Power equipment voltage and current stabilization and surge suppression cooperative control method

Abstract

The invention relates to the technical field of power electronic control, and discloses a voltage and current stabilization and surge suppression cooperative control method for power equipment, and the method comprises the following steps: dynamically segmenting an output current frequency band, and remodeling the output impedance characteristics of a system; predicting and controlling a hierarchical optimization control target based on a multi-time scale model; dynamically switching an energy feedback path; identifying aging parameters of the device on line and adaptively adjusting a control strategy; global stability verification and fault-tolerant protection: verifying the stability of the control system based on the Lyapunov stability theory, and triggering a protection mechanism when the magnetic flux is saturated or the aging of the device exceeds the limit. By means of dynamic impedance frequency domain segmentation, the impedance characteristic of the system is divided into a low-frequency voltage stabilization area and a high-frequency surge suppression area according to frequency bands, the contradiction that in a traditional method, it is difficult to consider steady-state precision and transient surge suppression at the same time due to the single impedance characteristic is solved, and collaborative optimization of voltage stabilization control and surge energy blocking is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power electronic control, and specifically provides a collaborative control method for voltage and current stabilization and surge suppression of power equipment. Background Art

[0002] In the fields of voltage stabilization and surge suppression of power equipment, traditional methods usually adopt linear control strategies with fixed impedance characteristics or feedback regulation mechanisms on a single time scale. When facing broadband disturbances, due to the inability of the impedance characteristics to be dynamically adjusted according to the operating conditions, there are inherent contradictions between the steady-state voltage accuracy and the transient surge suppression performance: the voltage stabilization requirement in the low-frequency band requires reducing the output impedance to reduce the voltage deviation, while the surge suppression in the high-frequency band requires increasing the impedance to block the transient energy transfer. The fixed impedance design is difficult to meet these two conflicting requirements simultaneously.

[0003] In addition, the existing technologies generally use static thresholds to judge surge events and trigger the action of protection devices, which are prone to false triggering or response delays under complex dynamic operating conditions, causing frequent actions of key devices such as MOVs and accelerating their aging. At the control strategy level, the conventional single-layer optimization architecture is difficult to coordinate multi-time scale control objectives (such as millisecond-level voltage tracking and microsecond-level surge suppression), resulting in dynamic mismatch between surge energy distribution and voltage stabilization control. Especially when there are sudden changes in non-linear loads or device parameter drifts, the system is prone to cumulative steady-state deviation and even instability.

[0004] The existing solutions lack the ability to online identify device aging parameters and a global stability guarantee mechanism, cannot real-time correct control parameters to adapt to the device degradation state, and may cause cascading failures due to hardware saturation or control mismatch under extreme surge impacts. The above defects seriously restrict the long-term reliable operation of power equipment in industrial complex environments. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technologies, the present invention provides a collaborative control method for voltage and current stabilization and surge suppression of power equipment, which solves the problems of mutual restriction between steady-state accuracy and transient suppression ability, device life loss, and system instability in traditional solutions.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A collaborative control method for voltage and current stabilization and surge suppression of power equipment, including: Dynamically divide the output current frequency band and reshape the system output impedance characteristics. By real-time analyzing the load current spectrum, set a low impedance in the low-frequency band to suppress voltage fluctuations, and set a high impedance in the high-frequency band to absorb surge energy; Based on the multi-time scale model predictive control, hierarchically optimize the control objectives, optimize the voltage tracking accuracy on the slow time scale, optimize the surge energy distribution on the fast time scale, and coordinate the conflicting requirements of voltage stabilization and surge suppression through the coupled objective function; Dynamically switch the energy feedback path. By monitoring the bus surge voltage in real time, trigger the boost or buck mode of the bidirectional DC / DC converter to feedback the surge energy to the supercapacitor for storage or release it to the regulated bus. Online identify the device aging parameters and adaptively adjust the control strategy. Update the MOV threshold drift amount through a recursive algorithm and dynamically correct the priority weight of surge suppression in the MPC objective function. Global stability verification and fault tolerance protection. Based on the Lyapunov stability theory, verify the stability of the control system and trigger the protection mechanism when the magnetic flux saturation or device aging exceeds the limit.

[0007] Preferably, the dynamic frequency band division and impedance reshaping include: performing fast Fourier transform analysis on the output current, dividing the low-frequency band ( ) and the high-frequency band ( ; Set low impedance characteristics in the low-frequency band to suppress voltage fluctuations, and set high impedance characteristics in the high-frequency band to absorb surge energy; dynamically match the target impedance frequency domain characteristics by injecting specific harmonics of the PWM modulation wave.

[0008] Preferably, the cut-off frequencies of the low-frequency band and the high-frequency band are adaptively adjusted according to the spectral energy distribution of the load current, satisfying , where is the fundamental frequency, is the coefficient corresponding to the energy ratio threshold.

[0009] Preferably, the multi-time scale MPC hierarchical optimization includes: Construct a state space model in the slow time layer, and the optimization objective is to minimize the voltage tracking deviation; Construct a surge energy distribution model in the fast time layer, and the optimization objective is to minimize the number of MOV operations; Coordinate the control instructions of the slow time layer and the fast time layer through the coupled objective function.

[0010] Preferably, the optimization period of the slow time layer is 1 ms, the optimization period of the fast time layer is 10 μs, and the weight coefficients of the objective functions of the two layers are dynamically adjusted according to the remaining capacity of the supercapacitor.

[0011] Preferably, the dynamic switching of the energy feedback path includes: When it is detected that the bus surge voltage exceeds the threshold, trigger the boost mode of the bidirectional DC / DC converter to feedback the surge energy to the supercapacitor; When the surge voltage is lower than the threshold, switch to the buck mode to release the supercapacitor energy to the regulated bus; The switching logic of the boost and buck modes is linked to the surge energy distribution result of the MPC.

[0012] Preferably, the duty cycle of the boost mode and the duty cycle of the buck mode are calculated in real time according to the input and output voltages and are restricted by the magnetic flux saturation constraint of the magnetically integrated inductor.

[0013] Preferably, the online identification of device aging parameters includes: Updating the threshold voltage drift of the MOV using the recursive least squares method ; According to dynamically modifying the weight coefficient of the surge suppression term in the MPC objective function to satisfy ).

[0014] Preferably, the forgetting factor of the RLS algorithm exponentially decays according to the system operation time, and the parameter update step size is constrained to , avoiding mutations.

[0015] Preferably, the global stability check includes: Constructing a Lyapunov function to verify the global asymptotic stability of the control system; When the magnetic flux exceeds the magnetic core saturation threshold, restricting the PWM duty cycle proportionally; When the MOV aging degree exceeds 20%, switching to the standby surge absorption branch and giving an alarm.

[0016] The present invention provides a method for coordinated control of voltage and current stabilization and surge suppression of power equipment. It has the following beneficial effects: 1. Through dynamic impedance frequency domain segmentation, the present invention divides the system impedance characteristics into a low-frequency voltage stabilization region and a high-frequency surge suppression region according to frequency bands, solving the contradiction that a single impedance characteristic in the traditional method is difficult to balance steady-state accuracy and transient surge suppression, and realizing the coordinated optimization of voltage stabilization control and surge energy blocking.

[0017] 2. Based on the hierarchical optimization architecture of multi-time scale model predictive control, the present invention solves the multi-objective conflict problem of voltage tracking, surge distribution, and device life management through the independent optimization and dynamic weight adjustment of steady-state control in the slow time layer and surge suppression in the fast time layer, significantly improving the control adaptability under complex working conditions.

[0018] 3. The present invention adopts a dynamic switching boost / buck mode combined with a magnetic flux saturation protection mechanism to ensure the efficient storage and release of surge energy, while avoiding hardware failure caused by inductor magnetic core saturation, improving the energy feedback efficiency and system hardware safety. Description of the Drawings

[0019] Figure 1 is a flowchart of the present invention. Detailed Implementation Modes

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Please refer to the attached Figure 1 , the embodiments of the present invention provide a coordinated control method for voltage stabilization, current stabilization and surge suppression of power equipment, including: Step 1: Dynamic impedance frequency domain segmentation and reshaping In this embodiment, the dynamic impedance frequency domain segmentation and reshaping technology realizes the optimization of the frequency domain characteristics of the output impedance of power equipment through the coordinated control of real-time spectrum analysis, frequency band dynamic division, target impedance characteristic design and modulation wave harmonic injection. The following will be fully elaborated from three dimensions: technical principle, implementation method and key algorithm.

[0022] Spectrum analysis and frequency band dynamic division: The output current of the power equipment is continuously sampled through a high-speed sampling circuit, and the sampling frequency needs to meet the Nyquist criterion to ensure coverage of the highest frequency components of the fundamental wave, harmonic wave and surge component of the load current. The sampled time-domain current signal is windowed, preferably a Hanning window to reduce spectrum leakage, and then its frequency domain components are calculated through a fast Fourier transform (FFT) Based on the cumulative distribution characteristics of spectrum energy, the cut-off frequency is dynamically determined , and the specific calculation method is: ; Where is a preset energy threshold (usually taken as 0.8 to 0.95), is the total spectrum energy, is the frequency point index.

[0023] In this way, the low-frequency band ( ) covers the main energy components of the load steady-state current, and the high-frequency band ( ) contains surge transient energy, realizing the dynamic adaptive segmentation of the frequency band.

[0024] Low-impedance characteristic of the low-frequency band: Within the low-frequency band, the target impedance characteristic is generated through the virtual impedance algorithm .

[0025] Among them, the virtual resistance is used to suppress the steady-state voltage deviation, and its value is determined by the following formula: ; is the proportionality coefficient, is the voltage tracking error.

[0026] Virtual inductor is used to filter out low-frequency harmonics, and its value is designed according to the inductor current ripple constraint: ; where is the switching frequency, is the peak value of the allowable inductor current ripple.

[0027] By adjusting the PWM duty cycle , the equivalent output impedance of the system approximates in the low-frequency band, thereby suppressing voltage fluctuations.

[0028] High-impedance characteristic in the high-frequency band: In the high-frequency band, the target impedance characteristic is set to , and the high-frequency impedance is increased to isolate the surge energy transfer.

[0029] To achieve impedance matching, specific sub-harmonic components are injected into the PWM modulation wave, where is the harmonic order, and are the optimization variables.

[0030] The harmonic parameters are solved through the following optimization model: ; where is the regularization coefficient, which is used to limit the harmonic amplitude to avoid over-modulation.

[0031] The optimization problem is solved by the sequential quadratic programming (SQP) algorithm, and the harmonic injection parameters are updated in real time in the digital signal processor (DSP).

[0032] Multi-band impedance controller: The frequency band division result, the target impedance parameters, and the harmonic injection command are passed to the multi-band impedance controller to generate the PWM modulation wave correction signal.

[0033] Preferably, a finite impulse response (FIR) filter is used to reconstruct the time domain of the frequency domain response of the target impedance, and its transfer function is designed as: ; where the filter coefficient Obtained by the inverse Fourier transform of the target impedance, it ensures that the time-domain modulated wave correction signal can accurately match the frequency-domain impedance characteristics.

[0034] The digital pulse width modulator dynamically adjusts the conduction timing of the switch tube to achieve frequency domain characteristic tracking of the system output impedance.

[0035] Step 2: Multi-timescale model predictive control In this embodiment, a hierarchical optimization control objective based on multi-timescale model predictive control (MPC) is implemented through a collaborative control architecture of slow and fast time layers. This architecture independently optimizes steady-state voltage tracking accuracy and transient surge energy distribution, reconciling conflicting requirements through coupled objective functions. The following fully explains the technical principles, implementation, and algorithm design.

[0036] Slow time layer voltage stabilization control is carried out within the slow time layer, and the state space model of the main circuit of the power equipment is constructed to output voltage and the inductor current is the state variable, the duty cycle of the switch tube With reference current is the control variable. The continuous state equation is discretized by the zero-order hold method to obtain the discrete state space expression: ; in 、 is the discretized system matrix, is the control variable. The optimization objective is defined as minimizing the voltage tracking deviation and the current steady-state error. The objective function is designed as: ; in 、 is the weight coefficient used to balance the optimization priority of voltage and current.

[0037] Preferably, the optimization period of the slow time layer is set to 1 ms to ensure the accuracy of steady-state control and the reasonable allocation of computing resources.

[0038] Fast time layer surge suppression In the fast time layer, a surge energy distribution model is built based on the transient characteristics of surge energy, and supercapacitors are used to absorb energy. Number of operations compared to metal oxide varistor (MOV) is the optimization variable.

[0039] The objective function is designed as: ; in 、 is a dynamically adjusted weight coefficient, which is used to balance the conflicting requirements of surge energy suppression and MOV life management.

[0040] The constraint conditions include the dynamic equation of the bus voltage: ; And the equivalent circuit model of the energy feedback current: ; Where is the equivalent resistance of the bidirectional DC / DC converter.

[0041] The optimization period of the fast time layer is set to 10 μs to meet the real-time requirements of surge suppression.

[0042] The hierarchical cooperation mechanism realizes the cooperation of the slow time layer and the fast time layer control instructions through the priority arbiter. When there is no surge event, the slow time layer instructions dominate the control output; when a surge voltage is detected, the priority of the fast time layer instructions is increased to ensure the real-time response of surge suppression.

[0043] The weight coefficients , , are dynamically adjusted according to the remaining capacity of the supercapacitor. The adjustment logic is: ; Where , are the reference weights, and the balanced management of energy storage and device life is realized through this adjustment strategy.

[0044] The model predictive control solves the optimization problems for the slow time layer and the fast time layer, and uses the sequential quadratic programming (SQP) algorithm for solution. Within the slow time layer, a control sequence for the next steps is generated through rolling horizon optimization, and the first control quantity is taken as the actual output.

[0045] Within the fast time layer, the finite element method is used to discretize the bus voltage field into a node voltage matrix , and the optimal energy distribution strategy is solved through the stiffness matrix . Its equation is: ; Where is the equivalent node force vector of the surge current.

[0046] Step 3: Dynamic switching of the energy feedback path In this embodiment, the dynamic switching of the energy feedback path realizes the efficient storage and release of surge energy by monitoring the bus surge voltage and the supercapacitor state in real time and triggering the boost or buck mode of the bidirectional DC / DC converter. The following elaborates on the technical implementation details from dimensions such as trigger logic, duty cycle calculation, closed-loop regulation, and protection mechanisms.

[0047] Boost mode trigger and energy storage When the bus surge voltage is detected trigger the boost mode of the bidirectional DC / DC converter to feedback the surge energy to the supercapacitor for storage.

[0048] Boost mode duty cycle It is calculated in real time according to the input-output voltage relationship, and the calculation formula is: ; where is the target voltage of the regulated bus, is the real-time bus voltage.

[0049] By adjusting the boost duty cycle, the charging current of the supercapacitor is controlled , and its expression is: ; In the formula is the on-resistance of the switching device, is the voltage across the supercapacitor.

[0050] Preferably, monitor in real time and set a redundant discharge circuit. When , divert part of the energy through the discharge resistor to avoid overvoltage risk.

[0051] Buck mode trigger and energy release When the surge event ends and , switch to the buck mode to release the energy stored in the supercapacitor to the regulated bus.

[0052] Buck mode duty cycle is dynamically determined by the input-output voltage ratio: ; To suppress the bus voltage fluctuation, a closed-loop proportional-integral (PI) regulator is introduced to dynamically correct the duty cycle, and its control law is: ; where , are the proportional and integral coefficients, which are designed through frequency-domain stability analysis to ensure the system's dynamic response and steady-state accuracy.

[0053] Mode switching logic and MPC: The switching instruction between the boost mode and the buck mode is linked to the surge energy distribution result of the multi-time-scale model predictive control (MPC).

[0054] When the optimization result of the fast time layer of the MPC indicates that the surge energy needs to be stored preferentially, the boost mode is forcibly locked; when the remaining capacity of the supercapacitor is lower than the threshold, the slow time layer of the MPC triggers the buck mode to release energy.

[0055] Preferably, a dead time is inserted during the mode switching process to avoid the switch tubes of the boost and buck circuits being directly short-circuited.

[0056] Magnetic flux constraint and hardware protection In the boost and buck modes, the magnetic flux of the magnetic integrated inductor is monitored in real time , and its calculation formula is: ; where , are the relative magnetic permeability of the magnetic core and the magnetic permeability of vacuum respectively, is the cross-sectional area of the magnetic core, , are the voltage stabilizing inductors.

[0057] And the number of turns of the surge inductor.

[0058] When , the duty cycle output is limited proportionally: ; Ensure that the magnetic core operates in the non-saturated region to avoid the degradation of the inductor performance.

[0059] Step 4: Online identification and adaptation of MOV parameters In this embodiment, the online identification of device aging parameters and the adaptive adjustment of the control strategy are realized through the coordinated action of recursive parameter identification, dynamic weight correction, supercapacitor parameter linkage calibration and stability guarantee mechanism, so as to realize the real-time tracking of the aging state of the metal oxide varistor (MOV) and the adaptive optimization of the control system. The technical details are fully elaborated from three dimensions: mathematical modeling, algorithm implementation and system integration.

[0060] Recursive least squares (RLS) identification of MOV aging parameters Based on the time series data of the MOV operating voltage and the surge current, a dynamic observation model is constructed, and the observation vector and the parameter vector are defined, where is the nominal threshold voltage of the MOV, is the dynamic resistance.

[0061] The update process of the recursive least squares method is implemented by the following formula: ; The Kalman gain Calculated as: ; Covariance matrix Updated to: ; Forgetting Factor Designed as a time-varying parameter, it decays exponentially: ; in is the initial value (typically 0.99), is a time constant used to balance the weight of new data and the accumulation effect of historical data to prevent parameter estimation from falling into local stability.

[0062] Adaptive adjustment of MPC surge suppression weights: MOV threshold voltage drift As a quantitative indicator of aging, is the RLS estimate.

[0063] Surge suppression weight coefficient The adjustment formula is designed as: ; This formula ensures that when the MOV threshold voltage decreases (aging increases), the priority of allocating surge energy to the supercapacitor is increased, reducing the number of MOV operations and thus extending its life.

[0064] In order to suppress the control oscillation caused by weight mutation, a first-order low-pass filter is introduced to To perform smoothing: ; in is the filter coefficient (preferably 0.1-0.3), and its physical meaning is the inertia characteristic that controls the change of weight.

[0065] Online calibration of supercapacitor equivalent series resistance (ESR) when (The threshold is set according to the MOV aging failure standard), the online identification of the supercapacitor ESR parameters is triggered.

[0066] The ESR calculation model is established based on the circuit equation of the charge and discharge process: ; in is the open circuit voltage (and state of charge (related) is the charge and discharge current.

[0067] By least squares fitting the real-time value, and the objective function is:[[]] ; The updated is used to correct the equivalent resistance of the energy feedback path in the MPC slow time layer state space model to ensure that the model is consistent with the actual hardware characteristics.

[0068] The parameter update step size constraint and stability guarantee mechanism impose constraints on the parameter update step size of the RLS algorithm to prevent parameter mutations caused by noise or abnormal data: ; where is the maximum allowable step size, which is set according to the statistical change rate of the MOV parameter history. When the step size exceeds the limit, the gradient projection method is used for smooth transition: ; This operation keeps the parameter update direction the same as the original direction, but limits the amplitude within the allowable range to avoid control instability caused by parameter jumps.

[0069] Dynamic forgetting factor: time-varying balances the parameter tracking speed and historical data accumulation to adapt to the aging and gradual change characteristics of MOV.

[0070] Weight filtering mechanism: low-pass filtering suppresses sudden weight changes to ensure smooth transition of control commands.

[0071] ESR calibration trigger logic: triggers the update of supercapacitor parameters when the MOV aging exceeds the limit to maintain model accuracy.

[0072] Gradient projection constraint: mathematical limiting and direction preservation take into account both the robustness and stability of parameter updates.

[0073] Step 5: Global stability verification and fault tolerance protection In this embodiment, the global stability verification and fault tolerance protection ensure the stable operation of the control system and the hardware safety within the full operating range through the synergistic effects of Lyapunov stability analysis, core magnetic flux saturation protection, and MOV aging branch switching mechanism. The following details the technical implementation from three dimensions: mathematical verification, real-time monitoring, and protection strategy.

[0074] Lyapunov global stability verification: Construct a Lyapunov function based on the state space model of the power equipment , where is the state variable, is a symmetric positive definite matrix.

[0075] By solving the matrix equation , verify the system matrix The stability of is the preset positive definite weight matrix.

[0076] If the equation has a positive definite solution , the system is judged to be globally asymptotically stable; otherwise, the control parameter reset instruction is triggered to restore the MPC objective function weight to the default value , and re-solve the Lyapunov equation. The core flux saturation protection monitors the magnetic flux of the magnetic integrated inductor in real time. , which is calculated based on Ampere's circuit law and the core material properties: ; in is the relative magnetic permeability of the core, is the vacuum permeability, is the effective cross-sectional area of the core, are the winding turns of the voltage stabilizing inductor and the surge inductor respectively.

[0077] when hour( is the core saturation flux density), and the PWM duty cycle output is limited proportionally: ; This operation limits the growth of the inductor current by shortening the on-time of the switch tube, forcing the core to desaturate, and avoiding control instability caused by degradation of the inductor characteristics.

[0078] MOV aging fault tolerance and backup branch switching: MOV threshold voltage drift based on online identification ,when When ≥0.2, determine MOV Severe aging.

[0079] Triggering the fault tolerance protection mechanism: 1. Cut off the control signal of the current MOV branch and isolate the fault branch through the solid-state relay (SSR); 2. Close the switch of the backup MOV branch to transfer the surge energy to the backup branch; 3. Send an alarm signal to the monitoring system to indicate the aging status of the MOV.

[0080] Preferably, a 10ms delay is introduced into the switching logic of the standby branch to avoid false triggering during surge transients.

[0081] Lyapunov function construction: mathematical verification based on the state-space model to ensure the theoretical support of the global stability of the control system.

[0082] Magnetic flux calculation model: Combines core physical parameters with electrical quantity monitoring to quantify saturation risk and dynamically limit duty cycle.

[0083] MOV aging fault-tolerant logic: triggers hardware branch switching based on parameter identification results to achieve fault isolation and redundancy protection.

[0084] Control parameter reset mechanism: restore default weights when stability check fails to avoid continuous mismatch of control strategy.

[0085] 1. Lyapunov equation: , the system stability is determined by the matrix positive definiteness.

[0086] 2. Magnetic flux calculation formula: Correlates inductor current, core parameters, and material properties to accurately quantify saturation state.

[0087] 3. Duty cycle limitation logic: Dynamically adjust the duty cycle according to the magnetic flux ratio to ensure that the core operates in the linear region.

[0088] 4. MOV aging threshold condition: ΔVdrift ≥ 0.2VMOV_nominal, with tolerance threshold set based on industry standards.

[0089] Through dynamic impedance frequency domain segmentation combined with multi-time scale model predictive control, the voltage regulation performance is optimized in the low-frequency band, the surge energy is suppressed in the high-frequency band, and the energy feedback path and device parameters are adaptively adjusted based on real-time operating conditions. At the same time, by utilizing Lyapunov stability verification, magnetic flux protection and MOV fault-tolerant switching mechanism, a collaborative system from local control to global stability is constructed, realizing integrated control of high-precision voltage regulation and surge suppression of power equipment under complex disturbances and device aging.

[0090] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for coordinated control of voltage stabilization, current stabilization and surge suppression of electrical equipment, characterized in that, Including: Dynamically divide the output current frequency band and reshape the system output impedance characteristics. By analyzing the load current spectrum in real time, set a low impedance in the low-frequency band to suppress voltage fluctuations, and set a high impedance in the high-frequency band to absorb surge energy; Based on the multi-time scale model predictive control, hierarchically optimize the control objectives. Optimize the voltage tracking accuracy on the slow time scale, optimize the surge energy distribution on the fast time scale, and coordinate the conflicting requirements of voltage regulation and surge suppression through the coupled objective function; Dynamically switch the energy feedback path. By monitoring the bus surge voltage in real time, trigger the boost or buck mode of the bidirectional DC / DC converter, and feedback the surge energy to the super capacitor for storage or release it to the regulated bus; Online identify the device aging parameters and adaptively adjust the control strategy. Update the MOV threshold drift through the recursive algorithm, and dynamically correct the priority weight of surge suppression in the MPC objective function; Global stability verification and fault tolerance protection. Verify the stability of the control system based on the Lyapunov stability theory, and trigger the protection mechanism when the magnetic flux saturation or device aging exceeds the limit.

2. A method for coordinated control of voltage stabilization, current stabilization and surge suppression of an electrical device according to claim 1, characterized in that, The dynamic frequency band segmentation and impedance reshaping includes: performing a fast Fourier transform analysis on the output current to divide the low-frequency band ( ) and the high-frequency band ( ; Set a low impedance characteristic in the low-frequency band to suppress voltage fluctuations, and set a high impedance characteristic in the high-frequency band to absorb surge energy; Dynamically match the target impedance frequency domain characteristics by injecting specific harmonics of the PWM modulation wave.

3. A method for collaborative control of voltage stabilization, current stabilization and surge suppression of a power device according to claim 1, characterized in that, The cut-off frequencies of the low-frequency band and the high-frequency band are adaptively adjusted according to the spectral energy distribution of the load current to satisfy , where is the fundamental frequency, is the coefficient corresponding to the energy ratio threshold.

4. A method for coordinated control of voltage and current stabilization and surge suppression of a power device according to claim 1, characterized in that, The hierarchical optimization of the multi-time scale MPC includes: Construct a state space model in the slow time layer, and the optimization objective is to minimize the voltage tracking deviation; Construct a surge energy distribution model in the fast time layer, and the optimization objective is to minimize the number of MOV operations; Coordinate the control instructions of the slow time layer and the fast time layer through the coupled objective function.

5. A method for coordinated control of voltage stabilization, current stabilization and surge suppression of a power device according to claim 4, characterized in that, The optimization period of the slow time layer is 1 ms, and the optimization period of the fast time layer is 10 μs. And the weight coefficients of the objective functions of the two layers are dynamically adjusted according to the remaining capacity of the super capacitor.

6. A method for coordinated control of voltage stabilization, current stabilization and surge suppression of a power device according to claim 1, characterized in that, The dynamic switching of the energy feedback path includes: When it is detected that the bus surge voltage exceeds the threshold, trigger the boost mode of the bidirectional DC / DC converter, and feedback the surge energy to the super capacitor; When the surge voltage is lower than the threshold, switch to the buck mode and release the super capacitor energy to the regulated bus; The switching logic of the boost and buck modes is linked with the surge energy distribution result of the MPC.

7. A method for coordinated control of voltage stabilization, current stabilization and surge suppression of an electrical device according to claim 6, characterized in that, The boost mode duty cycle and the buck mode duty cycle are calculated in real time according to the input and output voltages, and are restricted by the magnetic flux saturation constraint of the magnetically integrated inductor.

8. A coordinated control method for voltage and current stabilization and surge suppression of an electrical device according to claim 1, characterized in that, The online identification of device aging parameters includes: Update the threshold voltage drift of MOV using the recursive least squares method ; According to dynamically correct the weight coefficient of the surge suppression term in the MPC objective function to satisfy ).

9. A method for coordinated control of voltage and current stabilization and surge suppression of electric power equipment according to claim 1, characterized in that: The forgetting factor of the RLS algorithm Exponentially decays according to the system running time, and the parameter update step size is constrained to , avoiding mutations.

10. A method for coordinated control of voltage stabilization, current stabilization and surge suppression of a power device according to claim 1, characterized in that, The global stability verification includes: Construct a Lyapunov function to verify the global asymptotic stability of the control system; When the magnetic flux exceeds the magnetic core saturation threshold, limit the PWM duty cycle proportionally; When the MOV aging degree exceeds 20%, switch to the standby surge absorption branch and give an alarm.

Images