High-power rectification resonance control method and system for quick charger

Abstract

The invention provides a high-power rectification resonance control method and system for a quick charger, relates to the technical field of rectification resonance, and aims to obtain direct-current bus voltage, adjust output power through frequency modulation of a full-bridge LLC resonant converter, and perform voltage stability control adjustment in combination with feedback current data. Charging curve data analysis is carried out to acquire charging key nodes and resonance frequency nodes, device state data acquisition and analysis are carried out, a frequency conversion control loop is constructed, resonance frequency adjustment is carried out, after adjustment is completed, abnormal nodes are identified and combined analysis is carried out, optimal parameters of each load segment are acquired, and voltage fluctuation data are acquired. The method comprises the steps of carrying out fluctuation influence analysis, carrying out fluctuation suppression by adjusting a switching frequency and phase shift angle combination of a full-bridge LLC resonant converter, carrying out calculation through the full-bridge LLC resonant converter to obtain an equivalent voltage control instruction, carrying out mode buffer switching and obtaining mode control switching data, and realizing rectification resonance accurate control of the charger.

Description

Technical Field

[0001] This invention proposes a high-power rectifier resonant control method and system for fast chargers, relating to the field of rectifier resonant control technology, specifically to the field of high-power rectifier resonant control technology for fast chargers. Background Technology

[0002] The charging scheme of full-bridge LLC resonant converters is widely used due to their high efficiency. However, traditional schemes often employ single-parameter regulation, which easily deviates from the optimal resonance range under wide load conditions, leading to large efficiency fluctuations. Furthermore, grid voltage fluctuations can easily cause bus voltage deviations, further reducing charging stability. Abnormal state handling often relies on single-point threshold judgments, lacking combined analysis and early prediction of multiple abnormal nodes, resulting in delayed fault response and potential device damage or charging interruption. Charging mode switching often experiences voltage or current overshoot due to the lack of buffered control, and the disconnect between the front-stage rectification and the rear-stage LLC regulation makes it difficult to balance voltage matching and power transfer efficiency. Existing control methods are mostly based on offline simulation or static parameters, lacking real-time data-driven dynamic optimization capabilities and failing to adapt to complex and changing charging conditions. Summary of the Invention

[0003] This invention provides a high-power rectified resonant control method and system for a fast charger to solve the above-mentioned problems:

[0004] This invention proposes a high-power rectified resonant control method and system for a fast charger, the method comprising:

[0005] S1. Obtain the DC bus voltage through a three-phase full-bridge rectifier circuit, adjust the output power through frequency modulation of the full-bridge LLC resonant converter, and combine the feedback current data to perform voltage stability control and obtain voltage control data. Based on the voltage control data, perform charging curve data analysis to obtain the key charging node and resonant frequency node, and then obtain the control target node.

[0006] S2. Collect and analyze device status data according to the control target node, obtain device status analysis data, and then construct a frequency conversion control loop to adjust the resonant frequency. After the adjustment is completed, identify abnormal nodes and perform combined analysis to obtain load efficiency adjustment combined data.

[0007] S3. Obtain the optimal parameters for each load segment based on the load efficiency adjustment combination data, collect voltage fluctuation data, perform fluctuation impact analysis, suppress fluctuations by adjusting the combination of switching frequency and phase shift angle of the full-bridge LLC resonant converter, calculate the equivalent voltage control command through the full-bridge LLC resonant converter, adjust the switching drive signal of the full-bridge LLC resonant converter through PI control, perform mode buffer switching, and obtain mode control switching data.

[0008] Further, S1 includes:

[0009] The grid voltage is converted into DC bus voltage through a three-phase full-bridge rectifier circuit;

[0010] The output power is adjusted by frequency modulation through a full-bridge LLC resonant converter to obtain output power adjustment data.

[0011] The output current data of the full-bridge LLC resonant converter is collected based on the output power adjustment data to obtain the feedback current data;

[0012] A voltage and current dual feedback system is established by combining feedback current data with DC bus voltage.

[0013] For the voltage and current dual feedback system, the feedback current data is checked for deviation to obtain current deviation determination information;

[0014] The output of the full-bridge LLC resonant converter is adjusted based on the current deviation judgment information to obtain voltage control data.

[0015] Further, S1 includes:

[0016] Voltage and current data are collected through voltage control data acquisition and battery charging curve data are generated.

[0017] Feature analysis and annotation of battery charging curve data were performed to obtain multiple types of key charging nodes;

[0018] Efficiency analysis and annotation of battery charging curve data were performed to obtain multiple resonant frequency nodes;

[0019] The control target node is determined based on the key charging node and the resonant frequency node.

[0020] Further, S2 includes:

[0021] Based on the control target node, device status data is acquired to obtain device status acquisition data;

[0022] Data is collected from the fast charger to obtain the charger's collected data;

[0023] By combining data collected from the charger with data collected from the device status, device status analysis data is obtained for the fast charger.

[0024] The optimal operating point in the resonant region is determined based on the device state analysis data.

[0025] Construct a frequency conversion control loop based on the optimal operating point in the resonant region and the control target node;

[0026] The driving frequency of the full-bridge LLC resonant converter is adjusted by comparing the optimal resonant frequency node until the preset target frequency range is reached, thereby controlling the frequency conversion control loop to match the optimal resonant state.

[0027] Further, S2 includes:

[0028] When the frequency conversion control loop matches the optimal resonance state, abnormal data identification is performed on the device state analysis data to obtain abnormal identification data, and abnormal state nodes are marked according to the abnormal identification data.

[0029] Multiple abnormal state nodes are combined to obtain abnormal state combination information;

[0030] An abnormal state analysis command is triggered based on the abnormal state combination information to perform abnormal state analysis and obtain abnormal state analysis data.

[0031] A three-dimensional matching database consisting of resonant frequency, efficiency, and load is constructed based on the data collected from the charger.

[0032] Multiple resonant frequency combinations are obtained from the three-dimensional matching database to obtain multiple resonant frequency combination data.

[0033] Load efficiency analysis is performed on multiple resonant frequency combinations to obtain the optimal load efficiency combination.

[0034] The abnormal state analysis data and the optimal combination data of load efficiency ratio are analyzed to obtain the combined corresponding data.

[0035] Adjust the abnormal state node according to the combined data to obtain load efficiency adjustment combination data.

[0036] Furthermore, the step of triggering an anomaly analysis instruction based on the anomaly state combination information, performing anomaly state analysis, and obtaining anomaly state analysis data includes:

[0037] Establish a digital twin model, and use the digital twin model to virtually map the fast charger to obtain charger mapping data;

[0038] The abnormal state combination information is input into the digital twin model to obtain fault prediction data for a preset switching cycle;

[0039] A control instruction set is generated based on the preset switching cycle fault prediction data;

[0040] By combining the control instruction set with abnormal state analysis data, abnormal state analysis data can be obtained.

[0041] Further, S3 includes:

[0042] The input voltage fluctuation data and the output voltage fluctuation data of the full-bridge LLC resonant converter are obtained by combining load efficiency adjustment data; the ratio of the input voltage fluctuation data to the resonant frequency fluctuation data in the resonant region is obtained to obtain the input voltage influence coefficient; the ratio of the output voltage fluctuation data of the full-bridge LLC resonant converter to the resonant frequency fluctuation data in the resonant region is obtained to obtain the output voltage influence coefficient; the average value of the input voltage influence coefficient and the output voltage influence coefficient is obtained to obtain the comprehensive influence coefficient; a mode buffer control command is triggered based on the comprehensive influence coefficient; a mode buffer coefficient is generated based on the mode buffer control command; mode buffer control is performed based on the mode buffer coefficient to obtain mode buffer control data.

[0043] Further, S3 includes:

[0044] Generate voltage regulation commands based on mode buffer control data;

[0045] The target value of the bus voltage is obtained based on the output power demand according to the voltage regulation command;

[0046] The target value of the bus voltage and the actual value of the bus voltage are compared by a PI controller to obtain the bus voltage comparison result.

[0047] The rectifier PWM duty cycle adjustment signal is obtained based on the bus voltage comparison results.

[0048] The output of the rectifier circuit is adjusted according to the rectifier PWM duty cycle adjustment signal, thereby obtaining the bus voltage regulation data.

[0049] A voltage regulation command is generated based on the mode buffer control data. Based on the voltage regulation command, the target voltage gain required by the full-bridge LLC resonant converter is calculated using the output power demand and the current input voltage. The target voltage gain is compared with the actual voltage gain calculated using the output voltage and the bus voltage via a PI controller to obtain a gain comparison result. The switching frequency or phase shift angle adjustment signal of the full-bridge LLC resonant converter is obtained based on the gain comparison result. The output of the LLC converter is adjusted according to the switching frequency or phase shift angle adjustment signal to obtain output voltage regulation data.

[0050] Further, S3 includes:

[0051] After the output voltage is regulated, the fast charger performs charging stage data analysis to obtain charging stage analysis data; the charging mode is switched based on the charging node analysis data to obtain mode switching data; mode switching buffer control is performed based on the mode switching data to obtain mode switching buffer control data; and mode control is switched based on the mode switching buffer control data to obtain mode control switching data.

[0052] Furthermore, the system includes:

[0053] The stabilization and regulation module is used to obtain the DC bus voltage through the three-phase full-bridge rectifier circuit, regulate the output power through the full-bridge LLC resonant converter, and perform voltage stabilization control and regulation in combination with feedback current data to obtain voltage control data. Based on the voltage control data, the charging curve data is analyzed to obtain the key charging nodes and resonant frequency nodes, and then the control target nodes are obtained.

[0054] The anomaly analysis module is used to collect and analyze device status data based on the control target node, obtain device status analysis data, and then construct a frequency conversion control loop to adjust the resonant frequency. After the adjustment is completed, anomaly nodes are identified and combined analysis is performed to obtain load efficiency adjustment combination data.

[0055] The fluctuation switching module is used to obtain the optimal parameters for each load segment based on the load efficiency adjustment combination data, collect voltage fluctuation data, perform fluctuation impact analysis, suppress fluctuations by adjusting the combination of switching frequency and phase shift angle of the full-bridge LLC resonant converter, calculate the equivalent voltage control command through the full-bridge LLC resonant converter, and perform mode buffer switching by adjusting the switching drive signal of the full-bridge LLC resonant converter through PI control to obtain mode control switching data.

[0056] The beneficial effects of this invention are as follows: This method realizes closed-loop control of the entire process from voltage acquisition, power regulation, target determination, loop construction, anomaly handling, fluctuation suppression, and mode switching. It solves the problems of disconnect between power regulation and voltage stability, inaccurate fluctuation response, and abrupt mode switching in traditional control, improves the coordination and reliability of high-power fast charging process, and adapts to the high-efficiency charging needs under different load scenarios. Attached Figure Description

[0057] Figure 1 A schematic diagram of a high-power rectifier resonant control method for a fast charger;

[0058] Figure 2 This is a schematic diagram of current deviation feedback adjustment;

[0059] Figure 3 A schematic diagram for analyzing the control target. Detailed Implementation

[0060] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0061] In one embodiment of the present invention, a high-power rectifier resonant control method and system for a fast charger is proposed, the method comprising:

[0062] S1. Obtain the DC bus voltage through a three-phase full-bridge rectifier circuit, adjust the output power through frequency modulation of the full-bridge LLC resonant converter, and combine the feedback current data to perform voltage stability control and obtain voltage control data. Based on the voltage control data, perform charging curve data analysis to obtain the key charging node and resonant frequency node, and then obtain the control target node.

[0063] S2. Collect and analyze device status data according to the control target node, obtain device status analysis data, and then construct a frequency conversion control loop to adjust the resonant frequency. After the adjustment is completed, identify abnormal nodes and perform combined analysis to obtain load efficiency adjustment combined data.

[0064] S3. Obtain the optimal parameters for each load segment based on the load efficiency adjustment combination data, collect voltage fluctuation data, perform fluctuation impact analysis, suppress fluctuations by adjusting the switching frequency and phase shift angle combination of the full-bridge LLC resonant converter, calculate the equivalent voltage control command through the full-bridge LLC resonant converter, adjust the switching drive signal of the full-bridge LLC resonant converter through PI control, perform mode buffer switching, and obtain mode control switching data, such as... Figure 1 As shown.

[0065] The working principle and technical effect of the above technical solution are as follows: The AC power from the power grid is converted to DC bus voltage using a three-phase full-bridge rectifier circuit. The output power is then adjusted using the frequency modulation function of the full-bridge LLC resonant converter. Voltage is corrected in real time using feedback current data to obtain voltage control data. Based on this data, the charging curve is analyzed to extract key charging nodes and resonant frequency nodes, thus determining the control target. Device status data is collected and analyzed according to the control target to construct a frequency converter control loop to adjust the resonant frequency. Simultaneously, abnormal nodes are identified and analyzed in combination to generate combined load efficiency adjustment data. Based on this data, optimal parameters for each load segment are selected. Voltage fluctuation data is collected and its impact is analyzed. Fluctuations are suppressed by adjusting the switching frequency and phase shift angle of the LLC converter. The equivalent voltage control command is calculated, and the switching drive signal is adjusted via PI control to complete the mode buffer switching, obtaining mode control switching data.

[0066] This method achieves closed-loop control throughout the entire process, from voltage acquisition, power regulation, target determination, loop construction, anomaly handling, fluctuation suppression, and mode switching. It solves the problems of disconnect between power regulation and voltage stability, inaccurate fluctuation response, and abrupt mode switching in traditional control, improves the coordination and reliability of high-power fast charging, and adapts to the efficient charging needs under different load scenarios.

[0067] The specific definitions of the methods are illustrated below (for understanding purposes only; the actual working principle shall prevail):

[0068] The critical charging point is on the battery charging curve, marking the specific voltage or current threshold point where the charging state or mode needs to be changed.

[0069] When the battery voltage rises to the preset maximum charging voltage (e.g., 4.2V / cell), this node is triggered, and the charger switches from outputting the maximum constant current to outputting the constant voltage.

[0070] When the charging current decays to the cutoff current (e.g., 0.05C), the node is triggered, and the charging process ends.

[0071] The resonant frequency node refers to the switching frequency point at which an LLC resonant converter achieves optimal efficiency or a specific gain under specific load conditions.

[0072] The rated resonant frequency is determined by the resonant inductance (Lr) and resonant capacitance (Cr), and the natural frequency (fr=1 / (2π)) is the same as the natural frequency. At this frequency range, converters often achieve zero-voltage switching (ZVS), resulting in the highest efficiency.

[0073] The optimal efficiency frequency point under different loads refers to a frequency point that deviates from fr in order to maintain high efficiency under light or heavy loads.

[0074] The control target node is a set of operating points that needs to be tracked and maintained in the control system, which is jointly determined by the charging critical node and the resonant frequency node.

[0075] During the constant current phase, the control target node might be {4.2V, 10A, 95kHz}, indicating that the system needs to control the output current to 10A while adjusting the switching frequency to around 95kHz to maintain high efficiency.

[0076] Device status data are key electrical and thermal parameters that reflect the operating conditions and health status of power devices.

[0077] The temperature of the power switching transistor (MOSFET) and rectifier diode in thermal state (estimated by thermistor or sensor).

[0078] Electrical stress refers to the peak current / voltage of the switching transistor and the peak voltage of the resonant capacitor.

[0079] Drive status refers to the rise / fall time of the drive waveform, which is used to determine whether the drive is working properly.

[0080] An abnormal node refers to a fault or warning event that is marked when the device status data exceeds the preset safety range.

[0081] Overheating abnormality refers to MOSFET temperature >125℃.

[0082] An overcurrent anomaly is defined as a resonant current peak value >25A.

[0083] Resonance detuning anomaly refers to a significant deviation between the detected resonant waveform (such as current phase) and the drive signal, indicating a possible loss of ZVS.

[0084] Mode buffer switching refers to the transition process that uses a ramp function instead of instantaneous switching to avoid abrupt changes in output when switching between different charging modes (such as constant current CC, constant voltage CV) or different control modes (such as pure PFM, PFM+phase shift).

[0085] The mode refers to the charging mode, namely CC mode and CV mode.

[0086] The buffer ensures that when switching from CC mode to CV mode, the voltage command value does not jump instantaneously from the current value to 4.2V. Instead, it increases linearly to the target value over a period of 10-100 milliseconds, controlled by a ramp generator.

[0087] Similarly, control parameters (such as the integral term of a PI controller) may also be initialized or gradually varied during this process to prevent integral saturation and system oscillation.

[0088] Simplified calculation of the overall impact coefficient (comp):

[0089] Quantify the impact of input / output voltage fluctuations on the system for use in feedforward compensation.

[0090] Simplified formula: comp = k1*(ΔVin / Vinnom) + k2*(ΔVout / Voutnom)

[0091] ΔVin = Vininstant - Vinnom (Instantaneous input voltage - Rated input voltage)

[0092] ΔVout = Voutinstant - Vouttarget (Instantaneous output voltage - Target output voltage)

[0093] k1 and k2 are weighting coefficients, determined through experiments or simulations. For example, k1=0.7 and k2=0.3 indicates that more attention is paid to input voltage fluctuations.

[0094] The mechanism for obtaining the optimal parameters for each load segment can be the lookup table method, which is the most commonly used and efficient method in engineering.

[0095] During the research and development phase, an efficiency map or three-dimensional lookup table is established through extensive testing.

[0096] The controller monitors Iout and Vin in real time, and directly obtains Fswopt and Phaseopt under the current conditions by looking up a table (or using an interpolation algorithm), and sets them as the control target.

[0097] The prediction mechanism of the digital twin model is a simplified real-time circuit simulator running in a microprocessor.

[0098] It incorporates a simplified LLC equivalent circuit model (state-space equations or approximate transfer function).

[0099] Input the current state (such as Vin, Vout, Fsw) and the planned control commands (such as "reduce the frequency by 5kHz in the next cycle") into the model.

[0100] The model calculates the critical states one or more switching cycles in advance, such as:

[0101] If the prediction result is abnormal, the original control command is corrected or abandoned to avoid actual failure.

[0102] In one embodiment of the present invention, S1 includes:

[0103] The grid voltage is converted into DC bus voltage through a three-phase full-bridge rectifier circuit;

[0104] The output power is adjusted by frequency modulation through a full-bridge LLC resonant converter to obtain output power adjustment data.

[0105] The output current data of the full-bridge LLC resonant converter is collected based on the output power adjustment data to obtain the feedback current data;

[0106] A voltage and current dual feedback system is established by combining feedback current data with DC bus voltage.

[0107] For the voltage and current dual feedback system, the feedback current data is checked for deviation to obtain current deviation determination information;

[0108] The output of the full-bridge LLC resonant converter is adjusted based on the current deviation judgment information to obtain voltage control data, such as... Figure 2 As shown.

[0109] The working principle and technical effect of the above technical solution are as follows: the grid voltage is rectified and converted by a three-phase full-bridge rectifier circuit to output DC bus voltage; the full-bridge LLC resonant converter adjusts the output power according to frequency changes and records the output power adjustment data; based on this data, the output current of the LLC converter is collected to form feedback current data; combined with the feedback current data and the DC bus voltage, a voltage and current dual feedback system is built; it is determined whether the feedback current deviates from the preset target value and current deviation judgment information is generated; the output parameters of the LLC converter are adjusted according to this information to finally obtain voltage control data.

[0110] By implementing a dual feedback system to monitor the linkage between current and voltage, the regulation lag caused by single parameter feedback is avoided, ensuring the stable output of the LLC converter, preventing power fluctuations or increased device stress caused by current deviation, and providing a stable voltage basis for charging curve analysis.

[0111] In one embodiment of the present invention, S1 includes:

[0112] Voltage and current data are collected through voltage control data acquisition and battery charging curve data are generated.

[0113] Feature analysis and annotation are performed on battery charging curve data to obtain multiple types of key charging nodes; these key charging nodes include constant current charging key nodes and constant voltage charging key nodes, etc.

[0114] Efficiency analysis and annotation of battery charging curve data were performed to obtain multiple resonant frequency nodes;

[0115] The control target node is determined based on the key charging node and the resonant frequency node, such as... Figure 3 As shown (this diagram is for illustrative purposes only and does not represent the actual node locations).

[0116] The working principle and technical effect of the above technical solution are as follows: Based on voltage control data, voltage and current data during the charging process are collected synchronously to generate complete battery charging curve data; feature extraction is performed on the charging curve data to mark the key charging nodes corresponding to key stages such as constant current charging and constant voltage charging; at the same time, the charging efficiency at different resonant frequencies in the curve is analyzed to mark the resonant frequency nodes with better efficiency; the key charging nodes are combined with the resonant frequency nodes to determine the control target nodes that take into account both the needs of the charging stage and the efficiency requirements.

[0117] Identify the core control nodes during the charging process to avoid blindly adjusting the resonance parameters. This will ensure that subsequent control has a clear objective and that the charging process meets the battery's charging characteristics (such as the requirements of constant current and constant voltage stages) while maintaining a high resonance efficiency, thereby improving the adaptability and energy efficiency of charging.

[0118] In one embodiment of the present invention, S2 includes:

[0119] Based on the control target node, device status data is acquired to obtain device status acquisition data; the device status acquisition data includes temperature data, inductor current ripple data, and capacitor voltage ripple data, etc.

[0120] Data is acquired from the fast charger to obtain charger data; the charger data includes input and output current, bus voltage, and resonant voltage, etc.

[0121] By combining data collected from the charger with data collected from the device status, device status analysis data is obtained for the fast charger.

[0122] The optimal operating point in the resonant region is determined based on the device state analysis data.

[0123] Construct a frequency conversion control loop based on the optimal operating point in the resonant region and the control target node;

[0124] The driving frequency of the full-bridge LLC resonant converter is adjusted by comparing the optimal resonant frequency node until the preset target frequency range is reached, thereby controlling the frequency conversion control loop to match the optimal resonant state.

[0125] The working principle and technical effect of the above technical solution are as follows: Based on the parameter requirements of the control target node, the system collects status data such as device temperature, inductor current ripple, and capacitor voltage ripple, as well as operating data such as charger input / output current, bus voltage, and resonant voltage; combines the two types of data to comprehensively analyze the device's operating status, determine whether it is in a safe and efficient range, and determine the optimal operating point of the resonant region; based on the optimal operating point and the control target node, a frequency converter control loop is built; compared with the preset optimal resonant frequency node, the driving frequency of the LLC converter is adjusted until the frequency falls into the target region, so that the frequency converter control loop matches the optimal resonant state.

[0126] The system achieves coordinated analysis of device status and operating parameters, ensuring that the resonant system operates within the optimal range and avoiding efficiency degradation caused by abnormal device status or frequency deviation. The precise adjustment of the frequency conversion control loop enhances the system's adaptability to different loads and maintains a high-efficiency resonant state under all operating conditions.

[0127] In one embodiment of the present invention, S2 includes:

[0128] When the frequency conversion control loop matches the optimal resonance state, abnormal data identification is performed on the device state analysis data to obtain abnormal identification data, and abnormal state nodes are marked according to the abnormal identification data.

[0129] Multiple abnormal state nodes are combined to obtain abnormal state combination information;

[0130] An abnormal state analysis command is triggered based on the abnormal state combination information to perform abnormal state analysis and obtain abnormal state analysis data.

[0131] A three-dimensional matching database consisting of resonant frequency, efficiency, and load is constructed based on the data collected from the charger.

[0132] Multiple resonant frequency combinations are obtained from the three-dimensional matching database to obtain multiple resonant frequency combination data.

[0133] Load efficiency analysis is performed on multiple resonant frequency combinations to obtain the optimal load efficiency combination.

[0134] The abnormal state analysis data and the optimal combination data of load efficiency ratio are analyzed to obtain the combined corresponding data.

[0135] Adjust the abnormal state node according to the combined data to obtain load efficiency adjustment combination data.

[0136] The working principle and technical effects of the above technical solution are as follows: When the frequency converter control loop is in the optimal resonance state, abnormal data in the device status analysis data are screened, and the corresponding abnormal status nodes are marked; multiple abnormal nodes are combined to form abnormal status combination information; an abnormal status analysis command is triggered to trace the cause and scope of the abnormality and generate abnormal status analysis data; simultaneously, based on the data collected by the charger, a three-dimensional matching database of resonant frequency, efficiency, and load is constructed, multiple resonant frequency combinations are extracted, and the load efficiency of each combination is analyzed to obtain the optimal load efficiency combination data; the abnormal status analysis data is correlated with the optimal combination data, and the abnormal nodes are adjusted to adapt to the optimal efficiency combination, ultimately obtaining the load efficiency adjustment combination data. Through abnormal node combination analysis, misjudgment or omission caused by single abnormality judgment is avoided, improving the accuracy of fault identification; combining abnormality handling with efficiency optimization ensures that the system maintains high-efficiency operation while solving abnormal problems, and the generated load efficiency adjustment combination data provides accurate basis for subsequent optimization.

[0137] In one embodiment of the present invention, the step of triggering an abnormal state analysis instruction based on abnormal state combination information, performing abnormal state analysis, and obtaining abnormal state analysis data includes:

[0138] Establish a digital twin model, and use the digital twin model to virtually map the fast charger to obtain charger mapping data;

[0139] The abnormal state combination information is input into the digital twin model to obtain fault prediction data for a preset switching cycle;

[0140] A control instruction set is generated based on the preset switching cycle fault prediction data;

[0141] By combining the control instruction set with abnormal state analysis data, abnormal state analysis data can be obtained.

[0142] The working principle and technical effects of the above technical solution are as follows: A digital twin model consistent with the physical structure and operating logic of the fast charger is constructed; real-time operating data of the charger is input into the model to achieve a virtual mapping of the physical system; abnormal state combination information is input into the digital twin model to simulate the equipment operating state under different working conditions, predict the possible fault risks within a preset switching cycle, and generate fault prediction data; a corresponding control instruction set is designed based on the fault prediction data to deal with potential faults; the control instruction set is integrated with the actual abnormal state analysis data to supplement and improve the abnormal impact assessment and response plan, obtaining the final abnormal state analysis data.

[0143] The system replicates the key component parameters of the charger in a 1:1 ratio, including the IGBT model of the three-phase full-bridge rectifier, the inductance and capacitance values ​​of the full-bridge LLC resonant converter, and the junction temperature characteristic curves of the switching transistors, ensuring that the electrical characteristics of the virtual components are consistent with those of the physical components. It also incorporates actual control algorithms, including LLC frequency modulation logic, PI control parameters, and anomaly detection thresholds, to synchronize the control response of the virtual system with the physical system.

[0144] Industrial Ethernet (such as EtherCAT) is used to enable data interaction between physical devices and the digital twin platform, with transmission latency controlled to ≤1ms to avoid prediction errors caused by synchronization lag;

[0145] Electrical parameters (voltage, current) are synchronized at a frequency of 50kHz (consistent with the acquisition frequency), and device status data (temperature, ripple) are synchronized at a frequency of 1kHz, balancing synchronization accuracy with platform computing power consumption.

[0146] Input abnormal combination information: for example, Tj=98℃+ΔIr=6%, the model will call the built-in device failure model (such as the Arrhenius model) to calculate the remaining lifetime decay rate of the switching transistor under this state;

[0147] Using 10 switching cycles (approximately 20 μs, assuming a resonant frequency of 50 kHz) as the prediction window, the changing trends of current and temperature within this cycle are simulated. If it is predicted that Tj will reach 120℃ (over-temperature threshold) after 5 cycles, temperature warning fault prediction data is generated.

[0148] In response to temperature warnings, a set of instructions is generated to reduce output power by 10% and increase cooling fan speed, enabling early intervention of physical equipment.

[0149] Digital twins enable early prediction of faults, transforming passive handling into proactive prevention and avoiding equipment damage caused by the escalation of faults. The combination of virtual mapping and actual data improves the comprehensiveness and accuracy of anomaly analysis, providing a more scientific basis for subsequent control and regulation decisions.

[0150] In one embodiment of the present invention, S3 includes:

[0151] The input voltage fluctuation data and the output voltage fluctuation data of the full-bridge LLC resonant converter are obtained by combining load efficiency adjustment data; the ratio of the input voltage fluctuation data to the resonant frequency fluctuation data in the resonant region is obtained to obtain the input voltage influence coefficient; the ratio of the output voltage fluctuation data of the full-bridge LLC resonant converter to the resonant frequency fluctuation data in the resonant region is obtained to obtain the output voltage influence coefficient; the average value of the input voltage influence coefficient and the output voltage influence coefficient is obtained to obtain the comprehensive influence coefficient; a mode buffer control command is triggered based on the comprehensive influence coefficient; a mode buffer coefficient is generated based on the mode buffer control command; mode buffer control is performed based on the mode buffer coefficient to obtain mode buffer control data.

[0152] The working principle and technical effect of the above technical solution are as follows: Input voltage fluctuation data and output voltage fluctuation data of the LLC converter are extracted from the load efficiency regulation combination data; the ratio of input voltage fluctuation to resonant frequency fluctuation (input voltage influence coefficient) and the ratio of output voltage fluctuation to resonant frequency fluctuation (output voltage influence coefficient) are calculated; the average of the two coefficients is taken as the comprehensive influence coefficient to measure the overall impact of voltage fluctuation on the resonant system; the corresponding mode buffer control command is triggered according to the comprehensive influence coefficient to generate the mode buffer coefficient; the control parameters are adjusted according to the buffer coefficient to achieve mode buffer control and obtain mode buffer control data.

[0153] Take 10ms as the acquisition window and calculate the maximum voltage / frequency fluctuation within the window (e.g., when the input voltage drops from 220V to 200V, the fluctuation value is 20V; when the resonant frequency drops from 200kHz to 195kHz, the fluctuation value is 5kHz).

[0154] A sliding window acquisition method is used (the window data is updated every 1ms) to avoid random errors in a single sampling and ensure the representativeness of the fluctuating data.

[0155] The formula for the input voltage influence coefficient is the input voltage fluctuation value / resonant frequency fluctuation value. For example, 20V / 5kHz = 4V / kHz, which means that for every 1V fluctuation in the input voltage, the resonant frequency will fluctuate by 0.25kHz. The larger the coefficient, the stronger the interference of the input voltage on the resonant system.

[0156] The formula for the output voltage influence coefficient is the output voltage fluctuation value / resonant frequency fluctuation value. For example, 5V / 5kHz = 1V / kHz, which means that for every 1V fluctuation in output voltage, the resonant frequency fluctuates by 1kHz, reflecting the degree of influence of output load changes on the resonant state.

[0157] The comprehensive influence coefficient calculation includes, for example, an input voltage influence coefficient of 4V / kHz and an output voltage influence coefficient of 1V / kHz, with an average value of 2.5V / kHz, representing the overall influence level of voltage fluctuations on the resonant system;

[0158] Through offline experimental calibration, when the comprehensive influence coefficient is ≤1V / kHz, the fluctuation has little impact on the system, triggering weak buffering (buffer coefficient 0.2, small frequency adjustment); when 1V / kHz < coefficient ≤3V / kHz, medium buffering is triggered (buffer coefficient 0.5, medium frequency and phase shift angle adjustment); when the coefficient >3V / kHz, strong buffering is triggered (buffer coefficient 0.8, large parameter adjustment and reduced output power).

[0159] The impact of voltage fluctuations on the resonant system is quantified to avoid the blindness of buffer control; by adjusting the mode buffer coefficient, the impact of voltage fluctuations on the system is mitigated, the efficiency reduction or device damage caused by sudden changes in the resonant state is prevented, and the system's anti-fluctuation capability is improved.

[0160] In one embodiment of the present invention, S3 includes:

[0161] Generate voltage regulation commands based on mode buffer control data;

[0162] The target value of the bus voltage is obtained based on the output power demand according to the voltage regulation command;

[0163] The target value of the bus voltage and the actual value of the bus voltage are compared by a PI controller to obtain the bus voltage comparison result.

[0164] The rectifier PWM duty cycle adjustment signal is obtained based on the bus voltage comparison results.

[0165] The output of the rectifier circuit is adjusted according to the rectifier PWM duty cycle adjustment signal, thereby obtaining the bus voltage regulation data.

[0166] A voltage regulation command is generated based on the mode buffer control data. Based on the voltage regulation command, the target voltage gain required by the full-bridge LLC resonant converter is calculated using the output power demand and the current input voltage. The target voltage gain is compared with the actual voltage gain calculated using the output voltage and the bus voltage via a PI controller to obtain a gain comparison result. The switching frequency or phase shift angle adjustment signal of the full-bridge LLC resonant converter is obtained based on the gain comparison result. The output of the LLC converter is adjusted according to the switching frequency or phase shift angle adjustment signal to obtain output voltage regulation data.

[0167] The working principle and technical effect of the above technical solution are as follows: Voltage regulation commands are generated based on mode buffer control data. On one hand, the target value of the bus voltage is calculated in conjunction with the output power requirement. The PI controller compares the target value with the actual bus voltage to generate a rectifier PWM duty cycle adjustment signal, which adjusts the rectifier circuit output to obtain bus voltage regulation data. On the other hand, the target voltage gain of the LLC converter is calculated in conjunction with the output power requirement and the current input voltage. The PI controller compares the target gain with the actual voltage gain (calculated from the output voltage and the bus voltage) to generate a switching frequency or phase shift angle adjustment signal, which adjusts the LLC converter output to obtain output voltage regulation data.

[0168] For example, if the current charging power requirement of the battery is 200W and the rated input current of the resonant converter is 5A, considering a redundancy factor of 1.05 (to cope with power fluctuations), then the target value of the bus voltage Vbustarget = (200W × 1.05) / 5A = 42V.

[0169] The redundancy is used to offset the bus voltage deviation caused by grid voltage fluctuations (such as AC220V±10% fluctuation). Through experimental verification, a redundancy of 1.05 can ensure that the bus voltage is not lower than 95% of the target value within the 90-264VAC input range.

[0170] The target voltage gain is calculated as follows: output power requirement is 200W, current input voltage is AC220V (rectified bus voltage is approximately 311V). Therefore, the LLC converter needs to output voltage Vout = 200W / Iout (assuming Iout = 10A, then Vout = 20V). The target voltage gain Gtarget = Vout / Vbus = 20V / 311V ≈ 0.064.

[0171] The actual voltage gain is calculated by collecting the actual output voltage (e.g., 19.5V) and the bus voltage (e.g., 311V), resulting in Gactual = 19.5V / 311V ≈ 0.063. The difference between the two is used for PI control adjustment.

[0172] When tuning the proportional coefficient Kp, first set Ki to 0, then gradually increase Kp until the output voltage shows slight oscillation (such as fluctuation ±1%). At this point, Kp is the critical value (such as 0.8). Take 0.6-0.8 times the critical value as the final Kp (such as 0.5).

[0173] After Kp is determined, the integral coefficient Ki is gradually increased until the static error (the deviation between the target value and the actual value) is ≤0.5%, while ensuring no oscillation. Finally, Ki=0.1 is determined.

[0174] The step response test (such as suddenly increasing the output power from 100W to 200W) verified that the voltage regulation response time is ≤1ms and the overshoot is ≤2%, which meets the design requirements.

[0175] PI control enables precise regulation of the bus voltage and output voltage, avoiding charging instability caused by voltage deviation; at the same time, the rectifier circuit and LLC converter are adjusted to achieve synergistic optimization of the front-end rectification and the rear-end resonance, ensuring the voltage matching of the entire charging system and improving power transmission efficiency.

[0176] In one embodiment of the present invention, S3 includes:

[0177] After the output voltage is regulated, the fast charger performs charging stage data analysis to obtain charging stage analysis data; the charging mode is switched based on the charging node analysis data to obtain mode switching data; mode switching buffer control is performed based on the mode switching data to obtain mode switching buffer control data; and mode control is switched based on the mode switching buffer control data to obtain mode control switching data.

[0178] The working principle and technical effect of the above technical solution are as follows: After the output voltage is adjusted to a stable state, the data such as the change of SOC and the trend of current and voltage during the charging process are analyzed to determine the current charging stage (such as constant current stage or constant voltage stage), and charging stage analysis data is generated; based on the stage analysis data, it is determined whether the charging mode needs to be switched, and mode switching data is generated; based on the mode switching data, a buffer control strategy is designed to mitigate the parameter abrupt changes during mode switching, and mode switching buffer control data is obtained; the mode switching operation is executed according to the buffer control data to ensure a smooth switching process, and finally, mode control switching data is obtained.

[0179] It achieves accurate judgment and smooth switching of charging modes, avoids current or voltage overshoot during mode switching, and protects the battery and charging devices; the application of buffer control strategy improves the coordination of mode switching and maintains charging efficiency and safety during the switching process.

[0180] According to one embodiment of the present invention, the system includes:

[0181] The stabilization and regulation module is used to obtain the DC bus voltage through the three-phase full-bridge rectifier circuit, regulate the output power through the full-bridge LLC resonant converter, and perform voltage stabilization control and regulation in combination with feedback current data to obtain voltage control data. Based on the voltage control data, the charging curve data is analyzed to obtain the key charging nodes and resonant frequency nodes, and then the control target nodes are obtained.

[0182] The anomaly analysis module is used to collect and analyze device status data based on the control target node, obtain device status analysis data, and then construct a frequency conversion control loop to adjust the resonant frequency. After the adjustment is completed, anomaly nodes are identified and combined analysis is performed to obtain load efficiency adjustment combination data.

[0183] The fluctuation switching module is used to obtain the optimal parameters for each load segment based on the load efficiency adjustment combination data, collect voltage fluctuation data, perform fluctuation impact analysis, suppress fluctuations by adjusting the combination of switching frequency and phase shift angle of the full-bridge LLC resonant converter, calculate the equivalent voltage control command through the full-bridge LLC resonant converter, and perform mode buffer switching by adjusting the switching drive signal of the full-bridge LLC resonant converter through PI control to obtain mode control switching data.

[0184] The working principle and technical effect of the above technical solution are as follows: The AC power from the power grid is converted to DC bus voltage using a three-phase full-bridge rectifier circuit. The output power is then adjusted using the frequency modulation function of the full-bridge LLC resonant converter. Voltage is corrected in real time using feedback current data to obtain voltage control data. Based on this data, the charging curve is analyzed to extract key charging nodes and resonant frequency nodes, thus determining the control target. Device status data is collected and analyzed according to the control target to construct a frequency converter control loop to adjust the resonant frequency. Simultaneously, abnormal nodes are identified and analyzed in combination to generate combined load efficiency adjustment data. Based on this data, optimal parameters for each load segment are selected. Voltage fluctuation data is collected and its impact is analyzed. Fluctuations are suppressed by adjusting the switching frequency and phase shift angle of the LLC converter. The equivalent voltage control command is calculated, and the switching drive signal is adjusted via PI control to complete the mode buffer switching, obtaining mode control switching data.

[0185] This method achieves closed-loop control throughout the entire process, from voltage acquisition, power regulation, target determination, loop construction, anomaly handling, fluctuation suppression, and mode switching. It solves the problems of disconnect between power regulation and voltage stability, inaccurate fluctuation response, and abrupt mode switching in traditional control, improves the coordination and reliability of high-power fast charging, and adapts to the efficient charging needs under different load scenarios.

[0186] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A high power rectifier resonant control method for fast chargers, characterized by, The method comprises: S1, obtaining a DC bus voltage through a three-phase full-bridge rectifier circuit, adjusting output power through a full-bridge LLC resonant converter, combining feedback current data to perform voltage stability control adjustment, obtaining voltage control data, and performing charging curve data analysis according to the voltage control data to obtain a charging key node and a resonant frequency node, and then obtaining a control target node; S2, collecting and analyzing device state data according to the control target node to obtain device state analysis data, and then constructing a variable frequency control loop to adjust the resonant frequency, identifying abnormal nodes after the adjustment is completed and performing combined analysis to obtain load efficiency adjustment combination data; S3, obtaining optimal parameters of each load section according to the load efficiency adjustment combination data, collecting voltage fluctuation data, performing fluctuation influence analysis, suppressing fluctuations by adjusting the switching frequency and phase shift angle combination of the full-bridge LLC resonant converter, calculating the equivalent voltage control command through the full-bridge LLC resonant converter, adjusting the switching drive signal of the full-bridge LLC resonant converter through PI control, performing mode buffer switching, and obtaining mode control switching data.

2. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S1 comprises: convert the grid voltage into a DC bus voltage through a three-phase full-bridge rectifier circuit; adjust the output power through a full-bridge LLC resonant converter according to frequency modulation to obtain output power adjustment data; collect the output current data of the full-bridge LLC resonant converter according to the output power adjustment data to obtain feedback current data; establish a voltage and current double feedback system through the feedback current data combined with the DC bus voltage; determine whether the feedback current data deviates from the voltage and current double feedback system to obtain current deviation determination information; adjust the output of the full-bridge LLC resonant converter according to the current deviation determination information to obtain voltage control data.

3. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S1 comprises: collect voltage and current data through the voltage control data and generate battery charging curve data; perform feature analysis and labeling on the battery charging curve data to obtain multiple types of charging key nodes; perform efficiency analysis and labeling on the battery charging curve data to obtain multiple resonant frequency nodes; determine the control target node according to the charging key node combined with the resonant frequency node.

4. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S2 comprises: collect device state data according to the control target node to obtain device state collection data; collect data from the fast charger to obtain charger collection data; perform device state analysis on the fast charger through the charger collection data combined with the device state collection data to obtain device state analysis data; determine the optimal operating point of the resonant region according to the device state analysis data; construct a variable frequency control loop according to the optimal operating point of the resonant region combined with the control target node; adjust the drive frequency of the full-bridge LLC resonant converter compared to the optimal resonant frequency node until the preset target frequency region is reached, and then control the variable frequency control loop to match the optimal resonant state.

5. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S2 comprises: when the variable frequency control loop matches the optimal resonant state, identify abnormal data from the device state analysis data to obtain abnormal identification data, and label abnormal state nodes according to the abnormal identification data; Combine the plurality of abnormal state nodes to obtain abnormal state combination information; Trigger abnormal state analysis instructions according to the abnormal state combination information, perform abnormal state analysis, and obtain abnormal state analysis data; According to the charging collector data, a three-dimensional matching database composed of resonance frequency, efficiency and load is constructed; According to the three-dimensional matching database, a plurality of resonance frequency combinations are obtained, and a plurality of resonance frequency combination data are obtained; Load efficiency analysis is performed on the plurality of resonance frequency combination data to obtain load efficiency optimal combination data; The abnormal state analysis data and the load efficiency optimal combination data are analyzed, and combination corresponding data is obtained; According to the combination corresponding data, the abnormal state node is adjusted to obtain load efficiency adjustment combination data.

6. The method of claim 5, wherein the high power rectifier resonant control method of the fast charger is characterized by, The abnormal state analysis data is obtained by triggering abnormal state analysis instructions according to the abnormal state combination information, including: A digital twin model is established, the fast charger is virtually mapped through the digital twin model, and charger mapping data is obtained; The abnormal state combination information is input into the digital twin model to obtain preset switch cycle fault prediction data; Control instruction set is generated according to the preset switch cycle fault prediction data; The control instruction set and the abnormal state analysis data are combined to obtain the abnormal state analysis data.

7. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S3 includes: Input voltage fluctuation data and output voltage fluctuation data of the full-bridge LLC resonant converter are obtained through load efficiency adjustment combination data; The ratio of input voltage fluctuation data to resonant frequency fluctuation data in the resonant region is obtained to obtain input voltage influence coefficient; The ratio of output voltage fluctuation data of the full-bridge LLC resonant converter to resonant frequency fluctuation data in the resonant region is obtained to obtain output voltage influence coefficient; The average of the input voltage influence coefficient and the output voltage influence coefficient is obtained to obtain the comprehensive influence coefficient; Mode buffer control instructions are triggered according to the comprehensive influence coefficient; Mode buffer coefficient is generated according to mode buffer control instructions, and mode buffer control is performed according to the mode buffer coefficient to obtain mode buffer control data.

8. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S3 includes: Voltage regulation instructions are generated according to mode buffer control data; Bus voltage target value is obtained through output power demand according to voltage regulation instructions; The bus voltage target value and the bus voltage actual value are compared through the PI controller to obtain the bus voltage comparison result; The rectifier PWM duty cycle regulation signal is obtained according to the bus voltage comparison result; The output of the rectifier circuit is adjusted according to the rectifier PWM duty cycle regulation signal, and then the bus voltage regulation data is obtained; Voltage regulation instructions are generated according to mode buffer control data; According to the voltage regulation instructions, the target voltage gain required by the full-bridge LLC resonant converter is calculated through the output power demand and the current input voltage; The gain comparison result is obtained by comparing the target voltage gain with the actual voltage gain calculated by the output voltage and the bus voltage through the PI controller; The switch frequency or phase shift angle regulation signal of the full-bridge LLC resonant converter is obtained according to the gain comparison result; According to the switching frequency or the phase-shifting angle regulation signal of the full-bridge LLC resonant converter, the output of the LLC converter is regulated, and then output voltage regulation data is obtained.

9. The method of claim 1, wherein the high power rectifier resonant control method of the fast charger is characterized by, The S3 comprises: After the output voltage regulation is completed, the fast charger is analyzed for charging phase data to obtain charging phase analysis data; According to the charging node analysis data, the charging mode is switched to obtain mode switching data; According to the mode switching data, mode switching buffer control is performed to obtain mode switching buffer control data; According to the mode switching buffer control data, mode control switching is performed to obtain mode control switching data.

10. A high power rectifier resonant control system for a fast charger, characterized by, The system comprises: The stable regulation module is used to obtain the DC bus voltage through the three-phase full-bridge rectifier circuit, to adjust the output power through the full-bridge LLC resonant converter, to perform voltage stability control regulation combined with the feedback current data, to obtain voltage control data, to analyze the charging curve data according to the voltage control data to obtain the charging key node and the resonant frequency node, and to further obtain the control target node. The abnormality analysis module is used to collect and analyze the device state data according to the control target node to obtain device state analysis data, to further construct a variable frequency control loop, to perform resonant frequency regulation, to identify abnormal nodes after the regulation is completed, and to perform combined analysis to obtain load efficiency regulation combined data. The fluctuation switching module is used to obtain the optimal parameters of each load section according to the load efficiency regulation combined data, to collect voltage fluctuation data, to perform fluctuation influence analysis, to perform fluctuation suppression through the combination of the switching frequency and the phase-shifting angle of the full-bridge LLC resonant converter, to obtain equivalent voltage control instructions through the full-bridge LLC resonant converter calculation, to adjust the switching drive signal of the full-bridge LLC resonant converter through PI control, to perform mode buffer switching, and to obtain mode control switching data.

Images