MMC-LLC converter control method for high voltage direct current-low voltage direct current conversion

Abstract

This invention discloses a control method for an MMC-LLC converter used in high-voltage DC-to-low-voltage DC conversion, comprising: constructing an equivalent circuit model of a multi-submodule series structure; establishing a dual-channel ordered submodule capacitor voltage equalization strategy based on the equivalent circuit model; and constructing a resonant current loss optimization and output voltage stabilization control method in the LLC resonant cavity, consisting of equivalent K-value adjustment, resonant frequency control, bridge arm voltage envelope constraint, and free adjustment of duty cycle of some submodules; and establishing a reverse prediction model for the mode switching process based on the system dynamic characteristics under input voltage fluctuations, and constructing a mode switching impact suppression strategy. This invention comprehensively utilizes multi-submodule modeling, resonant current optimization and control, and mode prediction techniques, achieving voltage control and energy regulation without relying on hardware compensation and complex feedforward models, effectively reducing voltage impact during mode switching and expanding the soft-turn-on range.

Description

Technical Field

[0001] This invention relates to the field of power electronic conversion and power conversion control technology, and in particular to a control method for an MMC-LLC converter used for high voltage DC-to-low voltage DC conversion. Background Technology

[0002] With the development of applications such as DC power grids, submarine observation networks and high-voltage power distribution, there is a strong demand for efficiently and safely converting high-voltage DC (8-16kV) into medium- and low-voltage DC (such as 375V) for use by end-user equipment.

[0003] Using an LLC resonant converter combined with modular multilevel converter (MMC) technology can simultaneously offer the advantages of efficient soft switching and high-voltage sharing. However, several problems need to be addressed in engineering implementation: 1) When multiple sub-modules are connected in series, the capacitor voltages of the sub-modules are prone to imbalance, affecting module safety and lifespan; 2) When input voltage fluctuations or load changes cause the K-value of the sub-modules to switch (mode switching), the bridge arm voltage often experiences transient surges; 3) It is difficult for the LLC resonant cavity to maintain a soft-turn-on range over a wide input range, leading to increased switching losses and resonant current stress; 4) The reference potential of the sub-modules floats with the series connection position, requiring reliable isolated power supply and isolated sampling schemes for driving and sampling; 5) The control unit needs to support independent phase and duty cycle dynamic adjustment of multiple PWM channels to achieve coordinated control of voltage equalization and resonant modulation.

[0004] Existing solutions mostly rely on a single voltage equalization strategy or fixed frequency / duty cycle control, which makes it difficult to simultaneously meet the requirements of voltage equalization speed, smooth mode switching, and minimized resonant loss under the aforementioned constraints. Therefore, a holistic control method and implementation scheme are urgently needed to solve these problems. Summary of the Invention

[0005] The problem to be solved by this invention is to provide a control method for MMC-LLC converters used in high-voltage DC-to-low-voltage DC conversion. By using multi-submodule equivalent modeling, dual-channel sorting voltage equalization, resonant current loss optimization, mode switching reverse prediction and isolated power supply / sampling design, the method achieves fast voltage equalization, mode switching impact suppression, soft switching extension and reliable driving and sampling of floating submodules, thereby improving system efficiency and operational stability.

[0006] This invention adopts the following technical solution: a control method for an MMC-LLC converter for high-voltage DC-to-low-voltage DC conversion, comprising the following steps:

[0007] Step 1: Construct an equivalent circuit model of a multi-submodule series structure: Based on the MMC-LLC converter, mathematical model the equivalent output voltage of multiple series submodules, dynamically model the LLC resonant inductor, resonant capacitor and equivalent K value, and generate an equivalent circuit model including bridge arm voltage, capacitor voltage and resonant current.

[0008] Step 2: Based on the equivalent circuit model, establish a dual-channel sorting sub-module capacitor voltage equalization strategy and generate a sub-module input priority table for fast capacitor voltage equalization during the positive and negative half-cycles of the bridge arm.

[0009] Step 3: Based on the equivalent circuit model, construct a resonant current loss optimization and output voltage stabilization control method in the LLC resonant cavity, which consists of equivalent K value adjustment, resonant frequency control, bridge arm voltage envelope constraint, and free adjustment of duty cycle of some sub-modules.

[0010] Step 4: Based on the system dynamic characteristics under input voltage fluctuations, establish a reverse prediction model for the mode switching process; predict the sub-module equivalent voltage and K value changes in the next cycle based on the input voltage change; deduce the optimal control conditions based on the circuit model characteristics, and adjust them together with feedback control.

[0011] Step 5: Based on the reverse prediction model, construct a mode transition impact suppression strategy, determine whether to enter the switching state based on the input voltage and call the reverse prediction model, and exit the reverse prediction after the circuit is stable and the variable weight parameters reach the set threshold.

[0012] Step 6: Construct an isolated drive and isolated sampling structure suitable for multi-submodule high-voltage floating potential environments to achieve voltage control and energy regulation, and reduce voltage surges during mode transitions.

[0013] As a preferred embodiment, the MMC-LLC converter includes: an upper bridge arm and a lower bridge arm, an LLC resonant module, an isolation transformer, a rectifier output stage, and a DSP controller.

[0014] The upper and lower bridge arms are each composed of multiple sub-modules connected in series. The midpoint of the upper and lower bridge arms is connected to one end of the LLC resonant cavity, and the other end is connected to the zero point of the input voltage. The LLC resonant module includes a resonant inductor, a resonant capacitor, and a magnetizing inductor to achieve resonant energy transfer. The isolation transformer is connected to the output terminal of the LLC resonant module to achieve electrical isolation and voltage conversion. The rectifier output stage is connected to the secondary side of the isolation transformer to rectify the AC voltage into a DC output voltage. The DSP controller is used to collect system operating parameters and generate control signals to control the on and off of the switching transistors of each sub-module, thereby realizing the operation control of the MMC-LLC converter.

[0015] The multiple series-connected sub-modules include: sub-module capacitors, power switching transistors, drive circuits, isolation power supply circuits, and sampling circuits.

[0016] The power switching transistors and submodule capacitors form a half-bridge submodule structure, with the submodule capacitors connected in parallel across the two power switching transistors. The driving circuit is used to drive the power switching transistors to turn on and off. The isolation power supply circuit is used to provide isolated power to the submodule driving circuit. The sampling circuit is used to collect the submodule capacitor voltage and related operating parameters. Multiple submodules are connected in series to form a bridge arm structure, which is used to realize the voltage division and voltage synthesis functions of the MMC-LLC converter system.

[0017] As a preferred embodiment, in step 1, the equivalent circuit model is formed by mathematically modeling the equivalent output voltage of multiple series sub-modules and dynamically modeling the LLC resonant inductor, resonant capacitor and equivalent K value, thereby forming a unified analysis model that can simultaneously characterize the bridge arm voltage, capacitor voltage and resonant current.

[0018] Among them, the voltages u of the upper and lower bridge arms p u n and the corresponding bridge arm current i p i n Together with the DC component I of the bridge arm current dc With excitation current I m The following relationship exists between them:

[0019] ;

[0020] ;

[0021] ;

[0022] in, DC input voltage, This is the equivalent input voltage across the resonant cavity. This represents the DC component of the current in the upper and lower bridge arms. For resonant current, For transmission power, For transformer magnetizing inductance, This refers to the switching frequency.

[0023] LLC resonant mode not only enables soft switching conditions for power devices, but also allows for voltage matching between different inputs and outputs by adjusting the switching frequency to change the resonant cavity voltage gain.

[0024] Ignoring the quasi-square wave ramp range of the MMC unit output, the gain of the LLC unit... Consistent with the half-bridge structure, it can be represented as:

[0025] ;

[0026] ;

[0027] in, The resonant frequency is determined by... and Decide; Normalized frequency; This is the quality factor, which is related to the resonant cavity parameters and the equivalent load. This is the ratio of the resonant inductance to the magnetizing inductance.

[0028] As a preferred embodiment, in step 2, the dual-channel sorting voltage equalization strategy first sorts the submodules based on the magnitude of the change in capacitor voltage before and after the previous cycle, and then sorts them based on the voltage level of the submodules in the current cycle. By merging these two sorting results, a submodule activation priority table is generated, and the activation order of the submodules is controlled according to this priority, thereby achieving balanced adjustment in each control cycle.

[0029] In addition, this priority table is used in the positive and negative half-cycles of the bridge arm, which enables the capacitor voltage to be quickly balanced throughout the entire operating cycle.

[0030] As a preferred embodiment, in the resonant current loss optimization method in step 3, the direction of energy flow in the resonant cavity is changed by adjusting the equivalent K value, and the resonant point position of LLC is controlled by frequency shift.

[0031] In addition, the allowable voltage swing is adjusted according to the constraints of the voltage envelope between bridge arms, and some sub-modules are allowed to freely change their duty cycle within a set range, so as to achieve the lowest possible loss operation of the power switching devices.

[0032] The input DC voltage and output AC voltage of the MMC unit satisfy the following relationship:

[0033] ;

[0034] The submodule capacitor voltage U can be further calculated. c and AC output voltage amplitude U ac The expression:

[0035] ;

[0036] ;

[0037] in, U in It is the DC input voltage.u p u n These are the voltages of the upper and lower bridge arms, respectively, and N is the number of bridge arm sub-modules. It is the amplitude of the MMC AC output voltage. It is the voltage of the submodule capacitor:

[0038] The voltage gain relationship from input to output can be expressed as:

[0039] ;

[0040] ;

[0041] In the formula, n is the transformer turns ratio, which is a hardware circuit parameter. K To minimize the number of submodules required, f s This refers to the switching frequency.

[0042] Based on this, find the submodule with the largest voltage value in each of the upper and lower bridge arms. v cm and the submodule with the smallest voltage value v cn ,in m , n For any submodule number, the voltage envelopes of the upper and lower bridge arms within the current control cycle are obtained by tracking these two values ​​in real time. The voltage situation is evaluated based on the envelopes, and the voltage difference between the upper and lower bridge arms is obtained. Obtained through control This allows for the correction of the duty cycle during module control.

[0043] ;

[0044] in, D 0 The default duty cycle is 0.5.

[0045] Simultaneously, a new control module is added, allowing its duty cycle to vary freely, not just fine-tuning around D0. This control sub-module primarily regulates the current during sub-module switching to minimize switching losses. Specifically, the required voltage and current are obtained based on the following equations, and then the value of D is determined in reverse:

[0046] ;

[0047] Where n is the primary-to-secondary voltage ratio of the transformer. U o This is the DC output voltage. u Cr This is the voltage across the resonant capacitor. i rFor the resonant inductor current, L r It is an equivalent resonant inductor, composed of the MMC bridge arm inductance and the transformer magnetizing inductance.

[0048] As a preferred embodiment, in step 4, the reverse prediction model predicts the sub-module equivalent voltage and K value change trend in the next cycle by analyzing the change in input voltage, and combines the inherent parameters of the circuit and the characteristics of the model to deduce the optimal control conditions, and adjusts them together with feedback control.

[0049] The reverse model is described as follows:

[0050] ;

[0051] ;

[0052] Where G is the LLC unit gain, f r , f s , f n , Q , λ The definition is consistent with the above. The optimal switching frequency of the K value obtained based on this model can significantly reduce the inrush voltage during module switching.

[0053] As a preferred option, in step 5, in order to mitigate the voltage surge that may occur during mode transition, the strategy determines whether to call the reverse prediction model by judging whether the input voltage enters the switching state, and introduces variable weight parameters in each control cycle.

[0054] When entering the switching state, the variable weights are reassigned and continuously multiplied by the decay coefficient in the subsequent control loop, so that they gradually decrease, thereby ensuring that the prediction model plays a dominant role in the switching phase, while the feedback control regains dominance in steady-state operation.

[0055] The weighting relationship satisfies:

[0056] ;

[0057] ;

[0058] in, f To ultimately control the frequency, f s To predict the control frequency, f PID For feedback control frequency.

[0059] As a preferred embodiment, in step 6, the isolated drive and isolated sampling structure consists of an isolated DC / DC drive power supply, a single-input complementary output PWM drive structure, and a high-precision differential sampling link.

[0060] The input terminal of the isolated DC / DC drive power supply is connected to the system's low-voltage DC power supply, including a DC / DC isolated power supply module and an input filter capacitor. The output terminal of the DC / DC isolated power supply module is connected to the drive and functional circuits of each sub-module, providing mutually isolated power supplies for the functional implementation of the power switch drive circuit.

[0061] The input terminal of the single-input complementary output PWM drive structure is connected to the PWM control signal output by the DSP controller. It includes a PWM drive chip and a drive isolation circuit. The output terminal of the PWM drive chip is connected to the gate of the two power switching transistors of the sub-module respectively, and is used to generate complementary drive signals to control the conduction and turn-off of the power switching transistors, thereby realizing the connection and bypass of the capacitor of the sub-module.

[0062] The input end of the high-precision differential sampling data structure is connected to the two ends of the submodule capacitor, the two ends of the upper and lower bridge arms, the two ends of the input filter capacitor, the two ends of the output load, and the two ends of the bridge arm current sampling resistor. It includes a differential sampling amplifier circuit and an isolation amplifier circuit. The output end of the differential sampling amplifier circuit is connected to the analog sampling interface of the DSP controller to achieve high-precision acquisition of the submodule capacitor voltage and system operating parameters.

[0063] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0064] The method of this invention can achieve the following: significantly improved submodule capacitor voltage equalization speed, significantly reduced bridge arm voltage surge during mode switching, expanded soft turn-on range of LLC resonant cavity, stable and reliable submodule driving and sampling under floating potential, and DSP-based multi-channel PWM control fully supports high-precision adjustment of phase and duty cycle, thereby stably outputting 375V in the 8-16 kV input range and improving the overall efficiency and reliability of the system. Attached Figure Description

[0065] Figure 1 This is the overall flowchart of the MMC-LLC converter control method of the present invention;

[0066] Figure 2 This is a block diagram of the MMC-LLC converter circuit structure of the present invention;

[0067] Figure 3 This is a schematic diagram of the submodule structure of the present invention;

[0068] Figure 4 This is the gain curve of the LLC resonant cavity of this invention as a function of frequency;

[0069] Figure 5 This is a flowchart of the dual-channel sorting and equalization process of the present invention;

[0070] Figure 6 This is a schematic diagram of the voltage situation of a bridge arm submodule without bridge arm voltage envelope constraint in an embodiment of the present invention;

[0071] Figure 7 This is a schematic diagram of the voltage status of the bridge arm submodule under bridge arm voltage envelope constraint according to an embodiment of the present invention;

[0072] Figure 8 This is a complex voltage impulse waveform diagram of module switching in an embodiment of the present invention without the inclusion of a reverse prediction model;

[0073] Figure 9 The complex voltage impulse waveform diagram for module switching when the inverse prediction model is added to the embodiment of the present invention. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the application will be further described in detail below with reference to the accompanying drawings. The described embodiments are only a part of the embodiments involved in this invention. All non-innovative embodiments based on these embodiments by other researchers in the art are within the protection scope of this invention. Furthermore, the step numbers in the embodiments of this invention are only set for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0075] In one embodiment of the present invention, a control method for an MMC-LLC converter for high-voltage DC-to-low-voltage DC conversion is provided, such as... Figure 1 As shown, it includes the following steps:

[0076] Step 1: Construct an equivalent circuit model of a multi-submodule series structure.

[0077] like Figure 2 As shown, the MMC-LLC converter of the present invention includes multiple series sub-modules of the upper and lower bridge arms, an LLC resonant cavity, an isolation transformer, and a rectifier output stage. Figure 3 The internal structure and topology of the submodules are shown. Under high-voltage DC input conditions (8–16 kV), multiple submodules are connected in series to provide the necessary voltage divider structure for the system, and capacitor voltage equalization, resonance regulation, and mode switching are achieved through an integrated control algorithm.

[0078] In this embodiment, the equivalent output voltages of multiple sub-modules are mathematically modeled to establish the bridge arm voltages. u p ,u n With the corresponding bridge arm current i p , i n and the DC component of the bridge arm current I dc Excitation current I m The relationship between them can be modeled as follows:

[0079] ;

[0080] ;

[0081] ;

[0082] in, U in For input DC voltage, P o For transmission power, L m For magnetizing inductance, f s This refers to the switching frequency.

[0083] This model is further combined with the parameters of the LLC resonant cavity to dynamically characterize the resonant inductance, resonant capacitance, and equivalent K value, forming a unified system model.

[0084] like Figure 4 As shown, ignoring the ramp region of the quasi-square wave output by the MMC, the LLC cell voltage gain G can be expressed as a function of the switching frequency, with the following expression:

[0085] ;

[0086] ;

[0087] in, f r The main resonant frequency, f n Here, λ is the normalized frequency, and λ is the inductance ratio.

[0088] Step 2: Based on the equivalent model, establish a dual-channel sorting submodule capacitor voltage equalization strategy.

[0089] like Figure 3 As shown, each MMC submodule includes power devices, capacitors, isolation drivers, and differential sampling circuitry. In this embodiment, fast voltage equalization is achieved through dual-channel sorting, including:

[0090] 1) Sort the changes in capacitor voltage in the previous cycle;

[0091] 2) Sort the absolute values ​​of the capacitor voltages in the current cycle;

[0092] The system generates the priority of sub-module deployment based on the two sorting results.

[0093] like Figure 5 As shown, this strategy can quickly balance the capacitor voltage during the positive and negative half-cycles of the bridge arm, enabling autonomous redistribution of energy in the submodules and ensuring stable operation of the system under large voltage swing conditions.

[0094] Step 3: Construct a method for optimizing resonant current loss and stabilizing output voltage in the LLC resonant cavity.

[0095] In this embodiment, the optimization of the resonant current and the stability of the output voltage are achieved through the adjustment of the equivalent K value, frequency offset control, voltage envelope constraint, and free adjustment of the duty cycle.

[0096] First, based on the LLC model, the relationship between the AC output voltage and the input voltage of the MMC unit can be obtained as follows:

[0097] ;

[0098] This leads to the derivation of the capacitor voltage. U c With output amplitude U ac The expression:

[0099] ;

[0100] ;

[0101] in, The DC voltage modulation ratio of the MMC is determined by the number of switching sub-modules N and K, and is expressed as:

[0102] ;

[0103] The expression for the voltage gain from input to output is:

[0104] ;

[0105] In the formula, N is the number of bridge arm sub-modules, n is the transformer turns ratio (a hardware circuit parameter), and K is the minimum number of sub-modules required. f s This refers to the switching frequency.

[0106] Based on this, find the submodule with the largest voltage value in each of the upper and lower bridge arms. v cm and the submodule with the smallest voltage value vcn Where m and n are arbitrary submodule numbers, the voltage envelopes of the upper and lower bridge arms within the current control cycle are obtained by tracking these two values ​​in real time. The voltage situation is evaluated based on the envelopes, and the voltage difference between the upper and lower bridge arms is obtained. Obtained through control This allows for the correction of the duty cycle during module control.

[0107] ;

[0108] in, D 0 The default duty cycle is 0.5 in this embodiment.

[0109] Specifically, the envelope constraint effect achieved by the method of this invention is for example... Figure 6 and Figure 7 As shown, a significant difference can be observed.

[0110] Simultaneously, a new control module is added, allowing its duty cycle to vary freely, not just fine-tuning around D0. This control sub-module primarily regulates the current during sub-module switching to minimize switching losses. Specifically, the required voltage and current are obtained based on the following equations, and then the value of D is determined in reverse:

[0111] ;

[0112] Where n is the primary-to-secondary voltage ratio of the transformer. U o This is the DC output voltage. u Cr This is the voltage across the resonant capacitor. i r For the resonant inductor current, L r It is an equivalent resonant inductor, composed of the MMC bridge arm inductance and the transformer magnetizing inductance.

[0113] The above mechanism can keep LLC in the soft-on region, improve conversion efficiency, and reduce switching losses.

[0114] Step 4: Establish a reverse prediction model for mode switching based on input voltage changes.

[0115] like Figure 8 and Figure 9 As shown in the comparison, when the input voltage fluctuation causes the resonant mode to switch, this embodiment achieves early adjustment through the inverse prediction model to eliminate transient impacts.

[0116] Specifically, it includes:

[0117] 1) Predict the changes in the equivalent voltage and K value of the submodule in the next cycle based on the changes in the input voltage;

[0118] 2) The optimal control frequency is derived by reverse calculation based on the resonance model.

[0119] As shown in this embodiment, the reverse prediction model satisfies the following relationship:

[0120] ;

[0121] ;

[0122] Where G is the LLC gain, f n Q and λ have the same meaning as mentioned above.

[0123] This method allows for the prediction of the optimal operating frequency near the switching point in advance, significantly reducing voltage surges during mode switching.

[0124] Step 5: Construct a mode transition shock suppression strategy based on the inverse prediction model.

[0125] In this embodiment, during mode switching, the system quickly determines whether the switching range has been entered based on the input voltage and invokes the inverse prediction module. Variable weight parameters are incorporated during the adjustment process. , The value is reassigned during module switching, and continuously compared with the attenuation coefficient in the control loop. The multiplication factor gradually decreases over time, meaning that the backpropagation model only plays a dominant role during module switching, while feedback control dominates during steady-state control.

[0126] ;

[0127] ;

[0128] In the formula, f is the final control frequency. f s For reverse prediction frequency, f PID For feedback control frequency.

[0129] This embodiment uses a variable weight switching mechanism to enable predictive control to dominate the transient process and feedback control to dominate the steady-state process, achieving a flexible transition and avoiding switching shocks.

[0130] Step 6: Construct an isolated drive and differential sampling structure suitable for floating potentials of multiple submodules.

[0131] Since each submodule in the MMC architecture has a different reference potential, and the reference potential increases step by step as the position rises, it is necessary to configure isolated power supplies for each submodule to ensure that the drive signals can work normally under different reference potentials.

[0132] Same reference Figure 3 As shown, the reference potential of the MMC submodules is highly floating due to their different positions. Therefore, this embodiment uses isolated DC / DC power supplies (such as B1212S and B0505S) to provide independent drive power, and uses IR21844 to implement complementary PWM drive to reduce the number of PWM pins of the controller.

[0133] In addition, to ensure data accuracy, improve anti-interference capability and reduce sampling error, differential isolation sampling chips (such as AMC1311 and AMC1301) are used to obtain capacitor voltage and resonant current in high electromagnetic interference environments, thereby achieving high anti-interference and high reliability sampling.

[0134] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products or devices.

[0135] In summary, the MMC-LLC converter control method for high-voltage DC-to-low-voltage DC conversion provided by this invention can be implemented through various means. The above description is only a preferred embodiment of this invention. It should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technology.

Claims

1. A control method for an MMC-LLC converter used for high-voltage DC-to-low-voltage DC conversion, characterized in that, Includes the following steps: Step 1: Construct an equivalent circuit model of a multi-submodule series structure: Based on the MMC-LLC converter, mathematical model the equivalent output voltage of multiple series submodules, dynamically model the LLC resonant inductor, resonant capacitor and equivalent K value, and generate an equivalent circuit model including bridge arm voltage, capacitor voltage and resonant current. Step 2: Based on the equivalent circuit model, establish a dual-channel sorting sub-module capacitor voltage equalization strategy and generate a sub-module input priority table for fast capacitor voltage equalization during the positive and negative half-cycles of the bridge arm. Step 3: Based on the equivalent circuit model, construct a resonant current loss optimization and output voltage stabilization control method in the LLC resonant cavity, which consists of equivalent K value adjustment, resonant frequency control, bridge arm voltage envelope constraint, and free adjustment of duty cycle of some sub-modules. Step 4: Based on the system dynamic characteristics under input voltage fluctuations, establish a reverse prediction model for the mode switching process; predict the sub-module equivalent voltage and K value changes in the next cycle based on the input voltage change; deduce the optimal control conditions based on the circuit model characteristics, and adjust them together with feedback control. Step 5: Based on the reverse prediction model, construct a mode transition impact suppression strategy, determine whether to enter the switching state based on the input voltage and call the reverse prediction model, and exit the reverse prediction after the circuit is stable and the variable weight parameters reach the set threshold. Step 6: Construct an isolated drive and isolated sampling structure suitable for multi-submodule high-voltage floating potential environments to achieve voltage control and energy regulation, and reduce voltage surges during mode transitions.

2. The MMC-LLC converter control method according to claim 1, characterized in that: The MMC-LLC converter described in step 1 includes: an upper bridge arm and a lower bridge arm, an LLC resonant module, an isolation transformer, a rectifier output stage, and a DSP controller; The upper and lower bridge arms are each composed of multiple sub-modules connected in series. The midpoint of the upper and lower bridge arms is connected to one end of the LLC resonant module, and the other end of the LLC resonant module is connected to the zero point of the input voltage. The LLC resonant module includes a resonant inductor, a resonant capacitor, and a magnetizing inductor, used to achieve resonant energy transfer. The isolation transformer is connected to the output terminal of the LLC resonant module to achieve electrical isolation and voltage transformation; The rectifier output stage is connected to the secondary side of the isolation transformer and is used to rectify AC voltage into DC output voltage. The DSP controller is used to collect system operating parameters and generate control signals to control the on and off of the switching transistors of each submodule, thereby realizing the operation control of the MMC-LLC converter. The multiple series-connected sub-modules include: sub-module capacitors, power switching transistors, drive circuits, isolation power supply circuits, and sampling circuits; The power switching transistors and submodule capacitors form a half-bridge submodule structure, with the submodule capacitors connected in parallel across the two power switching transistors; the driving circuit is used to drive the power switching transistors to turn on and off; the isolated power supply circuit is used to provide isolated power to the submodule driving circuit; the sampling circuit is used to collect the submodule capacitor voltage and related operating parameters. Multiple sub-modules are connected in series to form a bridge arm structure, which is used to realize the voltage division and voltage synthesis functions of the MMC-LLC converter system.

3. The MMC-LLC converter control method according to claim 2, characterized in that, In the equivalent circuit model described in step 1, the voltages u of the upper and lower bridge arms are... p u n With the corresponding current i p i n and the DC component of the bridge arm current I dc and excitation current I m satisfy: ； ； ； In the formula, DC input voltage, This is the equivalent input voltage across the resonant cavity. This represents the DC component of the current in the upper and lower bridge arms. For resonant current, For transmission power, For transformer magnetizing inductance, This refers to the switching frequency.

4. The MMC-LLC converter control method according to claim 3, characterized in that, The equivalent circuit model described in step 1 achieves soft switching of the power devices in the circuit sub-module through LLC resonant operation mode. By changing the switching frequency of the sub-module, the voltage gain of the LLC resonant cavity is changed to achieve input-output voltage matching. Ignoring the ramp region of the quasi-square wave output by the MMC unit, the voltage gain G of the LLC unit has the same expression as that of the half-bridge LLC, which is: ； ； in, f r The main resonant frequency of the LLC circuit is determined by the main resonant inductor. L r and resonant capacitor C r Decide, f n To normalize the switching frequency, Q The quality factor is related to the resonant cavity parameters and the equivalent load. R eq related, λ is the inductance coefficient, and is the resonant inductance. L r And excitation inductance L m The ratio of .

5. The MMC-LLC converter control method according to claim 1, characterized in that, The voltage equalization strategy for dual-channel sorting described in step 2 includes: Step 2.1: Sort the submodule capacitor voltage changes from largest to smallest before and after the previous cycle; Step 2.2: Sort the submodule capacitor voltages in ascending order according to the current cycle. Step 2.3: Generate a submodule deployment priority table based on the double sorting results; Step 2.4: Control the order in which sub-modules are activated according to the priority table.

6. The MMC-LLC converter control method according to claim 1, characterized in that, Step 3, the optimization of resonant current loss, includes the following methods: Step 3.1: Change the direction of energy flow in the resonant cavity by adjusting the equivalent K value; Step 3.2: Adjust the position of the LLC resonant point using frequency offset control; Step 3.3: Adjust the allowable voltage swing of each bridge arm according to the voltage envelope constraint between bridge arms; Step 3.4: Allow the duty cycle of some sub-modules to vary freely within a set range to achieve the lowest possible switching loss of the switching transistor.

7. The MMC-LLC converter control method according to claim 4, characterized in that, The output voltage stabilization control method described in step 3 includes: The DC input voltage and AC output voltage of the MMC unit satisfy the following equation: ； in, U in It is the DC input voltage. u p , u n These are the voltages of the upper and lower bridge arms, N Number of bridge arm sub-modules; It is the amplitude of the MMC AC output voltage. It is the voltage of the submodule capacitor: ； ； The input-to-output voltage gain is expressed as: ； In the formula, U o This is the DC output voltage. This refers to the DC voltage modulation ratio of the MMC. n The transformer turns ratio is a hardware circuit parameter. K To minimize the number of submodules required, f s The switching frequency; Find the submodule with the largest voltage value in each of the upper and lower bridge arms. and the submodule with the smallest voltage value Where m and n are arbitrary submodule number values, which are determined by real-time tracking. and The value is used to obtain the voltage envelopes of the upper and lower bridge arms within the current control cycle. The voltage situation is evaluated based on the envelopes, and the voltage difference between the upper and lower bridge arms is obtained. Obtained through control Correct the duty cycle during module control: ； in, D 0 This is the default duty cycle size; A new control module is added to regulate the current during submodule switching, minimizing switching losses. The required voltage and current are obtained based on the following equation, and then the duty cycle D is determined in reverse: ； in, n This is the ratio of the primary to secondary voltages of the transformer. This is the voltage across the resonant capacitor. i r For the resonant inductor current, L r It is an equivalent resonant inductor, composed of the MMC bridge arm inductance and the transformer magnetizing inductance.

8. The MMC-LLC converter control method according to claim 7, characterized in that, In the reverse prediction model described in step 4, the model back-calculation formula is as follows: ； ； In the formula, G For the voltage gain of the LLC cell, f r The main resonant frequency of the LLC circuit is determined by the main resonant inductor. L r and resonant capacitor C r Decide, f s To control the switching frequency, f n The normalized switching frequency is given by Q, which is the quality factor, related to the resonant cavity parameters and the equivalent load. R eq related; λ is the inductance coefficient, and is the equivalent resonant inductance. L r And excitation inductance L m The ratio; Based on the optimal switching frequency of the K value obtained from the inverse prediction model, the inrush voltage is reduced during module switching.

9. The MMC-LLC converter control method according to claim 1, characterized in that, The mode transition shock suppression strategy described in step 5 incorporates variable weight parameters during the adjustment process. , The value is reassigned during module switching, and continuously compared with the attenuation coefficient in the control loop. The multiplication factor gradually decreases over time, allowing the inverse prediction model to play a dominant role during module switching, while feedback control dominates during steady-state control, resulting in the final control frequency f: ； ； In the formula, f s For reverse prediction frequency, f PID For feedback control frequency.

10. The MMC-LLC converter control method according to claim 1, characterized in that, The isolated drive and isolated sampling structure described in step 6 includes: an isolated DC / DC drive power supply, a single-input complementary output PWM drive structure, and a high-precision differential sampling data structure; The isolated DC / DC drive power supply has its input end connected to the system's low-voltage DC power supply and includes a DC / DC isolated power supply module and an input filter capacitor; its output end is connected to the drive and functional circuits of each sub-module, providing mutually isolated power supplies for the functional implementation of the power switch drive circuit. The single-input complementary-output PWM drive structure has an input terminal connected to the PWM control signal output by the DSP controller, including a PWM drive chip and a drive isolation circuit; the output terminal is connected to the gates of two power switching transistors of the submodule respectively, generating complementary drive signals to control the conduction and turn-off of the power switching transistors, thereby realizing the connection and bypass of the submodule capacitor. The high-precision differential sampling data structure has its input terminals connected to the two ends of the submodule capacitor, the two ends of the upper and lower bridge arms, the two ends of the input filter capacitor, the two ends of the output load, and the two ends of the bridge arm current sampling resistor. It includes a differential sampling amplifier circuit and an isolation amplifier circuit. The output terminal is connected to the analog sampling interface of the DSP controller to realize the acquisition of the submodule capacitor voltage and system operating parameters.

Images