A power supply and power conversion control system and method thereof
The intelligent power conversion control method using multi-topology path reconstruction and edge computing solves the problems of high efficiency and high power density of traditional power supplies under load changes, and realizes efficient, stable and reliable operation of power supplies under complex operating conditions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 湖南鹏耀科技有限公司
- Filing Date
- 2026-03-10
- Publication Date
- 2026-06-05
AI Technical Summary
Traditional switching power supplies struggle to achieve high efficiency and high power density under varying load conditions. Static control strategies cannot respond to dynamic factors in real time, resulting in low efficiency in light-load ranges and increased costs due to redundant design. They also fail to balance stability and reliability under complex operating conditions.
A power conversion control method employing multi-topology path reconstruction, edge computing, and intelligent prediction is used to generate real-time power conversion control commands through dynamic topology configuration, load characteristic parameter set modeling, multi-objective collaborative scheduling, and harmonic suppression, thereby optimizing energy efficiency and enhancing load response capability.
Improve average efficiency across the entire load range, reduce energy waste, enhance load surge response capability, reduce losses, prevent power supply overheating, extend service life, reduce size and cost, and ensure stability.
Smart Images
Figure CN122159674A_ABST
Abstract
Description
Technical Field
[0001] This invention proposes a power supply and its power conversion control system and method, belonging to the field of power electronics and intelligent management technology. Background Technology
[0002] In the field of power supply technology, traditional switching power supply (SMPS) power conversion control technology faces many bottlenecks. Most power supplies use a single topology, such as only LLC or Buck topology. This fixed mode makes it difficult to meet the performance requirements under different load conditions. It is difficult to achieve high efficiency under light loads, and it cannot guarantee high power density under heavy loads, resulting in limited overall power supply performance.
[0003] In terms of control strategies, existing technologies mostly adopt static PWM frequency and duty cycle adjustment methods, controlling according to preset curves. However, in actual operation, dynamic factors such as sudden load changes, harmonic injection, and component aging and degradation frequently occur. Static control strategies cannot respond to these changes in real time, which greatly reduces the stability and reliability of the power supply under complex operating conditions.
[0004] In terms of energy efficiency optimization, traditional solutions only focus on peak efficiency, neglecting the energy waste problem in the 5%-30% light load range, where efficiency often falls below 80%, resulting in significant energy loss. Furthermore, to suppress transient overshoot, existing technologies often rely on redundant designs such as increasing output capacitors and loop compensation, which not only sacrifices the power supply's size advantage but also increases cost. Therefore, developing a novel power conversion control method that breaks with convention and deeply integrates topology reconfigurability, intelligent prediction, and collaborative control is urgently needed. Summary of the Invention
[0005] This invention provides a power supply and its power conversion control system and method to solve the problems mentioned in the background section above: This invention proposes a power conversion control method for a power supply, the method comprising: S1. Divide the power conversion requirements into multiple topology paths and generate dynamic topology configuration data; construct a composite power conversion architecture based on the dynamic topology configuration data, and deploy edge computing nodes to form an intelligent topology reconfiguration network; S2. Real-time modeling of load behavior is performed through intelligent topology reconfiguration network to generate a load characteristic parameter set; multi-objective collaborative scheduling is performed based on the load characteristic parameter set to dynamically adjust the topology path combination and switching frequency, complete the acquisition of multi-source power signals and generate raw power conversion data; harmonic suppression and aging compensation processing is performed on the raw power conversion data to generate purified power signal data. S3. Extract transient response features from purification power signal data to obtain load mutation response data and harmonic injection response data respectively; evaluate the dynamic stability of harmonic injection response data using load mutation response data, perform topology switching threshold calibration, and generate topology reconfiguration control command data. S4. The composite power conversion architecture is reconstructed using topology reconfiguration control command data to generate real-time power conversion topology data; energy efficiency optimization analysis is performed on the real-time power conversion topology data in combination with load characteristic parameter set to generate multi-objective collaborative control parameters; power loss prediction is performed on the multi-objective collaborative control parameters to obtain power loss prediction data. S5. Fit the full-load efficiency curve based on the power loss prediction data to generate an intelligent efficiency optimization model; perform topology dynamic reconstruction and precise output power adjustment based on the intelligent efficiency optimization model to generate power conversion control commands; perform risk warning and energy efficiency optimization processing based on the power conversion control commands to generate intelligent operation data of the power system.
[0006] The present invention proposes a system for implementing a power conversion control method for a power supply as described above, the system comprising: Node deployment module: Divides power conversion requirements into multiple topology paths and generates dynamic topology configuration data; constructs a composite power conversion architecture based on the dynamic topology configuration data, and deploys edge computing nodes to form an intelligent topology reconfiguration network; Compensation processing module: Real-time modeling of load behavior is performed through intelligent topology reconstruction network to generate load characteristic parameter set; Multi-objective collaborative scheduling is performed based on load characteristic parameter set to dynamically adjust topology path combination and switching frequency, complete multi-source power signal acquisition and generate raw power conversion data; Harmonic suppression and aging compensation processing are performed on raw power conversion data to generate purified power signal data; Topology switching module: Extracts transient response features from purification power signal data to obtain load mutation response data and harmonic injection response data; performs dynamic stability evaluation on harmonic injection response data using load mutation response data, calibrates topology switching threshold, and generates topology reconfiguration control command data. The loss prediction module performs topology path reconstruction on the composite power conversion architecture using topology reconstruction control command data to generate real-time power conversion topology data; it performs energy efficiency optimization analysis on the real-time power conversion topology data in conjunction with load characteristic parameter sets to generate multi-objective collaborative control parameters; and it performs power loss prediction on the multi-objective collaborative control parameters to obtain power loss prediction data. Intelligent Operation Module: Based on power loss prediction data, it performs full-load efficiency curve fitting to generate an intelligent efficiency optimization model; based on the intelligent efficiency optimization model, it performs topology dynamic reconstruction and precise output power adjustment to generate power conversion control commands; based on the power conversion control commands, it performs risk warning and energy efficiency optimization processing to generate intelligent operation data of the power system.
[0007] The present invention provides a power supply comprising the system described above.
[0008] The beneficial effects of this invention are as follows: By integrating multi-topology paths, edge AI load prediction, and multi-objective collaborative control, the power conversion control method of this power supply achieves significant results. It improves the average efficiency across the entire load range, enabling the power supply to operate efficiently under various working conditions and reducing energy waste. Simultaneously, it shortens the transient response time to ≤20 μs, greatly enhancing the power supply's ability to respond quickly to load changes and ensuring stable output. It reduces power loss during the conversion process, preventing excessive heat generation due to excessive losses and extending the power supply's lifespan. It reduces redundant designs such as output capacitors and loop compensation, lowering the power supply's size and cost while maintaining stability. It avoids the problem of traditional static control strategies failing to respond to dynamic load changes in real time, preventing performance degradation or even damage caused by untimely control. Attached Figure Description
[0009] Figure 1 This is a diagram illustrating the steps of the method described in this invention; Figure 2 This is a system module diagram of the present invention; Figure 3 This is a flowchart of step S1 as described in this invention; Figure 4 This is a flowchart of step S2 of the present invention. Detailed Implementation
[0010] 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.
[0011] One embodiment of the present invention, such as Figure 1 As shown, a power conversion control method for a power supply includes: S1. Divide the power conversion requirements into multiple topology paths and generate dynamic topology configuration data; construct a composite power conversion architecture based on the dynamic topology configuration data. The composite power conversion architecture includes LLC, Buck and synchronous rectification topologies, and deploy edge computing nodes to form an intelligent topology reconfiguration network. S2. Real-time modeling of load behavior is performed through intelligent topology reconfiguration network to generate a load characteristic parameter set; multi-objective collaborative scheduling is performed based on the load characteristic parameter set to dynamically adjust the topology path combination and switching frequency, complete the acquisition of multi-source power signals and generate raw power conversion data; harmonic suppression and aging compensation processing is performed on the raw power conversion data to generate purified power signal data. S3. Extract transient response features from purification power signal data to obtain load mutation response data and harmonic injection response data respectively; evaluate the dynamic stability of harmonic injection response data using load mutation response data, perform topology switching threshold calibration, and generate topology reconfiguration control command data. S4. The composite power conversion architecture is reconstructed using topology reconfiguration control command data to generate real-time power conversion topology data; energy efficiency optimization analysis is performed on the real-time power conversion topology data in combination with load characteristic parameter set to generate multi-objective collaborative control parameters; power loss prediction is performed on the multi-objective collaborative control parameters to obtain power loss prediction data. S5. Fit the full-load efficiency curve based on the power loss prediction data to generate an intelligent efficiency optimization model; perform topology dynamic reconstruction and precise output power adjustment based on the intelligent efficiency optimization model to generate power conversion control commands; perform risk warning and energy efficiency optimization processing based on the power conversion control commands to generate intelligent operation data of the power system.
[0012] The working principle and effects of the above technical solution are as follows: Through multi-topology path combination and dynamic reconfiguration design, the energy efficiency of power conversion can be significantly improved, power loss during device operation can be reduced, and adaptability under different load conditions can be enhanced. The deployment of edge computing nodes accelerates load behavior modeling and response speed, reduces signal processing lag, and avoids operational instability caused by load abrupt changes or harmonic injection. Harmonic suppression and aging compensation processing can purify power signals, avoid signal deviations caused by noise interference and device aging, and reduce the probability of system failure. Topology switching threshold calibration further enhances operational stability and prevents voltage fluctuations caused by abnormal topology switching. The intelligent efficiency optimization model realizes precise coordination between topology reconfiguration and power regulation, which can ensure efficient operation across the entire load range and accurately match output requirements. The dual processing of risk warning and energy efficiency optimization avoids the expansion of overcurrent, overvoltage, and efficiency anomalies, reduces unnecessary energy waste, and improves the overall reliability and intelligence level of the power system, adapting to complex and ever-changing power conversion scenarios.
[0013] One embodiment of the present invention, such as Figure 3 As shown, S1 includes: S11. Collect input and output parameters and load condition information of the power conversion scenario, and generate power conversion demand characteristic data. S12. Perform multi-topology path partitioning on the power conversion demand characteristic data to generate dynamic topology configuration data; S13. Based on the dynamic topology configuration data, build a composite power conversion architecture, integrate LLC Buck and synchronous rectification topology modules, and generate composite topology hardware architecture data. S14. Deploy edge computing units on the physical nodes corresponding to the composite topology hardware architecture data to build an intelligent topology reconfiguration network and generate basic data for the operation of the topology reconfiguration network.
[0014] The working principle and effects of the above technical solution are as follows: Accurately collecting input / output parameters and load condition information allows power conversion demand characteristics to better match actual operating scenarios, preventing subsequent topology configuration from deviating from actual needs and reducing operational problems caused by configuration deviations from the outset. By dividing the demand characteristic data into multiple topology paths, the generated dynamic topology configuration data enhances architecture adaptability and reduces the limitations of a single topology path in handling complex operating conditions. Integrating multiple topology modules to build a composite architecture not only meets the needs of different power conversion scenarios but also improves the flexibility of architecture operation, avoiding the efficiency or stability shortcomings of a single topology module that restrict overall performance. Deploying edge computing units on physical nodes to build a reconfigurable network accelerates data processing and command transmission speed, reduces latency issues caused by centralized computing, avoids power conversion imbalances caused by untimely topology reconfiguration responses, lays a solid foundation for subsequent intelligent control, and comprehensively improves the rationality of power supply architecture construction and operational reliability.
[0015] In one embodiment of the present invention, step S12 includes: The power conversion demand characteristic data is decomposed into operating condition dimension to generate operating condition characteristic decomposition data; Based on the decomposition of working condition features, topology path type matching is carried out to generate a candidate set of topology paths; Perform pathway combination suitability analysis on the candidate set of topological pathways to generate topological pathway combination data; Dynamically adapt the path parameters to the topology path combination data to generate a topology path parameter set; Integrate topology path combination data and topology path parameter set to generate dynamic topology configuration data.
[0016] The working principle and effects of the above technical solution are as follows: Decomposing power conversion demand characteristic data according to operating conditions allows for more detailed and comprehensive demand analysis, avoiding the mismatch between topology paths and actual operating conditions caused by general analysis, and reducing the frequency of subsequent configuration adjustments. Based on the decomposed data, topology path type matching is performed, generating a candidate set that broadens the adaptability range, enhances coverage of complex operating conditions, and reduces the limitations of single path selection. Combination and matching analysis of the candidate set can filter out path combinations that better meet the requirements, balancing power conversion efficiency and stability while adapting to dynamic changes in different operating conditions. Dynamic adaptation of path parameters optimizes parameter matching, avoiding the problem of fixed parameters being unable to adapt to operating condition fluctuations, and reducing operational losses caused by parameter imbalances. Finally, integrating the combined data and parameter set generates dynamic topology configuration data, improving the accuracy and relevance of the configuration, avoiding configuration deviations that could hinder subsequent architecture construction, providing a reliable configuration basis for composite power conversion architectures, and overall improving the rationality and adaptability of topology configuration.
[0017] One embodiment of the present invention, such as Figure 4 As shown, S2 includes: S21. Call the basic data for topology reconstruction network operation, start the load perception module of intelligent topology reconstruction network, carry out real-time modeling of load behavior, and generate a set of load characteristic parameters. S22. Perform multi-objective collaborative scheduling based on load characteristic parameter set, dynamically adjust topology path combination and switching frequency parameters, and generate topology scheduling execution data; S23. Initiate a multi-source power signal acquisition process for topology scheduling execution data, acquire power signals from the input side, output side, and topology nodes, and generate raw power conversion data; S23. Perform harmonic suppression processing on the original power conversion data to filter out high-frequency harmonics and low-frequency noise components, and generate harmonic-suppressed power data. S25. Perform device aging compensation processing on the power data after harmonic suppression to correct power signal deviation and generate purified power signal data.
[0018] The working principle and effects of the above technical solution are as follows: The load sensing module initiates real-time modeling, accurately capturing dynamic load changes. The generated load characteristic parameter set more closely matches the actual operating state, avoiding modeling lag that could lead to a disconnect between topology scheduling and load demand, and reducing efficiency losses caused by scheduling deviations. Multi-objective collaborative scheduling is performed based on the parameter set, dynamically adjusting the topology path and switching frequency. This adapts to changes in different load conditions and optimizes the power conversion rhythm, reducing ineffective energy consumption. Multi-source signal acquisition covers input / output and topology nodes, making the raw data more comprehensive and complete, avoiding the omission of key signals from single-dimensional acquisition, and reducing deviations in subsequent data processing. Harmonic suppression filters out noise components, reducing harmonic interference to equipment, preventing power signal distortion caused by noise, and protecting devices to extend their lifespan. Device aging compensation corrects signal deviations, generating more accurate purified power signal data, avoiding signal offset caused by aging from affecting subsequent stability assessments, and providing reliable data support for topology control. The entire process is optimized layer by layer, improving the cleanliness and stability of power conversion, enhancing the system's adaptability to load and device states, and ensuring efficient and controllable operation.
[0019] In one embodiment of the present invention, step S22 includes: S221. Perform multi-objective dimension analysis on the load characteristic parameter set to generate multi-objective scheduling characteristic data; S222. Based on multi-objective scheduling feature data, perform topology path combination optimization to generate optimized topology path combination data; S223. Perform dynamic switching frequency matching on the preferred topology path combination data to generate a dynamic switching frequency parameter set; S224. Integrate the optimal topology path combination data with the dynamic switching frequency parameter set to generate topology scheduling execution data.
[0020] The working principle and effects of the above technical solution are as follows: Multi-dimensional analysis of the load characteristic parameter set can uncover core requirements under different operating conditions, generating more targeted scheduling characteristic data. This avoids missing key information in single-dimensional analysis and reduces bias in subsequent scheduling decisions. Based on the analyzed data, topology path combination optimization is performed to select more adaptable combination schemes that balance power conversion efficiency and operational stability while adapting to dynamic load changes, reducing energy waste caused by unreasonable combinations. Dynamic switching frequency matching is performed on the optimized combinations, breaking the limitations of fixed frequencies and ensuring precise matching between frequency parameters and path operating states. This avoids device overheating and efficiency loss caused by frequency imbalance and enhances the system's responsiveness to load fluctuations. Finally, the combination data and frequency parameters are integrated to generate scheduling execution data, ensuring the uniformity and executability of instructions, avoiding data fragmentation that could lead to disjointed scheduling actions, and reducing potential faults during execution. The entire process progressively optimizes the scheduling logic, improving the accuracy of topology scheduling and enhancing the system's adaptability to complex load conditions. This makes the power conversion process more efficient and controllable, further laying a solid foundation for subsequent signal processing and stability improvements.
[0021] In one embodiment of the present invention, S223 includes: Extract pathway operation characteristics from the data of the preferred topological pathway combinations to generate pathway characteristic characterization data; Based on the analysis of path characteristic characterization data, load adaptation frequency requirements are analyzed to generate frequency requirement interval data. Adjust the initial frequency value for the frequency demand range data to generate candidate switching frequency parameters; Verify the fit between candidate switching frequency parameters and path characteristics, and generate a dynamic switching frequency parameter set.
[0022] The working principle and effects of the above technical solution are as follows: The operating characteristics of the optimal topology path combination are extracted, and the generated characterization data accurately reflects the actual operating status of the path, avoiding deviations in frequency demand analysis due to ambiguity in understanding path characteristics, and reducing the blindness of subsequent parameter settings. Based on the characteristic data, the load-adaptive frequency requirements are decomposed, and the generated interval data can define a reasonable range, avoiding excessively high or low frequency settings that exceed the path's carrying capacity, and preventing devices from overheating and experiencing increased losses due to improper frequency. The initial frequency is adjusted within the frequency demand range to generate candidate parameters, not limited to fixed values, allowing flexible adaptation to dynamic changes in the path, reducing the problem of fixed frequencies being difficult to match with operating condition fluctuations, and enhancing the flexibility of parameter adaptation. Verifying the compatibility between candidate parameters and path characteristics eliminates incompatible parameters, ensuring that the final generated dynamic switching frequency parameter set accurately matches the path's operating requirements, avoiding operational jitter and efficiency decline caused by poor adaptation. The entire process revolves around accurately matching the frequency between path characteristics and load requirements, improving the rationality of switching frequency parameters, reducing device damage and energy waste caused by frequency deviations, providing reliable frequency support for efficient topology scheduling, and making the power conversion process more stable and controllable.
[0023] In one embodiment of the present invention, step S3 includes: S31. Extract transient change characteristics from the purification power signal data, conduct response analysis of load change scenarios, and generate load change response data; S32. Conduct response analysis on the purification power signal data under the harmonic injection scenario, and generate harmonic injection response data; S33. Based on the load mutation response data, perform dynamic stability verification on the harmonic injection response data to generate topology operation stability assessment data. S44. Perform topology switching threshold calibration on the topology operation stability assessment data, correct the threshold parameter deviation, and generate topology reconfiguration control command data.
[0024] The working principle and effects of the above technical solution are as follows: Transient change characteristics are extracted from the purification power signal to accurately capture signal fluctuations during load mutations. The generated load mutation response data accurately reflects the system's response capability, avoiding operational imbalances caused by insufficient mutation response analysis and untimely topology switching. Simultaneous analysis of the response status in harmonic injection scenarios supplements the system performance under different interference conditions, making the response data more comprehensive, reducing evaluation bias caused by single-scenario analysis, and avoiding misjudgments of stability due to ignoring harmonic effects. The stability of the harmonic injection response is verified based on the load mutation response data, and cross-validation under actual operating conditions can improve the accuracy of topology operational stability assessment, avoiding evaluation distortion based on partial data. The topology switching threshold is calibrated based on the evaluation data, correcting original parameter deviations and ensuring the threshold setting closely matches the actual operating characteristics of the system, avoiding frequent switching and switching lag caused by excessively high or low thresholds, and reducing device wear and power loss. The entire process, from multi-scenario response analysis to threshold calibration, strengthens stability control at every level. This not only enhances the system's ability to cope with load changes and harmonic interference, but also provides precise instruction support for topology reconfiguration, avoiding system failures caused by unstable operation and ensuring reliable and controllable power conversion throughout the entire process.
[0025] In one embodiment of the present invention, S34 includes: Extract threshold correlation features from topology operation stability assessment data to generate threshold calibration base data; Based on the threshold calibration baseline data analysis, the existing threshold parameter deviation trend is analyzed to generate threshold deviation trend data; Adjust the initial threshold parameter based on the threshold deviation trend data to generate a candidate threshold parameter set; Verify the topology operation stability corresponding to the candidate threshold parameter set, and generate calibrated threshold parameters; After integrating the calibrated threshold parameters, topology reconstruction control command data is generated.
[0026] The working principle and effects of the above technical solution are as follows: Threshold correlation features are extracted from the topology operational stability assessment data. The generated calibration baseline data accurately anchors the core influencing factors of the threshold, avoiding incomplete feature extraction leading to a lack of effective calibration basis and reducing parameter disorder caused by blind adjustments. Based on the baseline data analysis, existing threshold deviation trends are clearly captured, preventing threshold adjustments based solely on single deviations and avoiding exacerbation of reverse deviations, thus reducing the probability of calibration errors. A candidate parameter set is generated to adjust the initial threshold according to the deviation trend, breaking the limitations of fixed-value adjustments and allowing parameters to adapt to changes in system operational trends, enhancing the dynamic adaptability of the threshold and reducing the problem of the threshold being out of sync with the actual operating state. Verifying the topology stability corresponding to the candidate parameters allows for the selection of optimal calibration parameters, preventing the use of incompatible parameters that could cause topology switching anomalies and preventing potential problems such as operational jitter and efficiency decline. The calibrated thresholds are integrated to generate topology reconstruction control commands, ensuring a high degree of consistency between the commands and calibration results, avoiding command execution deviations caused by scattered parameters, and reducing topology reconstruction errors. The entire process revolves around precise calibration based on deviation patterns, which not only improves the rationality and accuracy of threshold parameters but also enhances the reliability of control commands, laying a solid foundation for topology stability reconfiguration and avoiding system operation risks caused by unstable thresholds.
[0027] In one embodiment of the present invention, step S4 includes: S41. Based on the topology reconfiguration control command data, perform topology path reconfiguration operation on the composite power conversion architecture, adjust the on / off state of the topology module, and generate real-time power conversion topology data. S42. Combine the load characteristic parameter set to conduct energy efficiency optimization analysis on the real-time power conversion topology data, calculate the energy efficiency index of different topology combinations, and generate energy efficiency analysis result data. S43. Optimize control parameters based on energy efficiency analysis results to generate multi-objective collaborative control parameters; S44. Perform power loss prediction calculations on multi-objective collaborative control parameters, analyze the power loss trend under different operating conditions, and generate power loss prediction data.
[0028] The working principle and effects of the above technical solution are as follows: Adjusting the on / off state of topology modules based on topology reconfiguration control commands completes path reconfiguration and generates real-time topology data. This allows the composite architecture to dynamically adapt to load and stability requirements, avoiding operational bottlenecks caused by topology rigidity and reducing efficiency losses due to architecture-operating-condition disconnect. Energy efficiency optimization analysis is conducted based on load characteristic parameters, accurately calculating energy efficiency indicators for different topology combinations. This ensures the analysis results are more closely aligned with actual operating scenarios, avoiding one-sided optimization that ignores load impact and preventing energy efficiency improvements without sufficient stability. Control parameters are optimized based on the energy efficiency analysis results. The generated multi-objective collaborative parameters balance power conversion efficiency and operational stability while adapting to fluctuations in different operating conditions, enhancing parameter versatility and adaptability, and reducing the disadvantages of single-objective optimization. Power loss prediction is performed on the collaborative parameters to anticipate loss trends under different operating conditions, preventing losses from exceeding limits without timely adjustment, reducing ineffective energy waste, and providing accurate data support for subsequent full-load efficiency optimization. The entire process, from topology reconstruction to loss prediction, forms a closed-loop optimization, which not only improves the energy efficiency and operational stability of power conversion, but also reduces potential faults and energy losses, allowing the system to remain highly efficient and controllable under complex operating conditions, thus laying a solid foundation for the final intelligent operation data generation.
[0029] In one embodiment of the present invention, step S5 includes: S51. Based on power loss prediction data, collect efficiency sample data across the entire load range, perform full-load efficiency curve fitting, and generate an intelligent efficiency optimization model. S52. Call the intelligent efficiency optimization model to perform topology dynamic reconstruction operation, adjust the topology path combination, and generate topology dynamic reconstruction result data; S53. Perform precise output power adjustment on the topology dynamic reconstruction result data, correct the output power deviation, and generate power adjustment execution data; S54. Integrate the topology dynamic reconfiguration result data with the power regulation execution data to generate power conversion control commands; S55. Conduct risk warning monitoring on power conversion control commands, identify overcurrent, overvoltage and efficiency abnormality risks, and generate risk warning data. S56. Combine risk warning data to carry out energy efficiency optimization processing, further optimize control parameters, and generate intelligent operation data of the power system.
[0030] The working principle and effects of the above technical solution are as follows: Based on power loss prediction data, samples are collected across the entire load range, and an intelligent model is generated by fitting an efficiency curve. This allows optimization to cover all operating conditions, avoiding uneven overall energy efficiency caused by optimizing only local loads and reducing the problem of sudden efficiency drops during load switching. The model is invoked to perform dynamic topology reconstruction, adjusting the path combination to match real-time requirements. This maintains the minimum power conversion efficiency while flexibly responding to load fluctuations, enhancing the system's adaptability under complex operating conditions. The reconstruction results are precisely adjusted to regulate output power, correcting deviations to make the output more stable and preventing power fluctuations from affecting downstream electrical equipment, reducing equipment failures caused by unstable output. Reconstruction and adjustment data are integrated to generate control commands, ensuring consistent command logic, avoiding information fragmentation that could lead to action conflicts, and reducing the probability of command execution errors. Risk warnings are provided for the commands, identifying overcurrent, overvoltage, and efficiency anomalies in advance, preventing potential problems from escalating and damaging components, and reducing downtime and maintenance costs. By combining early warning data with energy efficiency optimization and further refining control parameters, the system's operating energy efficiency is improved and its safety protection capabilities are enhanced. The resulting intelligent operating data enables the power supply to maintain a highly efficient, stable, and safe operating state throughout the process, adapting to diverse power needs.
[0031] In one embodiment of the present invention, S52 includes: The intelligent efficiency optimization model is invoked to extract topology adaptation rules and generate topology reconstruction constraint data. Based on topology reconstruction constraint data analysis, the current topology path operation status is analyzed to generate path status evaluation data; Adjust the topology path combination pattern based on the path status assessment data to generate candidate topology reconstruction schemes; Verify the energy efficiency adaptability of candidate topology reconfiguration schemes and select the optimal reconfiguration scheme; The optimal reconstruction scheme is integrated to generate topology dynamic reconstruction result data.
[0032] The working principle and effects of the above technical solution are as follows: The intelligent efficiency optimization model is invoked to extract topology adaptation rules. The generated constraint data can define reasonable reconfiguration boundaries, avoiding arbitrary adjustments to topology combinations due to fuzzy rules, and reducing energy waste and device wear caused by ineffective reconfiguration attempts. Based on the constraint data, the current path operating status is analyzed, accurately capturing the real-time performance of the path. The generated evaluation data can avoid the problem of adjusting combinations based on distorted states, reducing operational jitter caused by reconfiguration deviations. The evaluation data is used to adjust the topology combination mode to generate candidate solutions, breaking the limitations of fixed combinations, enhancing the adaptability to different load conditions, and avoiding the shortcomings of a single solution in balancing energy efficiency and stability. The energy efficiency adaptability of the candidate solutions is verified, and the optimal solution is selected to ensure that the reconfiguration can improve power conversion efficiency while maintaining topology operational stability, avoiding efficiency decline or potential faults due to poor adaptation. The optimal solution is integrated to generate reconfiguration result data, ensuring data integrity and consistency, avoiding fragmented solutions that lead to disjointed subsequent execution actions, and reducing operational risks caused by instruction conflicts. The entire process revolves around optimizing rule constraints and energy efficiency screening layer by layer, which not only improves the accuracy and rationality of topology dynamic reconstruction, but also strengthens the system's adaptability and provides reliable support for subsequent power regulation.
[0033] One embodiment of the present invention, such as Figure 2 As shown, a system for implementing the power conversion control method of a power supply as described above is provided, the system comprising: Node deployment module: Divides power conversion requirements into multiple topology paths and generates dynamic topology configuration data; constructs a composite power conversion architecture based on the dynamic topology configuration data, the composite power conversion architecture includes LLC, Buck and synchronous rectification topologies, and deploys edge computing nodes to form an intelligent topology reconfiguration network; Compensation processing module: Real-time modeling of load behavior is performed through intelligent topology reconstruction network to generate load characteristic parameter set; Multi-objective collaborative scheduling is performed based on load characteristic parameter set to dynamically adjust topology path combination and switching frequency, complete multi-source power signal acquisition and generate raw power conversion data; Harmonic suppression and aging compensation processing are performed on raw power conversion data to generate purified power signal data; Topology switching module: Extracts transient response features from purification power signal data to obtain load mutation response data and harmonic injection response data; performs dynamic stability evaluation on harmonic injection response data using load mutation response data, calibrates topology switching threshold, and generates topology reconfiguration control command data. The loss prediction module performs topology path reconstruction on the composite power conversion architecture using topology reconstruction control command data to generate real-time power conversion topology data; it performs energy efficiency optimization analysis on the real-time power conversion topology data in conjunction with load characteristic parameter sets to generate multi-objective collaborative control parameters; and it performs power loss prediction on the multi-objective collaborative control parameters to obtain power loss prediction data. Intelligent Operation Module: Based on power loss prediction data, it performs full-load efficiency curve fitting to generate an intelligent efficiency optimization model; based on the intelligent efficiency optimization model, it performs topology dynamic reconstruction and precise output power adjustment to generate power conversion control commands; based on the power conversion control commands, it performs risk warning and energy efficiency optimization processing to generate intelligent operation data of the power system.
[0034] According to one embodiment of the present invention, a power supply includes the system as described above.
[0035] 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 power conversion control method for a power supply, characterized in that, The method includes: S1. Divide the power conversion requirements into multiple topology paths and generate dynamic topology configuration data; construct a composite power conversion architecture based on the dynamic topology configuration data, and deploy edge computing nodes to form an intelligent topology reconfiguration network; S2. Real-time modeling of load behavior is performed through intelligent topology reconfiguration network to generate a load characteristic parameter set; multi-objective collaborative scheduling is performed based on the load characteristic parameter set to dynamically adjust the topology path combination and switching frequency, complete the acquisition of multi-source power signals and generate raw power conversion data; harmonic suppression and aging compensation processing is performed on the raw power conversion data to generate purified power signal data. S3. Extract transient response features from purification power signal data to obtain load mutation response data and harmonic injection response data respectively; evaluate the dynamic stability of harmonic injection response data using load mutation response data, perform topology switching threshold calibration, and generate topology reconfiguration control command data. S4. The composite power conversion architecture is reconstructed using topology reconfiguration control command data to generate real-time power conversion topology data; energy efficiency optimization analysis is performed on the real-time power conversion topology data in combination with load characteristic parameter set to generate multi-objective collaborative control parameters; power loss prediction is performed on the multi-objective collaborative control parameters to obtain power loss prediction data. S5. Fit the full-load efficiency curve based on the power loss prediction data to generate an intelligent efficiency optimization model; perform topology dynamic reconstruction and precise output power adjustment based on the intelligent efficiency optimization model to generate power conversion control commands; perform risk warning and energy efficiency optimization processing based on the power conversion control commands to generate intelligent operation data of the power system.
2. The power conversion control method for the power supply according to claim 1, characterized in that, S1 includes: S11. Collect input and output parameters and load condition information of the power conversion scenario, and generate power conversion demand characteristic data. S12. Perform multi-topology path partitioning on the power conversion demand characteristic data to generate dynamic topology configuration data; S13. Based on the dynamic topology configuration data, build a composite power conversion architecture, integrate LLC Buck and synchronous rectification topology modules, and generate composite topology hardware architecture data. S14. Deploy edge computing units on the physical nodes corresponding to the composite topology hardware architecture data to build an intelligent topology reconfiguration network and generate basic data for the operation of the topology reconfiguration network.
3. The power conversion control method for the power supply according to claim 1, characterized in that, The S2 includes: S21. Call the basic data for topology reconstruction network operation, start the load perception module of intelligent topology reconstruction network, carry out real-time modeling of load behavior, and generate a set of load characteristic parameters. S22. Perform multi-objective collaborative scheduling based on load characteristic parameter set, dynamically adjust topology path combination and switching frequency parameters, and generate topology scheduling execution data; S23. Initiate a multi-source power signal acquisition process for topology scheduling execution data, acquire power signals from the input side, output side, and topology nodes, and generate raw power conversion data; S23. Perform harmonic suppression processing on the original power conversion data to filter out high-frequency harmonics and low-frequency noise components, and generate harmonic-suppressed power data. S25. Perform device aging compensation processing on the power data after harmonic suppression to correct power signal deviation and generate purified power signal data.
4. The power conversion control method for the power supply according to claim 3, characterized in that, S22 includes: S221. Perform multi-objective dimension analysis on the load characteristic parameter set to generate multi-objective scheduling characteristic data; S222. Based on multi-objective scheduling feature data, perform topology path combination optimization to generate optimized topology path combination data; S223. Perform dynamic switching frequency matching on the preferred topology path combination data to generate a dynamic switching frequency parameter set; S224. Integrate the optimal topology path combination data with the dynamic switching frequency parameter set to generate topology scheduling execution data.
5. The power conversion control method for the power supply according to claim 4, characterized in that, S223 includes: Extract pathway operation characteristics from the data of the preferred topological pathway combinations to generate pathway characteristic characterization data; Based on the analysis of path characteristic characterization data, load adaptation frequency requirements are analyzed to generate frequency requirement interval data. Adjust the initial frequency value for the frequency demand range data to generate candidate switching frequency parameters; Verify the fit between candidate switching frequency parameters and path characteristics, and generate a dynamic switching frequency parameter set.
6. The power conversion control method for the power supply according to claim 1, characterized in that, The S3 includes: S31. Extract transient change characteristics from the purification power signal data, conduct response analysis of load change scenarios, and generate load change response data; S32. Conduct response analysis on the purification power signal data under the harmonic injection scenario, and generate harmonic injection response data; S33. Based on the load mutation response data, perform dynamic stability verification on the harmonic injection response data to generate topology operation stability assessment data. S44. Perform topology switching threshold calibration on the topology operation stability assessment data, correct the threshold parameter deviation, and generate topology reconfiguration control command data.
7. The power conversion control method for the power supply according to claim 1, characterized in that, The S4 includes: S41. Based on the topology reconfiguration control command data, perform topology path reconfiguration operation on the composite power conversion architecture, adjust the on / off state of the topology module, and generate real-time power conversion topology data. S42. Combine the load characteristic parameter set to conduct energy efficiency optimization analysis on the real-time power conversion topology data, calculate the energy efficiency index of different topology combinations, and generate energy efficiency analysis result data. S43. Optimize control parameters based on energy efficiency analysis results to generate multi-objective collaborative control parameters; S44. Perform power loss prediction calculations on multi-objective collaborative control parameters, analyze the power loss trend under different operating conditions, and generate power loss prediction data.
8. The power conversion control method for the power supply according to claim 1, characterized in that, The S5 includes: S51. Based on power loss prediction data, collect efficiency sample data across the entire load range, perform full-load efficiency curve fitting, and generate an intelligent efficiency optimization model. S52. Call the intelligent efficiency optimization model to perform topology dynamic reconstruction operation, adjust the topology path combination, and generate topology dynamic reconstruction result data; S53. Perform precise output power adjustment on the topology dynamic reconstruction result data, correct the output power deviation, and generate power adjustment execution data; S54. Integrate the topology dynamic reconfiguration result data with the power regulation execution data to generate power conversion control commands; S55. Conduct risk warning monitoring on power conversion control commands, identify overcurrent, overvoltage and efficiency abnormality risks, and generate risk warning data. S56. Combine risk warning data to carry out energy efficiency optimization processing, further optimize control parameters, and generate intelligent operation data of the power system.
9. A system for implementing the power conversion control method of a power supply as described in claim 1, characterized in that, The system includes: Node deployment module: Divides power conversion requirements into multiple topology paths and generates dynamic topology configuration data; constructs a composite power conversion architecture based on the dynamic topology configuration data, and deploys edge computing nodes to form an intelligent topology reconfiguration network; Compensation processing module: Real-time modeling of load behavior is performed through intelligent topology reconstruction network to generate load characteristic parameter set; Multi-objective collaborative scheduling is performed based on load characteristic parameter set to dynamically adjust topology path combination and switching frequency, complete multi-source power signal acquisition and generate raw power conversion data; Harmonic suppression and aging compensation processing are performed on raw power conversion data to generate purified power signal data; Topology switching module: Extracts transient response features from purification power signal data to obtain load mutation response data and harmonic injection response data; performs dynamic stability evaluation on harmonic injection response data using load mutation response data, calibrates topology switching threshold, and generates topology reconfiguration control command data. The loss prediction module performs topology path reconstruction on the composite power conversion architecture using topology reconstruction control command data to generate real-time power conversion topology data; it performs energy efficiency optimization analysis on the real-time power conversion topology data in conjunction with load characteristic parameter sets to generate multi-objective collaborative control parameters; and it performs power loss prediction on the multi-objective collaborative control parameters to obtain power loss prediction data. Intelligent Operation Module: Based on power loss prediction data, it performs full-load efficiency curve fitting to generate an intelligent efficiency optimization model; based on the intelligent efficiency optimization model, it performs topology dynamic reconstruction and precise output power adjustment to generate power conversion control commands; based on the power conversion control commands, it performs risk warning and energy efficiency optimization processing to generate intelligent operation data of the power system.
10. A power supply, characterized in that, The power source includes the system as described in claim 9.