A Capability Certificate-Based Buck System-Level Stability Control Method and System

By introducing a virtual impedance controller and capability certificate on the basis of the original control loop of the Buck-type load converter, the system-level stability problem of the load converter in the multi-converter DC power distribution system is solved. The compensation quantity follows the change of the system aggregate negative damping, ensuring that the stability control is completed within the single unit's capability range, and has an online identification and verification mechanism.

CN123073689APending Publication Date: 2026-11-10STATE GRID JIANGSU ELECTRIC POWER CO LTD NANTONG POWER SUPPLY BRANCH +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611429851.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-15
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In multi-converter single-bus DC distribution systems, existing technologies struggle to achieve system-level stabilization without accessing the internal parameters of other load converters, ensuring that the compensation amount follows the aggregate negative damping change formed by all parallel loads.

Method used

By introducing a virtual impedance controller into the original control loop of the selected Buck-type load converter, and combining the bus voltage sampling signal and duty cycle reference value, a basic transfer function template for the inner actuator is formed. The required compensation conductance is calculated using the synchronous identification dataset, and an additional duty cycle is generated through the capability certificate and execution parameter set to achieve system-level stabilization control.

Benefits of technology

It enables the compensation amount to follow the changes in the system's aggregate negative damping, ensuring that system-level stability control is completed within the capacity of a single unit. It avoids access to the internal parameters of other load converters and has an online identification and verification mechanism to ensure that the stability conclusion is confirmed by the online verification results after the system is put into operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123073689A_ABST
    Figure CN123073689A_ABST
Patent Text Reader

Abstract

This invention discloses a Buck system-level stability control method and system based on capability certificates, belonging to the field of DC power distribution system stability control. The method introduces a virtual impedance controller into the original control loop of the selected Buck-type load converter; it applies identification disturbances within the candidate bus oscillation frequency band to form a synchronous identification dataset; based on this, it identifies the system's equivalent drive point impedance and aggregated load admittance, obtaining the dominant oscillation angular frequency, coherence coefficient, and effective frequency band; it calculates the required minimum compensation conductance, forming a compensation demand set; it calculates the maximum achievable virtual conductance based on the selected converter current and duty cycle, forming a stability capability certificate; after comparison, it determines the execution conductance and constructs the desired virtual admittance, generating and limiting the additional duty cycle. This invention can improve the stability of a multi-converter single-bus system by modifying only a single Buck converter.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of novel power system stability control technology, specifically relating to a Buck system-level stability control method and system based on capability certificates. Background Technology

[0002] In DC power distribution systems, load converters typically employ feedback control to maintain a constant output voltage, exhibiting constant power load characteristics at the common bus port. The low-frequency negative damping of the constant power load interacts with the output impedance of the source converter at the common bus port, easily inducing bus voltage oscillations, which can lead to system instability in severe cases. Therefore, improving the negative damping characteristics of the constant power load is crucial for enhancing the stability of DC power distribution systems.

[0003] To reduce line losses, the DC bus voltage is typically higher than the load's rated voltage. Buck-type load converters are widely used in DC power distribution systems due to their simple structure and high efficiency. Regarding the stability problem of Buck-type constant power loads, existing literature has proposed a scheme to connect an equivalent parallel virtual impedance on the input side of the load converter. For example, the 2022 IEEE TPEL paper "A Power Adaptive Impedance Reshaping Strategy for Cascaded DC System With Buck-Type Constant Power Load" discloses a power adaptive impedance reshaping strategy. This strategy constructs a parallel virtual impedance on the load side for Buck-type constant power loads, simplifies the controller to a second-order bandpass form using the Buck open-loop characteristics, and makes it independent of load power, circuit parameters, and current sensors, reshaping the input impedance only near the oscillation frequency. Furthermore, Chinese invention patent CN115714371B discloses a multi-port converter stability control method based on virtual impedance. It reveals a scheme that improves the stability of the entire system in multi-port systems such as photovoltaic, energy storage, and load converters by implementing virtual impedance reshaping only on a specific frequency band of a single energy storage converter.

[0004] The aforementioned prior art demonstrates that improving the negative damping characteristics of Buck-type constant power loads through virtual impedance reshaping and implementing stability control only for a single converter has become a known technical approach in this field. However, in existing multi-converter single-bus DC distribution systems, the number and types of load converters are numerous and diverse. Furthermore, due to commercial confidentiality or dynamic performance requirements, most load converters do not disclose their internal controllers and power parameters. Implementing virtual impedance control for each load converter is difficult to achieve and costly to modify. In addition, existing technologies typically set the virtual impedance gain based on the selected converter's own power or fixed parameters, without considering the changes in the aggregate negative damping formed by all parallel loads as loads are added, removed, and power changes. They also fail to assess the actual compensation capability that a single converter can provide. When the required compensation exceeds the converter's current, duty cycle, ripple, or thermal margin constraints, the theoretical virtual impedance cannot be physically realized, and the system may still become unstable. Therefore, how to make the compensation amount follow the changes in the aggregate negative damping of the system and achieve stabilization within the capacity of a single unit, without modifying only a single Buck-type load converter or accessing the internal parameters of other load converters, has become an urgent technical problem to be solved. Summary of the Invention

[0005] The purpose of this invention is to provide a Buck system-level stabilization control method and system based on capability certificates, in order to solve the problem of how to make the compensation amount follow the aggregate negative damping change formed by all parallel loads in a multi-converter single-bus DC distribution system, under the condition of only modifying a single Buck-type load converter and without accessing the internal parameters of other load converters, and to achieve system-level stabilization within the physical constraints of a single converter.

[0006] The technical problem solved by this invention is achieved through the following technical solution: Firstly, a capability certificate-based Buck system-level stability control method is provided, which includes: Based on the original control loop of the selected Buck-type load converter, the bus voltage sampling signal is introduced into the duty cycle reference value through the virtual impedance controller, so that the selected Buck-type load converter is equivalent to a virtual impedance that is positively resistive only near the dominant oscillation frequency in parallel at the common bus port, forming the basic transfer function template of the inner actuator. An amplitude-limited identification perturbation is applied within the candidate oscillation frequency band of the bus, and the bus voltage, inductor current, steady-state duty cycle, ripple, temperature and limiting status are collected simultaneously. The port identification current is estimated to form a synchronous identification dataset. Based on the synchronous identification dataset, frequency domain demodulation is performed and frequency points with insufficient signal-to-noise ratio are removed to obtain the dominant oscillation angular frequency, coherence coefficient and effective frequency band corresponding to the retained frequency points. Within the effective frequency band, the damping deficit is calculated frequency by frequency and the maximum value is taken. After taking into account the robust margin, the minimum compensation conductance is obtained to form the compensation demand set. At the frequency and bandwidth specified in the compensation requirement set, the maximum achievable virtual conductance is calculated based on the current and duty cycle of the selected Buck-type load converter, forming a stability capability certificate with a validity period. The required minimum compensation conductance is compared with the maximum achievable virtual conductance to determine the execution conductance and record the projection residual. Then, a bandpass-type expected virtual admittance is constructed to form an execution parameter set. An additional duty cycle is generated based on the basic transfer function template and the execution parameter set, superimposed on the original control loop and limited to form an execution record set. Re-identify the aggregated load admittance after compensation, and verify the positive real margin and the generalized Nyquist minimum distance. If they meet the requirements, maintain the current execution parameter set and continue running. If they do not meet the requirements, update the compensation requirement set, stability capability certificate and execution parameter set. If two consecutive verifications fail, or if the execution record set shows duty cycle saturation, output ripple exceeding limits, device temperature exceeding limits, or the stability certificate expires or the margin is negative, the additional duty cycle is cancelled, and the system switches to a backup Buck-type load converter with a valid stability certificate; if there is no backup device, the load is reduced in stages according to load priority.

[0007] Furthermore, including: The basic transfer function template for forming the inner actuator includes: Define the open-loop transfer function from the duty cycle small signal to the Buck input current small signal when the small signal disturbance on the output side is zero, and define the open-loop transfer function from the duty cycle small signal to the output voltage small signal when the small signal disturbance on the input side is zero; The input port compensation current, formed by the open-loop transfer function from the duty cycle output of the virtual impedance controller to the input current, is equal to the product of the desired virtual admittance and the bus voltage disturbance. The transfer function of the virtual impedance controller can be solved from the port current balance relationship. Within the dominant oscillation frequency and its vicinity, the transfer function of the virtual impedance controller is simplified to a second-order bandpass form with the dominant oscillation frequency as the center frequency. The second-order bandpass form is discretized by a bilinear transform to obtain the basic transfer function template. The center frequency of the basic transfer function template is updated by the subsequently identified dominant oscillation angular frequency, and the gain is adjusted online by the subsequently output execution conductance.

[0008] Furthermore, including: The synchronous acquisition of bus voltage, inductor current, steady-state duty cycle, ripple, temperature, and limiting status, along with estimation of port identification current, forms a synchronous identification dataset, including: Within the candidate oscillation frequency band, a disturbance with limited amplitude is superimposed on the duty cycle reference value of the selected Buck-type load converter. The amplitude of the disturbance is limited by the upper limit of output ripple, duty cycle margin, and current margin. Simultaneously acquire bus voltage, inductor current, steady-state duty cycle, output voltage ripple, device temperature, and limiting status; based on the first-order linearization relationship of the Buck input current near the operating point, estimate the port identification current flowing into the selected Buck-type load converter, expressed as: ; In the formula, For steady-state duty cycle, For steady-state inductor current, The small-signal component of the inductor current is obtained from existing sampling. For duty cycle perturbation small signal, The imaginary unit, For the first m The angular frequency corresponding to each disturbance frequency point; The bus voltage, inductor current, steady-state duty cycle, port identification current, ripple, temperature, limiting status, and timestamp are summarized to form the synchronization identification dataset.

[0009] Furthermore, the process of obtaining the dominant oscillation angular frequency, coherence coefficient, and effective frequency band includes: At each known disturbance frequency point applied, the bus voltage time-domain sequence and the port identification current time-domain sequence in the synchronous identification dataset are subjected to discrete Fourier transform or phase-locked demodulation respectively to obtain the small-signal frequency domain component of the bus voltage and the small-signal frequency domain component of the port identification current at that frequency point. The coherence coefficient of each frequency point is calculated based on the small-signal frequency domain components of the bus voltage and the port identification current. The coherence coefficient characterizes the degree of linear correlation between the bus voltage and the port identification current at the same frequency point. When the coherence coefficient is less than the confidence threshold, the frequency point is eliminated. For the remaining frequency points, the positive direction of the port identification current is defined as flowing from the bus into the selected Buck-type load converter. The negative value of the ratio of the small-signal frequency domain component of the bus voltage to the small-signal frequency domain component of the port identification current is used as the system equivalent drive point impedance of the frequency point. The source converter output impedance is subtracted from the system equivalent drive point impedance to obtain the aggregated load admittance of the frequency point. The dominant oscillation angular frequency of the retained frequency point is determined, and the coherence coefficient corresponding to the frequency point is used as the coherence coefficient of mode identification. The range of continuous frequency points centered on the dominant oscillation angular frequency and with the real part of the aggregated load admittance being negative is taken as the effective frequency band.

[0010] Furthermore, this includes: calculating the damping deficit frequency-by-frequency within the effective frequency band and taking the maximum value; then, after factoring in the robustness margin, obtaining the required minimum compensation conductance to form a compensation demand set, including: Within the effective frequency band, calculate the real part of the sum of the source port admittance and the aggregate load admittance frequency by frequency. Using the condition that this real part is not lower than the target positive real margin, calculate the damping deficit frequency by frequency. Take the maximum value among all frequency points and include the robust margin to obtain the required minimum compensation conductance, expressed as: ; In the formula, For the target positive margin; For robust margin, The output impedance of the source converter, For aggregate load admittance, For effective frequency band, For robust margin; The required minimum compensation conductance, dominant oscillation angular frequency, effective compensation bandwidth, target positive real margin, robust margin, coherence coefficient, and timestamp are summarized to form the compensation requirement set.

[0011] Furthermore, this includes: calculating the maximum achievable virtual conductance based on the current and duty cycle of the selected Buck-type load converter at the frequency and bandwidth specified in the compensation demand set, including: Using the duty cycle to input current transfer function and duty cycle to output voltage transfer function in the average small-signal model of the selected Buck-type load converter, the above margins are converted into the maximum allowable compensation current. Take the minimum value among the maximum compensation currents and convert it to the maximum achievable virtual conductance based on the design bus oscillation amplitude, that is: ; Then, according to the design bus oscillation amplitude Calculated maximum achievable virtual conductance: ; in, This is the input current margin. The rated current of the input-side devices of the selected Buck-type load converter. The current input current, The duty cycle margin is determined by the difference between the current duty cycle and the upper and lower limits. To allow output ripple, This is a margin for heat loss. These are the transfer functions from input current to duty cycle and from output voltage to duty cycle, respectively. To design the bus oscillation amplitude, To allow output ripple, This represents the steady-state value of the bus voltage.

[0012] Further, it includes: comparing the required minimum compensation conductance with the maximum realizable virtual conductance, determining the execution conductance and recording the projection residual, thereby constructing a bandpass-type desired virtual admittance to form an execution parameter set, including: When the stabilization capability certificate is valid, and the maximum achievable virtual conductance is... Not less than the required minimum compensation conductance season Otherwise, order the execution. The difference between the two is denoted as the projection residual. ; Construct the bandpass-type desired virtual admittance, expressed as: ; in, To compensate for the actual conductance, For the desired virtual admittance bandwidth, From the set of compensation demands, This represents the projection residual.

[0013] Furthermore, it includes: generating an additional duty cycle based on the basic transfer function template and the execution parameter set, superimposing it onto the original control loop and limiting it to form an execution record set, including: The bus voltage oscillation component is converted into the desired compensation current via the actual desired virtual admittance, and then divided by the transfer function of the input current to the duty cycle of the selected Buck-type load converter to obtain the additional duty cycle, expressed as: ; in, The small-signal transfer function from the input current to the duty cycle of the selected Buck-type load converter; The additional duty cycle is added to the original control loop's duty cycle and projected onto the upper and lower limits of the duty cycle to obtain the actual injected duty cycle command, expressed as: ; in, To project the variable within the parentheses to the minimum duty cycle and maximum duty cycle Projection operator between, This represents the oscillating component of the bus voltage. This indicates the duty cycle of the original control loop output, i.e., the duty cycle signal of the original voltage loop and current loop of the selected Buck-type load converter when there is no additional virtual impedance control. During execution, the projected residual, temperature rise, and output ripple are continuously accumulated. The actual injected additional duty cycle, number of saturation occurrences, peak duty cycle, peak output ripple, device temperature rise, projected residual, and execution timestamp are summarized to form the execution record set. The execution record set serves as the execution-side data source for subsequent review and rollback.

[0014] Furthermore, this includes: re-identifying the compensated aggregate load admittance, verifying the positive real margin and the generalized Nyquist minimum distance, including: The compensated aggregate load admittance is recalculated using the synchronization data during the execution record set; Within the effective frequency band, calculate the minimum real part of the sum of the source port admittance and the compensated aggregated load admittance, using this as the positive real margin. Also, calculate the minimum amplitude of the preset Nyquist distance function within the effective frequency band, using this as the generalized Nyquist minimum distance, i.e.: ; In the formula, For the aggregate admittance after compensation; For positive real margin, For generalized Nyquist minimum distance, For the angular frequency variable within the effective frequency band, To take the real part of the complex number, This is the effective frequency band.

[0015] Secondly, the present invention also provides a Buck system-level stabilization control system based on capability certificates, comprising: The inner virtual impedance control module is used to introduce the bus voltage sampling signal into the duty cycle reference value through the virtual impedance controller on the basis of the original control loop of the selected Buck-type load converter, so that the selected Buck-type load converter is equivalent to a virtual impedance with positive resistance only near the dominant oscillation frequency in parallel at the common bus port, forming the basic transfer function template of the inner actuator. The bounded identification perturbation module is used to apply amplitude-limited identification perturbations within the candidate oscillation frequency band of the bus, synchronously collect bus voltage, inductor current, steady-state duty cycle, ripple, temperature and limiting status, and estimate port identification current to form a synchronous identification dataset. The aggregation admittance and deficit solution module is used to perform frequency domain demodulation based on the synchronous identification dataset and remove frequency points with insufficient signal-to-noise ratio to obtain the dominant oscillation angular frequency, coherence coefficient and effective frequency band corresponding to the retained frequency points. Within the effective frequency band, the damping deficit is calculated frequency by frequency and the maximum value is taken. After taking into account the robust margin, the minimum compensation conductance is obtained to form the compensation demand set. The capability certificate and projection execution module is used to calculate the maximum realizable virtual conductance based on the current and duty cycle of the selected Buck-type load converter at the frequency and bandwidth specified in the compensation requirement set, compare the required minimum compensation conductance with the maximum realizable virtual conductance, determine the execution conductance and record the projection residual, and then construct a bandpass-type expected virtual admittance to form an execution parameter set; generate an additional duty cycle based on the basic transfer function template and the execution parameter set, superimpose it on the original control loop and limit it to form an execution record set; The margin verification module is used to re-identify the aggregated load admittance after compensation, verify the positive real margin and the generalized Nyquist minimum distance. If they meet the criteria, the current execution parameter set is maintained and the operation continues. If they do not meet the criteria, the compensation requirement set, stability capability certificate and execution parameter set are updated. The exception handling module is used to cancel the additional duty cycle and switch to a backup Buck-type load converter with a valid stability capability certificate when two consecutive verifications fail, or when the execution record set shows duty cycle saturation, output ripple exceeding the limit, device temperature exceeding the limit, or the stability capability certificate expires or the margin is negative; if there is no backup device, the load is reduced in stages according to load priority.

[0016] Thirdly, the present invention also provides a computer device, comprising: one or more processors; The processor is used to store one or more programs; When the one or more programs are executed by the one or more processors, the steps of the Buck system-level stability control method based on capability certificates as described in the first aspect are implemented.

[0017] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed, it implements the steps of the Buck system-level stability control method based on capability certificates as described in the first aspect.

[0018] The advantages and positive effects of this invention are: (1) This application identifies the system equivalent drive point impedance and aggregated load admittance of the common bus port online, and calculates the minimum compensation conductance required by the system from the real part deficit of the aggregated load admittance in the effective frequency band. This makes the compensation amount follow the aggregated negative damping change formed by all parallel loads, rather than just following the power change of the selected Buck itself. Thus, the inner virtual impedance control is extended from natural adaptation to the power of the selected Buck itself to system-level adaptation to the aggregated negative damping of all parallel loads.

[0019] (2) In this application, the input current margin, duty cycle margin, allowable output ripple and heat loss margin of the selected Buck-type load converter are uniformly converted into the maximum realizable virtual conductance, forming a stabilization capability certificate with a validity period. The converter is only allowed to undertake the system stabilization task when the certificate is valid and the maximum realizable virtual conductance is not less than the required minimum compensation conductance. The required compensation amount is projected onto the single-machine realizable domain and the projection residual is recorded, thereby avoiding the theoretical virtual impedance from exceeding the realizable capability of the physical converter.

[0020] (3) After the compensation is implemented, this application re-identifies the aggregated load admittance after compensation, calculates the positive real margin and the generalized Nyquist minimum distance; if the standard is met, the current execution parameters remain unchanged; if the standard is not met, the compensation requirement set, stability capability certificate and execution parameter set are updated according to the projection residual and the reason for exceeding the limit in the execution record set. When the verification fails twice in a row, the additional duty cycle is cancelled, the alternative Buck-type load converter is switched or the load is reduced in stages according to the load priority, so that the stability conclusion is confirmed by the online verification result after the implementation.

[0021] (4) The apparent driving point impedance identified and modified in this application includes the closed-loop coupling relationship of the interconnected system, rather than the local open-loop input impedance of the unmodified load converter itself. The impedance change in the experimental verification has a clear measurement port and physical causal explanation, maintaining physical causal consistency.

[0022] (5) This application selects only one Buck-type load converter from all Buck-type load converters as the selected Buck-type load converter, introduces a virtual impedance controller on the basis of its original voltage loop and current loop, and uses its existing bus voltage sampling, inductor current sampling and duty cycle information to complete identification, capability assessment and execution. There is no need to access the internal parameters of other load converters, nor is there a need to add sensors for other loads.

[0023] (6) This application solves the virtual impedance controller transfer function through the port current balance relationship, and simplifies it into a second-order bandpass form with the dominant oscillation frequency as the center frequency in the dominant oscillation frequency and its vicinity. The gain of this second-order bandpass form is only related to the bus voltage sampling coefficient, the isolation transformer turns ratio, the steady-state duty cycle and the virtual impedance gain constant, and has no direct relationship with the voltage loop controller parameters and load power. No additional power sampling and calculation circuit is required. Its center frequency and gain are updated by the subsequent online identification results and the execution conductance, so that the inner actuator can follow the online identification results of the system-level closed loop.

[0024] Therefore, this application enables the online identification system to aggregate negative damping, calculate the required compensation amount, evaluate the single-machine capability, project to the realizable domain, verify the stability margin after implementation, and update parameters or perform safe rollback in case of failure through a system-level closed-loop data flow. These steps are sequentially connected to form an inseparable data chain, and the overall stability effect is guaranteed by the system-level closed loop. Attached Figure Description

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] Figure 1 This is a schematic diagram of the amplitude-frequency characteristics of the bandpass filter of the virtual impedance controller described in an embodiment of the present invention; Figure 2 This is a control structure diagram of a Buck-type load converter using system-level power adaptive stabilization control as described in an embodiment of the present invention; Figure 3 This is the Bode plot of the equivalent input impedance of converters #3 and #4 before and after the addition of the stabilization control described in this embodiment of the invention; Figure 4 The stabilization control described in this embodiment of the invention includes the transfer function before and after its addition. Bode plot; Figure 5 The stabilization control described in this embodiment of the invention includes the transfer function before and after its addition. and Bode plot; Figure 6 This is a flowchart of the system-level power adaptive stabilization control method based on the capability certificate of a single Buck-type load converter and a aggregated damping closed loop, as described in an embodiment of the present invention. Detailed Implementation

[0027] 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.

[0028] This embodiment provides a system-level power adaptive stabilization control method based on the capability certificate of a single Buck-type load converter and a convergent damping closed loop. This embodiment focuses on a multi-converter single-bus DC power distribution system. The system includes a source converter and several load converters connected in parallel to a common bus. The load converters include Buck-type load converters, whose main circuit consists of a switching transistor, a freewheeling diode, a filter inductor, and a filter capacitor. The input side is connected to the common bus, and the output side is connected to the load. The load converter itself is equipped with a dual-loop controller, including an outer voltage loop and an inner current loop: the outer voltage loop collects the output voltage and compares it with a voltage reference value, generating a reference value for the inner current loop via a voltage regulator; the inner current loop collects the inductor current and compares it with a current reference value, generating a duty cycle signal via a current regulator, which is then used to drive the switching transistor via PWM modulation. This dual-loop controller is the original control loop of the load converter. This invention selects one Buck-type load converter from all Buck-type load converters as the selected Buck-type load converter. This selected Buck-type load converter has an outer voltage loop, an inner current loop, and PWM modulation. The remaining load converters do not expose their internal controllers and power parameters. The digital controller sampling frequency is... The sampling period is The bus voltage is Buck output voltage is The selected Buck load power is The dominant oscillation frequency of the system is This implementation achieves system-level adaptive stabilization through the coordinated operation of the inner virtual impedance control step S100 and the closed loop of the outer steps S200 to S900.

[0029] Specifically, such as Figure 6 As shown, the method includes the following steps: Step S100: Inner Layer Virtual Impedance Control Configuration In this step, without changing the internal structure of the original voltage and current loops of the selected Buck-type load converter, the sampled signal of the common bus voltage is introduced into the duty cycle reference value through a virtual impedance controller, so that the selected converter is equivalently connected in parallel with a desired virtual impedance at the common bus port. This virtual impedance only exhibits positive resistivity at and near the system's dominant oscillation frequency, which is used to compensate for the low-frequency negative damping on the input side of the selected converter, while it is approximately open-circuited in other frequency bands, thus not affecting the normal dynamic characteristics of the selected converter. The center frequency of the virtual impedance is updated by the subsequent online identification results, and the gain is adjusted online by the subsequent system-level closed loop.

[0030] In a preferred embodiment of this invention, this step is implemented as follows: The selected Buck-type load converter is the Buck-type load converter selected to undertake the stabilization task in a multi-converter single-bus DC distribution system. The existing voltage loop refers to the voltage outer loop already configured on the selected Buck-type load converter, which collects the Buck output voltage. and the voltage reference value Comparison, via voltage regulator Generate a current inner loop reference value. The existing current loop refers to the current inner loop already configured in the selected Buck-type load converter, which collects inductor current. And compared with the current reference value, after passing through the current regulator Duty cycle signal Then through the PWM modulator The driving switching transistor. The voltage loop and current loop described above constitute the original dual-loop control structure of the selected Buck-type load converter. Figure 2 This is a control structure diagram of a Buck-type load converter using system-level power adaptive stabilization control according to the present invention. The diagram shows the bus voltage sampling, voltage loop, current loop, virtual impedance controller, and PWM modulation stage.

[0031] While maintaining the original voltage loop and current loop structure, the bus voltage sampling signal is passed through a virtual impedance controller. Introducing a duty cycle reference value, designing parallel virtual impedance This makes it at the system instability frequency The frequency bands in and around exhibit positive resistivity, while other frequency bands are approximately infinite. In this embodiment, the parallel virtual impedance is represented as: (1) In the formula, This is the virtual impedance gain constant, which is determined based on the required degree of negative damping compensation, and is usually taken to be around 1. The load power of the selected Buck is obtained through real-time sampling; This is the DC bus voltage, determined according to the system voltage level. The dominant oscillation frequency of the system is obtained through online identification in subsequent step S300. The quality factor is determined based on the oscillation bandwidth requirements and must meet them. It covers the dominant oscillation frequency band without affecting low-frequency dynamics.

[0032] This embodiment Figure 1 This is a schematic diagram of the amplitude-frequency characteristics of the bandpass filter of the virtual impedance controller of the present invention, showing the desired virtual admittance. The ratio of open-loop transfer functions and the amplitude-frequency characteristics of their product.

[0033] Implement parallel virtual impedance in the control loop Controller transfer function The detailed derivation is as follows: The Buck average small signal model is composed of Figure 2 The main circuit of the Buck-type load converter shown is obtained through state averaging and linearization, specifically including: Within one switching cycle, for Figure 2 The switching transistors and freewheeling diodes of the Buck-type load converter shown are used for state averaging, using duty cycle. d Replacing the switching function, an average circuit model is obtained; in this average circuit model, the switching transistor branch is equivalent to a current source controlled by the duty cycle, the freewheeling diode branch is equivalent to a voltage source controlled by the duty cycle, and the filter inductor... L Filter capacitor C and its equivalent series resistance R L , R C If the load remains unchanged, it is equivalent to a constant power load. Small-signal perturbations are applied to each variable in the average circuit model near the steady-state operating point, and second-order and higher-order terms are ignored to obtain a linearized Buck average small-signal equivalent circuit model; in the small-signal equivalent circuit model, the duty cycle small-signal... By applying a controlled source to the input and output ports, the small signal of the bus voltage... Acting on the input port, output voltage small signal and output current small signal Applying to the output port; in the Buck average small-signal equivalent circuit model, the small-signal disturbance on the output side is set to zero, eliminating the intermediate variables of inductor current and capacitor voltage, thus obtaining the open-loop transfer function from the duty cycle small signal to the Buck input current small signal. By setting the small-signal disturbance on the input side to zero and eliminating the intermediate variables of inductor current and capacitor voltage, the open-loop transfer function from the duty cycle small signal to the output voltage small signal is obtained. Similarly, the transfer function from the input current to the duty cycle used in subsequent S500 and S700 circuits can be obtained from the Buck average small-signal equivalent circuit model. and the transfer function from output voltage to duty cycle .

[0034] In a preferred embodiment, to avoid the lack of a source for the open-loop transfer function, a definition is provided. Let the open-loop transfer function from the duty cycle small signal to the Buck input current small signal be defined when the small-signal disturbance on the output side is zero. The open-loop transfer function from the duty cycle small signal to the output voltage small signal is given when the input-side small-signal disturbance is zero. Both are derived from... Figure 2The Buck average small-signal model shown is obtained by eliminating the intermediate variables of inductor current and capacitor voltage, and is expressed as: (2) In the formula, The small-signal component of the Buck input current; For small signal components with duty cycle; This refers to the small-signal component of the bus voltage. This refers to the small-signal component of the Buck output voltage. This represents the small-signal component of the Buck output current; the vertical bar with a subscript indicates the value of this small-signal component under the condition that the corresponding disturbance is set to zero. Specifically, This indicates that the small signal disturbance on the output side is set to zero (i.e. When =0), the duty cycle small signal Small signal to Buck input current The open-loop transfer function; This indicates that the small signal disturbance on the input side is set to zero (i.e. When =0), the duty cycle small signal small signal to Buck output voltage The open-loop transfer function.

[0035] The duty cycle compensation amount output by the controller acts on This generates the input port compensation current; setting this compensation current equal to the product of the desired virtual admittance and the bus voltage disturbance, the input port current balance relationship can then be used to solve for the compensation current. .

[0036] therefore, and It is not a newly introduced independent parameter, but... Figure 2 The two open-loop port transfer functions of the average small-signal model. From this, the controller transfer function is obtained as: (3) Among them, the numerator coefficient and denominator coefficient The expansion is as follows: (4) (5) in, This is the power of a constant power load; The equivalent series resistance of the output filter capacitor; L is the equivalent series resistance of the output filter inductor; C is the output filter inductor; and C is the output filter capacitor. For the turns ratio of the isolation transformer; This is the voltage loop proportionality coefficient; The voltage loop integral coefficient; For PWM modulator gain; This refers to the output voltage. This refers to the bus voltage. This represents the voltage sampling coefficient.

[0037] Because within the control bandwidth of a Buck-type load converter, there are Furthermore, considering that the oscillation frequency of a DC power distribution system is generally much smaller than the cutoff frequency of the load converter... Therefore, in the frequency range Within, combining equations (3) and (5), we can approximate the following: (6) Since the equivalent series resistance of capacitor C and inductor L satisfies: And less than The low frequency band has ,therefore, It can be further simplified to a second-order low-pass filter, with a cutoff frequency of The bandpass filter characteristic at the center frequency is expressed as: (7) The cutoff frequency of this second-order low-pass filter is Low-frequency gain is .because Therefore, This means The center frequency is less than The cutoff frequency. Therefore and The product still has the property of The bandpass filtering characteristics are for the center frequency. Therefore, substituting equations (1) and (7) into equation (6) reveals that within the frequency range... Within, approximately: (8) It can be seen from the above formula that: controller Only with voltage sampling coefficient Isolation transformer parameters Duty cycle D, constant and system oscillation frequency Related. Compared with the simplified expression (3), the order of the transfer function is significantly reduced, and its correlation with system parameters is lower, making it more feasible. It is important to note that the above... It is not directly related to the voltage loop controller parameters and load power, and has a natural power adaptive characteristic, so there is no need to add additional power sampling and calculation circuits.

[0038] Furthermore, when this Buck-type load converter is selected to undertake the stabilization task of the entire multi-converter single-bus DC power distribution system, in order to compensate for the aggregate negative damping formed by all parallel loads, the gain of the inner virtual impedance controller needs to be adjusted according to the total load power of the system. With the power of the selected Buck itself The ratio is extended, that is, in the frequency range. Internally, the controller transfer function can be expressed as: (9) in, This represents the total power of all parallel load converters. This equation shows that the inner-layer virtual impedance control inherently possesses system-level adaptive capability that scales proportionally to power. In the subsequent closed-loop steps S200–S900, the required compensation at the system level... The online projection will be performed via step S600. And replace the fixed gain parameter in this step to achieve online adaptive adjustment.

[0039] When implementing the above virtual impedance control strategy in a digital controller, it is necessary to convert the transfer function in the continuous time domain... Transform to the Z-domain. Use the bilinear transformation formula: (10) Substituting the above equation into the simplified version The discrete-time transfer function can be obtained. : (11) In the formula, The sampling period is [percentage]. This step yields [the following information]. As the basic transfer function template for the inner actuator, its center frequency The gain will be updated online via subsequent system-level closed-loop updates. Specifically, the dominant oscillation angular frequency identified in step S300... This will return to the previous step and replace the initial center frequency; the execution conductance output by step S600 The fixed gain in this step will be replaced. When the closed loop is put into operation in steps S200 to S900, step S700 will generate the actual additional duty cycle based on the bandpass filter structure established in this step and the online parameters given in step S600.

[0040] Between steps S100 and S200: Step S100 only configures the inner execution channel, and step S200 does not allocate an additional duty cycle during the identification phase. Only apply identification perturbation To avoid conflicts, the bus voltage and inductor current collected in step S200 provide data for the identification in step S300. The identification in step S300... Return to step S100 to update the center frequency, and step S600 outputs... Used for online adjustment of the gain in step S100.

[0041] Step S200: Bounded identification of disturbances and generation of synchronization data packets Within the candidate oscillation frequency band, the control loop of the selected load converter applies an amplitude-limited excitation signal to the duty cycle reference value, causing the common bus port to generate identifiable voltage and current responses. Simultaneously, using the existing bus voltage sampling, inductor current sampling, and duty cycle information of the selected converter, the bus voltage, inductor current, steady-state duty cycle, output ripple, device temperature, and limiting status are recorded synchronously. Based on the first-order linearized relationship between the input current, inductor current, and duty cycle, the port identification current flowing into the selected converter is estimated, thereby forming an identification data package containing frequency domain response data, operating point information, and device status information. This provides a unique data source for the subsequent identification of the system's equivalent drive point impedance and aggregated load admittance.

[0042] In a preferred embodiment of this invention, this step is implemented as follows: Based on offline impedance analysis, historical operating data, or identification results from the previous control cycle, candidate oscillation frequency bands where dominant oscillations may occur at the common bus port are pre-determined. Within these candidate oscillation frequency bands, the controller superimposes amplitude-limited multi-frequency or frequency-sweeping disturbances onto the duty cycle reference value of the selected Buck. Synchronously acquire bus voltage Buck inductor current Steady-state duty cycle Output voltage ripple, device temperature, and limiting status are used to form a synchronous identification dataset. The bus port identification current is estimated using the first-order linearized relationship of the Buck input current. (12) In the formula, The steady-state duty cycle is determined by the operating point; The steady-state inductor current is determined by the load power and the output voltage; The small-signal component of the inductor current is obtained from existing sampling. The duty cycle disturbance is a small signal, known internally by the controller. The disturbance amplitude is limited by the upper limit of output ripple, duty cycle margin, and current margin, typically ranging from 1% to 5% of the steady-state duty cycle; the disturbance frequency band covers the candidate oscillation frequency band. , The imaginary unit, For the first m The angular frequency corresponding to each disturbance frequency point These represent the port identification current, inductor current, and duty cycle disturbance, respectively, on the [number]th [period]. m The small signal frequency domain components at each perturbation frequency point. The synchronization identification dataset package output in step 200 contains... as well as These data are then fed into step S300 for calculation. and The only source of data. Therefore, the consecutive parameters from step S200 to step S300 are: And timestamp.

[0043] Step S300: Identification of system equivalent drive point impedance and aggregated load admittance This step uses discrete Fourier transform or phase-locked demodulation to extract the amplitude and phase of each frequency point based on the same frequency voltage and port current in the previous level synchronization data packet, and uses the coherence coefficient to eliminate frequency points with insufficient signal-to-noise ratio. In this embodiment, discrete Fourier transform or phase-locked demodulation is performed on the same-frequency bus voltage and port identification current in the synchronous identification dataset to obtain each disturbance frequency point. Voltage frequency domain components at the location and current frequency domain components The frequency domain demodulation refers to: each known disturbance frequency point applied in step S200 At this point, the time-domain sampling sequence of the bus voltage and the time-domain sequence of the port identification current are subjected to discrete Fourier transform or phase-locked demodulation, respectively, to obtain the small-signal frequency domain components of the bus voltage and the port identification current. The results of the frequency domain demodulation are used to subsequently calculate the coherence coefficient, system equivalent drive point impedance, and aggregate load admittance at each frequency point.

[0044] The coherence coefficient at each frequency point is calculated using the amplitude-squared coherence function, which is to take the square of the cross-power spectral magnitude of the bus voltage and the port identification current, and divide it by the product of the bus voltage self-power spectral density and the port identification current self-power spectral density. This is expressed as: ,in, yes conjugate, k For the same frequency point, the sampling sequence number, when When that frequency point is removed, it will not be included in subsequent calculations; The coherence coefficient confidence threshold is determined based on the required identification accuracy.

[0045] For the retained frequency points, the positive direction of the port identification current is specified as flowing from the bus into the selected Buck-type load converter, and the system equivalent drive point impedance and aggregate load admittance are calculated.

[0046] In this embodiment, under the specified positive current direction, the system equivalent drive point impedance of the common bus port is calculated, and the influence of the source converter output impedance is subtracted from the impedance to obtain the aggregate load admittance formed by all parallel loads. At the same time, the dominant oscillation angular frequency, effective bandwidth and identification confidence are output as the basis for subsequent damping deficit calculation and closed-loop enable judgment.

[0047] In a preferred embodiment of this invention, this step is implemented as follows: Discrete Fourier or phase-locked loop demodulation is performed on the same-frequency voltage and port current in the synchronous identification dataset to obtain the coherence coefficient. Frequency points with insufficient signal-to-noise ratio are eliminated. (Specification) If the positive direction is from the busbar into the selected Buck, then: (13) In the formula, The output impedance of the source converter can be obtained from offline identification results or source port testing in this cycle. S300 output mode identification result set = { timestamp} The coherence coefficient confidence threshold is determined based on the required identification accuracy; only If the pattern recognition result set is in the correct state, then proceed to step S400; otherwise, return to step S200 to adjust the perturbation frequency or sampling duration.

[0048] The pattern recognition results output in step S300 are in the set. Step S400 calculates the center frequency of the damping deficit frequency by frequency. The steps are S400 calculation. Direct input, This is the enable condition for step S400. This refers to the integration frequency band in step S400. If... The pattern recognition result set is returned to step S200, forming a confidence loop between steps S200 and S300. Therefore, the coherent parameters from step S300 to step S400 are: .

[0049] Among the retained frequency points, the angular frequency corresponding to the frequency point where the real part of the aggregate load admittance is negative and the absolute value is the largest is taken as the dominant oscillation angular frequency, the coherence coefficient of this frequency point is taken as the coherence coefficient of mode identification, and the range of continuous frequency points centered on the dominant oscillation angular frequency where the real part of the aggregate load admittance is negative is taken as the effective frequency band.

[0050] In this embodiment, the dominant oscillation angular frequency, aggregated load admittance, coherence coefficient, effective frequency band, and timestamp are summarized to form a pattern identification result set. Specifically, the dominant oscillation angular frequency is used in S400 to determine the center frequency when calculating the damping deficit frequency-by-frequency within the effective frequency band; the aggregated load admittance is used in S400 to calculate the real part deficit of the sum of the source port admittance and the aggregated load admittance; the coherence coefficient is used to determine the reliability of the current identification result; the pattern identification result set only proceeds to step S400 if the coherence coefficient is not less than the reliability threshold; the effective frequency band is used in S400 to limit the frequency-by-frequency calculation range of the damping deficit; and the timestamp is used for validity period alignment in S400, S500, and S600, and for time synchronization during review in S800. If the coherence coefficient is less than the reliability threshold, the process returns to S200, adjusting the perturbation frequency or extending the sampling duration to re-form a synchronized identification dataset.

[0051] Step S400: Solving for system damping deficit and required compensation The technical content of this step can be summarized as follows: Based on the aggregated load admittance and source port admittance identified in the previous stage, calculate the real part of the sum of the source port admittance and aggregated load admittance frequency by frequency within the effective oscillation frequency band; with the real part not being lower than the target positive real margin as the design condition, calculate the damping deficit frequency by frequency, and take the maximum value among all frequency points as the minimum compensation conductance that a single selected converter needs to provide; at the same time, take into account the robustness margin of coverage identification error and parameter drift, and output a compensation requirement package containing the required compensation conductance, center angular frequency, compensation bandwidth, target positive real margin, robustness margin, identification confidence and timestamp, to provide a requirement benchmark for subsequent single-unit capability assessment and realizable domain projection.

[0052] In a preferred embodiment of this invention, this step is implemented as follows: The real part of the sum of the source port admittance and the aggregate load admittance is not less than the target positive real margin. For design conditions, in Intra-frequency calculation: (14) In the formula, The target positive real margin is determined based on the system stability margin requirements; For robustness margin, it is determined based on identification error and parameter drift, and is usually taken as... 20% to 30%; To compensate for the effective bandwidth, it is determined based on the oscillation bandwidth, and is usually taken as... .

[0053] The minimum compensation conductance, dominant oscillation angular frequency, effective compensation bandwidth, target positive real margin, robustness margin, coherence coefficient, and timestamp required in this embodiment are summarized to form the compensation requirement set, which is represented as follows: Then, proceed to steps S500 and S600. These are the S500 capability assessment benchmark and the S600 comparison benchmark; The steps are S500 for evaluating the transfer function frequency and S600 for evaluating the bandpass center frequency. Step S500 evaluates the frequency band and S600 determines... The basis; and Used for verification in step S800; The timestamp is used to determine credibility in steps S500 and S600; it is used for certificate validity and verification synchronization. Therefore, the consecutive parameters from step S400 to step S500 are: .

[0054] Step S500: Generation of a single Buck stabilization certificate This step receives the compensation requirement package from the previous level. Under the frequency and bandwidth specified in the compensation requirement, it evaluates the current input current margin, duty cycle margin, allowable output ripple, and thermal loss margin of the selected converter. Using the duty cycle to input current transfer function and duty cycle to output voltage transfer function in the average small-signal model of the selected converter, the above margins are uniformly converted into the maximum allowable compensation current, and the minimum value is taken as the upper limit of the compensation current that can be injected into the converter. Then, it is converted into the maximum realizable virtual conductance according to the design bus oscillation amplitude, forming a capability certificate containing the maximum realizable virtual conductance, various margins, validity period, and equipment identification. When the certificate expires, the temperature exceeds the limit, or any quantity is negative, the converter is prohibited from undertaking the system stabilization task.

[0055] In a preferred embodiment of this invention, this step is implemented as follows: Step S500 is to compensate for the demand set. To evaluate frequency points, with To evaluate the frequency band, with This is a reliability enable condition. Based on the input current margin. Duty cycle margin Allow output ripple Heat loss margin And the corresponding port transfer function, converted into the maximum allowable compensation current, and the minimum value is taken: (15) Then, according to the design bus oscillation amplitude Calculated maximum achievable virtual conductance: (16) in, The input current margin is determined by the difference between the device's rated current and the current, and is expressed as: ,in, The rated current of the input-side devices of the selected Buck-type load converter. The current input current is determined by sampling the inductor current and the steady-state duty cycle. Calculated.

[0056] The duty cycle margin is determined by the difference between the current duty cycle and its upper and lower limits; the upper and lower limits are known constraints of the switching device, and the current duty cycle is known internally by the controller. Therefore... It can be calculated directly, for example: ,in, This represents the current steady-state duty cycle. and These are the minimum and maximum duty cycles allowed for the switching device, respectively.

[0057] To allow for output ripple, the value is determined based on the required output voltage accuracy, typically ranging from 1% to 5% of the output voltage.

[0058] The thermal dissipation margin is determined by the maximum junction temperature, the current temperature, and the thermal resistance. The maximum junction temperature is a value known from the device datasheet, the current junction temperature is obtained from a temperature sensor, and the thermal resistance is a known value from the device package. ,in T represents the maximum allowable junction temperature of the device, while T represents the current junction temperature, obtained from a temperature sensor. This represents the thermal resistance of the device.

[0059] The frequency points are obtained from the Buck average small-signal model described above, and are derived from the compensation demand set. Sure; The design bus oscillation amplitude is determined based on the allowable fluctuation range of the bus voltage, and the effective period is determined based on the control cycle and the system dynamic response time. This refers to the steady-state or large-signal value of the bus voltage.

[0060] Step S500 outputs the stabilization capability certificate. .

[0061] The Buck function is prohibited if the certificate expires, the temperature exceeds the limit, or any quantity is negative. Step S500 is the key link between steps S400 and S600, and it compensates for the demand set. and Evaluation is conducted under the same frequency and bandwidth. The stabilization capability certificate output from step S500 is then processed in step S600 and compared with the compensation requirement set. Comparison: If Step S600 If insufficient, let And record If the certificate is invalid or the margin is negative, the S600 output capacity is insufficient, triggering the alternative Buck or load reduction.

[0062] Step S600: Project compensation requirements onto a single-machine implementable domain. This step simultaneously receives the compensation requirement package and capability certificate from the previous level, compares the minimum compensation conductance required by the system with the maximum realizable virtual conductance of a single selected converter; when the capability certificate is valid and the realizable conductance is not less than the required compensation conductance, the required compensation conductance is directly used as the execution conductance; when only degraded operation is allowed, the smaller value between the required compensation conductance and the realizable conductance is taken as the execution conductance, and the projection residual between the two is recorded; based on the execution conductance, center angular frequency, and compensation bandwidth, a second-order bandpass form of the desired virtual admittance is constructed, and an execution parameter package containing the desired virtual admittance, execution conductance, center angular frequency, bandwidth, projection residual, duty cycle limit, and validity period is output, providing direct instructions for subsequent additional duty cycle generation.

[0063] In a preferred embodiment of this invention, this step is implemented as follows: Compare and When identification is credible and season If only degraded operation is allowed, then let Record the projection residual Otherwise, the output capability is insufficient. It is expected that the virtual admittance will adopt a... Centered on, with a bandwidth of The second-order bandpass form: (17) In the formula, For actual compensation conductivity; For the desired virtual admittance bandwidth, according to Determined, usually taken ; From the set of compensation demands; The projection residual; the upper and lower limits of the duty cycle are determined according to the limitations of the switching devices; the validity period is aligned with the timestamp of the compensation demand set and the stabilization capability certificate. In this embodiment, the gain, center frequency and bandwidth of the expected virtual admittance are determined by the online results of S300 and S400, rather than being fixed offline.

[0064] The execution conductance output in this step and center angular frequency This will be used to replace step S100 Fixed parameters. Specifically, Replace the fixed gain coefficient in step S100. Replace the initial center frequency in step S100, so that the inner actuator follows the online identification result of the system-level closed loop. Step S600 outputs the set of execution parameters, represented as: The process is then moved to step S700. Step S600 is the demand and capacity matching node: the compensation demand set is provided. Stability Capability Certificate Provided And certificate validity. In the set of execution parameters. and It is the direct input for generating the additional duty cycle in step S700; This is the basis for step S800 to determine whether to update or switch; and Used for bandpass filter implementation in step S700. Therefore, the coherent parameters from steps S500 to S600 are: and The comparison results show that the continuity parameters from step S600 to step S700 are: , , , Duty cycle limit.

[0065] Step S700: Execution of Limit-Sense Inner Layer Virtual Admittance The technical content of this step can be summarized as follows: Receive the desired virtual admittance, execution conductance, center angular frequency, bandwidth, projection residual, and duty cycle limiting parameters given by the previous stage; convert the oscillating component of the common bus voltage into the desired compensation current through the desired virtual admittance; then calculate the additional duty cycle based on the small-signal propagation relationship between the selected converter input current and the duty cycle; add the additional duty cycle to the original control loop duty cycle and limit it to obtain the actual injected duty cycle command; continuously monitor duty cycle saturation, device temperature rise, output ripple, and projection residual during execution, forming an execution record containing the actual injected duty cycle, saturation count, duty cycle peak value, output ripple peak value, device temperature rise, projection residual, and execution timestamp, providing execution-side data for subsequent stability margin verification and backoff judgment.

[0066] In a preferred embodiment of this invention, this step is implemented as follows: Step S700 receives a set of execution parameters from step S600, which includes the desired virtual admittance. Actual compensated conductivity , center angular frequency ,bandwidth Projection residuals Duty cycle upper and lower limits and validity period. Step S700 is based on the bandpass filter structure established in S100, i.e. The transfer function form will include the execution conductance in the set of execution parameters. As gain, center angular frequency As the center frequency and bandwidth The actual expected virtual admittance is generated, thereby combining the basic template of step S100 with the online parameters of step S600.

[0067] Specifically, step S700 first converts the bus voltage oscillation component The desired virtual admittance is converted into the desired compensation current through online updates. Then divide by the input current and duty cycle transfer function of the selected Buck. To obtain the additional duty cycle : (18) In the formula, The small-signal transfer function from the selected Buck input current to the duty cycle is obtained from the Buck average small-signal model. Duty cycle compared to the original controller Projected onto after addition The duty cycle instruction after limiting is expressed as: (19) in, To project the variable within the parentheses to the minimum duty cycle and maximum duty cycle The projection operator between them is equivalent to the saturation function.

[0068] If the expected duty cycle is not limited If the amplitude exceeds the limit, a saturation event is recorded, and the projected residual caused by the amplitude limit is recorded. Store the execution record in the execution record set.

[0069] During execution, step S700 continuously accumulates the projection residual, temperature rise, and output ripple, forming an execution record set. The execution record set includes: the actual injected additional duty cycle, the number of saturation occurrences, the peak duty cycle value, the peak output ripple value, the device temperature rise, and the projection residual. And the execution timestamp. Therefore, the actual additional duty cycle injected in the specific implementation is related to... Correspondingly, its execution effect is determined by Record.

[0070] In this embodiment, during the execution of the additional duty cycle injection into the original control loop, S700 continuously monitors and accumulates the following five types of execution-side data: First, the projection residual. Whenever Exceeding When within a range, press Calculate the projection residual caused by this clipping event, and sum the absolute values ​​of the projection residuals of all clipping events during the execution period to obtain the cumulative projection residual; if no clipping event occurs during the execution period, the cumulative projection residual is zero.

[0071] Secondly, the number of times saturation occurs. Whenever the projection operator... Limit to or When the saturation occurrence count is incremented by one, the cumulative saturation occurrence count during the execution period reflects the frequency with which the additional duty cycle is limited.

[0072] Third, peak duty cycle. The actual injected duty cycle commands are continuously recorded during execution. The values ​​are calculated, and the maximum and minimum values ​​are taken as the duty cycle peak and duty cycle valley values, respectively. The duty cycle peak value is used to determine whether the switch is close to the duty cycle boundary during execution.

[0073] Fourth, peak output ripple and device temperature rise. During execution, the output voltage ripple and device temperature of the selected Buck-type load converter are continuously collected, and the maximum values ​​during execution are taken as the peak output ripple and device temperature rise, respectively. When the peak output ripple exceeds the allowable output ripple... Or the device temperature rise exceeds the maximum allowable junction temperature T max At that time, record the out-of-limit event and its corresponding timestamp.

[0074] Fifth, the actual additional duty cycle and execution timestamp of the injection. The actual additional duty cycle of the injection is continuously recorded during execution. It also timestamps the start and end times of the execution period.

[0075] The accumulated projected residuals, saturation occurrences, peak duty cycle, peak output ripple, device temperature rise, actual injected additional duty cycle, and execution timestamps are summarized to form an execution record set. This execution record set serves as the execution-side data source for subsequent review and rollback: In the subsequent review phase, if the positive real margin or generalized Nyquist minimum distance fails to meet the standard, the accumulated projected residuals and reasons for exceeding limits in the execution record set are used as the basis for updating the compensation requirement set, stability certificate, and execution parameter set; In the subsequent rollback phase, if duty cycle saturation, output ripple exceeding limits, or device temperature exceeding limits occur in the execution record set, the additional duty cycle is immediately cancelled, and a backup Buck-type load converter is switched or the load is reduced in stages according to load priority.

[0076] Therefore, in this embodiment, formula (8) gives Can be used as A simplified implementation within the Buck operating range, where the center frequency and gain are updated online by the execution parameter set. The execution record set output in step S700 enters step S800, serving as the sole execution-side data source for stability margin verification and backoff determination. If the execution record set records duty cycle saturation, ripple exceeding limits, or temperature rise exceeding limits, step S800 will update the compensation requirement set, stabilization capability certificate, and execution parameter set accordingly, or trigger a backoff. Therefore, the continuous parameters from step S600 to step S700 are: The limiting value, and the consecutive parameters from step S700 to step S800 are the projection residual, temperature rise, ripple, saturation event and execution timestamp in the execution record set.

[0077] Step S800: Post-implementation stability margin review and rollback The technical content of this step can be summarized as follows: After compensation is implemented, the aggregated load admittance of the common bus port is re-identified using the synchronization data re-acquired during the execution record, and the compensated positive real margin and generalized Nyquist minimum distance are calculated; the positive real margin is compared with the target positive real margin, the Nyquist minimum distance is compared with the minimum allowable distance, and the oscillation envelope attenuation rate is checked to see if it meets the requirements; if all are met, the current execution parameters are maintained; if any are not met, the compensation requirement package, capacity certificate, and execution parameter package are updated according to the projection residual and the reason for exceeding the limit in the execution record; if two consecutive reviews fail, the additional duty cycle is cancelled, and the switch is made to an alternative converter with a valid capacity certificate; if there is no alternative equipment, a tiered load reduction request is issued.

[0078] In a preferred embodiment of this invention, this step is implemented as follows: By repeating S300 using the synchronized data re-obtained from the execution record set, the compensated aggregate admittance is obtained. And calculate: (20) (twenty one) In the formula, For the aggregate admittance after compensation; For a positive real margin, it must be no less than ; The minimum generalized Nyquist distance must be no less than [a certain value]. ; The required stability margin is determined based on the stability margin requirement; the required oscillation envelope decay rate is determined based on the dynamic response requirement. For the angular frequency variable within the effective frequency band, To take the real part of the complex number, Ωosc is the effective frequency band, which comes from the pattern recognition result set.

[0079] In this embodiment, the updated objects are the compensation requirement set, the stabilization capability certificate, and the execution parameter set; the rollback strategy is to cancel the compensation and switch to the alternative Buck if two consecutive reviews fail, and to reduce the load in stages if there is no alternative.

[0080] when The parameters are maintained if the attenuation rate meets the requirements; otherwise, the compensation requirement set, stabilization capability certificate, and execution parameter set are updated with the projected residuals and out-of-limit reasons from the execution record set. Reversal occurs upon two consecutive failures. Switch to a backup Buck with a valid certificate; if no backup is available, issue a tiered load reduction request. Step S800 is the closed-loop endpoint and feedback start point, with inputs including execution records from the execution record set and compensation request sets. And re-identified If the target is met, the set of execution parameters remains unchanged; otherwise, the target is not met. The set of compensation requirements for exceeding limits, the stabilization capability certificate, and the set of execution parameters are updated to form a feedback chain from steps S800 to S400 to S500 to S600 to S700. Therefore, the consecutive parameters from steps S800 to S400 are the updated parameters. and re-identified The consecutive parameters from step S800 to step S600 are new. And limited amplitude status.

[0081] Step S900: Anomaly Handling and Graded Load Reduction The technical content of this step can be summarized as follows: When the capability certificate expires, the device temperature exceeds the limit, the duty cycle saturates, the output ripple exceeds the limit, or the stability margin verification fails continuously, the current additional duty cycle is immediately canceled, and the stabilization control of the selected converter is stopped; the list of alternative converters is queried, and a device with a valid capability certificate and sufficient current margin is selected to continue to undertake the stabilization task; if there is no alternative device, the power of non-critical loads is reduced step by step from low to high according to load priority, with a certain percentage reduction at each level, until the system returns to stability; at the same time, abnormal events, timestamps, projected residuals, temperature rise, and ripple data are recorded for subsequent analysis.

[0082] In a preferred embodiment of this invention, this step is implemented as follows: Cancel immediately in case of an anomaly ; Query the alternative Buck list and select equipment with valid certificates and sufficient margin; if no alternatives are available, issue tiered load reduction requests according to priority to reduce non-critical load power; record abnormal events, timestamps, projected residuals, temperature rise, and ripple data. Certificate expiration refers to exceeding the validity period of the stability capability certificate; temperature exceeding limits refers to exceeding the maximum junction temperature; duty cycle saturation refers to reaching... or Output ripple exceeds the limit. The load is reduced in stages according to the load priority, with each stage reducing the power by 5% to 10% until the system returns to stability.

[0083] To verify the validity of this application, the following experimental verification is provided: This embodiment is a preferred example of the above generalized implementation method. Specific values ​​are given for the parameters in each step, and experimental verification was performed based on this specific embodiment. The system parameters are as follows: digital controller sampling frequency. Sampling period Bus voltage Buck output voltage Rated power The dominant oscillation frequency of the system In this experiment, converter #3 was the selected Buck-type load converter, and converter #4 was the unmodified load converter.

[0084] Step S100: Inner Layer Virtual Impedance Control Configuration In this embodiment, the selected Buck-type load converter is a Buck-type load converter selected to undertake the stabilization task in a multi-converter single-bus DC power distribution system, which is already configured with an outer voltage loop and an inner current loop. The original voltage loop collects the Buck output voltage. The voltage reference value is compared with the current reference value, and a current inner loop reference value is generated by the voltage regulator. The original current loop collects the inductor current and compares it with the current reference value. The current regulator generates a duty cycle signal, which is then modulated by PWM to drive the switching transistor. While maintaining the original voltage and current loop structures unchanged, the virtual impedance gain constant is... Selected Buck load power Bus voltage Initial value of dominant oscillation frequency This value was updated by S300, quality factor. Effective bandwidth Bus voltage sampling coefficient Isolation transformer turns ratio steady-state duty cycle Sampling period The parallel virtual impedance is designed according to equation (1), and the controller is implemented according to equation (3). After discretization by bilinear transformation, the result is obtained. The result obtained in this step As the basic transfer function template for the inner actuator, its center frequency The gain will be updated online via a subsequent system-level closed loop.

[0085] Specifically, the dominant oscillation angular frequency identified by the S300 This will return to the previous step, replacing the initial center frequency; the execution conductance of the S600 output. This will replace the fixed gain in this step. When the closed loop of S200 to S900 is in operation, S700 will generate the actual additional duty cycle based on the bandpass filter structure established in this step, substituting the online parameters given by S600. Between S100 and S200: S200 only applies a perturbation during the identification phase. No additional duty cycle is required. The dominant oscillation frequency identified by S300 Returning to this step updates the center frequency; the S600 outputs the execution conductance. Used to adjust the gain of this step online.

[0086] In this embodiment, the original control loop refers to the original dual-loop control structure and its PWM modulation stage of the selected Buck-type load converter, including the original voltage loop, the original current loop and the PWM modulator; the original voltage loop and the original current loop constitute the original dual-loop control structure.

[0087] Step S200: Bounded identification of disturbances and generation of synchronization data packets In this embodiment, the steady-state duty cycle steady-state inductor current The disturbance amplitude is taken as follows: Disturbance frequency band coverage Synchronously acquire bus voltage Inductor current Steady-state duty cycle The output voltage ripple, device temperature, and limiting status are used to form a synchronization identification dataset. The port identification current is estimated using equation (12). The synchronization identification dataset output by S200 is fed into S300 and used to calculate the equivalent drive point impedance of the system. and aggregated load admittance The sole source of data. The continuous parameters from S200 to S300 are the small-signal bus voltage. Port identification current Disturbance frequency And timestamp.

[0088] Step S300: Identification of system equivalent drive point impedance and aggregated load admittance In this embodiment, the output impedance of the source converter Confidentiality threshold of coherence coefficient Dominant oscillation angular frequency The synchronous identification dataset is demodulated using Discrete Fourier or Phase-Locked Demodulation, calculated according to Equation (13). The output pattern identification result set is {dominant oscillation angular frequency}. Aggregate load admittance Coherence coefficient Effective frequency band Timestamp. Only coherence coefficient. Only when the pattern recognition result set is in the specified condition will it proceed to S400; otherwise, it will return to S200. The coherent parameter from S300 to S400 is the dominant oscillation angular frequency. Aggregate load admittance Coherence coefficient Effective frequency band .

[0089] Step S400: Solving for system damping deficit and required compensation In this embodiment, the target positive real margin Robust margin Compensation effective bandwidth =35.3Hz. Calculate the minimum compensation conductance frequency by frequency according to equation (14) and take the maximum value to obtain the required minimum compensation conductance. The set of output compensation requirements is {minimum required compensation conductance}. Dominant oscillation angular frequency Compensation effective bandwidth Target real margin Robust margin Coherence coefficient The timestamp is sent to S500 and S600. The coherent parameters from S400 to S500 are the required minimum compensation conductance. Dominant oscillation angular frequency Compensation effective bandwidth Target real margin Robust margin Coherence coefficient Timestamp.

[0090] Step S500: Generation of a single Buck stabilization certificate In this embodiment, the input current margin Duty cycle margin Allow output ripple Heat loss margin In this embodiment, the maximum junction temperature is used. Current temperature Thermal resistance have to Design bus oscillation amplitude Validity period Calculate the maximum compensation current according to formula (15). Then, the maximum achievable virtual conductance can be calculated using equation (16). Output stabilization capability certificate, represented as: {Maximum achievable virtual conductance} Input current margin Duty cycle margin Allow output ripple Heat loss margin Validity period Equipment identification. The stability certificate output by S500 is entered into S600, along with the minimum required compensation conductance in the compensation requirement set. Comparison. The coherent chain from S400 to S500 to S600 is: the set of compensation demands provides demand. and evaluation frequency The Stability Certification provides capabilities S600 completes the matching.

[0091] Step S600: Project compensation requirements onto a single-machine implementable domain. In this embodiment, the desired virtual admittance bandwidth is... Center angular frequency Duty cycle lower limit Duty cycle limit Validity period Compare the maximum achievable virtual conductance. With the required minimum compensation conductance :like , to execute conduction Otherwise, Record the projection residual The desired virtual admittance is constructed according to equation (17). The execution conductance output in this step is... and center angular frequency Will be used to replace S100 Fixed parameters. The set of output execution parameters is: The input is sent to S700. The coherent parameters from S600 to S700 are the desired virtual admittance. , Execution conductance , center angular frequency ,bandwidth Duty cycle lower limit Duty cycle limit .

[0092] Step S700: Execution of Limit-Sense Inner Layer Virtual Admittance In this embodiment, after receiving the set of execution parameters, S700, based on the bandpass filter structure established in S100, filters the execution conductance in the set of execution parameters. As gain, center angular frequency As the center frequency and bandwidth As bandwidth, generate the actual desired virtual admittance. Incorporate the bus voltage oscillation component. This desired virtual admittance is converted into the desired compensation current. Divide by Obtain additional duty cycle That is, calculate according to formula (18). Then calculate according to formula (19). The lower limit of duty cycle =0.05, upper limit of duty cycle =0.95. If saturation occurs, record the saturation event and the projected residual. During execution, the accumulated projected residual, temperature rise, and output ripple form an execution record set. In this embodiment, the execution record set includes the actual injected additional duty cycle, saturation count, peak duty cycle, peak output ripple, device temperature rise, and projected residual. and execution timestamp. The execution record set enters S800. The coherent parameters from S600 to S700 are the expected virtual admittance. , Execution conductance , center angular frequency ,bandwidth Duty cycle lower limit Duty cycle limit The coherent parameters from S700 to S800 are the projected residual, temperature rise, ripple, saturation event, and execution timestamp in the execution record set.

[0093] Step S800: Post-implementation stability margin review and rollback In this embodiment, the generalized Nyquist minimum distance threshold The reduction rate requires the peak ratio of adjacent cycles to be less than 0.9. Using the synchronized data re-acquired during the execution record set, repeat S300 to obtain the compensated aggregate admittance. Calculate the positive real margin according to equations (20) and (21). Minimum distance to Nyquist .when , And the parameters are maintained if the attenuation rate meets the requirements; otherwise, the projected residuals in the execution record set are used. Update the compensation requirement set, stabilization capability certificate, and execution parameter set according to the reason for exceeding the limit. Cancel the additional duty cycle upon two consecutive failures. Switch to the alternative Buck; if no alternative is available, reduce load in stages. The coherent parameters from S800 to S400 are the updated target positive real margin. Robust margin and re-identified aggregate load admittance The coherent parameters from S800 to S600 are the new projected residuals. And limited amplitude status.

[0094] Step S900: Anomaly Handling and Graded Load Reduction In this embodiment, "certificate expired" refers to a certificate that has exceeded its validity period. Temperature exceeding the limit refers to exceeding the maximum junction temperature. Duty cycle saturation refers to reaching the lower limit of the duty cycle. Or duty cycle upper limit Output ripple exceeding the limit indicates that the output ripple exceeds the allowable limit. The load is reduced in stages according to the load priority, with each stage reducing the power by 5% to 10% until the system returns to stability.

[0095] To verify the technical effectiveness of this embodiment, an experiment was conducted based on the specific parameters described above. The experiment will... Figure 2 The parallel virtual impedance control strategy shown is added to converter #3. The virtual impedance controller is designed according to equation (8), and the isolation transformer turns ratio is... Virtual impedance gain constant This is to completely suppress the negative damping characteristics of the converter load at the corresponding power.

[0096] Figure 3 The Bode plots for the equivalent input impedance of converters #3 and #4 before and after the implementation of the stabilization control in this invention comprise four sub-plots. The top and bottom left plots show the amplitude-frequency and phase-frequency characteristics of the equivalent input impedance of converter #3 before and after the implementation of the stabilization control, respectively; the top and bottom right plots show the amplitude-frequency and phase-frequency characteristics of the equivalent input impedance of converter #4 before and after the implementation of the stabilization control, respectively. It should be noted that, to more clearly compare the changes in equivalent input impedance before and after the implementation of the control strategy, the input capacitance of the two converters is not considered in the plots. In this embodiment, the "equivalent input impedance of the ports where #3 and #4 are located" refers to the apparent drive point impedance on the common bus side, including the closed-loop coupling relationship of the interconnected system; since the internal controller of #4 is not modified, it is not included in the calculation. Figure 3 The results are interpreted as a change in the open-loop input impedance of converter #4. The amplitude-frequency characteristics show that before the control strategy was implemented, both converters #3 and #4 exhibited significant impedance peaks and negative damping ranges around 353Hz; after the control strategy was implemented, these peaks were significantly suppressed, and the impedance amplitude became more gradual across the entire frequency band. The phase-frequency characteristics show that before the control strategy was implemented, converters #3 and #4 exhibited phases close to or exceeding ±180° around 353Hz, displaying significant negative damping characteristics; after the control strategy was implemented, the phase in this frequency band was pulled back to within ±90°, exhibiting positive damping characteristics. These results indicate that after the control strategy was implemented, the negative damping characteristics of the equivalent input impedance of converters #3 and #4 in the 353Hz and surrounding frequency bands were significantly suppressed.

[0097] Figure 4 Adding forward and backward transfer functions to the stabilization control of this invention The Bode plot consists of four subplots. The top left and top right subplots show the Bode plots before and after the addition of stabilizing control under condition 2, while the bottom left and bottom right subplots show the Bode plots before and after the addition of stabilizing control under condition 3. Each subplot contains two curves: amplitude-frequency response and phase-frequency response. The amplitude-frequency response shows that before the control strategy was implemented, the system exhibited significant resonance peaks in both condition 2 and condition 3; after the control strategy was implemented, these resonance peaks were significantly reduced, and the system's amplitude-frequency response became flatter. The phase-frequency response shows that before the control strategy was implemented, the system had insufficient phase margin in both condition 2 and condition 3; after the control strategy was implemented, the phase margin was significantly improved. Figure 4 It can be seen that after the control strategy was added, the system met the requirements in both operating conditions 2 and 3. ,Right now The Nyquist curves do not encircle the origin. Therefore, it can be inferred that the system can operate stably after adopting the proposed system-level power adaptive stabilization control. Figure 5 Adding forward and backward transfer functions to the stabilization control of this invention and The Bode plot. This plot is used to illustrate the effect of virtual impedance control on the voltage loop transfer function, further verifying the positive resistance compensation effect. Figure 5 It can be seen that after the addition of stabilization control, the resonance peak of the voltage loop transfer function near 353Hz is significantly suppressed, and the damping characteristics of the system are significantly improved.

[0098] In addition, this embodiment provides a control system obtained by a Buck system-level stabilization control method based on capability certificates. The system includes: The inner virtual impedance control module is used to introduce the bus voltage sampling signal into the duty cycle reference value through the virtual impedance controller on the basis of the original control loop of the selected Buck-type load converter, so that the selected Buck-type load converter is equivalent to a virtual impedance with positive resistance only near the dominant oscillation frequency in parallel at the common bus port, forming the basic transfer function template of the inner actuator. The bounded identification perturbation module is used to apply amplitude-limited identification perturbations within the candidate oscillation frequency band of the bus, synchronously collect bus voltage, inductor current, steady-state duty cycle, ripple, temperature and limiting status, and estimate port identification current to form a synchronous identification dataset. The aggregation admittance and deficit solution module is used to perform frequency domain demodulation based on the synchronous identification dataset and remove frequency points with insufficient signal-to-noise ratio to obtain the dominant oscillation angular frequency, coherence coefficient and effective frequency band corresponding to the retained frequency points. Within the effective frequency band, the damping deficit is calculated frequency by frequency and the maximum value is taken. After taking into account the robust margin, the minimum compensation conductance is obtained to form the compensation demand set. The capability certificate and projection execution module is used to calculate the maximum realizable virtual conductance based on the current and duty cycle of the selected Buck-type load converter at the frequency and bandwidth specified in the compensation requirement set, compare the required minimum compensation conductance with the maximum realizable virtual conductance, determine the execution conductance and record the projection residual, and then construct a bandpass-type expected virtual admittance to form an execution parameter set; generate an additional duty cycle based on the basic transfer function template and the execution parameter set, superimpose it on the original control loop and limit it to form an execution record set; The margin verification module is used to re-identify the aggregated load admittance after compensation, verify the positive real margin and the generalized Nyquist minimum distance. If they meet the criteria, the current execution parameter set is maintained and the operation continues. If they do not meet the criteria, the compensation requirement set, stability capability certificate and execution parameter set are updated. The exception handling module is used to cancel the additional duty cycle and switch to a backup Buck-type load converter with a valid stability capability certificate when two consecutive verifications fail, or when the execution record set shows duty cycle saturation, output ripple exceeding the limit, device temperature exceeding the limit, or the stability capability certificate expires or the margin is negative; if there is no backup device, the load is reduced in stages according to load priority.

[0099] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0100] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0101] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0102] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in a capability-certified Buck system-level stabilization control method.

[0103] The computer device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device may include, but is not limited to, a processor and memory.

[0104] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0105] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0106] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0107] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0108] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A Buck system-level stability control method based on capability certificates, characterized in that, The method includes: Based on the original control loop of the selected Buck-type load converter, the bus voltage sampling signal is introduced into the duty cycle reference value through the virtual impedance controller, so that the selected Buck-type load converter is equivalent to a virtual impedance that is positively resistive only near the dominant oscillation frequency in parallel at the common bus port, forming the basic transfer function template of the inner actuator. An amplitude-limited identification perturbation is applied within the candidate oscillation frequency band of the bus, and the bus voltage, inductor current, steady-state duty cycle, ripple, temperature and limiting status are collected simultaneously. The port identification current is estimated to form a synchronous identification dataset. Based on the synchronous identification dataset, frequency domain demodulation is performed and frequency points with insufficient signal-to-noise ratio are removed to obtain the dominant oscillation angular frequency, coherence coefficient and effective frequency band corresponding to the retained frequency points. Within the effective frequency band, the damping deficit is calculated frequency by frequency and the maximum value is taken. After taking into account the robust margin, the minimum compensation conductance is obtained to form the compensation demand set. At the frequency and bandwidth specified in the compensation requirement set, the maximum achievable virtual conductance is calculated based on the current and duty cycle of the selected Buck-type load converter, forming a stability capability certificate with a validity period. The required minimum compensation conductance is compared with the maximum achievable virtual conductance to determine the execution conductance and record the projection residual. Then, a bandpass-type expected virtual admittance is constructed to form an execution parameter set. An additional duty cycle is generated based on the basic transfer function template and the execution parameter set, superimposed on the original control loop and limited to form an execution record set. Re-identify the aggregated load admittance after compensation, and verify the positive real margin and the generalized Nyquist minimum distance. If they meet the requirements, maintain the current execution parameter set and continue running. If they do not meet the requirements, update the compensation requirement set, stability capability certificate and execution parameter set. If two consecutive verifications fail, or if the execution record set shows duty cycle saturation, output ripple exceeding limits, device temperature exceeding limits, or the stability certificate expires or the margin is negative, the additional duty cycle is cancelled, and the system switches to a backup Buck-type load converter with a valid stability certificate; if there is no backup device, the load is reduced in stages according to load priority.

2. The Buck system-level stability control method based on capability certificates according to claim 1, characterized in that, The basic transfer function template for forming the inner actuator includes: Define the open-loop transfer function from the duty cycle small signal to the Buck input current small signal when the small signal disturbance on the output side is zero, and define the open-loop transfer function from the duty cycle small signal to the output voltage small signal when the small signal disturbance on the input side is zero; The input port compensation current, formed by the open-loop transfer function from the duty cycle output of the virtual impedance controller to the input current, is equal to the product of the desired virtual admittance and the bus voltage disturbance. The transfer function of the virtual impedance controller can be solved from the port current balance relationship. Within the dominant oscillation frequency and its vicinity, the transfer function of the virtual impedance controller is simplified to a second-order bandpass form with the dominant oscillation frequency as the center frequency. The second-order bandpass form is discretized by a bilinear transform to obtain the basic transfer function template. The center frequency of the basic transfer function template is updated by the subsequently identified dominant oscillation angular frequency, and the gain is adjusted online by the subsequently output execution conductance.

3. The Buck system-level stability control method based on capability certificates according to claim 2, characterized in that, The synchronous acquisition of bus voltage, inductor current, steady-state duty cycle, ripple, temperature, and limiting status, along with estimation of port identification current, forms a synchronous identification dataset, including: Within the candidate oscillation frequency band, a disturbance with limited amplitude is superimposed on the duty cycle reference value of the selected Buck-type load converter. The amplitude of the disturbance is limited by the upper limit of output ripple, duty cycle margin, and current margin. Simultaneously acquire bus voltage, inductor current, steady-state duty cycle, output voltage ripple, device temperature, and limiting status; based on the first-order linearization relationship of the Buck input current near the operating point, estimate the port identification current flowing into the selected Buck-type load converter, expressed as: ; In the formula, For steady-state duty cycle, For steady-state inductor current, The small-signal component of the inductor current is obtained from existing sampling. For duty cycle perturbation small signal, The imaginary unit, For the first m The angular frequency corresponding to each disturbance frequency point; The bus voltage, inductor current, steady-state duty cycle, port identification current, ripple, temperature, limiting status, and timestamp are summarized to form the synchronization identification dataset.

4. The Buck system-level stability control method based on capability certificates according to claim 3, characterized in that, The process of obtaining the dominant oscillation angular frequency, coherence coefficient, and effective frequency band includes: At each known disturbance frequency point applied, the bus voltage time-domain sequence and the port identification current time-domain sequence in the synchronous identification dataset are subjected to discrete Fourier transform or phase-locked demodulation respectively to obtain the small-signal frequency domain component of the bus voltage and the small-signal frequency domain component of the port identification current at that frequency point. The coherence coefficient of each frequency point is calculated based on the small-signal frequency domain components of the bus voltage and the port identification current. The coherence coefficient characterizes the degree of linear correlation between the bus voltage and the port identification current at the same frequency point. When the coherence coefficient is less than the confidence threshold, the frequency point is eliminated. For the remaining frequency points, the positive direction of the port identification current is defined as flowing from the bus into the selected Buck-type load converter. The negative value of the ratio of the small-signal frequency domain component of the bus voltage to the small-signal frequency domain component of the port identification current is used as the system equivalent drive point impedance of the frequency point. The source converter output impedance is subtracted from the system equivalent drive point impedance to obtain the aggregated load admittance of the frequency point. The dominant oscillation angular frequency of the retained frequency point is determined, and the coherence coefficient corresponding to the frequency point is used as the coherence coefficient of mode identification. The range of continuous frequency points centered on the dominant oscillation angular frequency and with the real part of the aggregated load admittance being negative is taken as the effective frequency band.

5. The Buck system-level stability control method based on capability certificates according to claim 4, characterized in that, Within the effective frequency band, the damping deficit is calculated frequency-by-frequency and the maximum value is taken. After factoring in the robustness margin, the required minimum compensation conductance is obtained, forming a compensation requirement set, including: Within the effective frequency band, calculate the real part of the sum of the source port admittance and the aggregate load admittance frequency by frequency. Using the condition that this real part is not lower than the target positive real margin, calculate the damping deficit frequency by frequency. Take the maximum value among all frequency points and include the robust margin to obtain the required minimum compensation conductance, expressed as: ; In the formula, For the target positive margin; For robust margin, The output impedance of the source converter, For aggregate load admittance, For effective frequency band, For robust margin; The required minimum compensation conductance, dominant oscillation angular frequency, effective compensation bandwidth, target positive real margin, robust margin, coherence coefficient, and timestamp are summarized to form the compensation requirement set.

6. The Buck system-level stability control method based on capability certificates according to claim 5, characterized in that, At the frequency and bandwidth specified in the compensation demand set, the maximum achievable virtual conductance is calculated based on the current and duty cycle of the selected Buck-type load converter, including: Using the duty cycle to input current transfer function and duty cycle to output voltage transfer function in the average small-signal model of the selected Buck-type load converter, the above margins are converted into the maximum allowable compensation current. Take the minimum value among the maximum compensation currents and convert it to the maximum achievable virtual conductance based on the design bus oscillation amplitude, that is: ; Then, according to the design bus oscillation amplitude Calculated maximum achievable virtual conductance: ; in, This is the input current margin. The rated current of the input-side devices of the selected Buck-type load converter. The current input current, The duty cycle margin is determined by the difference between the current duty cycle and the upper and lower limits. To allow output ripple, This is a margin for heat loss. These are the transfer functions from input current to duty cycle and from output voltage to duty cycle, respectively. To design the bus oscillation amplitude, To allow output ripple, This represents the steady-state value of the bus voltage.

7. The Buck system-level stability control method based on capability certificates according to claim 6, characterized in that, The process involves comparing the required minimum compensation conductance with the maximum realizable virtual conductance to determine the execution conductance and record the projection residual. Based on this, a bandpass-type desired virtual admittance is constructed, forming an execution parameter set, including: When the stabilization capability certificate is valid, and the maximum achievable virtual conductance is... Not less than the required minimum compensation conductance season Otherwise, order the execution. The difference between the two is denoted as the projection residual. ; Construct the bandpass-type desired virtual admittance, expressed as: ; in, To compensate for the actual conductance, For the desired virtual admittance bandwidth, From the set of compensation demands, This represents the projection residual.

8. The Buck system-level stability control method based on capability certificates according to claim 7, characterized in that, The additional duty cycle generated based on the basic transfer function template and the execution parameter set is superimposed on the original control loop and limited to form an execution record set, including: The bus voltage oscillation component is converted into the desired compensation current via the actual desired virtual admittance, and then divided by the transfer function of the input current to the duty cycle of the selected Buck-type load converter to obtain the additional duty cycle, expressed as: ; in, The small-signal transfer function from the input current to the duty cycle of the selected Buck-type load converter; The additional duty cycle is added to the original control loop's duty cycle and projected onto the upper and lower limits of the duty cycle to obtain the actual injected duty cycle command, expressed as: ; in, To project the variable within the parentheses to the minimum duty cycle and maximum duty cycle Projection operator between, This represents the oscillating component of the bus voltage. This indicates the duty cycle of the original control loop output, i.e., the duty cycle signal of the original voltage loop and current loop of the selected Buck-type load converter when there is no additional virtual impedance control. During execution, the projected residual, temperature rise, and output ripple are continuously accumulated. The actual injected additional duty cycle, number of saturation occurrences, peak duty cycle, peak output ripple, device temperature rise, projected residual, and execution timestamp are summarized to form the execution record set. The execution record set serves as the execution-side data source for subsequent review and rollback.

9. The Buck system-level stability control method based on capability certificates according to claim 8, characterized in that, The process of re-identifying the compensated aggregate load admittance and verifying the positive real margin and generalized Nyquist minimum distance includes: The compensated aggregate load admittance is recalculated using the synchronization data during the execution record set; Within the effective frequency band, calculate the minimum real part of the sum of the source port admittance and the compensated aggregated load admittance, using this as the positive real margin. Also, calculate the minimum amplitude of the preset Nyquist distance function within the effective frequency band, using this as the generalized Nyquist minimum distance, i.e.: ; In the formula, For the aggregate admittance after compensation; For positive real margin, For generalized Nyquist minimum distance, For the angular frequency variable within the effective frequency band, To take the real part of the complex number, This is the effective frequency band.

10. A control system obtained based on the Buck system-level stabilization control method based on capability certificates as described in claim 1, characterized in that, include: The inner virtual impedance control module is used to introduce the bus voltage sampling signal into the duty cycle reference value through the virtual impedance controller on the basis of the original control loop of the selected Buck-type load converter, so that the selected Buck-type load converter is equivalent to a virtual impedance with positive resistance only near the dominant oscillation frequency in parallel at the common bus port, forming the basic transfer function template of the inner actuator. The bounded identification perturbation module is used to apply amplitude-limited identification perturbations within the candidate oscillation frequency band of the bus, synchronously collect bus voltage, inductor current, steady-state duty cycle, ripple, temperature and limiting status, and estimate port identification current to form a synchronous identification dataset. The aggregation admittance and deficit solution module is used to perform frequency domain demodulation based on the synchronous identification dataset and remove frequency points with insufficient signal-to-noise ratio to obtain the dominant oscillation angular frequency, coherence coefficient and effective frequency band corresponding to the retained frequency points. Within the effective frequency band, the damping deficit is calculated frequency by frequency and the maximum value is taken. After taking into account the robust margin, the minimum compensation conductance is obtained to form the compensation demand set. The capability certificate and projection execution module is used to calculate the maximum realizable virtual conductance based on the current and duty cycle of the selected Buck-type load converter at the frequency and bandwidth specified in the compensation requirement set, compare the required minimum compensation conductance with the maximum realizable virtual conductance, determine the execution conductance and record the projection residual, and then construct a bandpass-type expected virtual admittance to form an execution parameter set; generate an additional duty cycle based on the basic transfer function template and the execution parameter set, superimpose it on the original control loop and limit it to form an execution record set; The margin verification module is used to re-identify the aggregated load admittance after compensation, verify the positive real margin and the generalized Nyquist minimum distance. If they meet the criteria, the current execution parameter set is maintained and the operation continues. If they do not meet the criteria, the compensation requirement set, stability capability certificate and execution parameter set are updated. An exception handling module is used to cancel the additional duty cycle and switch to an alternative Buck-type load converter with a valid stability capability certificate when two consecutive verifications fail, or when the execution record set shows duty cycle saturation, output ripple exceeding the limit, device temperature exceeding the limit, or the stability capability certificate expires or the margin is negative. If there are no alternative devices, load will be reduced in stages according to load priority.

11. A computer device, characterized in that, include: One or more processors; The processor is used to store one or more programs; When the one or more programs are executed by the one or more processors, the steps of the Buck system-level stability control method based on capability certificates as described in any one of claims 1 to 9 are implemented.

12. A computer-readable storage medium, characterized in that, It contains a computer program, which, when executed, implements the steps of the Buck system-level stability control method based on capability certificates as described in any one of claims 1 to 9.

Citation Information

Patent Citations

  • Stability control method for multiport converters based on virtual impedance

    CN115714371B