An efficiency optimization method and system for bidirectional l-llc topology steady-state time domain analysis

Abstract

The application discloses an efficiency optimization method and system for steady-state time domain analysis of a bidirectional L-LLC topology, and the efficiency optimization method comprises the following steps: S1, according to the bidirectional L-LLC topology structure, a steady-state time domain analysis model based on three-phase shift frequency modulation is established; S2, according to the combination of three phase shift angles and switch frequency variables, a switch tube soft switch constraint condition is introduced with the minimization of inductance current effective value as an optimization target; S3, according to the excellent characteristics of the genetic algorithm for global optimization of parameters, the optimal modulation parameter combination under a light load working condition is obtained; and S4, according to the optimized modulation parameter combination, the efficiency improvement and soft switch performance optimization of the converter under the light load working condition are realized. The application can significantly improve the steady-state performance and transmission efficiency of the bidirectional L-LLC converter under the application of a switch cabinet power supply.

Description

Technical Field

[0001] This invention relates to the field of power electronic converter modeling, specifically to an efficiency optimization method and system for steady-state time-domain analysis of bidirectional L-LLC topology, which is particularly suitable for bidirectional energy conversion scenarios with wide voltage gain and wide load range, such as providing stable power supply to relays, fans, and monitoring systems in switchgear equipment. Background Technology

[0002] Bidirectional isolated DC-DC converters are widely used in power systems and industrial control switchgear equipment power supply applications due to their advantages such as electrical isolation, bidirectional power flow, and soft-switching capability. Among them, LLC resonant converters, with their excellent soft-switching characteristics, low electromagnetic interference, and high power density, have become one of the mainstream topologies for medium- and high-power bidirectional energy conversion. However, traditional LLC resonant converters have limited voltage gain range and asymmetrical soft-switching characteristics when operating in reverse, which limits their performance in bidirectional applications.

[0003] To address this issue, researchers have proposed adding an auxiliary inductor to the traditional LLC topology to form an L-LLC resonant converter. This topology typically employs frequency conversion modulation under medium to high load conditions, exhibiting good performance. However, under light load conditions, the voltage regulation capability decreases, the switching frequency increases significantly, leading to increased switching losses and a significant reduction in the overall converter efficiency.

[0004] To improve efficiency under light load, existing research often employs phase-shift modulation techniques, such as hybrid modulation combining single-phase-shift modulation and frequency conversion. While this improves performance under light load to some extent, it only has one degree of freedom for phase-shift control, making it difficult to achieve the optimal balance between reducing current stress and achieving full-range soft switching. Three-phase-shift modulation offers more control dimensions, allowing for more flexible adjustment of the current waveform, reducing circulating current losses, and expanding the soft-switching range. Therefore, in bidirectional L-LLC resonant converters, combining three-phase-shift modulation and frequency conversion modulation to establish an effective time-domain steady-state model and achieve high-efficiency operation under light load conditions has become a pressing technical problem to be solved. Summary of the Invention

[0005] The purpose of this invention is to propose an efficiency optimization method based on three-phase-shift frequency conversion modulation and genetic algorithm optimization. Through multi-degree-of-freedom collaborative combination and global optimization, the steady-state performance and transmission efficiency of bidirectional L-LLC converter under light load conditions are significantly improved.

[0006] To achieve the objectives of this invention, the following technical solution is adopted:

[0007] An efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis includes:

[0008] S1. Based on the bidirectional L-LLC topology, establish a steady-state time-domain analysis model based on three-phase-shift frequency conversion modulation;

[0009] S2. Based on the combination of three phase shift angles and switching frequency variables, with the optimization objective of minimizing the effective value of inductor current, soft switching constraints for the switching transistor are introduced.

[0010] S3. Based on the excellent characteristics of genetic algorithm for global parameter optimization, obtain the optimal combination of modulation parameters under light load conditions;

[0011] S4. Based on the optimized modulation parameter combination, the efficiency of the converter under light load conditions and the soft-switching performance are improved.

[0012] In some embodiments, establishing the steady-state time-domain analysis model in step S1 specifically includes:

[0013] S11: The resonant network of the bidirectional L-LLC resonant converter is equivalent to that of the resonant inductor. The resonant cavity is formed by the resonant capacitor Cr and the auxiliary inductor La, and combined with the transformer magnetizing inductance Lm;

[0014] S12: Establish a set of time-domain equations based on the circuit state equations and half-cycle symmetry; select the operating mode under three-phase-shift frequency conversion modulation, and write the voltage and current expressions for each time stage to establish a set of ten-element first-order steady-state time-domain equations characterizing the resonant inductor current and resonant capacitor voltage.

[0015] S13: Normalize the steady-state time-domain equations, set a normalization benchmark, and complete the normalization transformation of circuit variables to obtain a normalized steady-state time-domain analysis model.

[0016] In some embodiments, the ten-element linear steady-state time-domain equation set is as follows:

[0017]

[0018]

[0019]

[0020]

[0021]

[0022] In the formula, , , , , , , , , The instantaneous values ​​of current and voltage are represented by f, and the switching frequency is represented by T. s Indicates the switching period, C r and L r These are the resonant capacitor and the resonant inductor, respectively. D1 represents the phase angle shifted inward from the primary side, D2 represents the phase angle shifted inward from the secondary side, and D0 represents the phase angle shifted outward from the primary and secondary sides.

[0023] In some embodiments, establishing a steady-state time-domain model further includes a normalization process, wherein the normalization reference includes: using the resonant frequency f r Using the switching frequency as the reference, the input voltage V1 as the voltage reference, and the characteristic impedance Z as the reference. r Using impedance as a reference, and normalizing the current and power variables based on this reference, the specific calculation method is as follows:

[0024]

[0025] f s =1 / T s This represents the switching frequency, which is f after normalization. n = f s / f r ;

[0026] I² / n represents the converted output current, which is I after normalization. n =I2 / n / I BASE ;

[0027] i LrRMS This represents the effective value of the inductor current, which is i after normalization. LrRMSn .

[0028] In some embodiments, constructing a modulation parameter optimization model includes: establishing an objective function. The constraints include achieving soft switching of the switching transistor, and the specific formula is as follows:

[0029]

[0030] The g(·) function is used to characterize the critical current required to achieve ZVS, and its specific expression is:

[0031] , where C oss It is the output capacitor of the switching transistor, V ds It is the voltage that the switching transistor withstands when it is turned off, t d It's dead time.

[0032] In some embodiments, a genetic algorithm is used for global optimization, with the algorithm using the normalized effective value i of the resonant inductor current. LrRMSnThe goal is to minimize the modulation parameters, with the primary and secondary side arms ZVS as constraints. During the initialization phase, 100 individuals are randomly generated as combinations of modulation variables, and the maximum number of iterations is set to 200. The population is continuously updated through selection, crossover, and mutation operations, and the optimal modulation parameters are finally obtained after 200 generations of evolution.

[0033] In some embodiments, the global optimization method includes the following steps:

[0034] S10: Set the initial population size N to 1; randomly generate specific values ​​for the optimization variables and initialize the population;

[0035] S20: Calculate i of ZVS s1onn i s4onn i Q1onn and i Q4onn And calculate i LrRMSn ;

[0036] S30: Make a selection, i corresponding to a certain optimization variable. LrRMSn The smaller the value, the easier it is to be passed down.

[0037] S40: Randomly select a pair of optimization variables D0, D1, D2, f n Produce offspring;

[0038] S50: Controls the mutation operation of the corresponding variable for the individual; N=N+1, determine whether N>100 is true; if yes: end;

[0039] No: Proceed to step S20.

[0040] In some embodiments, the modulation method includes: a switching frequency f s Dynamic adjustment is achieved through closed-loop control; the three phase shift angles D1, D2, and D0 are looked up from a pre-stored optimized parameter table based on the load and voltage gain conditions.

[0041] An efficiency optimization system for bidirectional L-LLC topological steady-state time-domain analysis includes a sampling module, a calculation module, an optimization module, and a control execution module;

[0042] The sampling module is used to collect the primary input voltage, secondary output voltage, load current, and electrical parameters of the switching transistors of the bidirectional L-LLC resonant converter in real time.

[0043] The calculation module is electrically connected to the sampling module and is used to establish a steady-state time-domain analysis model based on three-phase-shift frequency conversion modulation, calculate voltage gain, load rate and normalized circuit variables, and construct a modulation parameter optimization model.

[0044] The optimization module is electrically connected to the calculation module and has a built-in genetic algorithm program for global optimization of the modulation parameter optimization model, to obtain the optimal combination of modulation parameters under light load conditions and generate an optimization parameter table.

[0045] The control execution module is electrically connected to the sampling module, calculation module, and optimization module, respectively. It is used to implement frequency conversion three-phase shift hybrid control by combining table lookup method and closed-loop control, generate drive signals to control the switching action of the bidirectional L-LLC resonant converter, and realize efficiency improvement and soft switching performance optimization.

[0046] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the program, implements the efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis.

[0047] The present invention provides an efficiency optimization method and system for bidirectional L-LLC topological steady-state time-domain analysis, which has the following advantages:

[0048] (1) This invention proposes a three-phase-shift frequency conversion hybrid modulation strategy for bidirectional L-LLC resonant converters. By introducing three control degrees of freedom—the inner phase shift angle of the primary side, the inner phase shift angle of the secondary side, and the outer phase shift angle—and combining them with the adjustment of the switching frequency, it provides a theoretical basis for realizing the power transmission of the converter.

[0049] (2) Based on the steady-state time-domain analysis model, with the minimization of the effective value of the resonant inductor current as the optimization objective, and taking into account the current constraint of zero-voltage turn-on of each switch, an optimization model for modulation parameters under light load conditions was constructed. Global optimization was performed using a genetic algorithm, which effectively reduced the current stress and optimized the soft-switching performance across the entire load range.

[0050] (3) The optimized modulation strategy proposed in this invention has strong engineering applicability and can be implemented by combining the lookup table method with frequency closed loop. It is suitable for application scenarios with high requirements for efficiency, power density and bidirectional operation performance, such as electric vehicle on-board chargers and energy storage systems.

[0051] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art based on the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a logic block diagram of an efficiency optimization method for bidirectional L-LLC topology steady-state time-domain analysis according to an embodiment of the present invention.

[0054] Figure 2 This is a topology diagram of the L-LLC converter involved in the embodiments of the present invention.

[0055] Figure 3 This is the ideal equivalent circuit diagram of the L-LLC converter topology involved in the embodiments of the present invention.

[0056] Figure 4 This is the voltage waveform at the midpoint of the primary and secondary bridge arms under three-phase-shift modulation according to an embodiment of the present invention.

[0057] Figure 5 This is a flowchart of the genetic algorithm used in the embodiments of the present invention.

[0058] Figure 6 This is a block diagram of the system control strategy for three-phase-shift frequency conversion modulation according to an embodiment of the present invention.

[0059] Figure 7 This diagram illustrates a comparison between calculated and simulated test values ​​of capacitor voltage and inductor current as a function of time in an embodiment of the present invention, with the switching frequency being 102kHz and the modulation variables (D1, D2, D0) = (0.398, 0.0567, 0.423).

[0060] Figure 8 shows the switching waveforms of the embodiment of the present invention with M=0.73, P=200W, V1=400V, and V2=250V. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0062] In this embodiment, "several" and "more than" refer to two or more. In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0063] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0064] <Example 1>

[0065] like Figures 1 to 7 As shown in the figure, this embodiment provides an efficiency optimization method for bidirectional L-LLC topology steady-state time-domain analysis, which includes the following steps:

[0066] S1. Based on the bidirectional L-LLC topology, a steady-state time-domain analysis model based on three-phase-shift frequency conversion modulation is established. Specifically, the structure diagram of the bidirectional L-LLC resonant converter is shown below. Figure 2 As shown, the resonant network is equivalent to a network consisting of a resonant inductor L. r Resonant capacitor C r With auxiliary inductor L a The resonant cavity is formed, and combined with the transformer magnetizing inductance L m Establish an ideal equivalent circuit diagram (such as...) Figure 3 A set of time-domain equations is established based on the circuit state equations and half-cycle symmetry; the operating mode under three-phase-shift frequency conversion modulation is selected, such as... Figure 4 As shown, the expressions for voltage and current at each time stage are listed respectively, and a system of equations for the resonant inductor current and capacitor voltage is established, specifically the following system of ten linear equations:

[0067]

[0068]

[0069]

[0070]

[0071]

[0072] In the formula, , , , , , , , , The instantaneous values ​​of current and voltage are represented by f, and the switching frequency is represented by T. s Indicates the switching period, C r and L r These are the resonant capacitor and the resonant inductor, respectively. D1 represents the inward phase shift angle of the primary side, D2 represents the inward phase shift angle of the secondary side, and D0 represents the outward phase shift angle of the primary and secondary sides. This technical solution accurately describes the steady-state operating characteristics of the bidirectional L-LLC resonant converter under three-phase-shift frequency modulation, providing a solid foundation for subsequent efficiency optimization. To simplify the model complexity and enhance the generality of the parameters, the model is normalized as follows:

[0073]

[0074] Circuit variables symbol Normalized variables Voltage gain <![CDATA[M=nV2 / V1]]> - Switching frequency <![CDATA[f s =1 / T s ]]> <![CDATA[f n = f s / f r ]]> Converted output current <![CDATA[I2 / n]]> <![CDATA[I n = I2 / n / I BASE ]]> Inductor current RMS value <![CDATA[i LrRMS ]]> <![CDATA[i LrRMSn ]]>

[0075] The normalized reference includes: voltage gain M = nV2 / V1, where n is the transformer turns ratio and V2 is the secondary output voltage; and the resonant frequency f r Using the switching frequency as the reference, the input voltage V1 as the voltage reference, and the characteristic impedance Z as the reference. r This serves as the impedance reference, and the current and power variables are normalized based on this reference. Specifically: f s =1 / T s The switching frequency is represented by fn = fs / fr after normalization; I² / n represents the converted output current, represented by In = I² / n / I. BASE I2 is the secondary output current, I BASE =V1 / Z r The reference current; i LrRMS The normalized variable i represents the effective value of the inductor current. LrRMSn f s f is the actual switching frequency. s =1 / T s T sThe switching cycle is defined. Normalization allows the model analysis to focus on the relative relationships between variables, rather than relying on specific circuit parameter values, facilitating a unified comparison and analysis of converter performance under different parameter configurations.

[0076] S2. Based on the combination of three phase shift angles and switching frequency variables, and with the minimization of the effective value of the inductor current as the optimization objective, a soft-switching constraint for the switching transistor is introduced. Through joint optimization of the three phase shift angles and switching frequency, with the goal of minimizing the effective value of the inductor current and the introduction of a soft-switching constraint, the effective value of the resonant inductor current can be minimized while ensuring zero-voltage / zero-current turn-on and turn-off of the switching transistor across the entire operating range. This significantly reduces conduction losses, decreases device current stress, widens the soft-switching operating range, and improves overall efficiency and power density, while simultaneously achieving optimal comprehensive performance in terms of efficiency, temperature rise, reliability, and dynamic performance. Specifically, constructing the modulation parameter optimization model includes: establishing the objective function... The constraints include achieving soft switching of the switching transistor:

[0077]

[0078] The g(·) function is used to characterize the critical current required to achieve ZVS, and its specific expression is shown below. Where C... oss It is the output capacitor of the switching transistor, V ds It is the voltage that the switching transistor withstands when it is turned off, t d It's dead time.

[0079]

[0080] Through the above technical solution, by introducing the output capacitor C of the switching transistor... oss Turn-off voltage V ds Dead time t d The critical current function g(·) is used to accurately mathematically represent the minimum inductor current required to achieve ZVS, transforming the soft-switching constraint from a qualitative judgment into a quantitative, calculable, and optimizable constraint. By establishing a modulation parameter optimization model that includes the objective function and the ZVS soft-switching constraint, the function g(·) representing the ZVS critical current is used, and the output capacitance C of the switching transistor is introduced. oss Turn-off voltage V ds Dead time t d It can accurately quantify the conditions for soft switching, minimize the effective value of inductor current while ensuring that the switching transistor can reliably achieve zero-voltage turn-on, reduce conduction loss and switching loss, improve overall efficiency and operational reliability, and the optimization results are highly matched with the actual device characteristics, making it highly feasible in engineering.

[0081] 1. S3. Based on the excellent characteristics of genetic algorithms in global parameter optimization, the optimal combination of modulation parameters under light load conditions is obtained. Specifically, a genetic algorithm is used for global optimization, with the algorithm using the normalized effective value of the resonant inductor current i... LrRMSn The objective is to minimize the ZVS of the primary and secondary bridge arms, constrained by the ZVS of the primary and secondary arms. The population size is set to 100, and an initial population is randomly generated, i.e., 100 sets of optimization variable combinations (D0, D1, D2, f) are randomly generated. n The algorithm then enters the iterative optimization phase. In each generation, the algorithm performs the following operations on each individual (i.e., each set of optimization variables): calculates the instantaneous current of the key switching transistor required to achieve ZVS under its corresponding operating condition, and further solves for the effective value of the resonant inductor current under that set of variables. This effective value serves as the individual's fitness, guiding the "selection" operation. That is, individuals with lower fitness (lower effective current value) have a higher probability of being retained and participating in reproduction. Afterward, the algorithm randomly selects two parent individuals through a "crossover" operation, exchanging some of their variables to generate new offspring individuals; and supplements this with a "mutation" operation, which randomly changes the value of a certain variable in the offspring with a small probability to maintain population diversity. The mutation operation is not performed on all individuals, but rather with a very low probability (e.g., 0.01-0.1), and only randomly selected individuals will mutate. This process is repeated until the set 200 generations of iterations are completed. Finally, the algorithm outputs the individual with the best fitness throughout the entire evolutionary process, which corresponds to the optimal combination of variables (D0, D1, D2, f) that achieves ZVS and minimizes the effective value of the inductor current. n The specific flowchart is as follows: Figure 5 As shown, the global optimization method includes the following steps:

[0082] S10: Set the initial population size N to 1; randomly generate specific values ​​for the optimization variables and initialize the population;

[0083] S20: Calculate i of ZVS s1onn i s4onn i Q1onn and i Q4onn And calculate i LrRMSn ;

[0084] S30: Make a selection, i corresponding to a certain optimization variable. LrRMSn The smaller the value, the easier it is to be passed down.

[0085] S40: Randomly select a pair of optimization variables D0, D1, D2, f n Produce offspring;

[0086] S50: Controls the mutation operation of the corresponding variable for the individual; N=N+1, determine whether N>100 is true; if yes: end;

[0087] No: Proceed to step S20.

[0088] The above technical solution employs a genetic algorithm to globally optimize the modulation parameters. With the goal of minimizing the effective value of the normalized resonant inductor current and constrained by the ZVS of the primary and secondary bridge arms, and through an initial population of 100 and a 200-generation iterative "selection-crossover-mutation" evolutionary process, it can traverse the entire solution space of multiple variable combinations and output the globally optimal parameter combination while ensuring reliable zero-voltage turn-on of the switching transistors. This effectively avoids local optima traps. Simultaneously, mutation operations maintain population diversity, ensuring the optimization results are adaptable to a wide range of operating conditions. Furthermore, the algorithm process is quantifiable and controllable, and the output parameters can be directly implemented in engineering. Ultimately, this achieves a dual reduction in converter conduction and switching losses, significantly improving overall efficiency and operational reliability.

[0089] S4. Based on the optimized modulation parameter combination, the converter's efficiency is improved and soft-switching performance is optimized under light load conditions. Specifically, the modulation strategy implementation includes: switching frequency f s The system employs a combination of closed-loop dynamic adjustment of the switching frequency and three-phase-shift output lookup based on load and voltage gain conditions. The switching frequency is adjusted in real-time via closed-loop control to ensure output voltage stability and dynamic response, while the three phase-shift angles obtain their optimal values ​​from a pre-stored optimized parameter table based on load and voltage gain conditions. This approach eliminates the need for complex online calculations, significantly reducing controller computational overhead. Simultaneously, it ensures the system operates reliably in a soft-switching state with minimal effective resonant inductor current across the entire operating range, effectively reducing losses, improving efficiency and power density. The control structure is simple, stable, and highly feasible in engineering applications.

[0090] like Figure 7 The figures show the numerical curves of the resonant inductor current and resonant capacitor voltage as a function of time. It can be seen that the calculation results and simulation results of the model proposed in this embodiment are basically consistent. The red and green solid lines represent the time-domain model expression curves established in this embodiment, while the blue and yellow dashed lines represent the simulation curves obtained using the PLECS simulation software. It can be seen that the waveforms of the two models match well, verifying the correctness of the numerical solution method.

[0091] Figure 8 shows the measured waveforms of the switching device under the conditions of M=0.73, P=200W, V1=400V, and V2=250V. Taking the primary side bridge arm as an example, Figure 8a shows the gate-source voltage v when S1 is turned on. gs and drain-source voltage v ds Waveform, Figure 8b The gate-source voltage v when S4 is turned on gs and drain-source voltage vds Waveform. As can be seen from the figure, the gate drive signals of both switching devices are at their respective drain-source voltages v. ds The fact that the switch was triggered after dropping to zero indicates that the body diode freewheeling and junction capacitance discharge processes had been completed before the switch was turned on, confirming the realization of ZVS.

[0092] Furthermore, the accuracy of the established time-domain model was verified using the effective value of the inductor current. The data in the table below shows that the results obtained by this method agree well with the PLECS simulation results, verifying the accuracy of the numerical calculation.

[0093] <![CDATA[D1]]> <![CDATA[D2]]> <![CDATA[D0]]> <![CDATA[MATLAB calculation result I LrRMS (A)]]> <![CDATA[PLECS simulation result I LrRMS (A)]]> Error Δ 0.1 0.1 0.2 4.8830 4.8962 0.27% 0.4 0.04 0.45 4.2442 4.2546 0.24% 0 0 0.2 4.6518 4.6643 0.27% 0.2 0.2 0.2 3.7201 3.7298 0.26%

[0094] In summary, this embodiment establishes a steady-state time-domain model using two-port network theory and basic circuit principles, combined with frequency conversion technology under three-phase modulation. Furthermore, this model ensures a certain level of accuracy, as verified through simulation and calculation. The model can accurately calculate the resonant inductor current and resonant capacitor voltage under various conditions.

[0095] <Example 2>

[0096] In this embodiment, the parts that are the same as in Embodiment 1 are given the same reference numerals, and the same text descriptions are omitted.

[0097] Compared to Embodiment 1, this embodiment provides an efficiency optimization system for steady-state time-domain analysis of a bidirectional L-LLC topology, including a sampling module, a calculation module, an optimization module, and a control execution module. The sampling module is used to collect the primary-side input voltage, secondary-side output voltage, load current, and switching transistor electrical parameters of the bidirectional L-LLC resonant converter in real time. The calculation module, electrically connected to the sampling module, is used to establish a steady-state time-domain analysis model based on three-phase-shift frequency conversion modulation, calculate voltage gain, load rate, and normalized circuit variables, and construct a modulation parameter optimization model. The optimization module, electrically connected to the calculation module, has a built-in genetic algorithm program for global optimization of the modulation parameter optimization model, obtaining the optimal modulation parameter combination under light load conditions and generating an optimized parameter table. The control execution module, electrically connected to the sampling module, calculation module, and optimization module respectively, is used to implement frequency conversion three-phase-shift hybrid control through a combination of table lookup and closed-loop control, generating drive signals to control the switching transistor operation of the bidirectional L-LLC resonant converter, achieving efficiency improvement and soft-switching performance optimization.

[0098] Through the above technical solution, a modular and hierarchical efficiency optimization system architecture was constructed. By coordinating the sampling module, calculation module, optimization module, and control execution module, a complete closed loop from data acquisition, model building, global optimization to control execution was achieved. The sampling module acquires the converter's operating status and device characteristic parameters in real time, providing an accurate data foundation for subsequent calculations and optimizations. The calculation module establishes a steady-state time-domain analysis model based on two-port network theory, completing normalization processing and optimizing the model construction. The optimization module uses a genetic algorithm to perform offline global optimization, generating an optimization parameter table covering a wide range of operating conditions. The control execution module achieves real-time application of the optimal modulation strategy with low computational overhead through a combination of table lookup and closed-loop control. Each module has clearly defined functions and interfaces, facilitating hardware and software co-design and engineering deployment. It also supports online updates and iterative optimization of the parameter table, exhibiting good scalability and adaptability.

[0099] Specifically, the sampling module includes a voltage sensor, a current sensor, and a parameter storage unit. The voltage sensor is configured at both the primary input and secondary output terminals to acquire the input voltage V1 and output voltage V2 in real time, respectively. The current sensor is configured at the secondary output terminal to acquire the load current I2 in real time. The parameter storage unit is used to pre-store the output capacitance C of the switching transistor. oss Dead time t d Key electrical parameters are recorded. The sampling module performs data acquisition at fixed intervals to ensure the accuracy of model calculations and optimized control.

[0100] The computation module incorporates a steady-state time-domain analysis model and a normalization operation unit. The steady-state time-domain analysis model, based on the two-port network method described in Example 1, establishes analytical expressions for the resonant inductor current and resonant capacitor voltage under three-phase-shift frequency modulation. The normalization operation unit uses the resonant frequency fr, input voltage V1, and characteristic impedance Zr as references to normalize the voltage gain M, switching frequency fs, converted output current I2 / n, and effective value of inductor current i. LrRMS After normalization, we get f n I n i LrRMSn Dimensionless variables. The calculation module further constructs a modulation parameter optimization model to normalize the effective value of the inductor current i. LrRMSn Using ZVS critical current constraints as the objective function, a standard optimization problem is formed, providing a mathematical basis for solving the problem using genetic algorithms.

[0101] The optimization module is implemented using an embedded processor or a host computer, and includes a built-in genetic algorithm library and optimization parameter table memory. The genetic algorithm program uses a population size of 100 and 200 iterations, and supports adjusting algorithm parameters according to actual operating conditions. The optimization module receives voltage gain M and load rate information transmitted from the calculation module, and calls the genetic algorithm to apply the modulation parameter combination (D0, D1, D2, f) under the current operating condition. n Perform a global optimization and output the result that makes i... LrRMSn The minimum optimal solution that satisfies the ZVS constraint.

[0102] The control execution module includes a digital signal processor (DSP) or a field-programmable gate array (FPGA), a drive circuit, and power switching transistors. Through the above modular system architecture, this embodiment achieves efficiency optimization and soft-switching performance improvement of the bidirectional L-LLC resonant converter under light load conditions.

[0103] <Example 3>

[0104] In this embodiment, the parts that are the same as in Embodiments 1 and 2 are given the same reference numerals, and the same text descriptions are omitted.

[0105] Compared to Embodiment 1 and Embodiment 2, this embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the efficiency optimization method for bidirectional L-LLC topology steady-state time-domain analysis.

[0106] The memory stores the program code, optimization parameter table, historical operating data, and system configuration information required to implement the efficiency optimization method, including but not limited to the steady-state time-domain analysis model algorithm for three-phase-shift frequency conversion modulation, the genetic algorithm program, the optimal modulation parameter combination lookup table for each operating condition, the switching transistor electrical parameter database, and the real-time sampling data buffer. The processor adopts a multi-core architecture design. Its core tasks include: executing real-time processing and filtering algorithms for sampled data; running the steady-state time-domain analysis model to quickly calculate voltage gain and load rate; calling the genetic algorithm for offline or online modulation parameter optimization; and generating precise drive signals to achieve hybrid three-phase-shift frequency conversion control. The processor has a built-in hardware floating-point unit and a dedicated digital signal processing instruction set to support complex matrix operations and iterative optimization calculations, ensuring that a single optimization solution is completed within milliseconds to meet real-time control requirements.

[0107] The electronic device further includes a communication interface module for data interaction with external host computers, monitoring systems, or other converter units. It supports multiple communication protocols such as CAN bus, RS485, and Ethernet, facilitating collaborative optimization and centralized monitoring during multi-machine parallel operation. The communication interface module also supports remote firmware upgrades and online updates of optimization parameter tables, enabling the system to continuously iterate and optimize strategies based on actual operating experience or device aging characteristics.

[0108] The electronic device also includes a human-machine interface, equipped with a display screen and input devices, for real-time display of the converter's operating status parameters, including input and output voltages, load current, switching frequency, phase shift angle values, estimated efficiency under current operating conditions, and soft-switching status indicators. Operators can manually set specific operating conditions for offline optimization calculations through the human-machine interface, or adjust key parameters of the genetic algorithm such as population size, number of iterations, crossover probability, and mutation probability to adapt to different optimization accuracy and computational timeliness requirements.

[0109] Through the above technical solution, this embodiment implements the efficiency optimization method in the form of an integrated hardware and software electronic device. By coordinating memory, processor, communication interface, and human-machine interface, a fully functional, reliable, and easily deployable intelligent control platform is constructed. This electronic device can be embedded as an independent controller in bidirectional L-LLC resonant converter products, or it can be expanded into a general-purpose resonant converter optimization control platform. It supports flexible configuration of various topologies and modulation strategies, possessing good engineering practical value and market prospects.

[0110] <Example 4>

[0111] In this embodiment, the parts that are the same as in embodiments one to three are given the same reference numerals, and the same text descriptions are omitted.

[0112] Compared to Embodiments 1 to 3, this embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis.

[0113] The computer-readable storage media includes two categories: non-volatile storage media and volatile storage media. Non-volatile storage media, in the form of flash memory, electrically erasable programmable read-only memory, or solid-state drives, are used to permanently store the core program code of the efficiency optimization method, the full-condition parameter lookup table generated by the genetic algorithm, the switching device characteristic database, and system configuration files. This type of media has power-loss data retention capability, ensuring that optimization parameters and control strategies are quickly restored after device restart, without needing to re-execute time-consuming offline optimization calculations. Volatile storage media, in the form of synchronous dynamic random access memory or double data rate synchronous dynamic random access memory, serve as temporary data cache and working memory during program execution, storing real-time sampled data, intermediate calculation results, and population individual information for the current iteration. Its high-speed read / write characteristics support the large amount of data exchange and parallel computing requirements during the genetic algorithm iteration process.

[0114] Through the above technical solution, this embodiment embeds the efficiency optimization method into a readable storage medium in the form of a computer program product, realizing the replicability, portability, and large-scale deployment of the technical solution. This storage medium can be circulated as an independent product, adaptable to various hardware platforms and processor architectures, lowering the technical implementation threshold and facilitating the rapid promotion of the efficiency optimization technology for bidirectional L-LLC resonant converters in a wide range of application fields such as new energy power generation, electric vehicles, and energy storage systems, generating significant economic and social benefits.

[0115] <Example 5>

[0116] In this embodiment, the parts that are the same as in embodiments one to four are given the same reference numerals, and the same text descriptions are omitted.

[0117] Compared to Embodiments 1 to 4, this embodiment provides a bidirectional L-LLC resonant converter, including a primary-side full-bridge circuit, an L-LLC resonant network, a high-frequency isolation transformer, and a secondary-side full-bridge circuit. It can realize bidirectional energy flow between the primary and secondary sides, and is suitable for wide voltage gain and bidirectional energy interaction application scenarios such as on-board charging of electric vehicles and energy storage systems.

[0118] Through the above technical solution, this embodiment deeply integrates the efficiency optimization method and system with the bidirectional L-LLC resonant converter body, constructing a high-efficiency, high-power-density, and high-reliability power conversion device. Under light load conditions, this converter significantly reduces conduction and switching losses, expands the soft-switching operating range, and improves overall efficiency and dynamic response performance through a frequency conversion three-phase-shift hybrid modulation strategy optimized by a genetic algorithm. It is particularly suitable for applications with stringent efficiency and reliability requirements, such as electric vehicle on-board chargers, energy storage converters, and fuel cell energy management systems, and possesses good market promotion value and industrialization prospects.

[0119] In the above embodiments one to five, during the working process, depending on the different working environments, some of the technical implementation methods of embodiments one to five can be combined or replaced.

[0120] The technical principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that these descriptions are merely for explaining the principles of the present invention and should not be construed as limiting the scope of protection of the present invention in any way. Based on this explanation, those skilled in the art can conceive of other specific embodiments or equivalent substitutions of the present invention without creative effort, and all such embodiments will fall within the scope of protection of the present invention.

Claims

1. An efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis, characterized in that, include: S1. Based on the bidirectional L-LLC topology, establish a steady-state time-domain analysis model based on three-phase-shift frequency conversion modulation; S2. Based on the combination of three phase shift angles and switching frequency variables, with the optimization objective of minimizing the effective value of inductor current, soft switching constraints for the switching transistor are introduced. S3. Based on the excellent characteristics of genetic algorithm for global parameter optimization, obtain the optimal combination of modulation parameters under light load conditions; S4. Based on the optimized modulation parameter combination, the efficiency of the converter under light load conditions and the soft-switching performance are improved.

2. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 1, characterized in that, Step S1, establishing the steady-state time-domain analysis model, specifically includes: S11: The resonant network of the bidirectional L-LLC resonant converter is equivalent to that of the resonant inductor. The resonant cavity is formed by the resonant capacitor Cr and the auxiliary inductor La, and combined with the transformer magnetizing inductance Lm; S12: Establish a set of time-domain equations based on the circuit state equations and half-cycle symmetry; select the operating mode under three-phase-shift frequency conversion modulation, and write the voltage and current expressions for each time stage to establish a set of ten-element first-order steady-state time-domain equations characterizing the resonant inductor current and resonant capacitor voltage. S13: Normalize the steady-state time-domain equations, set a normalization benchmark, and complete the normalization transformation of circuit variables to obtain a normalized steady-state time-domain analysis model.

3. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 2, characterized in that, The ten-element linear steady-state time-domain equation set is as follows: In the formula, , , , , , , , , The instantaneous values ​​of current and voltage are represented by f, and the switching frequency is represented by T. s Indicates the switching period, C r and L r These are the resonant capacitor and the resonant inductor, respectively. D1 represents the phase angle shifted inward from the primary side, D2 represents the phase angle shifted inward from the secondary side, and D0 represents the phase angle shifted outward from the primary and secondary sides.

4. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 3, characterized in that, Establishing a steady-state time-domain model also includes normalization processing, with the normalization reference including: using the resonant frequency f r Using the switching frequency as the reference, the input voltage V1 as the voltage reference, and the characteristic impedance Z as the reference. r Using impedance as a reference, and normalizing the current and power variables based on this reference, the specific calculation method is as follows: f s =1 / T s This represents the switching frequency, which is f after normalization. n = f s / f r ; I² / n represents the converted output current, which is I after normalization. n =I2 / n / I BASE ; i LrRMS This represents the effective value of the inductor current, which is i after normalization. LrRMSn .

5. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 4, characterized in that, Constructing a modulation parameter optimization model includes: establishing the objective function. The constraints include achieving soft switching of the switching transistor, and the specific formula is as follows: The g(·) function is used to characterize the critical current required to achieve ZVS, and its specific expression is: , where C oss It is the output capacitor of the switching transistor, V ds It is the voltage that the switching transistor withstands when it is turned off, t d It's dead time.

6. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 5, characterized in that, A genetic algorithm is used for global optimization. The algorithm uses the normalized effective value of the resonant inductor current i. LrRMSn The objective is to minimize the ZVS of the primary and secondary side arms, with the ZVS of the primary and secondary side arms as constraints. In the initialization phase, 100 individuals are randomly generated as combinations of modulation variables, and the maximum number of iterations is set to 200. The population is continuously updated through selection, crossover, and mutation operations, and the optimal modulation parameters are finally obtained after 200 generations of evolution.

7. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 6, characterized in that, The global optimization method includes the following steps: S10: Set the initial population size N to 1; randomly generate specific values ​​for the optimization variables and initialize the population; S20: Calculate i of ZVS s1onn i s4onn i Q1onn and i Q4onn And calculate i LrRMSn ; S30: Make a selection, i corresponding to a certain optimization variable. LrRMSn The smaller the value, the easier it is to be passed down. S40: Randomly select a pair of optimization variables D0, D1, D2, f n Produce offspring; S50: Controls the mutation operation of the corresponding variable for the individual; N=N+1, determine whether N>100 is true; if yes: end; No: Proceed to step S20.

8. The efficiency optimization method for bidirectional L-LLC topological steady-state time-domain analysis according to claim 7, characterized in that, Modulation methods include: switching frequency f s Dynamic adjustment is achieved through closed-loop control; the three phase shift angles D1, D2, and D0 are looked up from a pre-stored optimized parameter table based on the load and voltage gain conditions.

9. An efficiency optimization system for bidirectional L-LLC topological steady-state time-domain analysis, characterized in that, It includes a sampling module, a calculation module, an optimization module, and a control execution module; The sampling module is used to collect the primary input voltage, secondary output voltage, load current, and electrical parameters of the switching transistors of the bidirectional L-LLC resonant converter in real time. The calculation module is electrically connected to the sampling module and is used to establish a steady-state time-domain analysis model based on three-phase-shift frequency conversion modulation, calculate voltage gain, load rate and normalized circuit variables, and construct a modulation parameter optimization model. The optimization module is electrically connected to the calculation module and has a built-in genetic algorithm program for global optimization of the modulation parameter optimization model, to obtain the optimal combination of modulation parameters under light load conditions and generate an optimization parameter table. The control execution module is electrically connected to the sampling module, calculation module, and optimization module, respectively. It is used to implement frequency conversion three-phase shift hybrid control by combining table lookup method and closed-loop control, generate drive signals to control the switching action of the bidirectional L-LLC resonant converter, and realize efficiency improvement and soft switching performance optimization.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the efficiency optimization method for steady-state time-domain analysis of bidirectional L-LLC topology as described in any one of claims 1 to 8.

Images