Efficient rectification conversion system and control method thereof

Abstract

The invention relates to the technical field of power electronic conversion, and discloses a high-efficiency rectification conversion system and a control method thereof, and the system comprises the following modules: a detection module, an input compatible module, a hybrid power device module, a control module, a thermal management module and a frequency modulation module. The control method comprises the following steps: acquiring input power grid voltage, output current and load state parameters; calculating an input voltage phase difference; dynamically selecting the type and channel configuration of the silicon carbide or gallium nitride power device according to the load rate; outputting a device configuration instruction and a switching frequency control instruction; adjusting the cooling system through the evaluated thermal state of the conversion system; and the control module receives the control instruction output by the control module and is used for implementing a corresponding pulse width modulation strategy. According to the method, a multi-system power grid identification and self-adaptive topology switching strategy is adopted, the technical effect of automatically configuring the optimal rectification architecture under different input types is achieved, and the problems of response delay and poor compatibility of the method in the environment of variable power grid types are solved.

Description

Technical Field

[0001] This invention relates to the field of power electronic conversion technology, specifically to a high-efficiency rectifier conversion system and its control method. Background Technology

[0002] As critical loads such as information infrastructure, data centers, medical systems, and industrial automation become increasingly reliant on power continuity and stability, UPS (Uninterruptible Power Supply) systems, as a core component of backup power, play a crucial role in the entire power supply chain, particularly in scenarios with fluctuating input power and frequent load jumps. The rectifier module must not only achieve high-efficiency AC / DC energy conversion but also ensure rapid dynamic response and adaptability to a wide range of operating conditions.

[0003] In existing high-efficiency rectification and conversion technologies, the rectifier unit mostly adopts a fixed topology based on a three-phase fully controlled bridge or a six-pulse rectifier, and is driven by devices such as IGBTs, SiC, or MOSFETs. This type of structure is often used in industrial UPS and data center applications, and commonly employs fixed-frequency PWM control to achieve voltage regulation. The controller typically runs on a DSP platform, using a PI circuit or voltage feedforward method to complete the regulation.

[0004] However, existing UPS rectification technologies use a fixed input topology, which has weak compatibility under multi-standard input power grid conditions. When the input power grid switches, external intervention or reinitialization may be required. Furthermore, component selection is mostly based on static configuration, making it impossible to flexibly switch power branches or types according to load changes during actual operation, resulting in efficiency deviations in certain operating ranges. Therefore, this invention provides a high-efficiency rectification and conversion system and its control method to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a high-efficiency rectification and conversion system and its control method, which solves the technical limitations that may arise from insufficient adaptability of rectification topology, lack of flexibility in power device configuration, lag in control strategy response, and fixed modulation method in existing technologies.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a high-efficiency rectification and conversion system, comprising the following modules: The detection module is used to collect input grid voltage, output current, and load status parameters; An input compatibility module is used to identify whether the input power grid is single-phase or three-phase and switch to the corresponding rectifier topology. Hybrid power device modules are used to dynamically select between silicon carbide and gallium nitride based on the load level to reduce losses, based on the current rectifier topology; The control module is used to calculate operating parameters based on model predictive control algorithms, output device configuration instructions and switching frequency control instructions, and evaluate the thermal state of the conversion system. The thermal management module is used to adjust the cooling system based on the thermal state of the switching system as assessed by the control module in order to maintain the thermal stability of the system. The frequency modulation module receives control commands output by the control module and is used to implement corresponding pulse width modulation strategies.

[0007] Preferably, the detection module is further used to calculate the input phase difference. When the input voltage phase difference is less than 15 degrees, it is judged as a single-phase input; otherwise, it is judged as a three-phase input, and the judgment result is provided to the input compatibility module.

[0008] Preferably, the input compatibility module specifically includes: The Vienna rectifier module with a three-level structure is used for single-phase input mode; The neutral-point clamping type T-type three-level rectifier module is used for three-phase input mode; Furthermore, a neutral point controllable offset method is used to achieve zero-voltage switching between the two modes, with a transition time of no more than 8 milliseconds.

[0009] Preferably, the hybrid power device module includes: A set of silicon carbide MOSFET devices with a rated voltage of 1200 volts and an on-resistance of 80 milliohms; A set of gallium nitride HEMT devices with a rated voltage of 650 volts and an on-resistance of 25 milliohms.

[0010] Preferably, the control module performs optimization calculations using a model predictive control method, and its control objective satisfies the following constraints: The total harmonic distortion of the output current is no more than 5%; The power factor is not less than 0.99; The output voltage regulation rate is controlled within ±3%. The temperature rise of the device shall not exceed 50 degrees Celsius.

[0011] Preferably, the device configuration instructions output by the control module include: power device type selection instructions, device channel start / stop instructions, drive level setting instructions, and current load level identification signals; the switching frequency control instructions include: PWM frequency setting value, modulation mode selection flag, duty cycle adjustment range, and load rate threshold setting parameters.

[0012] Preferably, the thermal management module includes: The air-cooled submodule regulates the fan speed via a proportional-integral-derivative controller, which takes the device junction temperature as the input signal. The liquid cooling submodule uses flow feedforward control based on the temperature difference between the device case temperature and the coolant temperature to maintain thermal stability.

[0013] Preferably, the control module evaluates the thermal state of the conversion system based on a Foster-type thermal resistance network model, wherein the device junction temperature... The calculation satisfies the following expression: ; in, Device junction temperature; This refers to the temperature of the cooling system. This represents the current power loss of the device. For the first Thermal resistance; For the first First thermal time constant; The order of the selected model; The reference time variable is used for modeling.

[0014] Preferably, the pulse width modulation strategy includes the following parameters: The modulation frequency is set to 20 to 80 kHz for variable frequency control under light load conditions and 160 kHz for fixed frequency control under rated load conditions. The duty cycle is adjusted in real time according to the dynamic changes in the load to match the target output voltage; The modulation mode selection indicator is used to switch between fixed-frequency modulation and variable-frequency modulation; The switching is based on the load rate. When the load rate is between 10% and 30%, variable frequency modulation is automatically enabled. When it exceeds this range, it switches to fixed frequency modulation.

[0015] A highly efficient rectification and conversion control method is also provided, including the following steps: Collect input grid voltage, output current, and load status parameters; Calculate the input voltage phase difference, determine whether the grid input type is single-phase or three-phase, and switch to the corresponding rectifier topology; Based on the current rectification topology, the type of silicon carbide or gallium nitride power devices and the channel configuration are dynamically selected according to the load rate. The system calculates operating parameters based on model predictive control algorithms, outputs device configuration instructions and switching frequency control instructions, and evaluates the thermal state of the conversion system. The cooling system is adjusted based on the thermal state of the conversion system being evaluated; The system receives control commands output by the control module to implement corresponding pulse width modulation strategies.

[0016] This invention provides a high-efficiency rectification and conversion system and its control method. It has the following beneficial effects: 1. This invention employs a multi-system power grid identification and adaptive topology switching strategy, achieving the technical effect of automatically configuring the optimal rectifier architecture under different input types. Compared with existing rectifier schemes that rely on manual settings or only support fixed single-phase / three-phase modes, this invention solves the problems of slow response and poor compatibility in environments with varying power grid types.

[0017] 2. This invention introduces a dynamic scheduling mechanism for hybrid power devices based on load rate, enabling on-demand coordinated control of SiC and GaN devices. Existing technologies often employ only a single power device configuration, making it difficult to achieve a balance between energy efficiency and cost, and easily leading to low efficiency and high device heat generation under light or heavy loads.

[0018] 3. This invention constructs a model predictive control (MPC) framework with thermal state feedback, whose output drive frequency and power configuration are more consistent with real-time operating conditions, and also allows the cooling system to intervene earlier. Compared with traditional PI control or piecewise logic control methods, it solves the hidden dangers of response delay and thermal coupling hysteresis leading to system thermal breakdown.

[0019] 4. This invention employs a layered modulation architecture combined with a variable frequency PWM strategy. The modulation frequency can be dynamically adjusted according to the load, and the duty cycle can quickly track changes in commands. Compared with the fixed frequency, fixed dead-time PWM method used in existing systems, this not only improves control accuracy but also solves the problems of insufficient EMI suppression and modulation lag. Attached Figure Description

[0020] Figure 1 This is a system architecture diagram of the present invention; Figure 2 This is a schematic diagram of the input compatibility module of the present invention; Figure 3 This is a schematic diagram of the hybrid power device module of the present invention; Figure 4 This is a schematic diagram of the thermal management module of the present invention; Figure 5 This is a flowchart of the method steps of the present invention. Detailed Implementation

[0021] The technical solutions in 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.

[0022] Please see the appendix Figure 1 - Appendix Figure 4 This invention provides a high-efficiency rectification and conversion system, comprising the following modules: The detection module is used to collect input grid voltage, output current, and load status parameters; An input compatibility module is used to identify whether the input power grid is single-phase or three-phase and switch to the corresponding rectifier topology. Hybrid power device modules are used to dynamically select between silicon carbide and gallium nitride based on the load level to reduce losses, based on the current rectifier topology; The control module is used to calculate operating parameters based on model predictive control algorithms, output device configuration instructions and switching frequency control instructions, and evaluate the thermal state of the conversion system. The thermal management module is used to adjust the cooling system based on the thermal state of the switching system as assessed by the control module in order to maintain the thermal stability of the system. The frequency modulation module receives control commands output by the control module and is used to implement corresponding pulse width modulation strategies.

[0023] In this embodiment, the rectifier conversion system needs to be adaptable to multi-standard power grid scenarios, automatically identifying the input voltage type and driving the subsequent topology to perform matching switching. To achieve this function, the system first needs to possess a precise and reliable input detection capability to identify parameters such as the phase type, phase configuration, frequency stability, and voltage amplitude of the power grid input. Based on this identification result, subsequent modules of the system can sequentially activate or switch the rectifier path, voltage conversion strategy, and frequency modulation scheme. Therefore, the detection module, as the core function of the system, has a decisive impact on the overall system's response performance and compatibility.

[0024] In this embodiment, the detection module is configured at the input end of the rectifier conversion system and is used to collect and analyze the physical characteristics of the external power grid input signal in real time, including but not limited to the effective voltage value, instantaneous voltage waveform, power grid frequency, and multi-phase voltage phase difference.

[0025] Typically, the detection module integrates at least three voltage sampling channels, corresponding to phases A, B, and C in a three-phase input system. Each channel employs a high-speed analog-to-digital converter (ADC) with a sampling accuracy better than 12 bits and a sampling frequency exceeding 200 kSPS to meet the sampling resolution requirements in complex waveform scenarios.

[0026] In one possible implementation, the detection module includes voltage synchronization analysis logic. This logic obtains the phase difference between adjacent input voltage waveforms by performing Fourier transform (FFT) or sliding correlation analysis. Specifically, the system uses phase A as a reference to calculate the relative phase angle between phase B and phase C. When it detects... If the phase difference is less than 15° (i.e., approximately synchronous) and there is only a single effective voltage component, it is determined to be a single-phase power grid; if the phase difference is close to 120° and there is a stable three-component structure, it is identified as a three-phase input structure.

[0027] Alternatively, the detection module can perform floating tracking of the grid frequency in addition to voltage identification. In some embodiments, the voltage cycle can be extracted using a zero-crossing detection method to estimate the current grid frequency. This value is used to synchronize the internal clock frequency adjustment of the PWM module, thereby ensuring frequency domain alignment between the subsequent rectification and filtering control strategies.

[0028] To improve the system's robustness under disturbance conditions, in this embodiment, the detection module further integrates a fluctuation tolerance judgment algorithm. The system measures voltage amplitude fluctuation rate. (Defined as: the ratio of the maximum to the minimum voltage amplitude per unit time) Short-time averaging is performed. If If the threshold is exceeded (e.g., 15%), the system will trigger a downgrade operation flag, delaying the subsequent switching instruction to avoid system oscillation during the switching process.

[0029] Specifically, to achieve stable identification and judgment, the system introduces the following phase difference estimation formula: ; in, , The first At each sampling point, With phase The instantaneous voltage value; The number of sampling points used for sliding correlation is usually the number of points in one period (e.g., 256 points). Indicates phase With phase The voltage phase difference between them, in radians.

[0030] This formula is used to estimate the input phase difference angle and remains highly robust even under the influence of low-frequency noise.

[0031] In some embodiments, to enhance fault detection capabilities, the detection module further includes a set of undervoltage, overvoltage, and frequency drift judgment logic. When the input voltage remains below 85% of the rated value for more than 3 cycles, or the grid frequency drift exceeds ±3Hz, the system enters a protection state and temporarily suspends the startup of subsequent modules.

[0032] Furthermore, to reduce system standby power consumption, the detection module operates in a low-speed scanning mode in standby mode, retaining only the core voltage channel sampling and performing a full-channel fast sampling cycle once per second. When the voltage amplitude is detected to have returned to normal and the frequency to be stable, the system automatically wakes up the main control unit and enters the initialization state.

[0033] In this invention, the detection module is not only used for input recognition, but in some implementations, it also participates in zero-voltage switching (ZVS) determination and triggering. For example, when an extreme point (valley / peak) of the input voltage is detected, a ZVS trigger signal is issued, which, in conjunction with a neutral point voltage offset strategy, enables smooth switching of the subsequent topology.

[0034] For the input compatibility module, in this embodiment, after the detection module completes the determination of the input power grid type, the system automatically calls the input compatibility module to select a matching rectifier topology based on parameters such as the number of input phases, voltage amplitude and frequency, and to realize necessary topology cascading switching, neutral point management and signal isolation.

[0035] The input compatibility module includes two parallel rectifier paths, suitable for single-phase and three-phase input scenarios respectively. The two rectifier paths are electrically isolated from each other via a bidirectional MOS gate drive isolation circuit, and their activation and switching processes are uniformly scheduled by the main control unit.

[0036] Specifically, under single-phase input conditions, the system activates an improved Vienna rectifier topology. This structure employs a three-level neutral point clamping (NPC) method, which provides a stable bus output under low voltage fluctuations while reducing voltage stress on the switching transistors. In one possible implementation, this structure uses the following circuit configuration: Each bridge arm contains two main control switches and a neutral point clamping diode pair; The input terminal is connected to the phase voltage and the neutral line, and the output is connected to the main bus after DC filtering. The control strategy combines peak current control with voltage balance modulation.

[0037] When a three-phase input condition is detected, the system automatically activates the three-phase T-type three-level rectifier structure. This structure features higher voltage utilization and harmonic suppression capabilities, making it particularly suitable for high-power continuous power supply scenarios. Compared to the traditional two-level structure, the T-type three-level topology introduces an intermediate level conduction path, effectively reducing output voltage ripple.

[0038] In general, the system achieves a smooth transition between topologies through a neutral point potential control unit, avoiding capacitor residual voltage interference or current jumps. The core of this unit adopts a controllable neutral point offset strategy, adjusting the neutral point voltage before switching to approach the expected level of the target structure, thereby switching the main bridge channel under zero differential voltage conditions and ensuring the completion of the ZVS process.

[0039] In some embodiments, the switching logic is described by the following state transition function: ; in, for The topology selection state at any given time; The phase difference between phase voltages (provided by the detection module); This is the effective value of the input voltage; The current power grid frequency; This is the controllable offset completion time for the neutral point; if the limit is exceeded, a switching delay will be applied.

[0040] This state transition mechanism introduces multi-dimensional information for decision-making, does not rely on fixed threshold judgments, and has stronger adaptability.

[0041] As an alternative, to further enhance the system's grid adaptability, the input compatibility module can also integrate polarity detection and harmonic verification logic. Under certain non-standard grid inputs (such as industrial phase-biased systems), the system determines the power supply type by comparing harmonic amplitudes (e.g., the ratio of 3rd to 5th harmonic amplitudes), and then selects whether to enable the bridge filter submodule. This bridge filter module uses an LC parallel topology, with its rated resonant frequency set near the main interfering subharmonics, typically 150Hz or 250Hz, depending on the target grid standard.

[0042] Furthermore, in some implementations, the input compatibility module also has input redundancy detection capabilities. The system samples the voltage status of each input phase in real time. If any phase is lost for an extended period (more than two cycles), it can automatically degrade to single-phase operation mode and issue a main control derating control command to ensure that the rectifier system maintains stable output rather than entering an abnormal state.

[0043] In cold start scenarios, the input compatibility module prioritizes activating the single-phase rectifier path and sets the voltage soft-start rise slope to no more than 30V / ms to prevent input-side surge impact. After the voltage stabilizes, the control module then re-determines whether to enter the three-phase topology state based on real-time input parameters.

[0044] In this invention, the input compatibility module and the detection module achieve an information closed loop, that is, the detection result is directly used as the input topology control signal. At the same time, the topology module feeds back its conduction status, bridge arm current and capacitor voltage status to the main controller in real time during operation for verification, so as to prevent misidentification from causing repeated switching or logic instability.

[0045] In this embodiment, the control module is used for various types of input power supplies and adjustable output load levels. Traditional single-type power devices cannot simultaneously achieve optimal performance in terms of energy efficiency, response speed, and thermal management. Therefore, after the input topology is identified and connected, the system further utilizes a hybrid power device module to achieve dynamic load driving under different operating conditions. As the core power execution unit of the rectifier system, the structure, configuration, and control method of this module directly determine the system's energy conversion efficiency and electromagnetic compatibility.

[0046] The hybrid power device module consists of two types of switching devices with different physical characteristics: silicon-based MOSFET devices and gallium nitride (GaN) high electron mobility transistors (HEMT) devices. Generally, silicon-based MOSFETs have good voltage withstand capability and reliability, making them suitable for high-current, low-frequency switching applications; while GaN devices have advantages such as high breakdown voltage, high switching frequency, and low conduction loss, making them suitable for high-frequency detailed control tasks.

[0047] Specifically, the system schedules the two types of devices to work together based on the load state of the rectifier output stage and the input voltage amplitude through the following logic: When the input voltage is high and the output load is in a steady state, the GaN device is activated to perform the main frequency modulation task to reduce high frequency ripple. When the input voltage fluctuates drastically or the load current changes abruptly, the system prioritizes rectification control using silicon MOSFET devices to achieve strong surge current withstand capability.

[0048] In one possible implementation, the internal topology of the hybrid power device module is a modified T-type three-level bridge structure, wherein each bridge arm is configured with a pair of GaN master switches and a pair of MOSFET auxiliary clamping devices. The topology is as follows: GaN switches are responsible for the main bridge conduction path and perform PWM control at a high frequency (control frequency can reach 100kHz). MOSFET devices are used for bridge arm state holding and intermediate level clamping control, with a low control frequency (typically 20~30kHz). Both are managed by a synchronous controller to control their turn-on timing in order to achieve dynamic power sharing.

[0049] As an alternative, to achieve dynamic device scheduling across multiple operating modes, the system employs a PWM control method based on thermal model feedback. The system continuously monitors the instantaneous junction temperature of each device. The following criteria are used to determine whether to switch the main device on: ; in, This represents the current PWM switching frequency. This is the drain-source voltage of the switching device; This represents the load current flowing through the device; For conduction time; The ambient temperature of the device; For the thermal resistance of the device; This is the junction temperature of the device, used for dynamic scheduling and judgment. This represents the power consumption when the device is turned on.

[0050] Once a certain type of device If the temperature exceeds a safety threshold (e.g., 125°C), the system will reduce its duty cycle and increase the load sharing weight of another type of device, thereby achieving soft scheduling and reducing the risk of heat accumulation.

[0051] In some embodiments, the hybrid power device module also integrates phase reversal control logic to support reverse rectification mode under input polarity errors. By reconfiguring the phase sequence of the bridge arm control signals, the system can achieve input polarity adaptive rectification without changing the hardware connections, improving system deployment flexibility.

[0052] Specifically, in scenarios where the input load is dominated by inductive characteristics, the system introduces an energy feedback submodule to guide the inductive current fed back by the load into the bridge arm GaN device for absorption, and suppress the reverse spike through synchronous PWM waveform and DC bus energy storage capacitor.

[0053] In this embodiment, the gate drive of the GaN device adopts a bidirectional negative voltage pull-down control mode to suppress false turn-on caused by capacitive coupling interference. Its control voltage is generally set in the +5V / -3V range, providing stronger anti-interference capability.

[0054] In some special implementations, the system also considers the switching synchronization mismatch problem caused by input frequency drift. The hybrid power device module locks the grid input frequency through phase-locked loop (PLL) technology and locks the phase synchronously with the controller, thereby ensuring the phase consistency and period stability of the PWM drive signal.

[0055] In this embodiment, to ensure the stable operation of power devices under different operating conditions and to avoid efficiency decline or device aging caused by temperature rise, a thermal management module is set in the system design. This module is used to adjust the operating parameters of the cooling system in real time based on the dynamic evaluation results of the current thermal state of the conversion system by the control module, thereby maintaining the overall thermal stability of the rectifier system.

[0056] The thermal management module maintains an information linkage with the control module. The latter calculates the current thermal stress index based on the device's operating status model and feeds the evaluation results back to the thermal management module to achieve dynamic control of the cooling intensity. The temperature control here is not limited to simple fan start / stop control, but is built on device-level thermal modeling and zoned heat dissipation strategies, possessing higher adjustment accuracy and response flexibility.

[0057] In this embodiment, the thermal management module includes two parts: an air-cooling submodule and a liquid-cooling submodule. Each is controlled independently but possesses collaborative linkage capabilities. The switching or concurrent use of cooling mechanisms is determined by the control module based on the current thermal state of the system, generating control commands without manual intervention.

[0058] The air-cooled submodule includes a PWM fan, whose speed control is performed by a controller based on a proportional-integral-derivative (PID) algorithm. This controller uses the junction temperature of the power devices as a reference. The main feedback input is used to generate the fan drive voltage signal in conjunction with the set target temperature window.

[0059] In some embodiments, a feedforward term may be added to the PID controller to incorporate power losses from the power devices. As a disturbance correction factor, it can respond to junction temperature change trends in advance and shorten the cooling response lag time.

[0060] The liquid-cooled submodule primarily serves the heat dissipation needs under high load or long-term operation conditions. Specifically, this submodule detects the case temperature of the power devices. With coolant inlet temperature Real-time temperature difference Dynamic adjustment of the pump flow rate is performed.

[0061] Under normal circumstances, when When the temperature exceeds the preset range, the system increases the coolant circulation rate by increasing the speed of the liquid pump, thereby achieving a rapid drop in junction temperature.

[0062] In terms of thermal state assessment, the control module uses a Foster-type thermal resistance network model to predict the device junction temperature, and the predicted values ​​are directly used for cooling control strategy decisions. Specifically, the device junction temperature... The calculation formula is: ; in: This indicates the current junction temperature of the device, in degrees Celsius (°C). This indicates the current ambient temperature of the cooling system, in degrees Celsius (°C). The total power loss generated by the device at the current moment is expressed in watts (W), and is usually composed of various loss components such as turn-on loss, turn-off loss, and conduction loss. In the Foster network, the first... Thermal resistance, measured in degrees Celsius per watt (°C / W). Indicates the first The thermal time constant corresponding to the order, in seconds (s), can be derived from... It can be concluded that; The modeling order of the selected Foster network is typically between 3 and 6; the higher the value, the higher the fitting accuracy. The model reference time variable is in seconds (s), representing the time span since the current load conditions were established.

[0063] The model is implemented recursively on a practical embedded controller, updating the junction temperature estimate in real time based on the sampling period. In one specific implementation, to improve computational efficiency, the controller uses a differential approximation method to process the exponential decay term, adapting to the control requirements of hard real-time scenarios.

[0064] As an option, model parameters and This parameter can be obtained by fitting thermal shock test data provided by the device manufacturer, or by actual measurement using the pulse power method during the system debugging phase. This parameter can be programmed into non-volatile memory for long-term use.

[0065] In one possible implementation, the thermal management module has an abnormal temperature rise diagnosis function. When the estimated junction temperature exceeds the set limit (e.g., 125°C) and the duration exceeds the preset threshold (e.g., 3 seconds), a fault signal is issued to trigger the protection mechanism.

[0066] In this embodiment, to achieve refined drive control of power devices under different load conditions and flexibly adjust the system's operating rhythm according to the instructions output by the control module, a frequency modulation module is further provided. This module, acting as the execution end of the control chain, receives modulation parameters and logic instructions from the control module and generates specific pulse width modulation (PWM) waveforms accordingly, driving the power devices to complete the corresponding turn-on and turn-off behaviors.

[0067] The operating state of the frequency modulation module is closely related to the system load level. The preceding control module has optimized the solution based on the model predictive control (MPC) algorithm for the current load state and outputs control commands containing parameters such as frequency value, modulation mode, and duty cycle range. To ensure the accurate implementation of these control strategies, the frequency modulation module needs to have frequency conversion capability, dynamic duty cycle adjustment function, and mode switching logic support.

[0068] In this embodiment, the frequency modulation module dynamically generates control signals based on the PWM frequency setting value and modulation mode identifier issued by the control module.

[0069] Specifically, this module supports a modulation frequency range of 20kHz to 160kHz, covering all operating conditions from light load to rated load. When the load rate is between 10% and 30%, the system employs a variable frequency modulation strategy, dynamically selecting the frequency based on real-time load conditions, typically between 20kHz and 80kHz, to reduce drive losses under frequent switching. In the rated load or high load range, the frequency is fixed at 160kHz to optimize the harmonic structure of the output voltage waveform and reduce output ripple.

[0070] In some embodiments, modulation mode switching is not triggered by a fixed threshold, but rather by a hysteresis interval. For example, the system switches to fixed-frequency mode only when the load rate rises from below 30% to above 35%; if it drops below 25% again, it switches to variable-frequency mode. This effectively avoids EMI fluctuations or unstable device operation caused by frequent switching.

[0071] In one possible implementation, the duty cycle regulation strategy is not only affected by load changes, but also incorporates the bus voltage deviation as a regulation factor. The system can be set to adjust the duty cycle from 30% to 95%. Under grid fluctuations or load step conditions, the frequency modulation module can adjust the duty cycle in tandem to quickly bring the output voltage back to the target value.

[0072] In this embodiment, the frequency modulation module also includes a set of internal registers and lookup table logic for storing modulation reference templates under different operating conditions. For example, when the load rate is 15% and the input voltage is 220V, a reference configuration with a PWM frequency of 60kHz and a duty cycle of 75% is generated by looking up the table. The control module can then fine-tune the output based on this configuration without global optimization, thereby improving the control response speed.

[0073] To ensure compatibility between the modulation process and the system's thermal state, the frequency modulation module can also refer to the temperature rise information provided by the control module. When the device junction temperature approaches the warning value (such as 90% of the thermal capacity limit), the module can appropriately reduce the PWM frequency or narrow the duty cycle variation range to limit the further increase in switching power, thus forming a thermal-electric coupling control closed loop.

[0074] As an alternative, the frequency modulation module is implemented using a digital PWM controller (such as the built-in EPWM module of the TITMS320 series DSP), which has an independent high-resolution counter and dead-time insertion control unit, and can complete the modulation parameter update in nanoseconds.

[0075] Under normal circumstances, this module updates the frequency and duty cycle once per control cycle (typically 10μs) to ensure that the system maintains control continuity under rapid load fluctuations.

[0076] In some extended solutions, a phase modulation unit can be added to the frequency modulation module to enable a phase difference between the PWM signals of different switching transistors, thereby realizing multi-channel soft switching, resonant control or distributed modulation strategies, which can be used to further optimize electromagnetic compatibility or reduce current coupling between parallel branches.

[0077] The efficient rectification and conversion control method described below can be referred to in correspondence with the efficient rectification and conversion system described above.

[0078] Please see the appendix Figure 5 The present invention also provides a high-efficiency rectification and conversion control method, comprising the following steps: S1. Collect input grid voltage, output current and load status parameters; S2. Calculate the input voltage phase difference, determine whether the grid input type is single-phase or three-phase, and switch to the corresponding rectifier topology; S3. Based on the current rectifier topology, dynamically select the type of silicon carbide or gallium nitride power device and the channel configuration according to the load rate; S4. The operating parameters are calculated based on the model predictive control algorithm, and the device configuration instructions and switching frequency control instructions are output. The thermal state of the conversion system is also evaluated. S5. Adjust the cooling system by evaluating the thermal state of the conversion system; S6. Receive the control command output by the control module to implement the corresponding pulse width modulation strategy.

[0079] For step S1, the input voltage waveform, output branch current, and load feedback information of the rectifier front end are collected through a sensor network to form the basic input data required for control calculation. The load status can be obtained by comparing the load current with the preset rated power to determine the load factor, which is used for subsequent dynamic scheduling judgment. For step S2, the presence of three-phase synchronization characteristics in the power grid is determined by detecting the phase difference of the multi-channel voltage signals (e.g., performing Fourier transform on the phase angles of the AB, BC, and CA three-phase voltages). If the phase differences are all close to 120°, it is considered a three-phase input, and the system switches to three-phase rectification mode, with topology switching completed by relay groups or solid-state switches; otherwise, it maintains or switches to single-phase rectification mode. This determination is typically performed once per cycle to accommodate changes in power input type.

[0080] For step S3, the current system power level is determined based on the obtained load rate. Generally, when the load rate is higher than a preset threshold (e.g., 70%), silicon carbide (SiC) devices are activated first to obtain higher current carrying capacity and thermal stability; when the load rate is lower (e.g., below 30%), gallium nitride (GaN) devices are switched to take advantage of their high-frequency and low-loss characteristics. The multi-standard channel configuration can be combined in the manner of "SiC for high-voltage branches and GaN for low-voltage branches", and the specific switching is implemented through a logic control array.

[0081] For step S4, the collected state parameters are input to the MPC controller. Multiple sets of future system state trajectories are constructed within a given prediction window. The optimal switching frequency and device driving strategy are determined by solving for the minimum cost function. The cost function may include indicators such as voltage tracking error, current harmonic distortion rate, and device loss model. Simultaneously, the junction temperature of the power devices is estimated in real time using the embedded thermal modeling module, serving as a basis for thermal state assessment.

[0082] For step S5, if the calculated junction temperature exceeds the set temperature control window, the thermal management module dynamically adjusts the cooling method according to the temperature rise rate and the current heat dissipation capacity, such as increasing the fan PWM duty cycle, increasing the liquid cooling circulation rate, or executing a partial power current limiting strategy in extreme cases to prevent thermal runaway.

[0083] In step S6, the frequency modulation module constructs a corresponding PWM waveform signal to drive the selected power device based on the frequency setting value and duty cycle range issued by the control module. The modulation strategy may include fixed frequency, variable frequency PWM, or space vector modulation (SVPWM), and can automatically adapt the timing of the drive signal and dead-time logic according to the differences in the rectifier topology to ensure the synchronization and stability of the system operation.

[0084] The method in this embodiment can be used to execute the above system embodiment, and its principle and technical effect are similar, so it will not be described again here.

[0085] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-efficiency rectifier-converter system, characterized in that, Includes the following modules: The detection module is used to collect input grid voltage, output current, and load status parameters; An input compatibility module is used to identify whether the input power grid is single-phase or three-phase and switch to the corresponding rectifier topology. Hybrid power device modules are used to dynamically select between silicon carbide and gallium nitride based on the load level to reduce losses, based on the current rectifier topology; The control module is used to calculate operating parameters based on model predictive control algorithms, output device configuration instructions and switching frequency control instructions, and evaluate the thermal state of the conversion system. The thermal management module is used to adjust the cooling system based on the thermal state of the switching system as assessed by the control module in order to maintain the thermal stability of the system. The frequency modulation module receives control commands output by the control module and is used to implement corresponding pulse width modulation strategies.

2. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The detection module is further used to calculate the input phase difference. When the input voltage phase difference is less than 15 degrees, it is judged as a single-phase input; otherwise, it is judged as a three-phase input. The judgment result is then provided to the input compatibility module.

3. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The input compatibility module specifically includes: The Vienna rectifier module with a three-level structure is used for single-phase input mode; The neutral-point clamping type T-type three-level rectifier module is used for three-phase input mode; Furthermore, a neutral point controllable offset method is used to achieve zero-voltage switching between the two modes, with a transition time of no more than 8 milliseconds.

4. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The hybrid power device module includes: A set of silicon carbide MOSFET devices with a rated voltage of 1200 volts and an on-resistance of 80 milliohms; A set of gallium nitride HEMT devices with a rated voltage of 650 volts and an on-resistance of 25 milliohms.

5. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The control module performs optimization calculations using model predictive control methods, and its control objective satisfies the following constraints: The total harmonic distortion of the output current is no more than 5%; The power factor is not less than 0.99; The output voltage regulation rate is controlled within ±3%. The temperature rise of the device shall not exceed 50 degrees Celsius.

6. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The device configuration instructions output by the control module include: power device type selection instructions, device channel start / stop instructions, drive level setting instructions, and current load level identification signals; the switching frequency control instructions include: PWM frequency setting value, modulation mode selection flag, duty cycle adjustment range, and load rate threshold setting parameters.

7. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The thermal management module includes: The air-cooled submodule regulates the fan speed via a proportional-integral-derivative controller, which takes the device junction temperature as the input signal. The liquid cooling submodule uses flow feedforward control based on the temperature difference between the device case temperature and the coolant temperature to maintain thermal stability.

8. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The control module evaluates the thermal state of the conversion system based on a Foster-type thermal resistance network model, where the device junction temperature... The calculation satisfies the following expression: ； in, Device junction temperature; This refers to the temperature of the cooling system. This represents the current power loss of the device. For the first Thermal resistance; For the first First thermal time constant; The order of the selected model; The reference time variable is used for modeling.

9. The high-efficiency rectification and conversion system according to claim 1, characterized in that, The pulse width modulation strategy includes the following parameters: The modulation frequency is set to 20 to 80 kHz for variable frequency control under light load conditions and 160 kHz for fixed frequency control under rated load conditions. The duty cycle is adjusted in real time according to the dynamic changes in the load to match the target output voltage; The modulation mode selection indicator is used to switch between fixed-frequency modulation and variable-frequency modulation; The switching is based on the load rate. When the load rate is between 10% and 30%, variable frequency modulation is automatically enabled. When it exceeds this range, it switches to fixed frequency modulation.

10. A high-efficiency rectification and conversion control method, applied to a high-efficiency rectification and conversion system according to any one of claims 1-9, characterized in that, Includes the following steps: Collect input grid voltage, output current, and load status parameters; Calculate the input voltage phase difference, determine whether the grid input type is single-phase or three-phase, and switch to the corresponding rectifier topology; Based on the current rectification topology, the type of silicon carbide or gallium nitride power devices and the channel configuration are dynamically selected according to the load rate. The system calculates operating parameters based on model predictive control algorithms, outputs device configuration instructions and switching frequency control instructions, and evaluates the thermal state of the conversion system. The cooling system is adjusted based on the thermal state of the conversion system being evaluated; The system receives control commands output by the control module to implement corresponding pulse width modulation strategies.

Images