Real-time detection circuit of AC-DC power conversion circuit and real-time detection method of AC-DC power conversion circuit

Through deep learning real-time detection circuit combined with specific circuit topology, the problems of high detection cost and low control efficiency of traditional AC-DC power conversion circuits are solved, efficient real-time detection and mode switching are achieved, and system efficiency and stability are improved.

CN119093708APending Publication Date: 2024-12-06SHENZHEN AIXIESHENG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411253442.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The detection technology of traditional AC-DC power conversion circuits is costly and difficult to package, and the control method is difficult to adapt to complex application environments, and there is a lack of efficient real-time detection and adjustment methods.

Method used

Deep learning real-time detection circuit is adopted, combined with rectifying filtering circuit, active clamp flyback circuit, zero voltage switching circuit and LLC half-bridge resonant conversion circuit, and adjusting signals are generated by tracking voltage and feedback voltage, using pre-trained deep learning algorithms Determine the excitation current state and switch the working mode to optimize system efficiency.

Benefits of technology

It realizes more efficient and real-time AC-DC power conversion, reduces switching losses and heat generation, simplifies the circuit structure, improves the conversion efficiency and stability of the system under different load conditions, and reduces costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093708A_ABST
    Figure CN119093708A_ABST
Patent Text Reader

Abstract

The invention discloses a real-time detection circuit for an AC-DC (alternating current-direct current) power conversion circuit, which is characterized in that in the AC-DC power conversion circuit, an active clamping flyback circuit ACF is used for controlling energy input and recovering leakage inductance energy; the zero voltage switching circuit ZVS is used for reducing energy loss in the switching process; the LLC half-bridge resonant conversion circuit is used for providing a resonant network and outputting alternating current voltage to the rectifying and filtering circuit; the real-time detection circuit comprises a deep learning real-time detection circuit which is arranged on the primary side of the LLC half-bridge resonant conversion circuit and used for tracking voltage, generating an adjusting signal according to feedback voltage and tracking voltage obtained from direct-current voltage output by the rectification filter circuit and adjusting the value of the tracking voltage according to the adjusting signal and the feedback voltage. And the state of the excitation current is judged according to the change of the tracking voltage, finally, load judgment is performed according to the detected state of the excitation current, and the working mode of the AC-DC power conversion circuit is controlled and switched according to the load.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a real-time detection circuit of an AC-DC power conversion circuit and a real-time detection method of the AC-DC power conversion circuit. Background Art

[0002] The traditional detection technology for AC-DC power conversion circuits mainly detects the excitation current state of the excitation inductor on the primary side of the power converter by adding an additional pin and a sensing resistor. This method not only increases the overall cost of the controller, but also increases the difficulty of packaging. In addition, the traditional control method often adjusts the circuit based on pre-set parameters or simple feedback mechanisms. Such a control method is difficult to adapt to complex application environments.

[0003] There is an urgent need for a detection and adjustment method for an AC-DC power conversion circuit with higher control efficiency and more real-time control. Summary of the invention

[0004] In view of the above problems, the present invention provides an AC-DC power conversion circuit with higher efficiency and more real-time control.

[0005] The present invention provides a real-time detection circuit for an AC-DC power conversion circuit. The AC-DC power conversion circuit comprises: a rectifier filter circuit, an active clamp flyback circuit, a zero voltage switch circuit ZVS and an LLC half-bridge resonant conversion circuit, wherein the rectifier filter circuit is used to convert an AC voltage into a DC voltage; the active clamp flyback circuit ACF is connected to the zero voltage switch circuit ZVS, the zero voltage switch circuit ZVS is connected to the LLC half-bridge resonant conversion circuit, and the LLC half-bridge resonant conversion circuit is connected to the rectifier filter circuit; the active clamp flyback circuit ACF is used to control energy input and recover leakage inductance energy; the zero voltage switch circuit ZVS is used to reduce energy loss during switching. Loss; LLC half-bridge resonant conversion circuit, including a half-bridge switch network and a resonant network, for providing a resonant network, and for outputting an AC voltage to the rectifier and filter circuit; the real-time detection circuit includes a deep learning real-time detection circuit, which is arranged on the primary side of the LLC half-bridge resonant conversion circuit, for tracking the voltage, and generating an adjustment signal based on the feedback voltage and the tracking voltage obtained from the DC voltage output by the rectifier and filter circuit, and adjusting the value of the tracking voltage based on the adjustment signal and the feedback voltage, and then judging the state of the excitation current based on the change of the tracking voltage, and finally making a load judgment based on the detected state of the excitation current, and switching the working mode of the AC-DC power conversion circuit based on the load control.

[0006] In some embodiments, the operating modes include: continuous current mode, quasi-resonant mode, frequency hopping mode and energy-saving mode; wherein, under heavy load conditions, the AC-DC power conversion circuit switches the operating mode to a continuous current mode, in which the current is continuous throughout the switching cycle; under light load conditions, the AC-DC power conversion circuit switches the operating mode to a quasi-resonant mode, in which the switch tube Q4 of the zero voltage switching circuit ZVS is turned on under zero voltage conditions; under very light load or idle conditions, the AC-DC power conversion circuit switches the operating mode to a continuous frequency hopping mode, by skipping certain switching cycles to reduce the number of switching times, thereby further reducing power consumption; in idle or low power consumption states, the AC-DC power conversion circuit switches the operating mode to a continuous energy-saving mode.

[0007] In some embodiments, the deep learning real-time detection circuit tracks the voltage of the AC-DC power conversion circuit in operation through a capacitor disposed on the primary side of the LLC half-bridge resonant conversion circuit.

[0008] In some embodiments, the active clamp flyback circuit ACF includes a clamp switch Q1, a main switch Q2, a clamp capacitor C3, and a transformer including a primary magnetizing inductor L2 and a primary leakage inductor L1; wherein the main switch Q2 is used to control the input of energy; the clamp switch Q1 is used to be connected when the main switch Q1 is turned off, and the transformer is used to store and transfer energy; the clamp capacitor C3 is used to resonate with the primary magnetizing inductor L2 and the primary leakage inductor L1 to recover and transfer energy.

[0009] In some embodiments, the zero voltage switching circuit ZVS includes a switch tube Q4, and the current of the switch tube Q4 lags behind the voltage, so that before the switch tube Q4 is turned on, the voltage across the switch tube Q4 drops to zero.

[0010] In some embodiments, the LLC half-bridge resonant conversion circuit includes a half-bridge switch network and a resonant network; wherein the half-bridge switch network includes: a first MOS switch Q5 and a second MOS switch Q6, the first MOS switch Q5 and the second MOS switch Q6 are alternately turned on to generate the AC voltage required for resonance; the resonant network includes: a resonant inductor L6, resonant capacitors C6 and C7, and an excitation inductor L5, and the resonant network is used to realize a resonant operation during the process of turning on and off the switch tube Q4, so that the switch tube Q4 can be switched under zero voltage conditions.

[0011] In some embodiments, the rectification and filtering circuit includes: a diode D3, a diode D4 and an output capacitor C10, and the rectification and filtering circuit is used to convert the AC voltage output by the LLC half-bridge resonant conversion circuit into a DC voltage.

[0012] In some embodiments, it also includes: a voltage regulation filter circuit, which includes a bus capacitor, a first filter capacitor Cc and a switch tube component Qc, and is used for the control circuit to automatically adjust the state of the switch tube component Qc when the bus voltage reaches a certain threshold during the working cycle of the AC / DC power supply.

[0013] The present invention also discloses a real-time detection method for an AC-DC power conversion circuit, including: tracking the voltage of the AC-DC power conversion circuit; generating an adjustment signal based on the feedback voltage and the tracking voltage obtained from the DC voltage output by the rectifier and filter circuit; adjusting the value of the tracking voltage based on the adjustment signal and the feedback voltage; judging the state of the excitation current based on the change of the tracking voltage; using a pre-trained deep learning algorithm model to perform load judgment based on the detected state of the excitation current, and switching the working mode of the AC-DC power conversion circuit based on the load control.

[0014] In some embodiments, the AC-DC power conversion circuit includes: a rectifier and filter circuit, an active clamped flyback circuit, a zero voltage switching circuit ZVS and an LLC half-bridge resonant conversion circuit, wherein the rectifier and filter circuit is used to convert AC voltage into DC voltage; the active clamped flyback circuit ACF is connected to the zero voltage switching circuit ZVS, the zero voltage switching circuit ZVS is connected to the LLC half-bridge resonant conversion circuit, and the LLC half-bridge resonant conversion circuit is connected to the rectifier and filter circuit; the active clamped flyback circuit ACF is used to control energy input and recover leakage inductance energy; the zero voltage switching circuit ZVS is used to reduce energy loss during the switching process; the LLC half-bridge resonant conversion circuit includes a half-bridge switch network and a resonant network, which is used to provide a resonant network and to output AC voltage to the rectifier and filter circuit.

[0015] The technical solution of the embodiment of the present invention uses a deep learning algorithm to perform real-time detection and control of the power conversion process, which can save energy, reduce heat generation, simplify the circuit and reduce costs. It can also achieve different working modes such as heavy load, light load and idle, and improve the conversion efficiency and stability of the system under different load conditions by optimizing the system's switching logic and parameter settings.

[0016] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference numerals are used throughout the drawings to represent the same components. In the drawings:

[0018] Figure 1 A circuit structure diagram of an AC-DC power conversion circuit in some embodiments of the present invention;

[0019] Figure 2 A circuit structure diagram of a voltage regulating filter circuit in an AC-DC power conversion circuit in some embodiments of the present invention;

[0020] Figure 3 The figure is a flow chart of a real-time detection method for an AC-DC power conversion circuit in some embodiments of the present invention. DETAILED DESCRIPTION

[0021] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the terms "including" and "having" in the specification and claims of the present invention and the above-mentioned drawings and any variations thereof are intended to cover non-exclusive inclusions.

[0023] In the description of the embodiments of the present invention, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present invention, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0024] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments. In the description of the embodiments of the present invention, the term "multiple" refers to more than two (including two)

[0025] This application provides a method for solving the above problems. Figure 1 As shown, the present invention provides a real-time detection circuit for an AC-DC power conversion circuit. The AC-DC power conversion circuit includes: a rectifier filter circuit 10, an active clamp flyback circuit AFC20, a zero voltage switch circuit ZVS 30 and an LLC half-bridge resonant conversion circuit 40, wherein the rectifier filter circuit 10 is used to convert an AC voltage into a DC voltage; the active clamp flyback circuit AFC20ACF is connected to the zero voltage switch circuit ZVS 30, the zero voltage switch circuit ZVS 30 is connected to the LLC half-bridge resonant conversion circuit 40, and the LLC half-bridge resonant conversion circuit 40 is connected to the rectifier filter circuit 10; the active clamp flyback circuit AFCACF 20 is used to control the input of energy and recover leakage inductance energy; the zero voltage switch circuit ZVS 30, used to reduce energy loss during the switching process; the LLC half-bridge resonant conversion circuit 40, including a half-bridge switching network and a resonant network, used to provide a resonant network, and used to output an AC voltage to the rectifier and filter circuit 10; the real-time detection circuit 60 includes a deep learning real-time detection circuit, which is arranged on the primary side of the LLC half-bridge resonant conversion circuit 40, for tracking the voltage, and generating an adjustment signal based on the feedback voltage and the tracking voltage obtained from the DC voltage output by the rectifier and filter circuit 10, and adjusting the value of the tracking voltage based on the adjustment signal and the feedback voltage, and then judging the state of the excitation current based on the change of the tracking voltage, and finally performing load judgment based on the detected state of the excitation current, and switching the working mode of the AC-DC power conversion circuit according to the load control.

[0026] Based on the above description, the deep learning real-time detection circuit substantially includes: a tracking voltage generating unit, an adjustment signal generating unit, a discharge control unit and a detection logic unit. Specifically, the voltage tracking unit is used to generate a tracking voltage according to the working state of the power converter during the period when the second MOS switch Q6 on the primary side of the LLC half-bridge resonant converter is turned on, and use the voltage to track the excitation current state of the primary excitation inductor L2. The adjustment signal generating unit is used to generate an adjustment signal according to the feedback voltage and the tracking voltage, wherein the feedback voltage is obtained from the DC voltage output by the rectifier filter circuit 10; the discharge detection unit is used to discharge the voltage tracking unit according to the feedback voltage and the adjustment signal during the period when the second MOS switch Q6 is turned off, so as to adjust the value of the tracking voltage, so as to enable the tracking voltage unit to track the excitation current state more accurately. The detection logic unit determines the state of the excitation current according to the change of the tracking voltage, and transmits the detected result to the deep learning algorithm processing unit for the control of the power converter; the deep learning algorithm processing unit is used to make a load judgment according to the detected result and the information of the external sensor, and switch the working mode according to the load judgment.

[0027] In some embodiments, the working modes include: continuous current mode, quasi-resonant mode, frequency hopping mode and energy-saving mode; wherein, under heavy load conditions, the AC-DC power conversion circuit switches the working mode to the continuous current mode, in which the current is continuous throughout the switching cycle; under light load conditions, the AC-DC power conversion circuit switches the working mode to the quasi-resonant mode, in which the switch tube Q4 of the zero voltage switching circuit ZVS 30 is turned on under zero voltage conditions; under very light load or idle state, the AC-DC power conversion circuit switches the working mode to the continuous frequency hopping mode, by skipping certain switching cycles to reduce the number of switching times, thereby further reducing power consumption; in idle or low power consumption state, the AC-DC power conversion circuit switches the working mode to the continuous energy-saving mode.

[0028] In some embodiments, the deep learning real-time detection circuit 60 tracks the voltage of the AC-DC power conversion circuit in operation through a capacitor disposed on the primary side of the LLC half-bridge resonant conversion circuit 40 .

[0029] The technical solution of the embodiment of the present invention proposes a new topology of ACF+ZVS+LLC for power conversion. The LLC half-bridge resonant conversion circuit in the new topology can build a resonant network of the entire topology, and achieve high efficiency and high power density at the output stage through the resonant network. At the same time, the LLC low electromagnetic interference and relatively simple circuit structure can be used to increase the anti-interference ability of the overall circuit and reduce the overall complexity of the circuit. On this basis, by using a deep learning algorithm to perform real-time detection and control of the power conversion process, it can be more energy-efficient, reduce heat generation, simplify the circuit and reduce costs, and can achieve different working modes such as heavy load, light load and idle, by optimizing the system's switching logic and parameter settings to improve the conversion efficiency and stability of the system under different load conditions.

[0030] In some embodiments, the active clamp flyback circuit AFC 20 includes a clamp switch Q1, a main switch Q2, a clamp capacitor C3, and a transformer including a primary magnetizing inductor L2 and a primary leakage inductor L1; wherein the main switch Q2 is used to control the input of energy; the clamp switch Q1 is used to be connected when the main switch Q1 is turned off, and the transformer is used to store and transfer energy; the clamp capacitor C3 is used to resonate with the primary magnetizing inductor L2 and the primary leakage inductor L1 to recover and transfer energy.

[0031] In some embodiments, the active clamp flyback circuit ACF 20 includes a clamp switch Q1, a main switch Q2, parasitic capacitors, and a transformer including a primary magnetizing inductor L2 and a primary leakage inductor L1; wherein DQ2 and CQ2 are parasitic body diodes and parasitic output capacitors of the main switch Q2, and DQ1 and CQ1 are parasitic body diodes and parasitic output capacitors of the clamp switch Q1. Figure 1 As shown, one end of the clamp switch Q1 is connected to the capacitor C1, the capacitor C1 is connected to the voltage input terminal, the other end of the clamp switch Q1 is connected to the main switch Q2, and the other end of the main switch Q2 is grounded. The transformer includes a primary side inductor L2 and a leakage inductor L1, which are used to store and transfer energy. The main switch Q2 controls the input of energy, and the clamp switch is turned on when the main switch Q1 is turned off to recover the leakage inductance energy. The clamp capacitor C3 is used to resonate with the primary excitation inductor L2 and the primary leakage inductor L1 to recover and transfer energy.

[0032] In some embodiments, the zero voltage switching circuit ZVS 30 includes a switch tube Q4 , and the current of the switch tube Q4 lags behind the voltage, so that before the switch tube Q4 is turned on, the voltage across the switch tube Q4 drops to zero.

[0033] In some embodiments, the LLC half-bridge resonant conversion circuit 40 includes a half-bridge switch network and a resonant network; wherein the half-bridge switch network includes: a first MOS switch Q5 and a second MOS switch Q6, the first MOS switch Q5 and the second MOS switch Q6 are alternately turned on to generate the AC voltage required for resonance; the resonant network includes: a resonant inductor L6, resonant capacitors C6 and C7, and an excitation inductor L5, and the resonant network is used to achieve resonant operation during the process of turning on and off the switch tube Q4, so that the switch tube Q4 can be switched under zero voltage conditions.

[0034] LLC half-bridge resonant conversion circuit (such as Figure 1 In the circuit, the resonant inductor L6, the resonant capacitors C6 and C7, and the excitation inductor L5 realize resonant operation during the turn-on and turn-off of the switch tube Q4. This operation mode enables the switch tube Q4 to switch under zero voltage conditions (ZVS), thereby significantly reducing the switching loss. The half-bridge switch network (composed of two MOS Q5 and Q6) in the LLC half-bridge resonant conversion circuit implements soft switching technology, which means that the voltage of the switch tube Q4 has dropped to zero before it is turned on, thereby reducing the energy loss during the switching process. Due to the use of the resonant network, the LLC half-bridge resonant conversion circuit can operate at a higher switching frequency, which helps to improve the overall power density.

[0035] In some embodiments, the rectification and filtering circuit 10 includes: a diode D3, a diode D4 and an output capacitor C10. The rectification and filtering circuit 10 is used to convert the AC voltage output by the LLC half-bridge resonant conversion circuit 40 into a DC voltage.

[0036] In some embodiments, it also includes: a voltage regulating filter circuit 50, the voltage regulating filter circuit includes a bus capacitor, a first filter capacitor Cc and a switch tube component Qc, which is used in the working cycle of the AC / DC power supply. When the bus voltage reaches a certain threshold, the control circuit automatically adjusts the state of the switch tube component Qc (a total of 8 ACF working modes, Figure 2 List one) to optimize the voltage utilization of the bus capacitor and further improve the efficiency of the AC-DC power supply. Figure 1 and Figure 2 , Lp and Cr represent the total primary inductance of the transformer (Lp = L2 + L1) and the stray resonant capacitance (Cr = CQ1 + CQ2 + Cto). Lp refers to the sum of the primary magnetizing inductance (L2) and leakage inductance (L1) of the transformer, while Cr is the sum of the parasitic capacitance of the switch tubes Q2 and Q1 and other stray resonant capacitance. Figure 2 In the working mode shown, when Q1 is turned off, Lp and Cr will resonate, the excitation current will charge CQ1 and discharge CQc, thereby achieving a resonant rise in the Vsw voltage.

[0037] The present invention uses advanced thermal management technology according to the detected use environment, optimizes the heat dissipation structure and intelligently adjusts the fan speed to ensure that good heat dissipation performance can be maintained even under high power operation.

[0038] In some embodiments, third-generation semiconductor materials represented by gallium nitride are used to further improve power conversion efficiency and reduce energy consumption by taking advantage of their low switching loss, high frequency, and high power density. The circuit is further optimized based on the energy consumption and heat generation power of the switch tube circuit to further improve system integration and heat dissipation performance.

[0039] In summary, conversion efficiency: Compared with traditional AC-DC power converters, the system conversion efficiency proposed in this patent can be improved by at least 10%. Switching loss and heat dissipation problems: Through the new topology structure and control strategy of ACF+ZVS+LLC, the energy loss and heat dissipation problems in the switching process are effectively reduced. By using deep learning algorithms to detect and control the power conversion process in real time, it can be more energy-efficient, reduce heat generation, simplify circuits and reduce costs, and can achieve different working modes such as heavy load, light load and idle. By optimizing the system's switching logic and parameter settings, it can improve the system's conversion efficiency and stability under different load conditions. Furthermore, due to the use of new design methods and design materials, the integration of the system is greatly enhanced, making the overall system smaller.

[0040] like Figure 3As shown, the present invention also discloses a real-time detection method for an AC-DC power conversion circuit, comprising:

[0041] Step 10: Trace the Voltage of the AC-DC Power Conversion Circuit.

[0042] Step 20: Generate an adjustment signal according to the feedback voltage and the tracking voltage obtained from the DC voltage output by the rectifier and filter circuit.

[0043] Step 30: adjusting the value of the tracking voltage according to the adjustment signal and the feedback voltage; and determining the state of the excitation current according to the change of the tracking voltage.

[0044] Step 40: Use a pre-trained deep learning algorithm model to perform load judgment based on the detected state of the excitation current, and switch the working mode of the AC-DC power conversion circuit according to the load control.

[0045] In some embodiments, the AC-DC power conversion circuit includes: a rectifier and filter circuit 10, an active clamp flyback circuit AFC 20, a zero voltage switching circuit ZVS 30 and an LLC half-bridge resonant conversion circuit 40, wherein the rectifier and filter circuit 10 is used to convert an AC voltage into a DC voltage; the active clamp flyback circuit AFC 20 is connected to the zero voltage switching circuit ZVS 30, the zero voltage switching circuit ZVS 30 is connected to the LLC half-bridge resonant conversion circuit 40, and the LLC half-bridge resonant conversion circuit 40 is connected to the rectifier and filter circuit 10; the active clamp flyback circuit AFC20ACF is used to control the input of energy and recover leakage inductance energy; the zero voltage switching circuit ZVS 30 is used to reduce energy loss during the switching process; the LLC half-bridge resonant conversion circuit 40 includes a half-bridge switch network and a resonant network, which is used to provide a resonant network and to output an AC voltage to the rectifier and filter circuit 10.

[0046] In summary, conversion efficiency: Compared with traditional AC-DC power converters, the system conversion efficiency proposed in this patent can be improved by at least 10%. Switching loss and heat dissipation problems: Through the new topology structure and control strategy of ACF+ZVS+LLC, the energy loss and heat dissipation problems in the switching process are effectively reduced. By using deep learning algorithms to detect and control the power conversion process in real time, it can be more energy-efficient, reduce heat generation, simplify circuits and reduce costs, and can achieve different working modes such as heavy load, light load and idle. By optimizing the system's switching logic and parameter settings, it can improve the system's conversion efficiency and stability under different load conditions.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention. In particular, as long as there is no structural conflict, the various technical features mentioned in each embodiment can be combined in any way. The present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions that fall within the scope of the claims.

Claims

1. A real-time detection circuit for an AC-DC power conversion circuit, the AC-DC power conversion circuit comprising: A rectifier and filter circuit, an active clamp flyback circuit, a zero voltage switch circuit ZVS and an LLC half-bridge resonant conversion circuit, wherein the rectifier and filter circuit is used to convert an AC voltage into a DC voltage; the active clamp flyback circuit ACF is connected to the zero voltage switch circuit ZVS, the zero voltage switch circuit ZVS is connected to the LLC half-bridge resonant conversion circuit, and the LLC half-bridge resonant conversion circuit is connected to the rectifier and filter circuit; the active clamp flyback circuit ACF is used to control the input of energy and recover leakage inductance energy; the zero voltage switch circuit ZVS is used to reduce energy loss during the switching process; the LLC half-bridge resonant conversion circuit includes a half-bridge switch network and a resonant network, which is used to provide a resonant network and to output an AC voltage to the rectifier and filter circuit; The real-time detection circuit includes a deep learning real-time detection circuit, which is arranged on the primary side of the LLC half-bridge resonant conversion circuit, and is used to track the voltage, and generate an adjustment signal based on the feedback voltage and the tracking voltage obtained from the DC voltage output by the rectifier and filter circuit, and adjust the value of the tracking voltage based on the adjustment signal and the feedback voltage, and then judge the state of the excitation current based on the change of the tracking voltage, and finally perform load judgment based on the detected state of the excitation current, and switch the working mode of the AC-DC power conversion circuit according to the load control.

2. The real-time detection circuit for AC-DC power conversion circuit according to claim 1, characterized in that: The working modes include: continuous current mode, quasi-resonant mode, frequency hopping mode and energy-saving mode; wherein, under heavy load conditions, the AC-DC power conversion circuit is switched to a continuous current mode, in which the current is continuous throughout the switching cycle; under light load conditions, the AC-DC power conversion circuit is switched to a quasi-resonant mode, in which the switch tube Q4 of the zero voltage switching circuit ZVS is turned on under zero voltage conditions; under very light load or idle conditions, the AC-DC power conversion circuit is switched to a continuous frequency hopping mode, in which the number of switches is reduced by skipping certain switching cycles, thereby further reducing power consumption; in idle or low power consumption states, the AC-DC power conversion circuit is switched to a continuous energy-saving mode.

3. The real-time detection circuit for AC-DC power conversion circuit according to claim 2, characterized in that: The deep learning real-time detection circuit tracks the voltage of the AC-DC power conversion circuit in operation through a capacitor arranged on the primary side of the LLC half-bridge resonant conversion circuit.

4. The real-time detection circuit for AC-DC power conversion circuit according to claim 3, characterized in that: The active clamp flyback circuit ACF includes a clamp switch Q1, a main switch Q2, a clamp capacitor C3, and a transformer including a primary magnetizing inductor L2 and a primary leakage inductor L1; wherein, The main switch Q2 is used to control the input of energy; the clamp switch Q1 is used to be connected when the main switch Q1 is turned off, and the transformer is used to store and transfer energy; the clamp capacitor C3 is used to resonate with the primary excitation inductance L2 and the primary leakage inductance L1 to recover and transfer energy.

5. The real-time detection circuit for AC-DC power conversion circuit according to claim 4, characterized in that: The zero voltage switching circuit ZVS includes a switch tube Q4, the current of which lags behind the voltage, so that before the switch tube Q4 is turned on, the voltage across the switch tube Q4 drops to zero.

6. The real-time detection circuit for AC-DC power conversion circuit according to claim 5, characterized in that: The LLC half-bridge resonant conversion circuit includes a half-bridge switch network and a resonant network; wherein, The half-bridge switch network includes: a first MOS switch Q5 and a second MOS switch Q6, wherein the first MOS switch Q5 and the second MOS switch Q6 are alternately turned on to generate an AC voltage required for resonance; The resonant network includes: a resonant inductor L6, resonant capacitors C6 and C7, and an excitation inductor L5. The resonant network is used to achieve a resonant operation during the process of turning on and off the switch tube Q4, so that the switch tube Q4 can switch under zero voltage conditions.

7. The real-time detection circuit for AC-DC power conversion circuit according to claim 6, characterized in that: The rectification and filtering circuit comprises: a diode D3, a diode D4 and an output capacitor C10, and the rectification and filtering circuit is used to convert the AC voltage output by the LLC half-bridge resonant conversion circuit into a DC voltage.

8. The real-time detection circuit for AC-DC power conversion circuit according to claim 7, characterized in that: Also includes: A voltage regulating filter circuit, which includes a bus capacitor, a first filter capacitor Cc and a switch tube component Qc, is used to control the circuit to automatically adjust the state of the switch tube component Qc when the bus voltage reaches a certain threshold during the working cycle of the AC / DC power supply.

9. A real-time detection method for an AC-DC power conversion circuit, comprising: Tracking voltage in AC-DC power conversion circuits; generating an adjustment signal according to a feedback voltage and a tracking voltage obtained from a DC voltage outputted by a rectifying and filtering circuit; Adjust the value of the tracking voltage according to the adjustment signal and the feedback voltage; Determine the state of the excitation current based on the change of the tracking voltage; A pre-trained deep learning algorithm mode is used to perform load judgment according to the detected state of the excitation current, and the working mode of the AC-DC power conversion circuit is switched according to the load control.

10. The real-time detection method for AC-DC power conversion circuit according to claim 9, characterized in that: The AC-DC power supply conversion circuit includes: a rectifier and filter circuit, an active clamp flyback circuit, a zero voltage switch circuit ZVS and an LLC half-bridge resonant conversion circuit, wherein the rectifier and filter circuit is used to convert an AC voltage into a DC voltage; the active clamp flyback circuit ACF is connected to the zero voltage switch circuit ZVS, the zero voltage switch circuit ZVS is connected to the LLC half-bridge resonant conversion circuit, and the LLC half-bridge resonant conversion circuit is connected to the rectifier and filter circuit; the active clamp flyback circuit ACF is used to control energy input and recover leakage inductance energy; the zero voltage switch circuit ZVS is used to reduce energy loss during the switching process; the LLC half-bridge resonant conversion circuit includes a half-bridge switch network and a resonant network, which is used to provide a resonant network and to output an AC voltage to the rectifier and filter circuit.