Voltage regulating device for distribution network modular parallel and current sharing control method thereof

Abstract

The application relates to the field of power conversion, and discloses a voltage regulating and stabilizing device for distribution network modular parallel connection and a current sharing control method thereof. The system comprises a bypass unit, a monitoring unit and a plurality of parallel power modules, each power module comprises a power conversion circuit and a module controller, and the method comprises the following steps: S1, each module controller receives output currents of the remaining power modules and calibrates the output current of itself; S2, the monitoring unit calculates a current sharing error based on the calibrated output currents, compensates and corrects the current sharing error in combination with a topology characteristic compensation term, and independently generates a voltage regulating instruction; S3, the module controller adjusts the output of the power conversion circuit based on the corrected current sharing error and the voltage regulating instruction; S4, when any power module fails, the output is cut off, and a fault signal is transmitted to the monitoring unit; and S5, the monitoring unit controls the bypass unit to be put into bypass and recalculates the corrected current sharing error based on the non-fault power modules. The application can improve the current sharing control precision.

Description

Technical Field

[0001] This application relates to the field of power conversion, and in particular to a modular parallel voltage regulating and stabilizing device for distribution networks and its current sharing control method. Background Technology

[0002] With the development of medium and high power AC power conversion technology, multi-module parallel structures are widely used in various power supply systems. By operating multiple power modules in parallel, the system output capacity can be effectively increased, the power level can be expanded, and redundancy backup can be achieved, thereby improving the reliability and continuity of system operation.

[0003] In low-voltage distribution substations, voltage fluctuations and exceeding limits at the terminal are becoming increasingly prominent due to factors such as long power supply radii, thin wire diameters, and distributed photovoltaic backfeeding. Traditional mechanical voltage regulators suffer from slow response and low accuracy, and reactive power compensation has limited effect on active power voltage drop, making continuous, bidirectional voltage regulation difficult. While voltage regulation and stabilization devices based on power electronics technology can achieve rapid and continuous adjustment, they often employ a single-unit high-power design with fixed capacity, making expansion difficult. Furthermore, a single point of failure can cause the entire unit to shut down, failing to meet the distribution network's requirements for high reliability and flexible configuration.

[0004] The above problems can be solved by adopting a modular architecture with multiple modules in parallel, but the accuracy of current sharing control and the ability to coordinate fault handling among modules become the key factors restricting system performance.

[0005] Currently, multi-module parallel systems typically employ communication methods to achieve data interaction and coordinated control between modules. For example, Chinese patent application CN116388141A discloses a multi-module parallel power supply system that achieves signal synchronization and phase reversal control through inter-module network communication. However, in practical power distribution network applications, many challenges remain: inter-module information interaction often relies on a single CAN communication method, which is prone to delays and packet loss in environments with strong electromagnetic interference, resulting in insufficient real-time performance and accuracy of current sharing; power allocation between modules does not consider systematic deviations caused by differences in topology parameters, making it difficult to adapt to complex operating conditions; systems often adopt distributed bypass designs, leading to numerous fault points and significant coordination difficulties; and current sharing parameters are not updated in a timely manner during fault switching, easily causing overload or output fluctuations in the remaining modules. Summary of the Invention

[0006] To address the aforementioned technical problems, this application provides a modular parallel voltage regulating and stabilizing device for power distribution networks and its current sharing control method.

[0007] Firstly, the current sharing control method for a modular parallel voltage regulating and stabilizing device in a power distribution network provided in this application adopts the following technical solution: A current sharing control method for a modular parallel voltage regulating and stabilizing device in a power distribution network, the modular parallel voltage regulating and stabilizing device comprising a bypass unit, a monitoring unit, and multiple parallel power modules, each power module comprising a power conversion circuit and a module controller, the method comprising the following steps: S1, each module controller acquires the output current Ii of its power module, and simultaneously receives the output currents of the other power modules, and calibrates the output current Ii based on the output currents of the other power modules to obtain a calibrated output current Ii′, and transmits the output current Ii′ to the monitoring unit, where i=1,2,...,N, and N is the total number of power modules; S2, the monitoring unit calculates the average output current Iavg based on the output current Ii′ transmitted by each module controller, and calculates the current sharing error ΔIi=Ii′−Iavg for the i-th power module; the monitoring unit compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi related to the power conversion circuit of the i-th power module, obtaining... The corrected current sharing error ΔIi′ = ΔIi − ΔIcompi corresponding to the i-th power module is generated independently of the current sharing error ΔIi, and the corrected current sharing error ΔIi′ and the voltage regulation command Uref are sent to the module controller corresponding to the i-th power module; S3, each module controller performs decoupling control of current sharing regulation and voltage regulation based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref to regulate the output of the corresponding power conversion circuit; S4, when any module controller detects a fault in its power module, it cuts off the output and transmits a fault signal to the monitoring unit; S5, when the monitoring unit receives the fault signal, it controls the bypass unit to engage the bypass, and at the same time recalculates the average output current Iavg′ based on the number of remaining non-faulty power modules N−1 and the total output current of the system, and recalculates the corrected current sharing error ΔIi′ based on the recalculated average output current Iavg′ and sends it to the remaining non-faulty module controllers.

[0008] By adopting the above technical solution, in step S1, each module controller not only collects its own output current but also receives and calibrates the output current of other power modules, eliminating current information deviation caused by differences in the acquisition link and providing a reliable data foundation for subsequent current sharing calculations. In step S2, the monitoring unit introduces a topology characteristic compensation term related to the power conversion circuit to compensate and correct the current sharing error, improving the current sharing deviation caused by inherent topology characteristics and differences in module parameters. At the same time, the voltage regulation command is generated independently of the current sharing error, achieving decoupling between voltage regulation and current sharing control, and avoiding mutual interference between the two during the regulation process. Interference; In step S3, each module controller performs decoupling control based on the correction of current sharing error and voltage regulation command, ensuring the synchronous maintenance of current sharing accuracy and output voltage stability under wide voltage regulation range and load fluctuation scenarios; In steps S4 and S5, the rapid disconnection of faulty modules and the timely activation of bypass units are coordinated, and the average output current is recalculated and issued based on the remaining non-faulty modules, so that the system can smoothly transition to derating operation mode under fault conditions, which can prevent the remaining modules from overload or output fluctuation caused by the failure to update the current sharing parameters in time during the fault switching process, and improve the redundancy reliability and continuous power supply capability of the system.

[0009] Preferably, the monitoring unit stores the topology parameters of each power module; in step S2, the monitoring unit compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi related to the power conversion circuit of the i-th power module, including: the monitoring unit obtains the input voltage, switching frequency and the pre-stored topology parameters of the i-th power module; and calculates the topology characteristic compensation term ΔIcompi based on the input voltage, the switching frequency and the topology parameters.

[0010] By adopting the above technical solution, the monitoring unit calculates the topology characteristic compensation item based on the pre-stored topology parameters and the real-time input voltage and switching frequency, so that the compensation amount can reflect the current deviation of each power module under the current working state caused by the inherent characteristics of the topology structure, including systematic deviations caused by factors such as differences in bridge arm inductance, inconsistent filter capacitor parameters, and differences in the on-state voltage drop of switching devices. This achieves targeted correction of current sharing error and improves the accuracy of current sharing control.

[0011] Preferably, the monitoring unit also acquires the operating temperature of each of the power modules; the method further includes the step of: the monitoring unit dynamically adjusting the topology characteristic compensation term ΔIcompi based on the operating temperature.

[0012] By adopting the above technical solution, the monitoring unit dynamically adjusts the topology characteristic compensation item based on the operating temperature of each power module. This can compensate for the impact of parameter changes such as on-resistance drift and threshold voltage shift caused by temperature changes of power devices on the output current. This allows the current sharing control to maintain high accuracy even under long-term system operation and uneven temperature rise conditions, thereby improving the system's adaptability to changing operating conditions.

[0013] Preferably, in step S3, each module controller performs decoupled control of current sharing regulation and voltage regulation based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref to regulate the output of the corresponding power conversion circuit. This includes: the module controller inputs the corrected current sharing error ΔIi′ into a fuzzy PID controller to generate a current regulation command ΔIrefi; the module controller combines the current regulation command ΔIrefi and the voltage regulation command Uref to generate a PWM control signal to control the switching timing of the corresponding power conversion circuit.

[0014] By adopting the above technical solution, the module controller will adaptively adjust the fuzzy PID controller to correct the current sharing error. It can dynamically optimize the control parameters according to the system operating conditions, improve the response speed and control accuracy of current sharing regulation, and improve the problem of insufficient regulation performance of traditional fixed parameter controllers under dynamic conditions. At the same time, the current regulation command and voltage regulation command are combined to generate a PWM control signal, so that current sharing regulation and voltage regulation are executed independently. This decouples the two from the control structure, avoids mutual interference in the regulation process, and ensures stable output voltage and reliable current sharing effect.

[0015] Preferably, the module controller inputs the corrected current sharing error ΔIi′ into the fuzzy PID controller to generate a current adjustment command ΔIrefi. Specifically, this includes: using the corrected current sharing error ΔIi′ and its rate of change as input linguistic variables for the fuzzy PID controller; performing inference and defuzzification according to a preset fuzzy control rule table to obtain dynamic adjustment values ​​for the proportional gain, integral gain, and derivative gain; updating the current parameters of the fuzzy PID controller based on the dynamic adjustment values; and calculating and outputting the current adjustment command ΔIrefi based on the updated parameters.

[0016] By adopting the above technical solution, using the correction of the current sharing error and its rate of change as input for fuzzy inference and defuzzification processing, the dynamic adjustment of the proportional gain, integral gain and derivative gain can be accurately obtained, enabling the parameters of the fuzzy PID controller to match the current operating conditions in real time, further improving the adaptive capability and robustness of the current sharing control, and ensuring that fast, stable and overshoot-free current sharing regulation can be achieved in complex scenarios such as load fluctuations and parameter drift.

[0017] Preferably, in step S5, after recalculating the corrected current sharing error ΔIi′ based on the recalculated average output current Iavg′, the error is sent to the controller of the remaining non-faulty modules. Specifically, the monitoring unit smoothly transitions the average output current of the remaining non-faulty power modules to the recalculated average output current Iavg′ at a preset ramp rate, and synchronously updates the sent corrected current sharing error ΔIi′ during the transition process, so that the output current of the remaining non-faulty power modules smoothly takes over the total load of the system.

[0018] By adopting the above technical solution, after the faulty module is disconnected, the monitoring unit does not instantly switch the average output current of the remaining non-faulty modules to a new average current value. Instead, it smoothly transitions according to a preset ramp rate. During the transition, the current sharing error is updated and corrected synchronously and continuously sent down, so that the output current of the remaining non-faulty modules gradually takes over the load share originally borne by the faulty module at a controllable rate. This improves the output current surge and voltage transient fluctuation problem caused by the step change in average output current, and enhances the stability and continuity of the system output during the fault switching process.

[0019] Preferably, the modular parallel voltage regulation and stabilization device for the power distribution network further includes a high-frequency communication line, a low-frequency communication line, and a CAN communication line. In step S1, each module controller collects the output current Ii of its power module and simultaneously receives the output current of the other power modules. Based on the output current of the other power modules, the module controller calibrates the output current Ii to obtain the calibrated output current Ii′. Specifically, the module controller receives the real-time current synchronization signal of the other power modules through the high-frequency communication line, receives the reference current signal sent by the monitoring unit through the low-frequency communication line, and receives the real-time current and status information of the other power modules through the CAN communication line. The module controller corrects the high-frequency interference error based on the data transmitted through the high-frequency communication line, corrects the low-frequency delay error based on the data transmitted through the low-frequency communication line, and corrects the CAN packet loss error based on the data transmitted through the CAN communication line to obtain the calibrated output current Ii′.

[0020] By adopting the above technical solution, the module controller uses three communication methods—high-frequency communication line, low-frequency communication line, and CAN communication line—to obtain current information with different characteristics, and performs complementary calibration to address the inherent transmission defects of each communication method: the real-time current synchronization signal of the high-frequency communication line is used to correct the sampling error introduced by high-frequency interference; the reference current signal of the low-frequency communication line is used to correct the low-frequency offset error caused by transmission delay; and the real-time current and status information of the CAN communication line is used to correct the information loss error caused by data packet loss. The three complement each other, making the accuracy of the calibrated output current higher than that of the scheme relying on a single communication method, providing a highly reliable data foundation for subsequent current sharing calculation.

[0021] Preferably, the process of correcting CAN packet loss errors based on data transmitted through the CAN communication line to obtain the calibrated output current Ii′ includes: when the module controller does not receive real-time current from other power modules through the CAN communication line within a preset period, it is determined to be CAN packet loss; the module controller performs interpolation compensation based on the reference current signal received through the low-frequency communication line and the historical current data received through the CAN communication line in the previous period, to replace the missing real-time current in the calibration calculation, and obtains the calibrated output current Ii′.

[0022] By adopting the above technical solution, when packet loss occurs in the CAN communication line, the module controller does not directly discard the missing data or replace it with the default value. Instead, it uses the reference current signal transmitted by the low-frequency communication line and the historical current data received by the CAN communication line in the previous cycle for interpolation compensation. The estimated value replaces the missing real-time current in the calibration calculation, ensuring that the calibration process is uninterrupted and the accuracy is controllable under the condition of communication interference. This improves the communication robustness and the continuity of current sharing control of the system in the electromagnetic interference environment.

[0023] Secondly, the modular parallel voltage regulating and stabilizing device for power distribution networks provided in this application adopts the following technical solution: A modular parallel voltage regulation and stabilization device for a power distribution network, used to execute the current sharing control method as described in any of the first aspects above, includes: multiple power modules connected in parallel between the power grid and the load, each power module including a power conversion circuit, a module controller, and a communication circuit; the power conversion circuit is used to convert the input AC voltage into an output AC voltage; the module controller is used to acquire the output current Ii of its own power module, and receive the output current of the other power modules through the communication circuit to calibrate and obtain a calibrated output current Ii′, and decouple the current sharing regulation and voltage regulation based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref, and output a PWM signal to the power conversion circuit; when a fault is detected, the module controller cuts off the output of the power conversion circuit and sends... A fault signal; a monitoring unit, electrically connected to the module controller, which receives the output current Ii′ transmitted by each module controller and calculates the average output current Iavg and current sharing error ΔIi. It then compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi to obtain the corrected current sharing error ΔIi′, and generates the voltage regulation command Uref independently of the current sharing error ΔIi. The corrected current sharing error ΔIi′ and the voltage regulation command Uref are then sent to the corresponding module controller. The monitoring unit is also used to recalculate the corrected current sharing error ΔIi′ based on the remaining non-faulty power modules and send it out upon receiving the fault signal. A bypass unit, connected to the monitoring unit, is used to engage or disengage the bypass under the control of the monitoring unit.

[0024] By adopting the above technical solutions, the modular parallel voltage regulation and stabilization device for distribution networks achieves flexible expansion and redundant backup of system power through the parallel operation of multiple power modules. The module controller of each power module has local current acquisition, calibration, and decoupling control capabilities. The monitoring unit undertakes centralized decision-making functions for current sharing error calculation, topology compensation correction, and voltage regulation command generation. The bypass unit realizes bypass activation in fault conditions under the control of the monitoring unit. The three work together through communication circuits to form a control architecture that combines distributed acquisition and centralized decision-making. This ensures the rapid local response capability of each module and achieves accurate current sharing and unified scheduling at the system level, taking into account the real-time performance, accuracy, and reliability of the system. In addition, the modular parallel architecture of the device makes each power module a standardized independent unit. In case of failure, only a single module needs to be replaced to restore full-capacity operation. It supports online hot-swapping and does not require power outage maintenance of the entire unit. When the load of the distribution area increases, modules can be added to achieve flexible expansion without replacing the entire equipment. Each module adopts a unified topology and interface specification, which is suitable for mass production on automated production lines, which helps to reduce manufacturing costs and ensure product consistency.

[0025] Preferably, the communication circuits are interconnected via a high-frequency communication line, a low-frequency communication line, and a CAN communication line; the high-frequency communication line, the low-frequency communication line, and the CAN communication line are all connected to the monitoring unit, the bypass unit, and each of the module controllers; wherein, the high-frequency communication line is used to transmit a high-frequency synchronization signal so that the module controller can perform high-frequency interference error correction and synchronize the switching action of the bypass unit with the switching timing of the power conversion circuit; the low-frequency communication line is used by the monitoring unit to send a reference current signal to the module controller and a phase reference signal to the bypass unit; the CAN communication line is used for the interaction of real-time data, status information, and control commands between the monitoring unit, the module controller, and the bypass unit.

[0026] By adopting the above technical solution, a triple communication architecture consisting of high-frequency communication lines, low-frequency communication lines, and CAN communication lines simultaneously connects the monitoring unit, bypass unit, and each module controller, ensuring multi-channel protection for signal synchronization, reference transmission, and data interaction within the system. The high-frequency communication line not only serves to correct high-frequency interference errors in the module controller but also ensures that the switching action of the bypass unit is precisely synchronized with the switching sequence of the power conversion circuit, preventing current surges caused by bypass switching. The low-frequency communication line sends a reference current signal to the module controller and a phase reference signal to the bypass unit, ensuring the phase continuity of the output voltage when the bypass is activated. The CAN communication line undertakes the global interaction of real-time data, status information, and control commands, providing a complete information channel for coordinated system control and rapid fault response.

[0027] In summary, this application includes at least one of the following beneficial technical effects: 1. Through the coordinated efforts of multi-communication collaborative calibration, topology characteristic compensation correction, and fuzzy PID current sharing regulation, high-precision current sharing control of multi-module parallel systems under various operating conditions is achieved. This improves the current sharing deviation caused by communication transmission errors, module parameter differences, and inherent topology characteristics. At the same time, it decouples voltage regulation and current sharing control, prevents coupling interference between the two, and improves the output stability of the system under wide voltage regulation range and load fluctuation scenarios. 2. By combining a centralized bypass design with a ramp-smooth transition mechanism during fault switching and the collaborative protection of a triple communication architecture, rapid response and smooth transition under fault conditions are achieved, simplifying the system structure, reducing the number of fault points, and improving the system's redundancy reliability and continuous power supply capability. 3. Through a triple communication architecture of high-frequency communication lines, low-frequency communication lines and CAN communication lines, multi-channel complementary transmission and calibration of current information between modules are realized, overcoming the limitations of a single communication method in terms of real-time performance, stability and anti-interference. At the same time, the bypass switching action and the switching sequence of the power conversion circuit are precisely synchronized, ensuring the continuity and reliability of system control under communication interference conditions. 4. Through modular parallel architecture, each power module supports hot-swappable replacement and flexible capacity expansion, which is suitable for standardized mass production and has redundancy backup capability, improving the ease of operation and maintenance, capacity scalability and power supply reliability of the device in distribution network voltage regulation and stabilization scenarios. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall structure of the modular parallel voltage regulating and stabilizing device for power distribution networks provided in the embodiments of this application; Figure 2 This is a topology diagram of the three-arm AC / AC power conversion circuit of the power module provided in the embodiments of this application; Figure 3 This is a schematic diagram of the triple communication bus architecture provided in the embodiments of this application; Figure 4 This is a flowchart of the current sharing control method for the modular parallel voltage regulating and stabilizing device in the power distribution network provided in the embodiments of this application.

[0029] Explanation of reference numerals in the attached figures: 10. Bypass unit; 20. Monitoring unit; 30. Power module; 31. Module controller; 32. Power conversion circuit; 33. Communication circuit; 40. Communication gateway module. Detailed Implementation

[0030] The following is in conjunction with the appendix Figure 1-4 This application will be described in further detail.

[0031] It should be noted that, in the description of this application, "input terminal" and "output terminal" refer to the side of the converter system that receives AC power from the mains and the side that supplies AC power to the load, respectively. The numbering of each power module i=1,2,...,N is only for ease of explanation and does not constitute a limitation on the physical arrangement or priority of the power modules. The terms "high frequency" and "low frequency" used in this application are relative concepts. The signal frequency transmitted by the "high frequency communication line" is higher than the signal frequency transmitted by the "low frequency communication line," and the specific frequency value can be set according to the actual application scenario.

[0032] In existing multi-module AC / AC voltage regulator systems, an indirect AC-DC-AC conversion topology is typically used. This involves rectifying the input AC power to DC, then inverting it back to AC power for output. This method requires a DC intermediate stage, resulting in large size, numerous power devices, low conversion efficiency, and slow dynamic response. Furthermore, current sharing control in existing multi-module parallel systems often relies on a single CAN communication method for inter-module information exchange. In the strong electromagnetic interference environment of industrial sites, communication delays and packet loss are prone to occur, leading to insufficient real-time performance and accuracy of current sharing control. In addition, existing systems often employ a distributed bypass design, where each power module has its own independent bypass circuit. This not only increases system size, cost, and wiring complexity but also leads to more potential failure points, making coordinated control between bypass devices difficult. Regarding monitoring, existing systems are mostly local monitoring systems, lacking centralized information collection and cloud data upload capabilities. Maintenance personnel cannot remotely monitor the system's operating status in real time, resulting in low troubleshooting efficiency.

[0033] To address the aforementioned issues, this application provides a modular parallel voltage regulation and stabilization device for distribution networks and its current sharing control method. Through the parallel operation of multiple standardized power modules, continuous, bidirectional regulation and stabilization control of the output voltage of the distribution substation is achieved. Simultaneously, high-precision current sharing control ensures balanced power distribution among the modules. The device adopts a modular architecture, supporting flexible configuration of the number of modules according to the substation capacity requirements, facilitating on-site maintenance, replacement, and capacity expansion upgrades. Specifically, by employing a three-arm direct ACAC conversion topology, triple communication collaborative calibration, topology characteristic compensation correction, fuzzy PID current sharing regulation, and centralized bypass design, the multi-module parallel system achieves high-precision current sharing control, high-reliability fault handling, and remote intelligent operation and maintenance under various operating conditions.

[0034] This application discloses a modular parallel voltage regulating and stabilizing device for a power distribution network. (Refer to...) Figure 1 The modular parallel voltage regulation and stabilization device for the distribution network includes a bypass unit 10, a monitoring unit 20, and multiple parallel power modules 30. The system's input is connected to the AC power grid via a distribution transformer to receive AC input; the system's output provides AC output to the load. The outputs of the multiple power modules 30 are connected in parallel to form an AC parallel output structure, jointly supplying power to the load.

[0035] This device is applied to low-voltage distribution transformer terminal voltage management scenarios, connected in series between the low-voltage side outgoing line of the distribution transformer and the user load. When the terminal voltage is too low due to increased load, the power conversion circuit 32 adds a positive compensation voltage to raise the output voltage to the target value; when the terminal voltage is too high due to photovoltaic backfeed, a reverse compensation voltage is added to lower the output voltage to the target value, achieving bidirectional and continuous voltage regulation. In this embodiment, the device's voltage regulation range is ±15% of the rated voltage, the voltage regulation accuracy is better than ±1%, and the dynamic response time is less than 20ms. The voltage regulation control and current sharing control adopt a decoupled design, and the two do not interfere with each other.

[0036] In one embodiment, the bypass unit 10 is connected in parallel with a plurality of power modules 30 for enabling or disabling the bypass under the control of the monitoring unit 20.

[0037] Specifically, refer to Figure 1 The bypass unit 10 includes a manual bypass switch S1, an electric bypass switch S2, and an electronic switch S3. The manual bypass switch S1 is operated manually by maintenance personnel in maintenance or emergency situations to directly switch the system output to bypass power supply, ensuring uninterrupted power supply during equipment maintenance. The electric bypass switch S2 is driven by the monitoring unit 20 through control commands, enabling automatic bypass connection and disconnection. The electronic switch S3 uses a fast semiconductor switching device with millisecond-level switching speed, used to quickly connect the bypass in the event of a power module 30 failure, ensuring continuous power supply to the load.

[0038] Understandably, the bypass unit 10 adopts a centralized bypass design, meaning that the entire system uses only one bypass unit 10, connected in parallel with the output terminals of all power modules 30, rather than configuring a separate bypass circuit for each power module 30. This centralized bypass design simplifies the system structure, reduces the number of bypass devices and potential failure points, lowers system cost and wiring complexity, and simultaneously increases system power density. The rated power of the bypass unit 10 matches the total system power, ensuring that it can stably carry the entire system load when the bypass is activated.

[0039] It should be noted that the manual bypass switch S1, the electric bypass switch S2, and the electronic switch S3 can be selected according to different application scenarios and reliability requirements. In high-reliability scenarios requiring rapid fault switching, the electronic switch S3 serves as the primary bypass switching device; the electric bypass switch S2 acts as a backup for the electronic switch S3, providing bypass protection in the event of a failure of the electronic switch S3; and the manual bypass switch S1 serves as the final safety measure, requiring manual operation by maintenance personnel when all automatic bypass mechanisms fail. These three components form a hierarchical bypass protection system, improving the system's power supply reliability.

[0040] This device adopts a modular parallel architecture, with each power module 30 being a standardized, independently packaged unit, offering the following advantages in power distribution network applications: Firstly, it facilitates maintenance and hot-swapping. Each power module 30 is installed in a standardized cabinet using pluggable connectors. In case of failure, maintenance personnel can remove the faulty module and insert a spare module without shutting down the system. After the new module powers on and performs a self-test, the monitoring unit 20 automatically incorporates it into the current sharing control loop, smoothly distributing the load according to the ramp rate to achieve uninterrupted connection. Secondly, flexible capacity expansion. The cabinet has reserved expansion slots. When the load of the distribution area increases, only 30 power modules need to be added to increase the total capacity. There is no need to replace the entire equipment or rewire, which can realize phased investment and expansion on demand. Third, standardized production. Each power module 30 adopts a unified circuit topology, structural dimensions, and interface specifications, which is suitable for mass production using intelligent manufacturing methods such as automated surface mount technology, assembly, and testing. This helps improve product consistency and reduce manufacturing costs. Fourth, redundancy and reliability. The parallel operation of multiple modules gives the system N+1 redundancy capability. If any module fails and is disconnected, the remaining modules automatically take over all loads. With the help of centralized bypass protection, the continuity of power supply to the transformer area is guaranteed.

[0041] In one embodiment, each power module 30 includes a module controller 31, a power conversion circuit 32, and a communication circuit 33.

[0042] Specifically, refer to Figure 1 and Figure 2 The power conversion circuit 32 is used to convert the input AC voltage into the output AC voltage. In this embodiment, the power conversion circuit 32 adopts a three-bridge direct AC-AC conversion topology.

[0043] Reference Figure 2 The power conversion circuit 32 includes an input port 1, an output port 2, a first bridge arm, a second bridge arm, a third bridge arm, an inductor L1, an input filter capacitor C2, an output filter capacitor C1, and an output filter inductor L3.

[0044] The first bridge arm consists of transistors Q1 and Q2 connected in series, with Q1 located in the upper bridge arm and Q2 in the lower bridge arm. The second bridge arm consists of transistors Q5 and Q6 connected in series, with Q5 located in the upper bridge arm and Q6 in the lower bridge arm. The third bridge arm consists of transistors Q3 and Q4 connected in series, with Q3 located in the upper bridge arm and Q4 in the lower bridge arm. The upper ends of the three bridge arms are connected together to form a positive busbar, and the lower ends of the three bridge arms are connected together to form a negative busbar.

[0045] Understandably, the switching transistors Q1 through Q6 are all two-quadrant switches, each consisting of a power switching device (IGBT or MOSFET) and an anti-parallel diode. The use of two-quadrant switches allows each bridge arm's switch to withstand bidirectional voltage but only allows controlled current to flow in one direction. This reduces the commutation transient voltage oscillation problem present in traditional direct ACAC converter topologies, lowering the voltage stress on the switching transistors and reducing control complexity.

[0046] Inductor L1 is connected between the midpoint of the first bridge arm (the connection point of Q1 and Q2) and input port 1, with a current IL1 flowing through it. One end of input port 1 is connected to inductor L1, and the other end is connected to the midpoint of the second bridge arm (the connection point of Q5 and Q6). Input filter capacitor C2 is connected across input port 1 to filter out high-frequency harmonic interference from the input AC power supply and suppress voltage spikes on the input side. One end of output port 2 is connected to the midpoint of the third bridge arm via output filter inductor L3, and the other end is connected to the midpoint of the second bridge arm. Output filter capacitor C1 is connected across output port 2, forming an LC low-pass filter network with output filter inductor L3 to filter out switching harmonic components in the output voltage, ensuring the waveform quality of the output AC voltage.

[0047] Input port 1 receives input voltage Vin and input current Iin, and output port 2 outputs output voltage Vout and output current Iout.

[0048] Understandably, the three-arm direct AC-AC converter topology, compared to the indirect AC-DC-AC converter topology, eliminates the large-capacity DC bus capacitor and the two-stage rectifier / inverter circuitry required in the intermediate DC link. This results in advantages such as compact structure, fewer power devices, high power density, and high conversion efficiency. The three-arm topology uses only six switching transistors, fewer than four-arm or matrix converter topologies, making control relatively simple and suitable for standardized and modular designs in multi-module parallel operation scenarios.

[0049] It should be noted that the output voltage amplitude and frequency of the power conversion circuit 32 are adjusted by controlling the switching sequence of each switch in the three bridge arms through the module controller 31. The module controller 31 uses SPWM (sinusoidal pulse width modulation) to generate PWM control signals. By adjusting the amplitude and frequency of the modulation wave, the amplitude and frequency of the output voltage are changed, thereby realizing the AC / AC voltage conversion function.

[0050] In one embodiment, the module controller 31 is used to acquire the output current Ii of the power module 30 it is in, and to receive the output current of the other power modules 30 through the communication circuit 33 for calibration to obtain the calibrated output current Ii′. Based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref, the module controller 31 performs decoupled control of current sharing regulation and voltage regulation, and outputs a PWM signal to the power conversion circuit 32. When a fault is detected, the module controller 31 cuts off the output of the power conversion circuit 32 and sends a fault signal.

[0051] Specifically, the module controller 31 can use a high-performance digital signal processor (DSP) or a field-programmable gate array (FPGA) as its core control chip, possessing the capabilities of high-speed data acquisition, real-time control calculation, and multi-channel PWM signal generation. The module controller 31 acquires the output current Ii of the power module 30 in real time through a current sensor (such as a Hall sensor), the output voltage Ui through a voltage sensor, and the operating temperature Ti of the power device through a temperature sensor. The module controller 31 also has a local data storage function, which can temporarily store module operating data to prevent data loss during communication interruptions.

[0052] It is understood that the fault detection performed by the module controller 31 includes, but is not limited to, the following fault types: output overcurrent (output current exceeding a preset multiple of the rated value), output overvoltage (output voltage exceeding a preset percentage of the rated value), power device overtemperature (operating temperature exceeding a preset threshold), and open or short circuit of switching devices. When any fault is detected, the module controller 31 immediately blocks all PWM output signals, cuts off the output of the power conversion circuit 32 to prevent the fault from spreading to the rest of the system, and simultaneously sends a fault signal to the monitoring unit 20 and the module controllers 31 of the other power modules 30 through the communication circuit 33.

[0053] In one embodiment, the monitoring unit 20 is electrically connected to the module controller 31 of each power module 30, and is also connected to the bypass unit 10.

[0054] Specifically, the monitoring unit 20 uses a high-performance microcontroller as its core processing chip to receive the calibrated output current Ii′ transmitted from each module controller 31, and performs current sharing error calculation and topology characteristic compensation correction based on the output current of all power modules 30. The monitoring unit 20 internally stores the topology parameters of each power module 30, including the inductance values ​​of inductors L1 and L3, the capacitance values ​​of capacitors C1 and C2, and the on-resistance of each switching transistor.

[0055] The monitoring unit 20 calculates the average output current Iavg and current sharing error ΔIi of each power module 30, and compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi to obtain the corrected current sharing error ΔIi′. It also generates a voltage regulation command Uref independently of the current sharing error ΔIi, and sends the corrected current sharing error ΔIi′ and the voltage regulation command Uref to the corresponding module controller 31.

[0056] It is understandable that the monitoring unit 20 generates the voltage regulation command Uref independently of the current sharing error ΔIi. This means that the generation of the voltage regulation command Uref is not affected by the current sharing error ΔIi. The voltage regulation command Uref is jointly determined by the system's target output voltage, the actual output voltage feedback, and external voltage regulation commands (including local voltage regulation commands or remote voltage regulation commands). This design achieves decoupling of voltage regulation and current sharing control at the control loop level, avoiding disturbances to the output voltage during current sharing regulation and the impact on current distribution between modules during voltage regulation.

[0057] The monitoring unit 20 is also used to control the bypass unit 10 to engage the bypass when a fault signal is received, and to recalculate and correct the current sharing error ΔIi′ based on the number of remaining non-faulty power modules 30 and send it to the module controller 31 of the remaining non-faulty modules.

[0058] In one embodiment, reference is made to Figure 1 and Figure 3 The monitoring unit 20 is also connected to the communication gateway module 40. The communication gateway module 40 communicates with the monitoring unit 20 through the RS485 interface and establishes a connection with the cloud server through the 4G / 5G wireless communication module.

[0059] Specifically, the monitoring unit 20 comprehensively collects status information, analog information, and fault information from each power module 30 through triple communication. Status information includes the operating mode of each power module 30 (normal operation, standby, fault clearing, etc.) and the working status of switching devices; analog information includes the input voltage, output current, output voltage, and operating temperature of each power module 30; fault information includes the fault module number, fault type, and fault occurrence time. The monitoring unit 20 organizes and analyzes the collected information. When abnormal information is detected, it triggers a local alarm (such as indicator light flashing or buzzer alarm) and simultaneously uploads the information to the cloud server through the communication gateway module 40. The cloud server enables data storage, querying, analysis, and visualization, supporting maintenance personnel to remotely monitor system operation status, query historical data, and issue remote control commands through remote control terminals (computers, mobile phones, etc.).

[0060] It should be noted that the monitoring unit 20 can also integrate a GPS positioning module to realize device location positioning and time synchronization, which makes it easier for maintenance personnel to accurately locate the device installation location; at the same time, the monitoring unit 20 has an SOE (Sequence of Events) function, which records key events in the system operation process in chronological order, providing a time sequence basis for fault analysis and tracing.

[0061] In one embodiment, reference is made to Figure 3 The communication circuits 33 are interconnected through high-frequency communication lines, low-frequency communication lines, and CAN communication lines. The high-frequency communication lines, low-frequency communication lines, and CAN communication lines are all connected to the monitoring unit 20, the bypass unit 10, and the module controllers 31 of each power module 30, forming a triple communication bus architecture.

[0062] Specifically, the high-frequency communication line is used to transmit high-frequency synchronization signals. These signals include real-time current waveform information and switching timing information for each power module 30. The module controller 31 receives the real-time current synchronization signals from the remaining power modules 30 via the high-frequency communication line for high-frequency interference error correction. Simultaneously, the bypass unit 10 receives the switching timing signals from the power conversion circuit 32 via the high-frequency communication line, ensuring that the switching action of the bypass unit 10 is accurately synchronized with the switching timing of the power conversion circuit 32, preventing current surges caused by switching timing mismatches during bypass switching.

[0063] The low-frequency communication line is used to transmit low-frequency signals. The monitoring unit 20 sends reference current signals to each module controller 31 and a phase reference signal to the bypass unit 10 via the low-frequency communication line. The reference current signal is the target output current reference value for each module determined by the monitoring unit 20 based on the system load and the number of modules. The module controller 31 uses this value to correct low-frequency delay errors. The phase reference signal is used to ensure that the output voltage phase of the bypass unit 10 is consistent with the output voltage phase of the power module 30, ensuring the phase continuity of the output voltage when the bypass is activated and preventing load surges caused by phase abrupt changes.

[0064] The CAN communication line adopts the standard CAN2.0B protocol for the exchange of real-time data, status information, and control commands between the monitoring unit 20, the controllers of each module 31, and the bypass unit 10. The data transmitted through the CAN communication line includes the real-time output current Ii, output voltage Ui, operating temperature Ti, operating mode, and fault information of each power module 30, as well as the current sharing error correction ΔIi′, voltage adjustment command Uref, and remote control commands issued by the monitoring unit 20.

[0065] Understandably, the purpose of the triple communication bus architecture is to leverage the complementary advantages of each communication method: high-frequency communication lines have high transmission rates and can promptly transmit information about rapid changes in current waveforms; low-frequency communication lines have lower transmission rates but offer stable signal transmission and strong anti-interference capabilities, making them suitable for transmitting slowly changing reference signals; CAN communication lines have good anti-interference capabilities and error correction mechanisms, making them suitable for transmitting data frame format information, but their high-frequency real-time performance is inferior to dedicated high-frequency signal lines. The three work together to ensure that the system maintains real-time performance, stability, and integrity of communication under various electromagnetic environments.

[0066] It should be noted that the signal frequency of the high-frequency communication line can be determined based on the switching frequency of the power conversion circuit 32, and is typically set to a range of 1 to 10 times the switching frequency to ensure real-time tracking capability of current changes during the switching cycle. The signal frequency of the low-frequency communication line is typically set to a range of several to tens of times the power frequency to meet the stable transmission requirements of the power frequency current reference signal. The baud rate of the CAN communication line can be set according to the number of modules and the amount of data, typically using a baud rate from 250kbps to 1Mbps.

[0067] Reference Figure 4 This application also provides a current sharing control method for a modular parallel voltage regulating and stabilizing device in a distribution network applied to any of the above embodiments, which is described below in conjunction with... Figure 4 The flowchart shown provides a detailed explanation of each step.

[0068] After power-on, the system first performs system initialization. During the initialization phase, the monitoring unit 20 sets the rated parameters of each power module 30, including rated output current, rated output voltage, and switching frequency; sets communication parameters, including high-frequency signal frequency, low-frequency signal frequency, and CAN communication baud rate; and sets monitoring parameters, including overcurrent threshold, overvoltage threshold, overtemperature threshold, and alarm threshold. Each module controller 31 initializes its own control parameters, including the initial parameters of the fuzzy PID controller (initial values ​​of proportional gain Kp, integral gain Ki, and derivative gain Kd) and topology characteristic compensation coefficients. The bypass unit 10 is initialized to the exit state, and all power modules 30 are put into normal operation. The monitoring unit 20 starts the information acquisition function and cloud upload function, establishes a communication connection with the cloud server through the communication gateway module 40, and completes the overall system initialization.

[0069] After the system initialization is completed, it enters the parameter acquisition and communication interaction stage, with the two stages being executed in parallel.

[0070] Step S1: Each module controller 31 collects the output current Ii of the power module 30 it is in, and at the same time receives the output current of the other power modules 30. Based on the output current of the other power modules 30, it calibrates the output current Ii to obtain the calibrated output current Ii′, and transmits the output current Ii′ to the monitoring unit 20, where i=1,2,...,N, and N is the total number of power modules.

[0071] Specifically, each module controller 31 collects the output current Ii, output voltage Ui, and power device operating temperature Ti of its respective power module 30 in real time through a configured current sensor. In a preferred embodiment, each module controller 31 performs information exchange and calibration through a triple communication method, including: The module controller 31 receives the real-time current synchronization signal from the other power modules 30 through a high-frequency communication line. This signal carries the real-time current waveform information of each module in the form of a high-frequency carrier. The module controller 31 demodulates the received high-frequency signal and extracts the real-time current value of each module.

[0072] The module controller 31 receives a reference current signal sent by the monitoring unit 20 through a low-frequency communication line. This reference current signal is the target output current reference value of each module calculated and determined by the monitoring unit 20 based on the total system load and the number of online modules.

[0073] The module controller 31 receives the real-time current Ij (j≠i) and status information of the other power modules 30, as well as the control commands issued by the monitoring unit 20, through the CAN communication line.

[0074] Based on the data acquired through the aforementioned three communication methods, the module controller 31 performs multi-communication collaborative calibration, including: correcting high-frequency interference errors based on real-time current synchronization signals transmitted through high-frequency communication lines, i.e., comparing the real-time current information of each module carried in the high-frequency signal with the locally acquired values ​​to reduce sampling deviations introduced by high-frequency electromagnetic interference; correcting low-frequency delay errors based on reference current signals transmitted through low-frequency communication lines, i.e., using stable low-frequency reference signals to perform zero-point correction and gain correction on the locally acquired current values ​​to reduce gradual offsets caused by sensor temperature drift or long-term operation of the acquisition circuit; and correcting CAN packet loss errors based on real-time current and status information transmitted through CAN communication lines.

[0075] In one embodiment, the specific method for correcting CAN packet loss errors based on data transmitted through the CAN communication line is as follows: when the module controller 31 fails to receive real-time current from the other power modules 30 through the CAN communication line within a preset period (such as one CAN communication frame period), it determines that CAN packets have been lost. At this time, the module controller 31 performs interpolation compensation based on the reference current signal received through the low-frequency communication line and the historical current data received through the CAN communication line in the previous period, and uses the estimated value to replace the missing real-time current in the calibration calculation to obtain the calibrated output current Ii′.

[0076] Understandably, interpolation compensation can be achieved using methods such as linear interpolation or weighted averaging. For example, if the real-time current Ij(k) of the j-th power module 30 is not received in the k-th CAN communication cycle, the estimated value Ijest(k) can be calculated using the historical current data Ij(k-1) from the previous cycle and the low-frequency reference current signal Ibasej, according to the weighted formula Ijest(k) = α × Ij(k-1) + (1-α) × Ibasej, to replace the missing data. Here, α is a preset weighting coefficient, ranging from 0 to 1. This method ensures that the calibration process is uninterrupted and the accuracy loss is controllable even when CAN communication is interfered with.

[0077] After multi-communication collaborative calibration, the module controller 31 obtains the calibrated output current Ii′ and transmits Ii′ to the monitoring unit 20.

[0078] Step S2: The monitoring unit 20 calculates the average output current Iavg based on the output current Ii′ transmitted by each module controller 31, and calculates the current sharing error ΔIi=Ii′−Iavg for the i-th power module 30; the monitoring unit 20 compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi related to the power conversion circuit 32 of the i-th power module 30, and obtains the corrected current sharing error ΔIi′=ΔIi−ΔIcompi corresponding to the i-th power module 30, and generates a voltage adjustment command Uref independently of the current sharing error ΔIi, and sends the corrected current sharing error ΔIi′ and the voltage adjustment command Uref to the module controller 31 corresponding to the i-th power module 30.

[0079] Specifically, after receiving the calibrated output currents I1′, I2′, ..., IN′ from all N power modules 30, the monitoring unit 20 calculates the average output current: Iavg = (I1′ + I2′ + ... + IN′) / N.

[0080] For the i-th power module 30, the current sharing error is: ΔIi = Ii′ − Iavg. A positive ΔIi indicates that the output current of the module is higher than the average value, and its output needs to be reduced; a negative ΔIi indicates that the output current of the module is lower than the average value, and its output needs to be increased.

[0081] In one embodiment, the monitoring unit 20 acquires the input voltage Vin, switching frequency fs, and pre-stored topology parameters (including the inductance value of inductor L1, the capacitance value of capacitor C1, etc.) of the i-th power module 30, and calculates the topology characteristic compensation term ΔIcompi based on the input voltage, switching frequency, and topology parameters.

[0082] Specifically, the topology characteristic compensation term ΔIcompi is calculated using the following formula: ΔIcompi = KL × (Li − Lavg) + KC × (Ci − Cavg) + KR × (Roni − Ronavg), where: Li is the actual inductance value of the filter inductor in the power conversion circuit of the i-th power module, Lavg is the average inductance value of the filter inductors of all power modules; Ci is the actual capacitance value of the output filter capacitor of the i-th power module, Cavg is the average capacitance value of the output filter capacitors of all power modules; Roni is the on-resistance of the switching device of the i-th power module, Ronavg is the average on-resistance of all modules; KL, KC, and KR are compensation coefficients related to inductance deviation, capacitance deviation, and on-resistance deviation, respectively, and are taken as positive or negative according to the influence characteristics of parameter deviation on current. For example, since on-resistance is inversely proportional to current, KR takes a negative value. This is determined by the monitoring unit 20 based on the current input voltage Vin and switching frequency fs, either by looking up a table or a preset curve. Through this formula, the systematic current deviation caused by the discreteness of topology parameters can be separated from the current sharing error, achieving accurate compensation.

[0083] Understandably, the physical meaning of the topology characteristic compensation term ΔIcompi is that even if the control parameters of all power modules 30 are exactly the same, due to manufacturing tolerances in the power devices of each module (such as differences in the on-resistance of switching transistors and deviations in inductance values), and differences in the switching transient characteristics of the three-arm topology under different input voltages and switching frequencies, the actual output current of each module under the same control command will still have a systematic deviation. This systematic deviation is directly related to the topology and parameters of the power conversion circuit 32. Therefore, by introducing the topology characteristic compensation term ΔIcompi for correction, the systematic deviation caused by differences in topology characteristics and parameters can be separated from the current sharing error, so that the corrected current sharing error ΔIi′ only reflects random and dynamic deviations, thereby improving the accuracy of current sharing control.

[0084] In one embodiment, the monitoring unit 20 also acquires the operating temperature Ti of each power module 30 and dynamically adjusts the topology characteristic compensation term ΔIcompi based on the operating temperature Ti.

[0085] A specific implementation example is as follows: The monitoring unit 20 pre-stores the characteristic curve of the on-resistance of the power device as a function of temperature: Roni(T)=Roni(25℃)×[1+α×(Ti−25℃)], where: α is the temperature coefficient of on-resistance (typical value for MOSFET is about 0.003~0.005 / ℃, and IGBT uses equivalent conversion compensation for on-voltage drop, with the corresponding temperature coefficient α being 0.001~0.002 / ℃, which is also pre-stored in the monitoring unit parameter table); Ti is the current operating temperature of the i-th power module. The monitoring unit 20 updates the on-resistance Roni(T) at the current temperature in real time based on the real-time collected Ti, and substitutes it into the aforementioned ΔIcompi calculation formula to obtain the topology characteristic compensation term after temperature adaptation: ΔIcompi(T)=KL×(Li−Lavg)+KC×(Ci−Cavg)+KR×(Roni(T)−Ronavg(T)), thereby realizing the real-time dynamic adjustment of ΔIcompi with the operating temperature to compensate for the current sharing offset caused by temperature drift.

[0086] It is understandable that parameters such as the on-resistance and threshold voltage of power devices are temperature-dependent. As the operating temperature increases, the on-resistance of MOSFETs typically increases, and the on-voltage drop of IGBTs also changes. These changes cause temperature-dependent deviations in the actual output current of each power module 30 under the same control command. The monitoring unit 20 performs temperature correction on the topology parameters based on the current operating temperature Ti, according to the pre-stored power device temperature-parameter characteristic curves (or lookup tables), and then updates the topology characteristic compensation term ΔIcompi. This temperature adaptive adjustment enables the current sharing control to maintain high accuracy even under conditions of long-term system operation and uneven temperature rise among modules.

[0087] The monitoring unit 20 generates the voltage regulation command Uref independently of the current sharing error ΔIi. The voltage regulation command Uref is calculated by the monitoring unit 20 through the voltage regulation loop based on the system's target output voltage Vset, the actual measured output bus voltage Vout, and external voltage regulation commands (such as remote voltage regulation commands from a cloud server). The voltage regulation loop and the current sharing control loop operate independently within the monitoring unit 20. The voltage regulation command Uref is uniformly sent to the module controller 31 of all online power modules 30. Each module receives the same Uref, ensuring that the target output voltage value of all modules is consistent.

[0088] The monitoring unit 20 sends the corrected current sharing error ΔIi′ calculated for the i-th power module 30 and the unified voltage regulation command Uref to the module controller 31 of the i-th power module 30 through a triple communication method (preferably through the CAN communication line).

[0089] Step S3: Each module controller 31 decouples the current sharing regulation and voltage regulation based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref, so as to regulate the output of the corresponding power conversion circuit 32.

[0090] Specifically, the module controller 31 will input the correction current sharing error ΔIi′ into the built-in fuzzy PID controller to generate the current regulation command ΔIrefi.

[0091] In one embodiment, the specific working process of the fuzzy PID controller is as follows: The corrected flow-averaging error ΔIi′ and its rate of change dΔIi′ / dt are used as the two input linguistic variables of the fuzzy PID controller. The input linguistic variables are fuzzified, that is, the precise values ​​are mapped to a predefined fuzzy set. The fuzzy set usually includes seven fuzzy subsets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB).

[0092] Inference is performed based on a preset fuzzy control rule table. This fuzzy control rule table is pre-set based on control experience with the power conversion system. The inputs to the rule table are fuzzy quantities of the current sharing error and the rate of change of error, while the outputs are adjustments to the proportional gain Kp, integral gain Ki, and derivative gain Kd. For example, when both the current sharing error and the rate of change of error are positive (PB), it indicates a large current sharing deviation that is increasing. In this case, the proportional gain Kp should be increased to accelerate the response, the integral gain Ki should be decreased to avoid integral saturation, and the derivative gain Kd should be increased to provide predictive braking.

[0093] The inference results are converted into precise dynamic adjustments ΔKp, ΔKi, and ΔKd for the proportional gain, integral gain, and derivative gain by defuzzification operations (such as the centroid method). The current parameters of the fuzzy PID controller are then updated based on these dynamic adjustments: Kpnew = Kpbase + ΔKp, Kinew = Kibase + ΔKi, Kdnew = Kdbase + ΔKd, where Kpbase, Kibase, and Kdbase are the baseline parameters of the PID controller.

[0094] Based on the updated parameters, the fuzzy PID controller calculates the output current adjustment command ΔIrefi according to the PID control algorithm.

[0095] Understandably, the advantage of fuzzy PID controllers over traditional fixed-parameter PID controllers lies in the following: Traditional PID controllers maintain fixed parameters after tuning. When system conditions change (such as sudden load changes, input voltage fluctuations, or module activation / deactivation), the fixed parameters struggle to adapt to the changed system characteristics, resulting in slow adjustment speed, large overshoot, or poor steady-state accuracy. Fuzzy PID controllers, through fuzzy inference, dynamically adjust PID parameters in real time, enabling the controller to adapt to different operating conditions. This results in faster response speed and smaller overshoot during dynamic processes, while maintaining high control accuracy during steady-state operation.

[0096] The module controller 31 combines the current regulation command ΔIrefi and the voltage regulation command Uref to generate a PWM control signal, which controls the switching timing of the corresponding power conversion circuit 32.

[0097] Specifically, the module controller 31 uses the voltage regulation command Uref as the reference amplitude of the output voltage to determine the amplitude of the SPWM modulation wave; it superimposes the current regulation command ΔIrefi onto the modulation wave to make a slight adjustment to the amplitude of the modulation wave, thereby achieving current sharing regulation while realizing the target output voltage. Since the voltage regulation command Uref corresponds to a large amplitude adjustment of the output voltage, while the current regulation command ΔIrefi corresponds to a small amplitude correction of the current sharing regulation, there is a significant difference in their amplitude magnitudes. The current regulation has a smaller impact on the voltage waveform, thus achieving decoupling of voltage regulation and current sharing control at the physical execution level.

[0098] Step S4: When any module controller 31 detects a fault in the power module 30 it is connected to, it cuts off the output and transmits a fault signal to the monitoring unit 20.

[0099] Specifically, while performing current sharing control, each module controller 31 monitors the operating status parameters of its respective power module 30 in real time, including output current Ii, output voltage Ui, and power device operating temperature Ti. If any parameter exceeds a preset protection threshold (such as output current exceeding 150% of the rated value, output voltage exceeding 110% of the rated value, or operating temperature exceeding the preset safe temperature limit), the module controller 31 determines that the power module 30 has malfunctioned, immediately blocks the PWM output signal, and cuts off the output of the power conversion circuit 32.

[0100] While cutting off the output, module controller 31 sends a fault signal to monitoring unit 20 via the CAN communication line. The fault signal includes the faulty module number i, the fault type (overcurrent / overvoltage / overtemperature / device failure), and the timestamp of the fault occurrence. Module controller 31 also broadcasts the fault information to the module controllers 31 of the other power modules 30 via the CAN communication line, so that the other modules can be aware of the presence of a faulty module in the system before the adjustment command from monitoring unit 20 arrives, and can perform local pre-adjustment.

[0101] Step S5: When the monitoring unit 20 receives a fault signal, it controls the bypass unit 10 to engage the bypass. At the same time, it recalculates the average output current Iavg′ based on the number N−1 of the remaining non-faulty power modules 30 and the total output current of the system. Based on the recalculated average output current Iavg′, it recalculates the corrected current sharing error ΔIi′ and sends it to the controller 31 of the remaining non-faulty modules.

[0102] Specifically, after receiving a fault signal, the monitoring unit 20 executes the following collaborative control process: First, a local alarm is triggered, including illuminating the alarm indicator light of the corresponding faulty module and activating the buzzer. Simultaneously, the fault information (faulty module number, fault type, and fault time) is uploaded to the cloud server via the communication gateway module 40. Upon receiving the fault information, the cloud server sends a fault warning notification to the remote control terminal of the maintenance personnel.

[0103] Second, the monitoring unit 20 drives the electronic switch S3 of the bypass unit 10 to close via a control signal, thus activating the centralized bypass. The activation timing of the bypass unit 10 is synchronized with the switching timing of the remaining normally operating power modules 30 through a high-frequency communication line, ensuring that no phase change or current surge occurs at the moment of bypass activation. Simultaneously, the phase reference signal sent to the bypass unit 10 via the low-frequency communication line ensures that the phase of the bypass output voltage is consistent with the phase of the output voltage of the power modules 30.

[0104] Third, the monitoring unit 20 removes the faulty module from the online module list and recalculates the average output current based on the remaining N-1 non-faulty power modules 30: Iavg′=Itotal / (N-1), where Itotal is the current total output current of the system.

[0105] In one embodiment, when the monitoring unit 20 recalculates and corrects the current sharing error ΔIi′ and sends it to the controller 31 of the remaining non-faulty modules, it does not switch the average output current to Iavg′ instantaneously, but rather smoothly transitions the average output current of the remaining non-faulty power modules 30 to Iavg′ at a preset ramp rate.

[0106] Specifically, the monitoring unit 20 sets a preset ramp rate ΔIslope (in A / ms). During the transition period, the average output current value increases linearly from the original Iavg to Iavg′ at the rate of ΔIslope. During the transition, the monitoring unit 20 updates and calculates the corrected current sharing error ΔIi′ for each remaining module based on the current average output current value in each control cycle, and synchronously sends it to the corresponding module controller 31, so that the output current of the remaining non-faulty power modules 30 gradually increases at a controllable rate, smoothly taking over the load share originally borne by the faulty module.

[0107] Understandably, the value of the preset ramp rate ΔIslope needs to comprehensively consider the system's dynamic response capability and the load's requirements for power supply continuity. An excessively high ramp rate may cause the current in the remaining modules to change too rapidly, leading to transient fluctuations in the output voltage; an excessively low ramp rate will result in a long transition time, potentially causing insufficient output capability during the transition period. In a preferred embodiment, the preset ramp rate ΔIslope is set to ensure the transition time is within the range of 10ms to 100ms, balancing dynamic response speed and output stability.

[0108] After the transition is completed, the system enters a steady-state current sharing control state in which N-1 modules operate in parallel, and repeats the current sharing control process from steps S1 to S3.

[0109] It should be noted that during fault switching, the activation time of bypass unit 10 and the transition time of average output current are independent of each other. The electronic switch S3 in bypass unit 10 activates at millisecond speeds after receiving the control signal, providing immediate bypass protection for the system. The ramp transition of average output current is performed after bypass activation, ensuring a smooth and gradual load takeover by the remaining modules. During the transition, bypass unit 10 and the remaining power module 30 simultaneously supply power to the load. Once the remaining modules have fully taken over the load, monitoring unit 20 determines whether to deactivate the bypass based on the actual system operation.

[0110] While the system operates normally and handles faults, it continuously performs dynamic adaptive adjustments and remote maintenance. The monitoring unit 20 monitors system load changes and input voltage fluctuations in real time. When there are sudden load changes or input voltage fluctuations, it quickly updates the average output current Iavg. The fuzzy PID controllers of each module controller 31 dynamically adjust control parameters, and multi-communication collaborative calibration corrects transmission errors in real time, ensuring current sharing accuracy and output voltage stability. Simultaneously, the monitoring unit 20 dynamically adjusts the topology characteristic compensation term ΔIcompi based on the operating temperature Ti of each module to improve current sharing deviations caused by temperature changes. The monitoring unit 20 continuously uploads system operating information to the cloud server. The cloud server analyzes historical data, can predict potential faults in advance, and send early warning information. Maintenance personnel can issue remote commands (such as parameter calibration, module start / stop, voltage adjustment commands, etc.) through a remote control terminal. After receiving the commands, the monitoring unit 20 synchronously distributes them to each module controller 31, realizing remote maintenance.

[0111] In one embodiment, the effect of current sharing control is explained. Taking N=4 power modules 30 operating in parallel as an example, the rated total output current of the system is 100A, and the rated output current of each module is 25A. Under normal operating conditions, the output current Ii′ of each module controller 31 after triple communication collaborative calibration in step S1 is I1′=25.3A, I2′=24.7A, I3′=25.1A, and I4′=24.9A. In step S2, the monitoring unit 20 calculates the average output current Iavg=(25.3+24.7+25.1+24.9) / 4=25.0A, and the current sharing errors of each module are ΔI1=0.3A, ΔI2=-0.3A, ΔI3=0.1A, and ΔI4=-0.1A, respectively. After topology characteristic compensation correction, the corrected current sharing error ΔIi′ is further reduced. In step S3, the fuzzy PID controller of each module generates a current adjustment command based on the corrected current sharing error, adjusting the output current of each module to approach 25.0A. After several control cycles, the current sharing accuracy of the output current of each module can be controlled within ±2%.

[0112] When the fourth module in power module 30 fails, in step S4, the module controller 31 of the fourth module detects the fault, immediately cuts off the output, and sends a fault signal. In step S5, the monitoring unit 20 is bypassed, and the average output current Iavg′=100 / (4-1)≈33.3A is recalculated. The average output current of the remaining three modules is smoothly transitioned from 25.0A to 33.3A at a preset ramp rate. During the transition, the current sharing error is updated and corrected synchronously and sent out, so that the system smoothly switches to the parallel operation state of the three modules.

[0113] It is understood that the above values ​​are merely examples for illustrative purposes. In actual applications, parameters such as the number of modules, rated current, and current sharing accuracy can be set according to specific application scenarios and system specifications, and do not constitute a limitation on the scope of protection of this application.

[0114] It should be noted that the power conversion circuit 32 in this application is described in detail using a three-arm direct ACAC conversion topology, but the applicability of the current sharing control method is not limited to this specific topology. For multi-module parallel systems using other types of power conversion topologies (such as four-arm topology, full-bridge topology, half-bridge topology, etc.), as long as the calculation method of the topology characteristic compensation term ΔIcompi is adjusted accordingly to match the characteristic parameters of the corresponding topology, the current sharing control method described in this application is also applicable.

[0115] It should be noted that the execution of each step in the current sharing control method of this application is not strictly sequential. In actual operation, the data acquisition and calibration in step S1, the current sharing error calculation in step S2, and the module output adjustment in step S3 are executed cyclically with a fixed control cycle (such as every switching cycle or every few switching cycles) to form a closed-loop control; the fault detection in step S4 is executed synchronously in each control cycle; the fault handling in step S5 is event-triggered and is executed only when a fault signal is received. The parallel and cyclic execution of the above steps ensures the real-time control capability of the system.

[0116] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A current sharing control method for a modular parallel voltage regulating and stabilizing device in a power distribution network, characterized in that, The modular parallel voltage regulation and stabilization device for the distribution network includes a bypass unit (10), a monitoring unit (20), and multiple parallel power modules (30). Each power module (30) includes a power conversion circuit (32) and a module controller (31). The method includes the following steps: S1. Each module controller (31) collects the output current Ii of the power module (30) it is in, and at the same time receives the output current of the other power modules (30), and calibrates the output current Ii based on the output current of the other power modules (30) to obtain the calibrated output current Ii′, and transmits the output current Ii′ to the monitoring unit (20), where i=1,2,...,N, and N is the total number of power modules (30); S2. The monitoring unit (20) calculates the average output current Iavg based on the output current Ii′ transmitted by each module controller (31), and calculates the current sharing error ΔIi=Ii′−Iavg for the i-th power module (30); the monitoring unit (20) compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi related to the power conversion circuit (32) of the i-th power module (30), and obtains the corrected current sharing error ΔIi′=ΔIi−ΔIcompi corresponding to the i-th power module (30), and generates a voltage adjustment command Uref independently of the current sharing error ΔIi, and sends the corrected current sharing error ΔIi′ and the voltage adjustment command Uref to the module controller (31) corresponding to the i-th power module (30). S3. Each module controller (31) decouples current sharing regulation and voltage regulation based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref to regulate the output of the corresponding power conversion circuit (32); S4. When any of the module controllers (31) detects a fault in the power module (30), it cuts off the output and transmits a fault signal to the monitoring unit (20); S5. When the monitoring unit (20) receives the fault signal, it controls the bypass unit (10) to put it into bypass mode. At the same time, it recalculates the average output current Iavg′ based on the number N−1 of the remaining non-faulty power modules (30) and the total output current of the system. Based on the recalculated average output current Iavg′, it recalculates the corrected current sharing error ΔIi′ and sends it to the controller (31) of the remaining non-faulty modules.

2. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 1, characterized in that, The monitoring unit (20) stores the topology parameters of each power module (30); in step S2, the monitoring unit (20) compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi related to the power conversion circuit (32) of the i-th power module (30), including: The monitoring unit (20) acquires the input voltage, switching frequency and pre-stored topology parameters of the i-th power module (30); and calculates the topology characteristic compensation term ΔIcompi based on the input voltage, the switching frequency and the topology parameters.

3. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 2, characterized in that, The monitoring unit (20) also acquires the operating temperature of each power module (30); the method further includes the following steps: The monitoring unit (20) dynamically adjusts the topology characteristic compensation term ΔIcompi based on the operating temperature.

4. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 1, characterized in that, In step S3, each module controller (31) performs decoupled control of current sharing regulation and voltage regulation based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref, in order to regulate the output of the corresponding power conversion circuit (32), including: The module controller (31) inputs the corrected current sharing error ΔIi′ into the fuzzy PID controller to generate the current regulation command ΔIrefi; the module controller (31) combines the current regulation command ΔIrefi and the voltage regulation command Uref to generate a PWM control signal to control the switching timing of the corresponding power conversion circuit (32).

5. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 4, characterized in that, The module controller (31) inputs the corrected current sharing error ΔIi′ into the fuzzy PID controller to generate a current regulation command ΔIrefi, which specifically includes: Using the corrected current sharing error ΔIi′ and the rate of change of the corrected current sharing error ΔIi′ as the input linguistic variables of the fuzzy PID controller, inference and defuzzification are performed according to the preset fuzzy control rule table to obtain the dynamic adjustment amounts of proportional gain, integral gain and derivative gain; the current parameters of the fuzzy PID controller are updated based on the dynamic adjustment amounts, and the current regulation command ΔIrefi is calculated and output according to the updated parameters.

6. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 1, characterized in that, In step S5, the corrected current sharing error ΔIi′ is recalculated based on the recalculated average output current Iavg′ and then sent to the remaining non-faulty module controller (31), specifically including: The monitoring unit (20) smoothly transitions the average output current of the remaining non-faulty power modules (30) to the recalculated average output current Iavg′ at a preset ramp rate, and synchronously updates the corrected current sharing error ΔIi′ during the transition process, so that the output current of the remaining non-faulty power modules (30) smoothly takes over the total load of the system.

7. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 1, characterized in that, The modular parallel voltage regulation and stabilization device of the power distribution network also includes a high-frequency communication line, a low-frequency communication line and a CAN communication line; in step S1, each module controller (31) collects the output current Ii of its power module (30), and simultaneously receives the output current of the other power modules (30), and calibrates the output current Ii based on the output current of the other power modules (30) to obtain the calibrated output current Ii′, specifically including: The module controller (31) receives real-time current synchronization signals from the other power modules (30) through the high-frequency communication line, receives reference current signals sent by the monitoring unit (20) through the low-frequency communication line, and receives real-time current and status information of the other power modules (30) through the CAN communication line. The module controller (31) corrects high-frequency interference errors based on the data transmitted through the high-frequency communication line, corrects low-frequency delay errors based on the data transmitted through the low-frequency communication line, and corrects CAN packet loss errors based on the data transmitted through the CAN communication line to obtain the calibrated output current Ii′.

8. The current sharing control method for the modular parallel voltage regulating and stabilizing device in the distribution network according to claim 7, characterized in that, When the module controller (31) fails to receive the real-time current of the other power modules (30) through the CAN communication line within a preset period, it is determined to be a CAN packet loss; the module controller (31) performs interpolation compensation based on the reference current signal received by the low-frequency communication line and the historical current data received through the CAN communication line in the previous period, so as to replace the missing real-time current in the calibration calculation and obtain the calibrated output current Ii′.

9. A modular parallel voltage regulating and stabilizing device for a power distribution network, used to execute the current sharing control method as described in any one of claims 1-8, characterized in that, include: Multiple power modules (30) are connected in parallel between the power grid and the load. Each power module (30) includes a power conversion circuit (32), a module controller (31), and a communication circuit (33). The power conversion circuit (32) is used to convert the input AC voltage into the output AC voltage. The module controller (31) is used to collect the output current Ii of the power module (30) and receive the output current of the other power modules (30) through the communication circuit (33) to obtain the calibrated output current Ii′. Based on the received corrected current sharing error ΔIi′ and voltage regulation command Uref, the current sharing regulation and voltage regulation are decoupled and controlled, and a PWM signal is output to the power conversion circuit (32). When a fault is detected, the module controller (31) cuts off the output of the power conversion circuit (32) and sends a fault signal. The monitoring unit (20) is electrically connected to the module controller (31). The monitoring unit (20) is used to receive the output current Ii′ transmitted by each module controller (31) and calculate the average output current Iavg and the current sharing error ΔIi. It compensates and corrects the current sharing error ΔIi by combining the topology characteristic compensation term ΔIcompi to obtain the corrected current sharing error ΔIi′. It generates the voltage adjustment command Uref independently of the current sharing error ΔIi and sends the corrected current sharing error ΔIi′ and the voltage adjustment command Uref to the corresponding module controller (31). The monitoring unit (20) is also used to recalculate the corrected current sharing error ΔIi′ based on the remaining non-faulty power module (30) and send it when the fault signal is received. The bypass unit (10) is connected to the monitoring unit (20) and is used to enable or disable the bypass under the control of the monitoring unit (20).

10. The modular parallel voltage regulating and stabilizing device for power distribution networks according to claim 9, characterized in that, The communication circuit (33) is interconnected through a high-frequency communication line, a low-frequency communication line and a CAN communication line respectively; the high-frequency communication line, the low-frequency communication line and the CAN communication line are all connected to the monitoring unit (20), the bypass unit (10) and each of the module controllers (31). The high-frequency communication line is used to transmit high-frequency synchronization signals so that the module controller (31) can correct high-frequency interference errors and synchronize the switching action of the bypass unit (10) with the switching timing of the power conversion circuit (32); the low-frequency communication line is used by the monitoring unit (20) to send a reference current signal to the module controller (31) and a phase reference signal to the bypass unit (10); the CAN communication line is used for the interaction of real-time data, status information and control commands between the monitoring unit (20), the module controller (31) and the bypass unit (10).

Images