vehicle in an embodiment.

[0064] Figure 3 This is a schematic diagram illustrating broadband oscillation mode identification and risk ranking based on simulated data, as an example.

[0065] Figure 4 This is a schematic diagram illustrating the identification of the equivalent impedance of the access port and the comparison between the target impedance and the reshaped impedance in an example.

[0066] Figure 5 This is a schematic diagram illustrating the oscillation energy attenuation and safety margin before and after impedance reshaping in an embodiment. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the application will be further described in detail below with reference to the accompanying drawings. The described embodiments are only a part of the embodiments involved in this invention. All non-innovative embodiments based on these embodiments by other researchers in the art are within the protection scope of this invention. Furthermore, the step numbers in the embodiments of this invention are only set for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0068] This invention discloses a method for suppressing broadband oscillations in data centers using a mobile energy storage vehicle based on plug-and-play impedance reshaping. The process is as follows: Figure 1 As shown, it includes:

[0069] S1. Plug-and-play access and information acquisition of mobile energy storage vehicle: After the mobile energy storage vehicle is connected to the preset energy storage interface of the data center, it acquires access port information, data center operation status information and mobile energy storage vehicle's own operation constraint information. The data center operation status information includes at least the grid connection point status, UPS operation mode, key power distribution bus voltage level and key load power supply constraints.

[0070] S2. Data Acquisition, Preprocessing and Broadband Decomposition: Acquire voltage, current and power signals at the data center grid connection point, UPS output side, key power distribution bus and mobile energy storage vehicle access port, and perform time synchronization, outlier removal, filtering and broadband decomposition on the signals to form a multi-point synchronous measurement matrix.

[0071] S3. Wideband oscillation mode identification and selection of modes to be suppressed: Identify the wideband oscillation modes of the data center based on the measurement signals after wideband decomposition, obtain the dominant oscillation frequency, oscillation amplitude, damping ratio, node phase vector and node participation vector, and select the set of dominant modes to be suppressed based on the oscillation risk index;

[0072] S4. Online identification of port equivalent impedance or admittance: Based on the voltage disturbance frequency domain vector and current disturbance frequency domain vector at the access port of the mobile energy storage vehicle, the equivalent admittance matrix and equivalent impedance matrix observed by the data center from the access port are identified online.

[0073] S5. Target Impedance Generation and PCS Parameter Solving: Based on the set of dominant modes to be suppressed, the equivalent impedance matrix, and the data center security constraints, the controllable impedance and target impedance presented by the mobile energy storage vehicle PCS are generated, and the impedance reshaping control parameters are solved.

[0074] S6. Damping current command generation and PCS control output under safety constraints: Under the conditions of satisfying UPS operation authority, critical load power supply continuity, mobile energy storage vehicle charge status, PCS current limit, output power limit and bus voltage deviation constraints, the damping current command is generated according to the impedance reshaping control parameters and the mobile energy storage vehicle PCS output is controlled.

[0075] S7. Oscillation Energy Calculation and Closed-Loop Parameter Update: Continuously calculate the broadband oscillation energy index after impedance reshaping. When the broadband oscillation energy index is not lower than the oscillation energy threshold or a new dominant oscillation mode is identified, update the equivalent impedance identification result and impedance reshaping control parameters to form closed-loop suppression.

[0076] In a preferred embodiment, the data center includes a 10kV / 0.4kV distribution transformer, UPS, PDU, IT server load, cooling inverter load, and a preset interface for a 0.4kV mobile energy storage vehicle. The mobile energy storage vehicle is equipped with an energy storage battery, PCS, port measurement unit, communication unit, and the controller of this invention. The PCS supports the issuance of current control loop superimposed damping current commands and equivalent virtual impedance parameters.

[0077] In this embodiment, the data center and mobile energy storage vehicle have plug-and-play access, measurement points, and control links, as shown below. Figure 2 As shown. After the mobile energy storage vehicle is connected to the preset energy storage interface of the data center, the controller receives synchronous measurement information from the grid connection point, the power output side, the key bus and the access port. At the same time, combined with safety constraints such as operating permissions, state of charge, current and power limits, and bus voltage deviation, the controller completes mode recognition, port impedance identification, target impedance generation and damped current output.

[0078] In step S1, the mobile energy storage vehicle stops and connects to the preset energy storage interface.

[0079] The controller reads the interface number, port rated voltage, allowable current, UPS current operating mode, PDU critical load level, PCS rated power, mobile energy storage vehicle charge status, and communication status.

[0080] If the UPS bypass state or communication state does not meet the impedance reshaping permission, then only power-limited safety support is allowed; if the permission variables meet the safety feasible domain constraints, then entering the impedance reshaping mode is allowed.

[0081] In step S2, three-phase voltage and current are simultaneously collected at the grid connection point, UPS output side, key power distribution bus, and mobile energy storage vehicle access port.

[0082] The controller preprocesses the sampled data, including timestamp alignment, outlier removal, per-unitization, and frequency band decomposition, constructs a multi-point synchronous measurement matrix, and obtains measurement components in different frequency bands.

[0083] In step S3, the controller identifies low-frequency, subsynchronous or supersynchronous, and mid-to-high frequency control coupled oscillation modes from the broadband decomposition results, forms a mode set, and calculates risk indicators.

[0084] Broadband oscillation mode identification and risk ranking based on simulation data, such as Figure 3 As shown, the simulation data in this embodiment identifies typical dominant oscillation modes such as 7.5Hz, 47Hz, and 180Hz. Modes with risk indicators exceeding a preset risk threshold are selected into the set of dominant modes to be suppressed, so that the controller applies band-selective damping to the higher-risk frequency bands instead of applying the same control gain to all frequency bands.

[0085] In step S4, the controller uses the natural load disturbance of the access port, or the small-amplitude multi-sine detection signal injected by the PCS that does not affect the power supply of the critical load, to extract the frequency domain vector of the port voltage disturbance and the frequency domain vector of the current disturbance, and to identify the port equivalent admittance matrix and the port equivalent impedance matrix online.

[0086] If the detection process causes the bus voltage deviation to approach the upper limit, the detection is stopped and the most recent valid identification result is used.

[0087] In step S5, the controller generates the target impedance based on the set of dominant modes to be suppressed and constructs a controllable impedance model of the mobile energy storage vehicle PCS. Subsequently, the equivalent impedance of the port after connection is calculated, and the impedance reshaping control parameters are solved within the safe and feasible region.

[0088] Equivalent impedance identification of access ports, comparison of target impedance and reshaped impedance, etc. Figure 4 As shown, the controller generates a target impedance based on the identified port impedance and the dominant mode to be suppressed. By solving the PCS impedance reshaping control parameters, the reshaped impedance moves closer to the target impedance in the dominant mode frequency band and the mid-to-high frequency coupling band, thereby increasing positive damping and limiting the control effect in non-target frequency bands.

[0089] In step S6, the controller generates a damping current command and superimposes the command onto the PCS control loop.

[0090] If the current command exceeds the PCS limit, the system projects according to the safe and feasible domain and reduces the corresponding mode gain; if the state of charge of the mobile energy storage vehicle is close to the lower limit, the critical load support margin is reserved first, and the damping power upper limit is reduced.

[0091] In step S7, the controller continuously calculates the broadband oscillation energy index.

[0092] If the oscillation energy does not fall below the threshold after impedance reshaping, or if a new dominant mode appears, the control parameters are updated, and mode identification, port impedance identification, and target impedance solving are re-executed.

[0093] The oscillation energy attenuation and safety margin before and after impedance reshaping in this embodiment are as follows: Figure 5 As shown, after impedance reshaping, the broadband oscillation energy decays rapidly and falls below the oscillation energy threshold. At the same time, the safety margins of the mobile energy storage vehicle, such as the current ratio and state of charge, remain within the allowable range. This forms a closed-loop broadband oscillation suppression process after the mobile energy storage vehicle is plug-and-play connected.

[0094] It should be noted that this embodiment is only used to illustrate the relationship between data flow, control flow and parameter calculation, and does not constitute a limitation on the access voltage level, equipment capacity or specific engineering configuration.

[0095] In a preferred embodiment, a system for suppressing broadband oscillations in data centers based on plug-and-play impedance reshaping of a mobile energy storage vehicle is provided, comprising: an access identification module, a broadband sampling module, a mode identification module, a port impedance identification module, a target impedance generation module, a PCS parameter tuning module, a safety constraint execution module, and a closed-loop feedback module; each module is configured to perform the aforementioned method.

[0096] In a preferred embodiment, a mobile energy storage vehicle is also provided, including an energy storage battery, a PCS, an access port measurement unit, a communication unit, and a controller. The controller is used to execute the aforementioned method and send damping current commands to the PCS.

[0097] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for suppressing broadband oscillations in data centers using a mobile energy storage vehicle based on plug-and-play impedance reshaping, characterized in that, Includes the following steps: S1. After the mobile energy storage vehicle connects to the preset energy storage interface of the data center, it obtains the access port information, the data center operation status information and the mobile energy storage vehicle's own operation constraint information. The data center operation status information includes at least the grid connection point status, UPS operation mode, key power distribution bus voltage level and key load power supply constraints. S2. Collect voltage, current and power signals at the data center grid connection point, UPS output side, key power distribution bus and mobile energy storage vehicle access port, perform time synchronization, outlier removal, filtering and broadband decomposition to form a multi-point synchronous measurement matrix; S3. Identify the broadband oscillation modes of the data center based on the measurement signals after broadband decomposition, obtain the dominant oscillation frequency, oscillation amplitude, damping ratio, node phase vector and node participation vector, and select the set of dominant modes to be suppressed based on the oscillation risk index. S4. Based on the voltage disturbance frequency domain vector and current disturbance frequency domain vector at the access port of the mobile energy storage vehicle, identify online the equivalent admittance matrix and equivalent impedance matrix observed by the data center from the access port; S5. Based on the set of dominant modes to be suppressed, the equivalent impedance matrix, and the data center security constraints, generate the controllable impedance and target impedance presented by the mobile energy storage vehicle PCS, and solve the impedance reshaping control parameters to generate the target impedance, and solve the impedance reshaping control parameters that make the controllable impedance presented by the mobile energy storage vehicle PCS approach the target impedance. S6. Under the conditions of satisfying UPS operation authority, critical load power supply continuity, mobile energy storage vehicle state of charge, PCS current limit, output power limit and bus voltage deviation constraint, generate damping current command and control the output of mobile energy storage vehicle PCS according to the impedance reshaping control parameters. S7. Continuously calculate the broadband oscillation energy index after impedance reshaping. When the broadband oscillation energy index is not lower than the oscillation energy threshold or a new dominant oscillation mode is identified, update the equivalent impedance identification result and impedance reshaping control parameters to form a closed-loop suppression.

2. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping as described in claim 1, characterized in that, In step S2, the multi-point synchronous measurement matrix is ​​constructed as follows: ； Where k represents the sampling time sequence number; pcc represents the data center grid connection point; ups represents the UPS output side; bus represents the key power distribution bus; and port represents the mobile energy storage vehicle access port. This represents a multi-point synchronous measurement matrix. Represents the three-phase voltage vector. This represents the three-phase current vector, with the superscript T indicating transpose; The broadband decomposition is expressed as: ； Where q represents the broadband decomposition frequency band number, and Q represents the total number of frequency bands. This represents the measured component of the q-th frequency band; This represents the q-th wideband pass decomposition operator, whose frequency band covers low-frequency oscillations, subsynchronous or supersynchronous oscillations, and mid-to-high frequency control-coupled oscillations.

3. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 1, characterized in that, In step S3, the set of broadband oscillation modes is represented as follows: ； in, This represents the set of oscillation modes identified at time k. This represents the frequency of the m-th mode. Indicates modal amplitude, Indicates the modal damping ratio. Represents the node phase vector. This represents the node participation vector, where M represents the number of modes; The oscillation risk indicator is calculated as follows: ； in, Indicator of oscillation risk This indicates the avoidance of small positive quantities with a denominator of zero. This indicates the security weights related to UPS mode, IT load level, and access port security margin. Represents the L1 norm; Modes whose oscillation risk index exceeds a preset risk threshold will be incorporated into the set of dominant modes to be suppressed.

4. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 1, characterized in that, In step S4, the equivalent admittance matrix is ​​identified according to the following formula: ； in, This represents the equivalent admittance matrix observed by the data center from the access port of the mobile energy storage vehicle. This represents the frequency domain vector of the voltage disturbance at the access port at the angular frequency. This represents the frequency domain vector of the current disturbance at the angular frequency of the access port. Represents the Moore-Penrose generalized inverse; The equivalent impedance matrix is ​​obtained according to the following formula: ； in, This represents the equivalent impedance matrix observed by the data center from the access port of the mobile energy storage vehicle. Represents the equivalent admittance matrix; The voltage and current disturbances are generated by natural load disturbances or small-amplitude multisine detection signals injected by the mobile energy storage vehicle PCS.

5. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 1, characterized in that, In step S5, the external controllable impedance of the mobile energy storage vehicle PCS is expressed as: ； in, This indicates the controllable impedance presented to the outside world by the PCS in the mobile energy storage vehicle. Let s denote the identity matrix, and s denote the Laplace operator. Indicates virtual resistance. Represents virtual inductance. This represents the damping gain of the m-th mode. Indicates the damping filter factor. This represents the angular frequency of the m-th mode. This represents the modal orientation selection matrix. This represents the PCS impedance reshaping control parameter vector. This represents the set of dominant modes selected to participate in control. The damping effect is concentrated on the node direction corresponding to the dominant mode to be suppressed by using the mode direction selection matrix.

6. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 5, characterized in that, In step S5, the equivalent impedance of the port after connecting to the mobile energy storage vehicle is expressed by the following formula: ； in, This represents the equivalent impedance of the port after being connected to the mobile energy storage vehicle. This represents the equivalent impedance of a data center port. This indicates the controllable impedance of the PCS in a mobile energy storage vehicle at the corresponding frequency. Furthermore, by making the port equivalent impedance close to the target impedance, the positive damping in the dominant oscillation frequency band is improved.

7. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 6, characterized in that, In step S5, the impedance reshaping control parameters are solved as follows: ； in, This represents the optimized impedance reshaping control parameters. Indicates the safe and feasible domain. This represents the set of dominant oscillation frequency bands to be suppressed. Indicates frequency point weight, Indicates the target impedance. This indicates the parameters of the previous control cycle. This represents the penalty coefficient for parameter changes. Denotes the Frobenius norm. Represents the L2 norm; Furthermore, the safe and feasible domain is limited by the following formula: ； in, Indicates the output current of the PCS. Indicates the upper limit of PCS current. Indicates the output power of the mobile energy storage vehicle. This indicates the upper limit of the power of the mobile energy storage vehicle, and SOC indicates the state of charge of the mobile energy storage vehicle. and This indicates the lower and upper limits of the permissible state of charge. Indicates the bus voltage deviation. This indicates the upper limit of the bus voltage deviation. This variable represents the permission variable for the UPS to perform damping actions. It represents the infinite norm.

8. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 7, characterized in that, In step S6, the damping current command is generated by the following formula and superimposed on the current control loop or power control loop of the mobile energy storage vehicle PCS: ； in, This indicates a damping current command. and These represent the damping coefficients for voltage feedback and current feedback, respectively. This represents the broadband pass decomposition operator. and These represent the port measurement voltage vector and the measurement current vector, respectively. and These represent the port reference voltage vector and the reference current vector, respectively.

9. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 6, characterized in that, In step S7, the broadband oscillation energy index is calculated as follows: ； in, This represents a broadband oscillation energy index. This represents the energy weight of the m-th dominant mode. This represents the amplitude of the m-th mode at time k; When the broadband oscillation energy index is higher than the oscillation energy threshold, the impedance reshaping control parameters are updated according to the following formula: ； in, This indicates the impedance reshaping control parameters for the next control cycle. This indicates projection onto the safe and feasible region. Indicates the parameters of the current control cycle. This indicates that the step size is updated online. This represents the parameter update direction matrix obtained from sensitivity identification. This represents the oscillation energy threshold.

10. The method for suppressing broadband oscillations in data centers by a mobile energy storage vehicle based on plug-and-play impedance reshaping according to claim 1, characterized in that, When the UPS is in bypass mode, the PCS experiences current overrun, bus voltage deviation overrun, access port communication interruption, or oscillation energy continuously increases after impedance reshaping, the mobile energy storage vehicle PCS will be switched from impedance reshaping control mode to power-limited safety support mode. When multiple mobile energy storage vehicles are connected to a data center or data center campus at the same time, the damping task is allocated according to the node participation vector, the available power of each mobile energy storage vehicle, and the impedance sensitivity of each access port.

Images