A modular power system with low loss and high efficiency energy transmission

Abstract

The application discloses a low-loss high-efficiency energy transmission modular power supply system, and belongs to the technical field of power supplies.The system comprises a load state prediction module, which is used for monitoring the total load current of an output bus in real time and calculating a load current change rate, and is used for calculating a predicted load current on the basis of the total load current and the load current change rate; a system risk assessment module, which is used for quantitatively calculating a system instability risk index according to the predicted load current output by the load state prediction module and the total rated output current of a current working unit; and a self-adaptive marshalling decision module, which is used for generating switching decision instructions of increasing units, decreasing units or maintaining the same according to the system instability risk index calculated by the system risk assessment module.The application introduces a dynamic dimension of change rate, so that the whole control system can predict the load rising or falling in advance, and the key response time is obtained for subsequent assessment and decision.

Description

Technical Field

[0001] This invention relates to the field of power supplies, specifically to a modular power supply system with low loss and high efficiency energy transmission. Background Technology

[0002] In high-reliability applications such as industrial automation and data centers, multiple power modules are typically connected in parallel to increase total output power and provide N+M redundancy. In existing technologies, the activation and deactivation of parallel modules often rely on a hard switching logic based on a fixed load percentage. For example, when the total load current exceeds a preset threshold, such as 80%, of the total rated capacity of the current working unit, the system will activate a new standby unit; conversely, when the load drops below a certain lower threshold, a working unit will be deactivated to reduce energy consumption.

[0003] This type of fixed-threshold control strategy has inherent flaws. When the load fluctuates rapidly around the threshold, it leads to frequent switching of parallel units, generating unnecessary switching losses and stress. More seriously, when the load current rapidly increases and crosses the critical point of adding or removing units, the inherent response delay of the control system and the impedance mismatch at the moment of connecting a new unit can cause significant undershoot or overshoot in the output bus voltage. Such transient voltage fluctuations can seriously affect the stable operation of load equipment and even cause malfunctions in precision instruments. Although existing technologies can ensure the current distribution accuracy of each unit under steady state through current sharing bus technology, they are insufficient in suppressing transient voltages during dynamic switching. Therefore, how to achieve smooth and seamless switching of parallel units from N to N+1, or from N+1 to N operating states, has become a key technical bottleneck for improving the dynamic performance of modular power supply systems.

[0004] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a modular power supply system with low loss and high efficiency energy transmission to solve the problems mentioned in the background art.

[0006] The technical solution of the present invention includes:

[0007] The load condition prediction module is used to monitor the total load current of the output bus in real time and calculate the load current change rate. Based on the total load current and the load current change rate, the predicted load current is calculated.

[0008] The system risk assessment module is used to quantitatively calculate the system instability risk index based on the predicted load current output by the load state prediction module and the total rated output current of the current working unit.

[0009] The adaptive grouping decision module is used to generate switching decision instructions to add, remove, or keep the system unchanged based on the system instability risk index calculated by the system risk assessment module.

[0010] The dynamic pre-adjustment execution module is used to respond to the unit addition switching decision command generated by the adaptive grouping decision module. Based on the predicted load current and the current bus voltage, it calculates the target output voltage of the unit to be connected and then connects the unit to the bus after pre-adjusting it according to the target output voltage.

[0011] Preferably, the load state prediction module is specifically used for:

[0012] A linear prediction model is used to generate the predicted load current by adding the current total load current value to the product of the load current change rate and the preset prediction time domain.

[0013] Preferably, the system risk assessment module is specifically used for:

[0014] Set a safe operating area coefficient to define the safe operating area boundary under the total rated output current of the current working unit;

[0015] If the predicted load current does not exceed the safe operating area boundary, the system instability risk index is determined to be zero.

[0016] If the predicted load current exceeds the boundary of the safe operating area, calculate the ratio of the portion of the predicted load current exceeding the boundary of the safe operating area to the total rated output current to obtain the risk benchmark ratio.

[0017] Based on the risk benchmark ratio, the system instability risk index is generated by squaring the ratio and multiplying it by the risk weighting coefficient.

[0018] Preferably, before generating the handover decision instruction, the adaptive grouping decision module is specifically used for:

[0019] Determine the absolute value of the load current change rate; multiply the absolute value by the dynamic adjustment coefficient to obtain the threshold adjustment amount; subtract the threshold adjustment amount from the basic decision threshold to obtain the initial input threshold; take the larger value between the initial input threshold and the preset minimum threshold as the final dynamic input threshold.

[0020] Preferably, the adaptive grouping decision module is specifically used for:

[0021] If the system instability risk index is greater than the dynamic input threshold, a unit switching decision instruction will be generated.

[0022] If the system instability risk index is not greater than the dynamic input threshold, then determine whether the total load current is continuously lower than the safe working area boundary formed by N-1 working units, and whether the system instability risk index is continuously lower than the fixed cut-off threshold.

[0023] If both conditions for subsequent judgment are met and the duration exceeds the preset delay confirmation time, a unit reduction switching decision instruction is generated; otherwise, an instruction to maintain the current number of working units is generated.

[0024] Preferably, the dynamic pre-adjustment execution module, when calculating the target output voltage, is specifically used for:

[0025] Based on the predicted load current and the total number of working units after grid connection, determine the predicted load share that the unit to be connected should share; multiply the predicted load share by the preset virtual output impedance to obtain the voltage compensation value; add the voltage compensation value to the real-time monitored bus voltage to generate the target output voltage.

[0026] Preferably, the dynamic pre-adjustment execution module is further specifically used for:

[0027] Before the grid-connection relay of the unit to be connected closes, the no-load voltage of the unit to be connected is adjusted to the target output voltage by adjusting the internal PWM controller.

[0028] Preferably, the safe working area coefficient is a preset parameter characterizing the maximum allowable steady-state load rate of the system; the risk weighting coefficient is a preset parameter used to adjust the overall sensitivity of the system's instability risk index.

[0029] This invention provides a modular power supply system with low-loss and high-efficiency energy transmission, which has the following improvements and advantages compared with the prior art:

[0030] 1. This solution gains the ability to predict load change trends by introducing the dynamic dimension of the rate of change. This allows the entire control system to anticipate load increases or decreases in advance, providing crucial response time for subsequent assessment and decision-making. Existing technologies require waiting for the actual current to reach, for example, a threshold point of 80% of the total capacity before initiating the unit addition operation, at which point the bus voltage may have already begun to drop. The load status prediction module in this solution, by monitoring a large positive current change rate, quickly calculates a predicted load current that far exceeds the current value, thereby triggering subsequent response procedures in advance.

[0031] 2. This solution transforms the abstract risk of instability into a precise, non-linear quantitative indicator, providing an effective mathematical simulation of the physical behavior of a system when it is on the verge of instability. This contrasts sharply with the simple threshold comparisons in existing technologies, which can only determine whether or not a boundary has been crossed, while this solution can measure how much the boundary has been crossed and how dangerous the situation is.

[0032] 3. This solution realizes the dynamic nature of the decision threshold and the asymmetry of the decision logic, effectively suppressing switching oscillations and ensuring that the system only reduces the number of working units after confirming that the load has decreased to a stable trend. It completely solves the problem of frequent power module switching on and off due to small fluctuations near the load critical point in the existing technology, and significantly improves the system's operational stability and component lifespan.

[0033] 4. This solution eliminates the transient impact caused by grid connection operation, transforming the grid connection process from a severe electrical shock process into a smooth and imperceptible electrical state transition process. It solves the problem of bus voltage undershoot or overshoot caused by impedance mismatch and response delay in the existing technology, reduces the peak voltage fluctuation during dynamic switching by at least one order of magnitude, greatly improves the dynamic response speed and output voltage stability of the power supply system, and provides cleaner and more reliable power for high-precision load equipment. Attached Figure Description

[0034] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0035] Figure 1 This is a flowchart of a modular power supply system for low-loss and high-efficiency energy transmission according to the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0037] Example 1:

[0038] Please see Figure 1 The present invention provides a modular power supply system with low loss and high efficiency energy transmission, including: a load state prediction module, used to monitor the total load current of the output bus in real time and calculate the load current change rate, and to deduce the predicted load current based on the total load current and the load current change rate.

[0039] The system risk assessment module is used to quantitatively calculate the system instability risk index based on the predicted load current output by the load state prediction module and the total rated output current of the current working unit.

[0040] The adaptive grouping decision module is used to generate switching decision instructions to add, remove, or keep the system unchanged based on the system instability risk index calculated by the system risk assessment module.

[0041] The dynamic pre-adjustment execution module is used to respond to the unit addition switching decision command generated by the adaptive grouping decision module. Based on the predicted load current and the current bus voltage, it calculates the target output voltage of the unit to be connected and connects it to the bus after pre-adjusting the unit to be connected according to the target output voltage.

[0042] The modular power supply system disclosed in this embodiment solves the problem of bus voltage fluctuation caused by fixed threshold switching in the prior art through an integrated control system consisting of a load state prediction module, a system risk assessment module, an adaptive grouping decision module, and a dynamic pre-adjustment execution module. The system abandons the passive response hard switching logic and establishes a closed-loop control method that combines prediction, assessment, decision-making, and pre-adjustment. This method aims to achieve smooth dynamic grouping of power modules, suppress transient voltage overshoot and undershoot, and provide a stable and reliable power supply for application scenarios with stringent power quality requirements, such as data centers or precision manufacturing.

[0043] Example 2:

[0044] The load status prediction module is specifically used for:

[0045] A linear prediction model is used to add the current total load current value to the product of the load current change rate and the preset prediction time domain to generate the predicted load current.

[0046] The load condition prediction module aims to enable the system to anticipate load changes, providing a decision-making time margin for subsequent control. This module uses an integrated digital signal processor and a linear extrapolation algorithm based on first-order Taylor expansion to predict future load current. Its underlying logic is to overcome the inherent response delay of traditional control strategies. By introducing the rate of change of current, the system can sense the trend of load changes, shifting from passive response to proactive prediction. This module uses the following linear prediction formula to calculate the predicted load current:

[0047] ;

[0048] in To predict the load current (A), The current load current (A) is sampled by the high-precision Hall sensor at the current moment. The preset prediction time domain (s) is experimentally calibrated based on the inherent physical switching delay of the system hardware to achieve a balance between prediction accuracy and response speed. This calibration process typically involves multiple measurements of the time required for the power module to receive a command and for its grid-connected relay to fully close, and the average value is taken as the prediction time. The benchmark ensures that the predicted time can cover the delay of physical actions. For digital signal processors to perform continuous The current load current change rate (A / s) is obtained by performing differential calculation on the sampled values; the predicted load current is calculated. It is transmitted in real time to the system risk assessment module, serving as the core input for subsequent assessments;

[0049] This invention, by incorporating a load state prediction module, achieves the ability to anticipate load change trends; the linear prediction model employed by this module is embodied in the formula... The physical meaning lies in the fact that it is based on the currently measurable current value. and its rate of change For a very short prediction time domain of the future Subsequent load current Linear extrapolation is performed; the validity of this formula, based on a first-order Taylor expansion, is an effective approximation of the system's behavior in the short time domain; compared to existing techniques that rely solely on the current value. In this solution, the rate of change is introduced as a dynamic dimension, enabling the entire control system to anticipate load increases or decreases in advance, thus gaining crucial response time for subsequent evaluation and decision-making. For example, in a scenario where a data center server cluster initiates high-load computation, the load current will rise rapidly. Existing technologies require waiting for the current to actually reach a threshold point, such as 80% of the total capacity, before initiating the unit addition operation, at which point the bus voltage may have already begun to drop. The load status prediction module in this solution detects large positive current change rates. It can quickly calculate a predicted load current that is far greater than the current value. This allows for the early triggering of subsequent response procedures.

[0050] Example 3:

[0051] The system risk assessment module is specifically used for:

[0052] Set a safe operating area coefficient to define the safe operating area boundary under the total rated output current of the current working unit;

[0053] If the predicted load current does not exceed the safe operating area boundary, the system instability risk index is determined to be zero.

[0054] If the predicted load current exceeds the boundary of the safe operating area, calculate the ratio of the portion of the predicted load current exceeding the boundary of the safe operating area to the total rated output current to obtain the risk benchmark ratio.

[0055] Based on the risk benchmark ratio, the ratio is squared and multiplied by the risk weighting coefficient to generate the system instability risk index.

[0056] The safe operating area coefficient is a preset parameter characterizing the maximum allowable steady-state load rate of the system; the risk weighting coefficient is a preset parameter used to adjust the overall sensitivity of the system's instability risk index.

[0057] The system risk assessment module receives the predicted load current. Its function is to transform the abstract risk of load exceeding the system's carrying capacity into precise, quantifiable indicators that can be used for decision-making; to enable the system to respond more strongly to severe load shocks, this assessment mechanism uses a non-linear relationship to amplify the danger signal; system instability risk index. The calculation rule is defined as: when hour, ;when hour, ;in It is a dimensionless system instability risk index; To predict the load current (A); This represents the total rated output current (A) of the current N working units; This is a dimensionless safe operating area coefficient. Its value is determined based on a trade-off between system reliability and operational economy, using experimental data or referencing industry safety standards for the target application scenario. It characterizes the maximum allowable steady-state load rate of the system. For example, if industry standards require an N+1 redundant system to continue operating safely after a single module failure, and considering that the power supply module operates most efficiently at 80% load, then... The value can be set to around 0.8; This is a dimensionless risk weighting coefficient, whose value is calibrated during the system commissioning phase by applying a standard load step test to adjust the overall sensitivity of the risk index. A load shock considered to be at a critical threshold can be applied, for example, jumping from 50% load to 95% load within 0.1 seconds, and then adjusted accordingly. The value of makes the risk index calculated at this moment... It just reaches a preset threshold sufficient to trigger the unit-increment decision; this quadratic relationship makes the risk index... As the degree of load exceeding the limit increases, it increases rapidly and non-linearly, forming a strong alarm signal and transmitting it to the decision module to ensure that the system can make a timely and necessary response to truly threatening load shocks.

[0058] This invention, by setting up a system risk assessment module, transforms the abstract instability risk into a precise, non-linear quantitative indicator; the risk index model used in this module... Its essence is to predict the load current. Beyond the boundary of the safe work area The part is normalized and squared; the quadratic relationship in the formula is crucial, as it determines the system instability risk index. The growth rate is much faster than the extent of load exceeding limits, thus generating a stronger response signal to severe load shocks; this design is an effective mathematical simulation of the physical behavior of the system when it is on the verge of instability; this is in stark contrast to the simple threshold comparison in existing technologies, which can only determine whether or not the load has exceeded limits, while this solution can measure how much the load has exceeded limits and how dangerous the situation is.

[0059] Example 4:

[0060] Before generating handover decision instructions, the adaptive grouping decision module is specifically used for:

[0061] Determine the absolute value of the load current change rate; multiply the absolute value by the dynamic adjustment coefficient to obtain the threshold adjustment amount; subtract the threshold adjustment amount from the basic decision threshold to obtain the initial input threshold; take the larger value between the initial input threshold and the preset minimum threshold as the final dynamic input threshold.

[0062] The adaptive grouping decision module is specifically used for:

[0063] If the system instability risk index is greater than the dynamic input threshold, a unit switching decision instruction will be generated.

[0064] If the system instability risk index is not greater than the dynamic input threshold, then determine whether the total load current is continuously lower than the safe working area boundary formed by N-1 working units, and whether the system instability risk index is continuously lower than the fixed cut-off threshold.

[0065] If both conditions of the subsequent judgment are met and the duration exceeds the preset delay confirmation time, a unit reduction switching decision instruction is generated; otherwise, an instruction to maintain the current number of working units is generated.

[0066] The adaptive grouping decision-making module is the core control unit of the system. It adopts a decision-making mechanism based on a risk index and combined with dynamic hysteresis characteristics to achieve a balance between rapid response and oscillation prevention; the decision-making of additional units is based on dynamic input thresholds. Control, its calculation formula is:

[0067] ;

[0068] in The final dynamic input threshold is dimensionless. A fixed minimum threshold, a dimensionless parameter, is set to a positive minimum value to ensure logical robustness. The basic decision threshold is dimensionless and its value is based on the risk index corresponding to the maximum load rate allowed when the system reaches steady state. Configure; This is a dynamic adjustment factor (s / A), the value of which has been experimentally calibrated to ensure that, under the expected maximum load change rate, It can effectively reduce load changes and thus achieve a proactive response; during calibration, it can record different load change rates. The point at which the lower bus voltage begins to experience an unacceptable drop, and then reverse adjustment. Values ​​that ensure dynamic input thresholds at these rates of change. It can be triggered in advance, thereby suppressing voltage fluctuations within the required range; Let be the absolute value of the load current change rate (A / s); the decision rule for adding a unit is defined as: if If so, an instruction to increase the number of units is generated; the decision to decrease the number of units adopts an asymmetric delay confirmation strategy, the rule of which is: if the total load Continuously below the boundary of the safe working area consisting of N-1 units And risk index Persistently below a fixed resection threshold Exceeding the preset delay confirmation time Then a decrement instruction is generated; where The rated current (A) for a single module. The set value is much smaller than To constitute a decision lag interval, The setting is determined based on the duration of typical transient load dips in the target application scenario. For example, in a data center scenario, after a server cluster completes a high-intensity computing task, its load current may experience a dip of several seconds. It can be set to a value slightly longer than the typical low point time, such as 3-5 seconds, to prevent the reduction unit action from being triggered erroneously due to a brief load drop; this asymmetric decision logic avoids switching oscillations near the critical point.

[0069] This invention, by setting an adaptive grouping decision module, achieves dynamic decision thresholds and asymmetric decision logic, effectively suppressing switching oscillations; the decision to add units is based on a dynamic input threshold. The practical significance of this formula lies in the fact that the decision-making threshold is not static, but rather varies with the "degree of drastic change" in the load. Dynamic correlation; the faster the load changes, the higher the input threshold. The lower the threshold, the more sensitive the system response; at the same time, reducing the number of units introduces an independent fixed cut-off threshold. With delayed confirmation time This design establishes a lag interval for decision-making; it ensures that the system only reduces the number of working units after confirming that the load has decreased to a stable trend, thus completely solving the problem of frequent power module switching on and off due to small fluctuations near the load critical point in existing technologies, and significantly improving the system's operational stability and component lifespan.

[0070] Example 5:

[0071] The dynamic pre-adjustment execution module is specifically used in calculating the target output voltage for:

[0072] Based on the predicted load current and the total number of working units after grid connection, determine the predicted load share that the unit to be connected should share; multiply the predicted load share by the preset virtual output impedance to obtain the voltage compensation value; add the voltage compensation value to the real-time monitored bus voltage to generate the target output voltage.

[0073] The dynamic pre-adjustment execution module is also specifically used for:

[0074] Before the grid-connection relay of the unit to be connected is closed, the no-load voltage of the unit to be connected is adjusted to the target output voltage by adjusting the internal PWM controller.

[0075] The dynamic pre-adjustment execution module is the physical execution unit for achieving seamless grid connection. Its technical logic is rooted in the droop control principle. Through inverse calculation, it proactively sets the output voltage of the standby unit before it is physically connected to the bus, eliminating transient current surges caused by voltage differences and impedance mismatches at the moment of grid connection. The target output voltage of the standby unit... Solve using the following formula:

[0076] ;

[0077] in It is the target output voltage (V) of the unit to be incorporated; It is the current bus voltage (V) monitored in real time; The predicted total load current (A) is calculated by the prediction module; The number of units currently connected to the grid; This is a virtual output impedance (Ω) uniformly set for all power supply modules. This parameter value is preset in each module to define its droop characteristics and ensure stable current sharing among parallel modules. The value is determined based on the system's maximum allowable output voltage deviation and the module's rated current. Typically, its selection should ensure that the voltage drop caused by virtual impedance is within an acceptable range, for example, not exceeding 3%-5% of the rated voltage, throughout the module's transition from no-load to full-load. Upon receiving the increment command, the digital signal processor immediately calculates the value using this formula. It also instructs the pulse width modulation controller of the standby unit to adjust the switching duty cycle so that its open-circuit voltage reaches This pre-adjustment process is completed before the grid-connection relay closes; when the relay closes, the unit can smoothly assume its due load share, thereby suppressing any disturbances in the bus voltage and transforming the parallel connection process from a transient impact process into a smooth electrical state transition process.

[0078] This invention eliminates the transient impact caused by grid connection operations by setting up a dynamic pre-adjustment execution module. The core of this module lies in reversing the droop control principle to determine the target output voltage that the unit to be connected needs to achieve before grid connection. The derivation of this formula is rigorous, ensuring that the port voltage of the unit to be connected to the bus is exactly equal to the real-time bus voltage at the instant it connects to the bus. Meanwhile, its output current can smoothly handle the predicted load share allocated to it by the system. By pre-adjusting the internal PWM controller of the unit to be connected before the relay closes to achieve the target output voltage, this solution transforms the grid connection process from a severe electrical shock process into a smooth and imperceptible electrical state transition process. This directly solves the problem of bus voltage undershoot or overshoot caused by impedance mismatch and response delay in the prior art, reduces the peak voltage fluctuation during dynamic switching by at least one order of magnitude, greatly improves the dynamic response speed and output voltage stability of the power supply system, and provides cleaner and more reliable power for high-precision load equipment.

[0079] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A modular power supply system with low loss and high efficiency energy transmission, characterized in that, include: The load condition prediction module is used to monitor the total load current of the output bus in real time and calculate the load current change rate. Based on the total load current and the load current change rate, the predicted load current is calculated. The system risk assessment module is used to quantitatively calculate the system instability risk index based on the predicted load current output by the load state prediction module and the total rated output current of the current working unit. The adaptive grouping decision module is used to generate switching decision instructions to add, remove, or keep the system unchanged based on the system instability risk index calculated by the system risk assessment module. The dynamic pre-adjustment execution module is used to respond to the unit addition switching decision command generated by the adaptive grouping decision module. Based on the predicted load current and the current bus voltage, it calculates the target output voltage of the unit to be connected and connects it to the bus after pre-adjusting the unit to be connected according to the target output voltage. The system risk assessment module is specifically used for: Set a safe operating area coefficient to define the safe operating area boundary under the total rated output current of the current working unit; If the predicted load current does not exceed the safe operating area boundary, the system instability risk index is determined to be zero. If the predicted load current exceeds the boundary of the safe operating area, calculate the ratio of the portion of the predicted load current exceeding the boundary of the safe operating area to the total rated output current to obtain the risk benchmark ratio. Based on the risk benchmark ratio, the ratio is squared and multiplied by the risk weighting coefficient to generate the system instability risk index. Before generating the handover decision instruction, the adaptive grouping decision module is specifically used for: Determine the absolute value of the load current change rate; multiply the absolute value by the dynamic adjustment coefficient to obtain the threshold adjustment amount; subtract the threshold adjustment amount from the basic decision threshold to obtain the initial input threshold; take the larger value between the initial input threshold and the preset minimum threshold as the final dynamic input threshold. The adaptive grouping decision module is specifically used for: If the system instability risk index is greater than the dynamic input threshold, a unit switching decision instruction will be generated. If the system instability risk index is not greater than the dynamic input threshold, then determine whether the total load current is continuously lower than the safe working area boundary formed by N-1 working units, and whether the system instability risk index is continuously lower than the fixed cut-off threshold. If both conditions for subsequent judgment are met and the duration exceeds the preset delay confirmation time, a unit reduction switching decision instruction is generated; otherwise, an instruction to maintain the current number of working units is generated.

2. The modular power supply system for low-loss, high-efficiency energy transmission according to claim 1, characterized in that, The load state prediction module is specifically used for: A linear prediction model is used to generate the predicted load current by adding the current total load current value to the product of the load current change rate and the preset prediction time domain.

3. The modular power supply system for low-loss, high-efficiency energy transmission according to claim 1, characterized in that, The dynamic pre-adjustment execution module is specifically used for: calculating the target output voltage. Based on the predicted load current and the total number of working units after grid connection, determine the predicted load share that the unit to be connected should share; multiply the predicted load share by the preset virtual output impedance to obtain the voltage compensation value; add the voltage compensation value to the real-time monitored bus voltage to generate the target output voltage.

4. A modular power supply system for low-loss, high-efficiency energy transmission according to claim 3, characterized in that, The dynamic pre-adjustment execution module is also specifically used for: Before the grid-connection relay of the unit to be connected closes, the no-load voltage of the unit to be connected is adjusted to the target output voltage by adjusting the internal PWM controller.

5. A modular power supply system for low-loss, high-efficiency energy transmission according to claim 1, characterized in that, The safe working area coefficient is a preset parameter characterizing the maximum allowable steady-state load rate of the system; the risk weighting coefficient is a preset parameter used to adjust the overall sensitivity of the system's instability risk index.

Images