Variable frequency phase-shift hybrid modulation method for dual active bridge single-stage ac / dc converter

Abstract

The application discloses a frequency-shifting hybrid modulation method of a dual active bridge single-stage AC / DC converter, acquires a three-phase power grid voltage interval, a battery voltage interval and a power interval of a target OBC, discretely processes each interval and converts to obtain a plurality of groups of normalized data; based on each group of normalized data and pre-acquired converter circuit parameters, defines five modulation parameters to be solved, and combines Fourier series decomposition to calculate and form a converter circuit model; based on the converter circuit model, an optimization objective function is constructed and constraint conditions are generated, so as to form an optimization mathematical model according to the constraint conditions, and a Lagrange function is constructed to obtain KKT conditions. The application can combine sector odd-even branch interpolation to calculate five optimal modulation parameters under the current working condition and output to the control end of the converter, and at the same time, accurate power regulation and power factor correction are completed, so that the high efficiency performance advantage of the converter is fully exerted.

Description

Technical Field

[0001] This invention relates to the field of modulation technology for dual active bridge single-stage AC / DC converters, and more specifically, to a frequency conversion and phase shift hybrid modulation method for dual active bridge single-stage AC / DC converters. Background Technology

[0002] The on-board charger (OBC) plays a crucial role in electric vehicle charging systems, responsible for converting alternating current (AC) to direct current (DC) to charge the battery. The OBC's charging operation relies on a three-phase OBC circuit topology. The core circuit of the OBC is the three-phase matrix series resonant dual active bridge single-stage AC / DC converter, whose topology and modulation control strategy directly determine the OBC's core performance indicators, such as power factor, efficiency, power density, and overall cost.

[0003] However, due to the large number of control variables and complex operating modes of three-phase matrix series resonant dual active bridge single-stage AC / DC converters, the strong coupling between frequency control and multiple phase-shift control during modulation causes the power transfer characteristics and power factor correction function of the converter to be mutually coupled, making them impossible to characterize explicitly. Furthermore, the current lack of a coordinated control strategy between frequency and multiple phase shifts prevents the full utilization of the converter's high efficiency, making it difficult to simultaneously achieve precise power regulation and power factor correction within the wide input / output voltage and full power operating range required for actual OBC operation.

[0004] Based on this, this application proposes a frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter. Summary of the Invention

[0005] This invention aims to overcome the challenges posed by three-phase matrix series resonant dual active bridge single-stage AC / DC converters, which suffer from numerous control variables, complex operating modes, and strong coupling between frequency control and multiple phase-shift control during modulation. This coupling results in the power transfer characteristics and power factor correction function of the converter being mutually coupled and unable to be directly and explicitly characterized. Furthermore, the current lack of a coordinated control strategy between frequency and multiple phase shifts prevents the full utilization of the converter's high efficiency, making it difficult to simultaneously achieve precise power regulation and power factor correction within the wide input / output voltage and full power operating range required for actual OBC operation.

[0006] To achieve the above objectives, the present invention provides a frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter, comprising: Obtain the three-phase grid voltage range, battery voltage range, and power range of the target OBC, and discretize and convert each range to obtain several sets of standardized data. Based on each set of standardized data and pre-acquired converter circuit parameters, five modulation parameters to be solved are defined, and the converter circuit model is formed by combining Fourier series decomposition. An optimization objective function is constructed based on the converter circuit model, and constraints are generated to form an optimization mathematical model based on the constraints. A Lagrangian function is then constructed to obtain the KKT conditions. For each set of standardized data, configure five initial modulation parameters and an initial Lagrangian function. Combine the optimization mathematical model and KKT conditions to iterate in the SQP algorithm process to obtain the five optimal modulation parameters and generate a lookup table for all sets of standardized data. Obtain the current three-phase grid voltage, current DC side voltage, and current target power of the converter in the target OBC. Calculate the current per-unit data and input it into the lookup table to obtain the corresponding lower and upper bound indices. Based on whether the current per-unit data is located in an odd or even sector, after calling the odd sector or updating the even sector, the five optimal modulation parameters under the corresponding lower and upper bound indices are called to perform interpolation operations to obtain the five optimal modulation parameters of the current per-unit data and output them to the control terminal of the converter.

[0007] Optionally, the process of obtaining the three-phase grid voltage range, battery voltage range, and power range of the target OBC, discretizing each range, and converting it into several sets of standardized data specifically includes: Based on the product specifications of the target OBC, obtain the three-phase grid voltage range, battery voltage range, and power range; By performing discrete processing on the three-phase grid voltage range, battery voltage range, and power range through linear distance segmentation, several three-phase grid voltages, several DC voltages, and several target powers are obtained. Based on the working conditions of the target OBC, the corresponding three-phase grid voltages, DC voltages, and target powers are matched to form several sets of discrete data. Perform per-unit processing on each set of discrete data to obtain several sets of per-unit data, each including per-unit AC voltage, per-unit DC voltage, and per-unit target current.

[0008] Optionally, the step of performing per-unit processing on each set of discrete data to obtain several sets of per-unit data, each including per-unit AC voltage, per-unit DC voltage, and per-unit target current, specifically includes: The phase-locked loop algorithm is invoked to calculate the grid phase of the three-phase grid voltage in each set of discrete data and to determine the sector in which the three-phase grid voltage is located. Based on the three-phase grid voltage and its sector, as well as the corresponding DC voltage and target power, and combined with the pre-acquired converter circuit parameters, calculate the per-unit AC voltage, per-unit DC voltage and per-unit target current. Based on the per-unit AC voltage, per-unit DC voltage, and per-unit target current corresponding to each set of discrete data, several sets of per-unit data are obtained.

[0009] Optionally, the step of calculating the per-unit AC voltage, per-unit DC voltage, and per-unit target current based on the three-phase grid voltage and its sector, as well as the corresponding DC voltage and target power, combined with the pre-acquired converter circuit parameters, specifically includes: Obtain the transformer turns ratio based on the inherent design parameters of the converter. , converter series inductor Resonant capacitor And calculate the resonant frequency. To obtain the converter circuit parameters; Calculate the per-unit AC voltage based on the three-phase grid voltage and its sector. : ; According to DC voltage Calculate the per-unit DC voltage : ; Based on target power Calculate the per-unit target current : In the formula: Represented as a current reference value, and .

[0010] Optionally, based on each set of standardized data and pre-acquired converter circuit parameters, five modulation parameters to be solved are defined, and a converter circuit model is formed by combining Fourier series decomposition. Specifically, this includes: Based on each set of standardized data and converter circuit parameters, the duty cycle of the maximum AC side line voltage within half a switching cycle of the high-frequency square wave on the AC side of the converter in the OBC to be solved is defined. The duty cycle of the AC side intermediate line voltage during half a switching cycle of the high-frequency square wave on the AC side of the converter in the OBC. The duty cycle of the DC-side line voltage during half a switching cycle of the high-frequency square wave on the DC side of the converter. Phase shift parameter representing the proportion of the rising edge of the high-frequency square wave on the DC side from time 0 to the half-switching cycle duration. and per-unit switching frequency ; The high-frequency square wave voltage output from the AC side was analyzed using the Fourier series method. DC-side high-frequency square wave voltage referred to the AC side Perform harmonic expansion and calculate the series resonant inductor current. ; Based on the actual conduction timing of the switching devices in the converter, the series resonant inductor current is... Perform segmentation to obtain the first phase current on the AC side of the converter. Second phase current and the third phase current ; Based on series resonant inductor current Calculate the DC output current of the transformer primary side. This is to form a converter circuit model.

[0011] Optionally, the step of constructing an optimization objective function based on the converter circuit model and generating constraints, forming an optimization mathematical model based on the constraints, and constructing a Lagrangian function to obtain KKT conditions, specifically includes: Optimization target value inductor current is calculated based on converter circuit model. : In the formula: The coefficient of the k-th term in a Fourier series in complex exponential form; The coefficient of the k-th term of the Fourier series in complex exponential form The conjugate of complex numbers; The optimization objective is defined as minimizing the total conduction losses of all switching devices in the converter, combined with the optimization objective value of the inductor current. The optimization objective function is obtained. : In the formula: Let represent the vector of modulation parameters to be determined, and ; This represents taking the minimum value; Constraints are generated based on the converter circuit model: In the formula: Represented as the first phase voltage; Represented as the second phase voltage; Represented as the third-phase voltage; , and All are three-phase line currents; To form an optimized mathematical model; Based on the optimized mathematical model, a Lagrange function is constructed to obtain the KKT conditions.

[0012] Optionally, the step of configuring five initial modulation parameters and an initial Lagrangian function for each set of standardized data, and iterating through the SQP algorithm process using an optimized mathematical model and KKT conditions, yields five optimal modulation parameters and generates a lookup table for all sets of standardized data. Specifically, this includes: Configure five initial modulation parameters for each set of standardized data: In the formula: Represents the initial modulation parameter vector; , , , and These represent the five initial modulation parameters; And the initial Lagrange function; By combining the optimization mathematical model and KKT conditional inputs to the SQP algorithm process, five optimal modulation parameters corresponding to each set of normalized data are obtained to generate a lookup table that includes all sets of normalized data and their corresponding optimal modulation parameters.

[0013] Optionally, the step of obtaining the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC, calculating the current per-unit data, and inputting it into a lookup table to obtain the corresponding lower and upper bound indices specifically includes: Collect the current three-phase grid voltage, current DC side voltage, and current target power of the converter in the target OBC, and perform per-unit processing to obtain the current per-unit data and its sector. Input the current standardized data into the lookup table to locate the floating-point index. Round down the floating-point index to obtain the lower bound index. Obtain the upper bound index based on the index above the lower bound index.

[0014] Optionally, the method of collecting the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC, and performing per-unit processing to obtain the current per-unit data and its sector, also includes: Collect the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC; The power deviation value is obtained based on the difference between the current target power and the target power, and is input into the PI controller to output the power correction value; Based on the sum of the power correction value and the current target power, combined with the current three-phase grid voltage and the current DC side voltage, a per-unit processing is performed to obtain the current per-unit data and its corresponding sector.

[0015] Optionally, the step of performing interpolation operations based on the odd or even sector where the current per-unit data is located, after calling the odd sector or updating the even sector, by calling the corresponding lower and upper bound indices of the five optimal modulation parameters to obtain the five optimal modulation parameters of the current per-unit data and outputting them to the control terminal of the converter, specifically includes: Obtain the sector where the current per-unit data is located, and determine whether it is an odd or even sector; Based on the odd-numbered and even-numbered sectors, the corresponding interpolation algorithm is invoked. If the sector is odd-numbered, the interpolation weight is calculated based on the difference between the floating-point index and the lower bound index. The lookup table is used to extract and perform linear interpolation on the five optimal modulation parameters under each index to obtain the interpolation result. : In the formula: These are represented as the five optimal modulation parameters corresponding to the upper limit index; These are represented as the five optimal modulation parameters corresponding to the lower limit index; If the number of sectors is even, then obtain the updated lower bound index. : In the formula: Represented as the total number of indexes; Indicates the lower bound index; It then calls the limiting function to obtain the updated upper limit index and calculates the interpolation weights based on the difference between the upper limit index and the updated lower limit index. : In the formula: Represented as a floating-point index; This is represented as rounding down; The five optimal modulation parameters corresponding to the updated lower bound index and updated upper bound index are extracted from the lookup table and linear interpolation is performed to obtain the interpolation result. : Based on the interpolation results, obtain the five optimal modulation parameters of the current per-unit data and output them to the control terminal of the converter.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The proposed method for hybrid frequency conversion and phase shift modulation of a dual active bridge single-stage AC / DC converter achieves a unified dimensionless representation of samples under different operating conditions by discretizing and converting the OBC operating range to obtain several sets of standardized data. This reduces the magnitude differences between multiple variables and provides a computational basis for decoupling the coupling relationship between frequency and multiple phase shifts. Furthermore, a converter circuit model is established using Fourier series decomposition to accurately characterize the power transmission relationship under complex operating modes of the converter, achieving an explicit expression of the multi-control quantity action mechanism. Finally, an optimization mathematical model is constructed by establishing an optimization objective function, constraints, and KKT conditions, transforming the operating condition constraints of power regulation and power factor correction into mathematical constraints, and converting the coupling problem of multiple control quantities into a constrained numerical optimization problem. To solve the problem, the SQP algorithm is used iteratively to obtain five optimal modulation parameters corresponding to several sets of standardized data and generate a lookup table. This completes the offline collaborative optimization of frequency and multiple phase shift parameters, resulting in collaborative control parameters suitable for all operating conditions. Furthermore, by collecting the real-time operating conditions of the converter and performing standardized processing, the optimal modulation parameters under the corresponding index of the lookup table are extracted to complete the matching of the actual operating conditions with the offline pre-optimized parameters. Combined with sector odd-even branch interpolation, the five optimal modulation parameters under the current operating conditions are calculated and output to the control terminal of the converter. The decoupled collaborative control quantity is output in real time within a wide input and output voltage range and the full power operating range, while simultaneously completing precise power regulation and power factor correction, giving full play to the high efficiency performance advantage of the converter.

[0017] As can be seen from the above, the technical solution of this invention can effectively solve the problems of the large number of control quantities, complex operating modes, and strong coupling characteristics between frequency control and multiple phase shift control during modulation in three-phase matrix series resonant dual active bridge single-stage AC / DC converters. This results in the mutual coupling between the power transfer characteristics and the power factor correction function of the converter, making it impossible to directly and explicitly characterize them. Furthermore, the current lack of a coordinated control strategy between frequency and multiple phase shifts prevents the full utilization of the converter's high efficiency performance, leading to difficulties in simultaneously achieving precise power regulation and power factor correction functions within the wide input / output voltage and full power operating range required for actual OBC operation.

[0018] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0019] Figure 1This is a schematic flowchart of the frequency conversion phase shift hybrid modulation method of the dual active bridge single-stage AC / DC converter according to an embodiment of the present invention; Figure 2 The diagram shows the three-phase grid sinusoidal waveform of the converter in an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to more fully understand the technical solutions of the present invention, exemplary embodiments of the present invention will be described more comprehensively and in detail below with reference to the accompanying drawings. Obviously, the one or more embodiments of the present invention described below are merely one or more specific ways to implement the technical solutions of the present invention, and are not exhaustive. It should be understood that other ways belonging to a general inventive concept can be used to implement the technical solutions of the present invention, and should not be limited to the embodiments described exemplary. Based on one or more embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] Reference Figure 1 The embodiments of the present invention provide a frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter, including: Step S1: Obtain the three-phase grid voltage range, battery voltage range, and power range of the target OBC, discretize each range, and convert it to obtain several sets of standardized data. Step S2: Based on each set of standardized data and the pre-acquired converter circuit parameters, define five modulation parameters to be solved, and combine Fourier series decomposition to calculate and form a converter circuit model. Step S3: Construct an optimization objective function and generate constraints based on the converter circuit model, form an optimization mathematical model according to the constraints, and construct a Lagrangian function to obtain the Karush-Kuhn-Tucker (KKT) conditions; Step S4: Configure five initial modulation parameters and an initial Lagrangian function for each group of standardized data, and iterate in the process of optimizing the mathematical model and KKT conditional input sequence quadratic programming (SQP) algorithm to obtain the five optimal modulation parameters and generate a lookup table for all groups of standardized data. Step S5: Obtain the current three-phase grid voltage, current DC side voltage, and current target power of the converter in the target OBC, calculate the current per-unit data, and input it into the lookup table to obtain the corresponding lower limit index and upper limit index; Step S6: Based on whether the current per-unit data is located in an odd or even sector, after calling the odd sector or updating the even sector, call the five optimal modulation parameters under the corresponding lower and upper bound indices to perform interpolation operations, so as to obtain the five optimal modulation parameters of the current per-unit data and output them to the control terminal of the converter.

[0022] The proposed method for hybrid frequency conversion and phase shift modulation of a dual active bridge single-stage AC / DC converter achieves a unified dimensionless representation of samples under different operating conditions by discretizing and converting the OBC operating range to obtain several sets of standardized data. This reduces the magnitude differences between multiple variables and provides a computational basis for decoupling the coupling relationship between frequency and multiple phase shifts. Furthermore, a converter circuit model is established using Fourier series decomposition to accurately characterize the power transmission relationship under complex operating modes of the converter, achieving an explicit expression of the multi-control quantity action mechanism. Finally, an optimization mathematical model is constructed by establishing an optimization objective function, constraints, and KKT conditions, transforming the operating condition constraints of power regulation and power factor correction into mathematical constraints, and converting the coupling problem of multiple control quantities into a constrained numerical optimization problem. To solve the problem, the SQP algorithm is used iteratively to obtain five optimal modulation parameters corresponding to several sets of standardized data and generate a lookup table. This completes the offline collaborative optimization of frequency and multiple phase shift parameters, resulting in collaborative control parameters suitable for all operating conditions. Furthermore, by collecting the real-time operating conditions of the converter and performing standardized processing, the optimal modulation parameters under the corresponding index of the lookup table are extracted to complete the matching of the actual operating conditions with the offline pre-optimized parameters. Combined with sector odd-even branch interpolation, the five optimal modulation parameters under the current operating conditions are calculated and output to the control terminal of the converter. The decoupled collaborative control quantity is output in real time within a wide input and output voltage range and the full power operating range, while simultaneously completing precise power regulation and power factor correction, giving full play to the high efficiency performance advantage of the converter.

[0023] Specifically, the converter of the present invention is a three-phase matrix series resonant dual active bridge single-stage AC / DC converter, which includes a three-phase matrix bridge circuit composed of multiple sets of bidirectional switching devices, a transformer with a turns ratio of n:1, etc. Its specific structure and working principle are existing technologies, and the present invention will not elaborate further here.

[0024] In one embodiment, in step S1, the three-phase grid voltage range, battery voltage range, and power range of the target OBC are obtained. Each range is discretized and converted to obtain several sets of standardized data, specifically including: Step S11: Based on the product specifications of the target OBC, obtain the three-phase grid voltage range, battery voltage range, and power range; Step S12: Perform discrete processing on the three-phase grid voltage range, battery voltage range and power range by segmenting by linear distance to obtain several three-phase grid voltages, several DC voltages and several target powers. Match the corresponding three-phase grid voltages, DC voltages and target powers based on the working conditions of the target OBC to form several sets of discrete data. Step S13: Perform per-unit processing on each set of discrete data to obtain several sets of per-unit data, each including per-unit AC voltage, per-unit DC voltage, and per-unit target current.

[0025] In one specific embodiment, in step S13, per-unit processing is performed on each set of discrete data to obtain several sets of per-unit data, each including a per-unit AC voltage, a per-unit DC voltage, and a per-unit target current. Specifically, this includes: Step S131: Call the phase-locked loop algorithm to calculate the grid phase of the three-phase grid voltage in each set of discrete data, and determine the sector where the three-phase grid voltage is located; Step S132: Based on the three-phase grid voltage and its sector, as well as the corresponding DC voltage and target power, and combined with the pre-acquired converter circuit parameters, calculate the per-unit AC voltage, per-unit DC voltage and per-unit target current. Step S133: Based on the per-unit AC voltage, per-unit DC voltage and per-unit target current corresponding to each set of discrete data, obtain several sets of per-unit data.

[0026] In one specific embodiment, in step S131, a phase-locked loop algorithm is invoked to calculate the grid phase of the three-phase grid voltage in each set of discrete data and determine the sector where the three-phase grid voltage is located, specifically including: Step S1311: Call the phase-locked loop algorithm to calculate the grid phase of the three-phase grid voltage in each set of discrete data. : In the formula: It is expressed as the amplitude of the theoretical standard sinusoidal three-phase voltage; Step S1312: Based on grid phase Determine the sector where the three-phase power grid voltage is located.

[0027] Specifically, the phase-locked loop algorithm can be a second-order generalized integrator (SOGI) or a phase-locked loop (PLL), but this invention is not limited thereto.

[0028] It is worth noting that the sector is obtained by subdividing the phase interval of the three-phase grid voltage of the dual active bridge single-stage AC / DC converter into 12 sectors, and the magnitude relationship of the line voltage in each sector is fixed.

[0029] Figure 2 The diagram shows the three-phase grid sinusoidal waveform of the converter in an embodiment of the present invention. Figure 2 In the diagram, the vertical axis represents the phase grid voltage, and the horizontal axis represents the grid phase of the phase grid voltage. The red sine wave represents the voltage of the first phase. The blue sine wave represents the second phase voltage. The green sine wave represents the third-phase voltage. The gray area represents an independent sector, which is divided into 12 sectors from left to right.

[0030] In addition, the correspondence between the maximum line voltage and the intermediate line voltage of the three-phase power grid corresponding to the 12 sectors is as follows: Table 1 in, This is expressed as the line voltage on the AC side of the converter from the first phase to the third phase; This is represented as the line voltage on the AC side of the converter from the second phase to the third phase; This is expressed as the line voltage on the AC side of the converter from the second phase to the first phase; This is expressed as the line voltage on the AC side of the converter from the third phase to the first phase; This is expressed as the line voltage on the AC side of the converter from the third phase to the second phase; It is represented as the line voltage on the AC side of the converter from the first phase to the second phase.

[0031] This allows us to determine the maximum and intermediate line voltages of the three-phase power grid, thus providing a phase basis for per-unit processing.

[0032] In one specific embodiment, in step S132, based on the three-phase grid voltage and its sector, as well as the corresponding DC voltage and target power, and combined with the pre-acquired converter circuit parameters, the per-unit AC voltage, per-unit DC voltage, and per-unit target current are calculated, specifically including: Step S1321: Obtain the transformer turns ratio based on the inherent design parameters of the converter. , converter series inductor Resonant capacitor And calculate the resonant frequency. To obtain the converter circuit parameters; Step S1322: Calculate the per-unit AC voltage based on the three-phase grid voltage and its sector. : ; Step S1323: Based on DC voltage Calculate the per-unit DC voltage : ; Step S1324: Based on the target power Calculate the per-unit target current : In the formula: Represented as a current reference value, and .

[0033] Specifically, in this embodiment, the specific calculation items for the per-unit AC voltage, per-unit DC voltage, and per-unit target current are determined by the maximum line voltage and intermediate line voltage of the sector in which they are located, as detailed in Table 1 above.

[0034] In addition, resonant frequency for: .

[0035] By acquiring several sets of standardized data, the effects of differences in voltage and power levels under different operating conditions can be eliminated, facilitating subsequent optimization calculations of modulation parameters.

[0036] In one embodiment, in step S2, based on each set of standardized data and pre-acquired converter circuit parameters, five modulation parameters to be solved are defined, and a converter circuit model is formed by combining Fourier series decomposition, specifically including: Step S21: Based on each set of standardized data and converter circuit parameters, define the duty cycle of the maximum AC line voltage in the OBC of the converter during half a switching cycle of the high-frequency square wave on the AC side. The duty cycle of the AC side intermediate line voltage during half a switching cycle of the high-frequency square wave on the AC side of the converter in the OBC. The duty cycle of the DC-side line voltage during half a switching cycle of the high-frequency square wave on the DC side of the converter. Phase shift parameter representing the proportion of the rising edge of the high-frequency square wave on the DC side from time 0 to the half-switching cycle duration. and per-unit switching frequency ; Step S22: Analyze the high-frequency square wave voltage output from the AC side using the Fourier series method. DC-side high-frequency square wave voltage referred to the AC side Perform harmonic expansion and calculate the series resonant inductor current. ; Step S23: Based on the actual turn-on timing of the switching devices in the converter, adjust the series resonant inductor current. Perform segmentation to obtain the first phase current on the AC side of the converter. Second phase current and the third phase current ; Step S24: Based on the series resonant inductor current Calculate the DC output current of the transformer primary side. This is to form a converter circuit model.

[0037] Specifically, the duty cycle of the maximum line voltage on the AC side of the converter in the OBC during half a switching cycle of the high-frequency square wave on the AC side. The duty cycle of the AC side intermediate line voltage during half a switching cycle of the high-frequency square wave on the AC side of the converter in the OBC. The duty cycle of the DC-side line voltage during half a switching cycle of the high-frequency square wave on the DC side of the converter. Phase shift parameter representing the proportion of the rising edge of the high-frequency square wave on the DC side from time 0 to the half-switching cycle duration. and per-unit switching frequency The definition can fully characterize the modulation behavior of the five degrees of freedom of a three-phase matrix series resonant dual active bridge AC / DC converter.

[0038] By simultaneously adjusting the maximum line voltage duty cycle and intermediate line voltage duty cycle on the AC side, the line voltage duty cycle on the DC side, the phase shift parameters, and the per-unit switching frequency, the voltage excitation form of the resonant circuit can be flexibly adjusted under three-phase grid input conditions. This enables coordinated control of output power and device conduction losses, providing all the control degrees of freedom to be optimized for subsequent constraint optimization solutions aimed at minimizing conduction losses. This facilitates finding the optimal operating point suitable for each operating condition, thereby reducing the overall conduction loss of the converter and improving the overall efficiency of the OBC.

[0039] Among them, the per-unit switching frequency for: In the formula: It is expressed as the switching frequency, which is given in advance by the OBC hardware design parameters.

[0040] In one specific embodiment, in step S22, the high-frequency square wave voltage output from the AC side is analyzed using the Fourier series method. DC-side high-frequency square wave voltage referred to the AC side Perform harmonic expansion and calculate the series resonant inductor current. Specifically, it includes: Step S221: Analyze the high-frequency square wave voltage output from the AC side using the Fourier series method. DC-side high-frequency square wave voltage referred to the AC side Performing harmonic expansion yields: ; ; In the formula: Let be the Fourier decomposition coefficients, and ; Represented as a unit complex number; Represented as harmonic order; Indicates time; Step S222: Based on the high-frequency square wave voltage output from the AC side and the DC-side high-frequency square wave voltage referred to the AC side Calculate the current of the series resonant inductor. : ; in, Let be the coefficient of the k-th term in a complex exponential Fourier series, and .

[0041] Specifically, the high-frequency square wave voltage output from the AC side is analyzed using Fourier series. and the DC-side high-frequency square wave voltage referred to the AC side Perform harmonic expansion to solve for the series resonant inductor current based on harmonic impedance. This establishes a mathematical mapping relationship between the five modulation parameters and the electrical quantities of the resonant circuit of the converter, providing a reliable basis for circuit calculations and ensuring the accuracy of subsequent calculations of the optimal modulation parameters under the corresponding operating conditions of the OBC.

[0042] In one specific embodiment, in step S23, the series resonant inductor current is adjusted according to the actual turn-on timing of the switching devices in the converter. Perform segmentation to obtain the first phase current on the AC side of the converter. Second phase current and the third phase current Specifically, it includes: Step S231: Based on the actual turn-on timing of the switching devices in the converter, adjust the series resonant inductor current. By performing segmentation, several segments of series resonant inductor current are obtained. ; Step S232: Based on the current of each segment of the series resonant inductor Calculate the first phase current on the AC side of the converter. : ; Step S233: Calculate the phase current of the third phase. : ; Step S234: Calculate the second-phase current based on the inherent constraints of the three-phase circuit. : .

[0043] Specifically, the series resonant inductor current The waveform will vary in segments depending on the turn-on timing of the switching devices, with different switch turn-on times corresponding to different stages of inductor current change. Therefore, the series resonant inductor current... It is a segmented continuous non-sinusoidal waveform, which cannot be solved analytically as a whole. It needs to be divided into multiple segments according to the time nodes of the switch being turned on and off, and the corresponding circuit differential equations need to be established in each segment.

[0044] In one specific embodiment, in step S24, based on the series resonant inductor current... Calculate the DC output current of the transformer primary side. Specifically, it includes: Based on the conservation of magnetomotive force in the ampere-turn balance on both sides of the transformer, and based on the series resonant inductor current... Calculate the DC output current of the transformer primary side. : ; To form a converter circuit model.

[0045] Specifically, the DC output current on the primary side of the transformer It can provide current constraints to ensure that the modulation parameters obtained from subsequent solutions can meet the operating conditions of the target power or target current.

[0046] This results in a system that can accurately characterize the voltage and current constraints of the circuit, providing a reliable theoretical basis for subsequent solution processes.

[0047] In one embodiment, in step S3, an optimization objective function is constructed based on the converter circuit model, and constraints are generated to form an optimization mathematical model according to the constraints. A Lagrangian function is then constructed to obtain the KKT conditions, specifically including: Step S31: Calculate the target inductor current based on the converter circuit model. : In the formula: The coefficient of the k-th term of the Fourier series in complex exponential form The conjugate of complex numbers; Step S32: Define the optimization objective as minimizing the total conduction loss of all switching devices in the converter, and combine this with the optimization objective value of the inductor current. The optimization objective function is obtained. : In the formula: Let represent the vector of modulation parameters to be determined, and ; This represents taking the minimum value; Step S33: Generate constraints based on the converter circuit model: In the formula: , and All are three-phase line currents; To form an optimized mathematical model; Step S34: Based on the optimized mathematical model, construct the Lagrangian function to obtain the KKT conditions.

[0048] Specifically, this embodiment constructs an optimization mathematical model with the goal of minimizing the total conduction loss of the switching devices. This model effectively aligns with the actual working mechanism of a three-phase matrix series resonant dual active bridge single-stage AC / DC converter, accurately reflecting the loss of the switching devices and ensuring that the modulation parameters obtained from subsequent optimization solutions are engineering-feasible. This provides a reliable mathematical model to support the acquisition of a set of modulation parameters with optimal conduction loss.

[0049] In one specific embodiment, in step S34, a Lagrangian function is constructed based on the optimization mathematical model to obtain the KKT conditions, specifically including: Step S341: Construct the Lagrange function based on the optimized mathematical model: In the formula: , , , as well as Let them be the coefficients of the k-th term of the complex exponential Fourier series. Partial derivatives with respect to the five modulation parameters; Step S342: Obtain KKT conditions based on the Lagrangian function. : In the formula: Represented as the first One modulation parameter, and ; ; ; ; ; The coefficient of the k-th term of the Fourier series in complex exponential form The conjugate complex number of the partial derivatives with respect to the five modulation parameters.

[0050] Specifically, obtaining the KKT conditions can transform the constrained loss optimization problem into a set of equations that the optimal solution must satisfy. It clarifies the mathematical criteria that the modulation parameters must satisfy when obtaining the optimal solution, provides a convergence judgment basis for subsequent iterative optimization, facilitates verification of whether the solution is the optimal solution in the feasible region, avoids outputting invalid modulation parameters that do not meet circuit constraints, and ensures that the obtained modulation parameters not only meet the converter circuit operation constraints, but also achieve the optimization goal of minimizing conduction loss.

[0051] In one embodiment, in step S4, five initial modulation parameters and an initial Lagrangian function are configured for each group of standardized data. The optimization mathematical model and KKT conditions are then iterated within the SQP algorithm flow to obtain the five optimal modulation parameters and generate a lookup table for all groups of standardized data. Specifically, this includes: Step S41: Configure the five initial modulation parameters for each set of standardized data: In the formula: Represents the initial modulation parameter vector; , , , and These represent the five initial modulation parameters; And the initial Lagrange function; Step S42: Combine the optimization mathematical model and KKT conditions to iterate in the SQP algorithm process to obtain five optimal modulation parameters corresponding to each set of normalized data, so as to generate a lookup table including all sets of normalized data and the corresponding optimal modulation parameters.

[0052] Specifically, by combining the SQP algorithm with KKT conditions, the five optimal modulation parameters are iteratively solved for all group-standardized data and a lookup table is constructed. This allows the five modulation parameters with optimal conduction loss to be obtained under all operating conditions, providing complete pre-calculated data for subsequent online feedforward control and improving the overall efficiency of the converter.

[0053] In addition, the five initial modulation parameters and the initial Lagrange function are configured by performing an initial configuration on the first per-unit data using empirical values. Then, the five optimal modulation parameters obtained from the preceding per-unit data adjacent to the current per-unit data are selected as the five initial modulation parameters and the initial Lagrange multipliers configured for the current per-unit data to participate in the SQP algorithm process iteration.

[0054] In one embodiment, in step S5, the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC are obtained; the current per-unit data is calculated and input into a lookup table to obtain the corresponding lower limit index and upper limit index, specifically including: Step S51: Collect the current three-phase grid voltage, current DC side voltage, and current target power of the converter in the target OBC, and perform per-unit processing to obtain the current per-unit data and its sector. Step S52: Input the current per-unit data into the lookup table to locate the floating-point index, round down the floating-point index to obtain the lower limit index, and obtain the upper limit index based on the previous index of the lower limit index.

[0055] Specifically, the standardization process for obtaining the current standardized data is the same as the process for obtaining the aforementioned sets of standardized data, and will not be described in detail here.

[0056] By acquiring the current three-phase grid voltage, current DC side voltage, and current target power of the converter and completing per-unit processing and sector identification, and then determining the upper and lower limit indices based on the floating-point index and extracting the five optimal modulation parameters pre-stored in the lookup table, the pre-optimized modulation parameters can be quickly retrieved. This avoids the huge computational overhead brought about by large-scale online real-time numerical optimization, reduces the computing power burden on the controller, and provides reliable original lookup table data for subsequent sector branch linear interpolation, ensuring the response speed of online feedforward control.

[0057] It is worth noting that the current three-phase grid voltage and the current DC-side voltage can be acquired using voltage sensors and current sensors, respectively. The current target power is obtained based on the product of the current three-phase grid voltage and the current DC-side voltage.

[0058] In one specific embodiment, step S51 involves acquiring the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC, and performing per-unit processing to obtain the current per-unit data and its corresponding sector. This also includes: Collect the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC; The power deviation value is obtained based on the difference between the current target power and the target power, and is input into the proportional-integral (PI) controller to output the power correction value; Based on the sum of the power correction value and the current target power, combined with the current three-phase grid voltage and the current DC side voltage, a per-unit processing is performed to obtain the current per-unit data and its corresponding sector.

[0059] Specifically, the target power can be sent from the vehicle's BMS battery management system to the OBC, representing the charging power required by the battery.

[0060] By introducing a PI controller closed loop to obtain the power correction value, the current target power is corrected before participating in the per-unit processing. This can compensate for the steady-state error caused by hardware sampling, suppress the power output deviation caused by grid disturbances and non-ideal characteristics of devices, integrate the closed-loop correction effect into the per-unit input stage, improve the control accuracy of output power under all operating conditions, and ensure the stability of the charging power of the on-board charger.

[0061] In one embodiment, in step S6, based on whether the current per-unit data is located in an odd or even sector, after calling the odd sector or updating the even sector, the five optimal modulation parameters under the corresponding lower and upper bound indices are called to perform interpolation operations to obtain the five optimal modulation parameters of the current per-unit data and output them to the control terminal of the converter. Specifically, this includes: Step S61: Obtain the sector where the current per-unit data is located, and determine whether it is an odd sector or an even sector; Step S62: Based on the odd-numbered and even-numbered sectors, call the corresponding interpolation algorithm. If the sector is odd-numbered, the interpolation weight is calculated based on the difference between the floating-point index and the lower bound index. The lookup table is used to extract and perform linear interpolation on the five optimal modulation parameters under each index to obtain the interpolation result. : In the formula: These are represented as the five optimal modulation parameters corresponding to the upper limit index; These are represented as the five optimal modulation parameters corresponding to the lower limit index; If the number of sectors is even, then obtain the updated lower bound index. : In the formula: Represented as the total number of indexes; Indicates the lower bound index; It then calls the limiting function to obtain the updated upper limit index and calculates the interpolation weights based on the difference between the upper limit index and the updated lower limit index. : In the formula: Represented as a floating-point index; This is represented as rounding down; The five optimal modulation parameters corresponding to the updated lower bound index and updated upper bound index are extracted from the lookup table and linear interpolation is performed to obtain the interpolation result. : ; Step S63: Obtain the five optimal modulation parameters of the current per-unit data based on the interpolation results, and output them to the control terminal of the converter.

[0062] Specifically, by differentiating odd and even sectors and using differentiated indexing and interpolation weighting methods, it can adapt to the data characteristics of a mirrored lookup table under 12 sectors. By using pre-stored optimal modulation parameters for linear interpolation, there is no need to run complex SQP optimization solutions online, which greatly reduces the real-time calculation load of the converter controller. At the same time, it ensures that the modulation parameters obtained by interpolation under different sector operating conditions are accurate and reliable, thereby improving the real-time performance and stability of OBC feedforward control under all operating conditions.

[0063] Furthermore, the specific expression and principle of the limiting function are existing technologies, and will not be elaborated further here.

[0064] In one specific embodiment, in step S63, five optimal modulation parameters of the current per-unit data are obtained based on the interpolation results and output to the control terminal of the converter, specifically including: Based on the interpolation results, five optimal modulation parameters for the current per-unit data are obtained. These parameters are then combined with the current sector input pulse-width modulation (PWM) generator to generate the corresponding PWM drive waveform and output it to the converter's control terminal.

[0065] Specifically, by combining the five optimal modulation parameters with the current sector and feeding them into the PWM generator, a drive waveform that adapts to the current grid phase conditions can be generated, ensuring that the converter operates reliably according to the optimal modulation parameters.

[0066] While one or more embodiments of the present invention have been described above, those skilled in the art will recognize that the present invention can be implemented in any other form without departing from its spirit and scope. Therefore, the embodiments described above are illustrative and not restrictive, and many modifications and substitutions will be apparent to those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter, characterized in that, include: Obtain the three-phase grid voltage range, battery voltage range, and power range of the target OBC, and discretize and convert each range to obtain several sets of standardized data. Based on each set of standardized data and pre-acquired converter circuit parameters, five modulation parameters to be solved are defined, and the converter circuit model is formed by combining Fourier series decomposition. An optimization objective function is constructed based on the converter circuit model, and constraints are generated to form an optimization mathematical model based on the constraints. A Lagrangian function is then constructed to obtain the KKT conditions. For each set of standardized data, configure five initial modulation parameters and an initial Lagrangian function. Combine the optimization mathematical model and KKT conditions to iterate in the SQP algorithm process to obtain the five optimal modulation parameters and generate a lookup table for all sets of standardized data. Obtain the current three-phase grid voltage, current DC side voltage, and current target power of the converter in the target OBC. Calculate the current per-unit data and input it into the lookup table to obtain the corresponding lower and upper bound indices. Based on whether the current per-unit data is located in an odd or even sector, after calling the odd sector or updating the even sector, the five optimal modulation parameters under the corresponding lower and upper bound indices are called to perform interpolation operations to obtain the five optimal modulation parameters of the current per-unit data and output them to the control terminal of the converter.

2. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 1, characterized in that, The process of obtaining the three-phase grid voltage range, battery voltage range, and power range of the target OBC involves discretizing and converting each range to obtain several sets of standardized data, specifically including: Based on the product specifications of the target OBC, obtain the three-phase grid voltage range, battery voltage range, and power range; By performing discrete processing on the three-phase grid voltage range, battery voltage range, and power range through linear distance segmentation, several three-phase grid voltages, several DC voltages, and several target powers are obtained. Based on the working conditions of the target OBC, the corresponding three-phase grid voltages, DC voltages, and target powers are matched to form several sets of discrete data. Perform per-unit processing on each set of discrete data to obtain several sets of per-unit data, each including per-unit AC voltage, per-unit DC voltage, and per-unit target current.

3. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 2, characterized in that, The step of performing per-unit processing on each set of discrete data to obtain several sets of per-unit data, each including per-unit AC voltage, per-unit DC voltage, and per-unit target current, specifically includes: The phase-locked loop algorithm is invoked to calculate the grid phase of the three-phase grid voltage in each set of discrete data and to determine the sector in which the three-phase grid voltage is located. Based on the three-phase grid voltage and its sector, as well as the corresponding DC voltage and target power, and combined with the pre-acquired converter circuit parameters, calculate the per-unit AC voltage, per-unit DC voltage and per-unit target current. Based on the per-unit AC voltage, per-unit DC voltage, and per-unit target current corresponding to each set of discrete data, several sets of per-unit data are obtained.

4. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 3, characterized in that, The calculation of per-unit AC voltage, per-unit DC voltage, and per-unit target current, based on the three-phase grid voltage and its sector, as well as the corresponding DC voltage and target power, combined with pre-acquired converter circuit parameters, specifically includes: Obtain the transformer turns ratio based on the inherent design parameters of the converter. , converter series inductor Resonant capacitor And calculate the resonant frequency. To obtain the converter circuit parameters; Calculate the per-unit AC voltage based on the three-phase grid voltage and its sector. : ； According to DC voltage Calculate the per-unit DC voltage : ； Based on target power Calculate the per-unit target current : In the formula: Represented as a current reference value, and .

5. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 4, characterized in that, Based on each set of standardized data and pre-acquired converter circuit parameters, five modulation parameters to be solved are defined, and a converter circuit model is formed by combining Fourier series decomposition. Specifically, this includes: Based on each set of standardized data and converter circuit parameters, the duty cycle of the maximum AC side line voltage within half a switching cycle of the high-frequency square wave on the AC side of the converter in the OBC to be solved is defined. The duty cycle of the AC side intermediate line voltage during half a switching cycle of the high-frequency square wave on the AC side of the converter in the OBC. The duty cycle of the DC-side line voltage during half a switching cycle of the high-frequency square wave on the DC side of the converter. Phase shift parameter representing the proportion of the rising edge of the high-frequency square wave on the DC side from time 0 to the half-switching cycle duration. and per-unit switching frequency ; The high-frequency square wave voltage output from the AC side was analyzed using the Fourier series method. DC-side high-frequency square wave voltage referred to the AC side Perform harmonic expansion and calculate the series resonant inductor current. ; Based on the actual conduction timing of the switching devices in the converter, the series resonant inductor current is... Perform segmentation to obtain the first phase current on the AC side of the converter. Second phase current and the third phase current ; Based on series resonant inductor current Calculate the DC output current of the transformer primary side. This is to form a converter circuit model.

6. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 5, characterized in that, The process of constructing an optimization objective function and generating constraints based on the converter circuit model, forming an optimization mathematical model based on the constraints, and constructing a Lagrangian function to obtain KKT conditions specifically includes: Optimization target value inductor current is calculated based on converter circuit model. : In the formula: The coefficient of the k-th term in a Fourier series in complex exponential form; The coefficient of the k-th term of the Fourier series in complex exponential form The conjugate of complex numbers; The optimization objective is defined as minimizing the total conduction losses of all switching devices in the converter, combined with the optimization objective value of the inductor current. The optimization objective function is obtained. : In the formula: Let represent the vector of modulation parameters to be determined, and ; This represents taking the minimum value; Constraints are generated based on the converter circuit model: In the formula: Represented as the first phase voltage; Represented as the second phase voltage; Represented as the third-phase voltage; , and All are three-phase line currents; To form an optimized mathematical model; Based on the optimized mathematical model, a Lagrange function is constructed to obtain the KKT conditions.

7. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 6, characterized in that, The process involves configuring five initial modulation parameters and an initial Lagrangian function for each set of standardized data, iterating through the SQP algorithm using an optimized mathematical model and KKT conditions, to obtain the five optimal modulation parameters and generate a lookup table for all sets of standardized data. Specifically, this includes: Configure five initial modulation parameters for each set of standardized data: In the formula: Represents the initial modulation parameter vector; , , , and These represent the five initial modulation parameters; And the initial Lagrange function; By combining the optimization mathematical model and KKT conditional inputs to the SQP algorithm process, five optimal modulation parameters corresponding to each set of normalized data are obtained to generate a lookup table that includes all sets of normalized data and their corresponding optimal modulation parameters.

8. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 7, characterized in that, The process of obtaining the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC, calculating the current per-unit data, and inputting it into a lookup table to obtain the corresponding lower and upper bound indices specifically includes: Collect the current three-phase grid voltage, current DC side voltage, and current target power of the converter in the target OBC, and perform per-unit processing to obtain the current per-unit data and its sector. Input the current standardized data into the lookup table to locate the floating-point index. Round down the floating-point index to obtain the lower bound index. Obtain the upper bound index based on the index above the lower bound index.

9. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 8, characterized in that, The process of acquiring the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC, and performing per-unit processing to obtain the current per-unit data and its corresponding sector, also includes: Collect the current three-phase grid voltage, current DC-side voltage, and current target power of the converter in the target OBC; The power deviation value is obtained based on the difference between the current target power and the target power, and is input into the PI controller to output the power correction value; Based on the sum of the power correction value and the current target power, combined with the current three-phase grid voltage and the current DC side voltage, a per-unit processing is performed to obtain the current per-unit data and its corresponding sector.

10. The frequency conversion and phase shift hybrid modulation method for a dual active bridge single-stage AC / DC converter according to claim 9, characterized in that, The process involves, based on whether the current perimeter data resides in an odd or even sector, performing interpolation operations on the five optimal modulation parameters under the corresponding lower and upper bound indices after odd sector invocation or even sector update, to obtain the five optimal modulation parameters for the current perimeter data and output them to the converter's control terminal. Specifically, this includes: Obtain the sector where the current per-unit data is located, and determine whether it is an odd or even sector; Based on the odd-numbered and even-numbered sectors, the corresponding interpolation algorithm is invoked. If the sector is odd-numbered, the interpolation weight is calculated based on the difference between the floating-point index and the lower bound index. The lookup table is used to extract and perform linear interpolation on the five optimal modulation parameters under each index to obtain the interpolation result. : In the formula: These are represented as the five optimal modulation parameters corresponding to the upper limit index; These are represented as the five optimal modulation parameters corresponding to the lower limit index; If the number of sectors is even, then obtain the updated lower bound index. : In the formula: Represented as the total number of indexes; Indicates the lower bound index; It then calls the limiting function to obtain the updated upper limit index and calculates the interpolation weights based on the difference between the upper limit index and the updated lower limit index. : In the formula: Represented as a floating-point index; This is represented as rounding down; The five optimal modulation parameters corresponding to the updated lower bound index and updated upper bound index are extracted from the lookup table and linear interpolation is performed to obtain the interpolation result. : Based on the interpolation results, obtain the five optimal modulation parameters of the current per-unit data and output them to the control terminal of the converter.

Images