CN-120999891-B - Power supply modularization combination method and system based on standardized interface

CN120999891BCN 120999891 BCN120999891 BCN 120999891BCN-120999891-B

Abstract

The invention discloses a power supply modularization combination method and a system based on a standardized interface, wherein the method comprises the following steps of collecting data through a sensor; the method comprises the steps of integrating data by a central control hub, establishing a four-dimensional cooperative model, adjusting by a MOSFET driving circuit according to a control command of the central control hub, predicting the load change trend and the performance attenuation condition of each module of a current power supply system by an AI prediction current sharing module through a built-in machine learning model, dynamically adjusting the current distribution of each module based on a prediction result and combining a target current value in the command, wherein the system comprises the central control hub, a sensor unit, the AI prediction current sharing module and the MOSFET driving circuit.

Inventors

HAN JIN
JI YONGJIE

Assignees

太原永明恒动源电子有限公司
百信信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20250804

Claims (6)

1. The power supply modularization combination method based on the standardized interface is characterized by comprising the following steps: collecting output impedance data, current data, junction temperature data and fault signals through a sensor, and transmitting the data to a central control hub; the central control hub is used as a data fusion center, data fed back by the sensors are collected uniformly, a decision scheduling engine is built, a four-dimensional collaborative model is built, iterative optimization is carried out on the four-dimensional collaborative model through the decision scheduling engine, and a model optimization result is converted into a control instruction and fed back to the AI prediction current sharing module and the MOSFET driving circuit; The MOSFET driving circuit receives a control instruction issued by the central control hub, changes the output impedance and the current output of the power supply module by adjusting the conduction degree and the switching frequency of the MOSFET, and feeds back the executed state data to the central control hub; The AI prediction current sharing module predicts the load change trend of the current power supply system and the performance attenuation condition of each module through a built-in machine learning model, dynamically adjusts the current distribution of each module based on the prediction result and the target current value in the instruction, and feeds back the adjusted current distribution result to the central control hub for the optimization adjustment of the subsequent control strategy; the power supply system comprises a central control hub, a sensor unit, an AI prediction current sharing module and a MOSFET driving circuit; the current distribution of each module is dynamically adjusted in the following specific ways: When the load is predicted to be increased, the current output of each module is increased in advance, and when the load is reduced, the current of each module is reduced in proportion; Aiming at the module with the performance attenuation, according to the historical data and the current state, the current distribution proportion of the module is reduced, the reduced current is distributed to other modules, the overall current sharing effect and stability of the system are ensured, and meanwhile, the adjusted current distribution result is fed back to the central control hub; the machine learning model adopts a machine learning model of regression fusion of a long-term memory network and a support vector.
2. A standardized interface based power modular combination system for performing the standardized interface based power modular combination method of claim 1, wherein the sensor unit is configured to collect data and transmit the data to the central control hub, the sensor unit comprises a vector impedance sensor, a current sensor, a temperature sensor, and a fault detection sensor, the vector impedance sensor is configured to collect output impedance data of the power module, the current sensor is configured to collect real-time current data, the temperature sensor is configured to collect junction temperature data, and the fault detection sensor is configured to collect fault signals.
3. The standardized interface-based power modular combination system of claim 2 wherein the central control hub serves as a data fusion center for collecting data transmitted by the sensor units by means of clock synchronization technology, constructing a decision scheduling engine, and establishing a four-dimensional collaborative model.
4. The standardized interface-based power modular combination system of claim 2, wherein the four-dimensional collaborative model is configured to implement multi-objective dynamic balance by coupling and modeling four key indicators of a power output stability dimension, a module life dimension, an energy efficiency dimension, and a fault response dimension, and to build an instruction distribution system using a CANFD bus to form closed-loop control.
5. The standardized interface-based power modular combination system of claim 2 wherein the central control hub has a current sharing control algorithm built therein for calculating control parameters required for adjusting the output of the power module, thereby generating control commands for issuing to the MOSFET drive circuit, and changing the output impedance and current output of the power module by adjusting the MOSFET drive circuit.
6. The standardized interface-based power modular combination system of claim 5 wherein the current sharing control algorithm is: ; Wherein, the For the turn-on ratio adjustment amount of the kth iteration, For the adaptation of the step size, In order to output the impedance gradient vector, For the purpose of equalizing the flow of the weight coefficients, For the average output current of the group of modules, The current is output for the current module.

Description

Power supply modularization combination method and system based on standardized interface Technical Field The invention relates to the field of power modules, in particular to a power modular combination method and system based on a standardized interface. Background With the rapid development of fields such as new energy, industrial automation, data centers and the like, higher requirements are provided for flexibility, reliability and efficiency of a power supply system, the traditional power supply system has the problems of low modularization degree, complex fault maintenance and the like, diversified load requirements and intelligent management requirements are difficult to meet, the power supply system still has technical bottlenecks in aspects such as multi-module cooperative control, dynamic performance optimization, predictive maintenance and the like, for example, insufficient flow equalization control precision among modules leads to system efficiency reduction and shortened service life of the modules, the rapid reconstruction and self-adaptive adjustment capability when faults occur is limited, the system reliability is influenced, and in addition, with the deep fusion of industrial Internet and artificial intelligence technology, the power supply system needs to have intelligent decision capability of data driving so as to realize the functions of multi-target dynamic balance and full life cycle management. Disclosure of Invention The invention aims to solve the problems that the traditional power supply system has low modularization degree, complex fault maintenance and difficulty in meeting diversified load demands and intelligent management demands, and therefore provides a power supply modularization combination method and system based on standardized interfaces. The invention can realize the aim through the following technical scheme that the power supply modularization combination method and system based on the standardized interface comprise the following steps: collecting output impedance data, current data, junction temperature data and fault signals through a sensor, and transmitting the data to a central control hub; the central control hub is used as a data fusion center, data fed back by the sensors are collected uniformly, a decision scheduling engine is built, a four-dimensional collaborative model is built, iterative optimization is carried out on the four-dimensional collaborative model through the decision scheduling engine, and a model optimization result is converted into a control instruction and fed back to the AI prediction current sharing module and the MOSFET driving circuit; The MOSFET driving circuit receives a control instruction issued by the central control hub, changes the output impedance and the current output of the power supply module by adjusting the conduction degree and the switching frequency of the MOSFET, and feeds back the executed state data to the central control hub; And fourthly, the AI prediction current sharing module predicts the load change trend of the current power supply system and the performance attenuation condition of each module through a built-in machine learning model, dynamically adjusts the current distribution of each module based on a prediction result and a target current value in an instruction, and feeds back the adjusted current distribution result to a central control hub for the optimization adjustment of a subsequent control strategy. The invention also provides a power supply modularization combination system based on the standardized interface, and the method is applied to the power supply modularization combination system, and the system comprises a central control hub, a sensor unit, an AI prediction current sharing module and a MOSFET driving circuit. The sensor unit is used for collecting data and transmitting the data to the central control hub, and comprises a vector impedance sensor, a current sensor, a temperature sensor and a fault detection sensor, wherein the vector impedance sensor is used for collecting output impedance data of the power supply module, the current sensor is used for collecting real-time current data, the temperature sensor is used for collecting junction temperature data, and the fault detection sensor is used for collecting fault signals. The central control hub is used as a data fusion center, data transmitted by the sensor units are collected by means of clock synchronization technology, a decision scheduling engine is built, and a four-dimensional collaborative model is built. The four-dimensional collaborative model is characterized in that the four key indexes of the power supply output stability dimension, the module service life dimension, the energy efficiency dimension and the fault response dimension are subjected to coupling modeling, so that multi-target dynamic balance is realized, and an instruction distribution system is built by utilizing a CANFD bus to form closed-loop contr