Power storage wave carrier PCBA multi-module carrier synchronous control system

Abstract

The application relates to the technical field of electric power energy storage, in particular to a power storage wave carrier PCBA multi-module carrier synchronous control system, which comprises a data acquisition and processing module used for acquiring power storage system operation data; a central control and decision module used for receiving data, generating carrier-level current loop control instructions through a voltage outer loop control algorithm optimized by deep reinforcement learning; a current inner loop driving control module used for combining real-time current feedback data, generating pulse width modulation signals through a repetitive control composite algorithm and an SVPWM algorithm; and a power storage converter execution module used for performing bidirectional energy conversion between a power storage battery and a power grid based on the pulse width modulation signals; and a multi-module carrier synchronous module used for performing multi-PCBA module carrier same-frequency and same-phase through a distributed synchronization algorithm based on whole-process data. Therefore, the problems of insufficient data precision, poor algorithm adaptability and multi-module synchronization difficulty in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of power energy storage technology, specifically to a power energy storage waveborne PCBA multi-module carrier synchronization control system. Background Technology

[0002] With the increasing proportion of new energy power generation and the intensifying demand for grid peak shaving, power storage systems are rapidly developing towards larger capacity and higher power. Parallel operation of multiple PCBA modules has become the core solution for system expansion. Currently, most mainstream energy storage systems adopt a distributed control architecture, with each PCBA module independently completing data acquisition and command execution. While this can meet basic energy conversion needs, it has significant technical shortcomings in multi-module collaborative control: On the one hand, the data acquisition stage lacks a unified preprocessing mechanism, and parameters such as voltage and current are easily affected by grid harmonics and module temperature drift, resulting in insufficient accuracy of the raw data. Consequently, the current loop command output by the control algorithm deviates from the actual demand. On the other hand, traditional control relies heavily on local decision-making of a single module and has not built a cross-module carrier synchronization mechanism. The carrier frequency and phase of each module are prone to asynchrony due to differences in hardware parameters or communication delays, leading to problems such as increased circulating current between modules and uneven power distribution. In severe cases, this can even cause converter overload shutdown.

[0003] The shortcomings of existing technologies further restrict the performance improvement of energy storage systems: At the control algorithm level, most voltage outer-loop controls use fixed-parameter PID algorithms, which cannot optimize control strategies in real time according to dynamic operating conditions such as sudden changes in grid load and battery SOC changes. This results in slow system response speed, poor voltage stability, and difficulty in balancing voltage regulation accuracy and energy consumption control objectives. At the multi-module synchronization level, the traditional CAN bus or Ethernet communication speed is low, which cannot achieve high-speed interaction of carrier phase information. Furthermore, the lack of distributed synchronization algorithm support makes it difficult to perform real-time fusion correction of phase deviations between modules, leading to decreased power conversion efficiency and increased harmonic pollution during multi-module operation. These problems not only reduce the operational reliability and economy of energy storage systems but also fail to meet the requirements of the new generation of power grids for high response speed and high synchronization accuracy. There is an urgent need for an integrated control system that integrates high-precision data processing, intelligent decision optimization, and efficient carrier synchronization functions. Summary of the Invention

[0004] This application provides a multi-module carrier synchronization control system for power storage waveborne PCBA to solve problems such as insufficient data accuracy, poor algorithm adaptability, and difficulty in synchronizing multiple modules in the prior art.

[0005] The first aspect of this application provides a multi-module carrier synchronization control system for power storage carrier PCBAs, comprising: a data acquisition and processing module, a central control and decision-making module, a current inner loop drive control module, an energy storage converter execution module, and a multi-module carrier synchronization module; wherein, the data acquisition and processing module is used to acquire voltage, current, temperature, and grid parameter data during the operation of the power storage system; the central control and decision-making module is used to receive the data, construct a multi-objective reward function to optimize control parameters in real time through a voltage outer loop control algorithm optimized by deep reinforcement learning, and generate carrier-level current loop control commands; the current inner loop drive control module receives the carrier-level current loop control commands, combines real-time current feedback data, and generates pulse width modulation signals through a repetitive control composite algorithm and an SVPWM algorithm; the energy storage converter execution module, based on the pulse width modulation signals, uses a wide bandgap semiconductor device full-bridge topology and a dynamic energy buffer circuit to perform bidirectional energy conversion between the energy storage battery and the grid; the multi-module carrier synchronization module, based on the full-process data, uses a distributed synchronization algorithm to perform carrier phase information interaction and fusion correction between multiple PCBA modules through a high-speed fiber optic bus to achieve carrier synchronization of multiple PCBA modules.

[0006] Preferably, the data acquisition and processing module includes a parameter acquisition unit, a signal preprocessing unit, and a data distribution unit; wherein, the parameter acquisition unit acquires the terminal voltage and charging / discharging current of the energy storage battery pack, the input and output current and voltage of the energy storage converter, the operating temperature of each PCBA module, and the voltage, frequency, and phase signals of the grid side through voltage sensors, current sensors, temperature sensors, and a grid monitoring unit; the signal preprocessing unit is used to perform amplitude normalization and high-frequency interference suppression on the acquired data using an adaptive filtering algorithm; the data distribution unit is used to synchronously transmit the processed data to the central control and decision-making module and the multi-module carrier synchronization module.

[0007] Preferably, the central control and decision-making module includes a data receiving unit, a status analysis unit, and an algorithm calculation unit; wherein, the data receiving unit is used to receive the processed data for verification and caching; the status analysis unit determines the system operating status based on the received data, including normal operation, load surge, power grid anomaly, and other operating conditions; the algorithm calculation unit constructs a multi-objective reward function with voltage stability, response speed, and energy consumption as objectives through a voltage outer loop control algorithm optimized by deep reinforcement learning, optimizes control parameters in real time, and generates carrier-level current loop control commands.

[0008] Preferably, the current inner loop drive control module includes an instruction parsing unit, a feedback acquisition unit, an algorithm processing unit, and a PWM generation unit; wherein, the instruction parsing unit is used to parse the carrier-level current loop control instruction and extract the current control parameters; the feedback acquisition unit is used to acquire real-time current feedback data from the energy storage converter execution module; the algorithm processing unit performs calculations on the instruction parameters and feedback data using a repetitive control composite algorithm to calculate the pulse width modulation signal parameters; the PWM generation unit generates a pulse width modulation signal based on the pulse width modulation signal parameters using the SVPWM algorithm, and transmits it to the energy storage converter execution module after power amplification.

[0009] Preferably, the energy storage converter execution module includes a power conversion unit, an energy buffer unit, a filtering unit, and a status feedback unit; wherein, the power conversion unit adopts a wide bandgap semiconductor device full-bridge topology and performs energy conversion under pulse width modulation signal control; the energy buffer unit absorbs voltage and current fluctuations during the conversion process through a dynamic energy buffer circuit; the filtering unit filters the converted electrical energy to reduce harmonic content; and the status feedback unit collects converter operating parameters and feeds them back to the data acquisition and processing module.

[0010] Preferably, the multi-module carrier synchronization module includes a data interaction unit, an algorithm fusion unit, an instruction generation unit, and a synchronization execution unit; wherein, the data interaction unit receives the operating data, carrier phase information, and system full-process data of each PCBA module through a high-speed fiber optic bus; the algorithm fusion unit uses a distributed synchronization algorithm to perform fusion calculations on the received data to determine the carrier synchronization deviation of each module; the instruction generation unit generates carrier phase adjustment instructions based on the synchronization deviation; and the synchronization execution unit distributes the adjustment instructions to each PCBA module to perform multi-module carrier frequency and phase synchronization control.

[0011] The second aspect of this application provides a method for multi-module carrier synchronization control of a power storage carrier PCBA, comprising: acquiring voltage, current, carrier frequency, phase difference, and operating temperature parameter data of each module; filtering noise from the voltage, current, carrier frequency, phase difference, and operating temperature parameter data of each module using an adaptive filtering algorithm, constructing a multi-objective reward function to optimize control parameters in real time using a voltage outer loop control algorithm optimized by deep reinforcement learning, and generating a carrier-level current loop control command; extracting current control parameters based on the carrier-level current loop control command, calculating pulse width modulation signal parameters using a repetitive control composite algorithm using real-time current feedback data, and generating a pulse width modulation signal using an SVPWM algorithm; performing bidirectional energy conversion between the energy storage battery and the grid according to the pulse width modulation signal, obtaining converter operating status feedback data, and calculating the carrier synchronization deviation of each module using a distributed synchronization algorithm using the full-process data, generating a carrier phase adjustment command, and performing multi-module carrier synchronous frequency and phase control.

[0012] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the power storage wavecarrier PCBA multi-module carrier synchronization control method as described in the above embodiments.

[0013] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the power storage waveborne PCBA multi-module carrier synchronization control method as described in the above embodiments.

[0014] The fifth aspect of this application provides a computer program product, including a computer program or instructions, for implementing the power storage waveborne PCBA multi-module carrier synchronization control method as described in the above embodiments.

[0015] Therefore, this application has the following beneficial effects: This application embodiment utilizes a multi-sensor collaborative data acquisition module in its data acquisition and processing module. Combined with an adaptive filtering algorithm, amplitude normalization and high-frequency interference suppression are performed to effectively address grid harmonics and temperature drift interference, improving parameter acquisition accuracy and data reliability under complex operating conditions. The central control and decision-making module employs deep reinforcement learning to optimize the voltage outer-loop control algorithm, constructing a multi-objective reward function. This optimizes control parameters in real-time under scenarios such as load surges and grid anomalies, enhancing the adaptability and accuracy of control commands, improving system voltage stability and response speed, and reducing energy consumption. Finally, the current inner-loop drive control module utilizes a repetitive control composite algorithm, combined with command parsing and real-time current feedback data. The system generates high-precision pulse-width modulation (PWM) signals to enhance current control accuracy and provide precise drive support for energy conversion. Through the wide-bandgap semiconductor full-bridge topology and dynamic energy buffer circuit of the energy storage converter execution module, combined with high-precision PWM signal control, it performs efficient bidirectional energy conversion between the energy storage battery and the grid, absorbing voltage and current fluctuations and reducing harmonic content, thus improving power quality and energy conversion efficiency. Utilizing the distributed synchronization algorithm of the multi-module carrier synchronization module and a high-speed fiber optic bus, it performs efficient interaction and fusion correction of carrier phase information from each PCBA module, eliminating carrier phase deviations between modules, enhancing the parallel coordination of multiple modules, and improving operational stability. This solves the problems of insufficient data accuracy, poor algorithm adaptability, and difficulty in multi-module synchronization in existing technologies.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the structure of the power storage waveborne PCBA multi-module carrier synchronization control system provided according to an embodiment of this application; Figure 2 This is a schematic diagram of a data acquisition and processing module according to an embodiment of this application; Figure 3 This is a schematic diagram of a central control and decision-making module according to an embodiment of this application; Figure 4 This is a schematic diagram of a current inner loop drive control module according to an embodiment of this application; Figure 5 This is a schematic diagram of an energy storage converter execution module according to an embodiment of this application; Figure 6 This is a schematic diagram of a multi-module carrier synchronization module according to an embodiment of this application; Figure 7 This is a flowchart of a multi-module carrier synchronization control method based on a power storage waveborne PCBA according to an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] The following describes a multi-module carrier synchronization control system for a power storage waveborne PCBA according to an embodiment of this application, with reference to the accompanying drawings. Addressing the difficulty of multi-module synchronization mentioned in the background art, this application provides a multi-module carrier synchronization control system for a power storage waveborne PCBA. In this system, data is collaboratively acquired by multiple sensors in the data acquisition and processing module. Combined with an adaptive filtering algorithm, amplitude normalization and high-frequency interference suppression are performed, effectively addressing grid harmonics and temperature drift interference, and improving parameter acquisition accuracy and data reliability under complex operating conditions. The deep reinforcement learning of the central control and decision module optimizes the voltage outer loop control algorithm, constructing a multi-objective reward function. This optimizes control parameters in real time under scenarios such as load mutations and grid anomalies, enhancing the adaptability and accuracy of control commands, improving system voltage stability and response speed, and reducing energy consumption. The system is driven by a current inner loop control module. A repetitive control composite algorithm, combined with instruction parsing and real-time current feedback data calculation, generates a high-precision pulse-width modulation (PWM) signal, enhancing current control accuracy and providing precise drive support for energy conversion. Through the wide-bandgap semiconductor device full-bridge topology and dynamic energy buffer circuit of the energy storage converter execution module, coupled with high-precision PWM signal control, efficient bidirectional energy conversion between the energy storage battery and the grid is achieved, absorbing voltage and current fluctuations and reducing harmonic content, thus improving power quality and energy conversion efficiency. Utilizing the distributed synchronization algorithm of the multi-module carrier synchronization module and a high-speed fiber optic bus, efficient interaction and fusion correction of carrier phase information from each PCBA module are performed, eliminating carrier phase deviations between modules, enhancing multi-module parallel coordination, and improving operational stability. This solves the problems of insufficient data accuracy, poor algorithm adaptability, and difficulty in multi-module synchronization in existing technologies.

[0020] Figure 1 This is a schematic diagram of the structure of the power storage waveborne PCBA multi-module carrier synchronization control system provided in the embodiments of this application.

[0021] This application provides a multi-module carrier synchronization control system for power storage waveborne PCBA, the system 10 including: The system includes a data acquisition and processing module 100, a central control and decision-making module 200, a current inner loop drive control module 300, an energy storage converter execution module 400, and a multi-module carrier synchronization module 500.

[0022] The system comprises several modules: a data acquisition and processing module 100, which collects voltage, current, temperature, and grid parameter data during the operation of the power storage system; a central control and decision-making module 200, which receives data, constructs a multi-objective reward function to optimize control parameters in real time using a voltage outer loop control algorithm optimized by deep reinforcement learning, and generates carrier-level current loop control commands; a current inner loop drive control module 300, which receives carrier-level current loop control commands, combines real-time current feedback data, and generates pulse width modulation signals using a repetitive control composite algorithm and an SVPWM algorithm; an energy storage converter execution module 400, which performs bidirectional energy conversion between the energy storage battery and the grid based on the pulse width modulation signal, using a wide bandgap semiconductor device full-bridge topology and a dynamic energy buffer circuit; and a multi-module carrier synchronization module 500, which uses distributed synchronization algorithms based on full-process data and a high-speed fiber optic bus to perform carrier phase information interaction and fusion correction between multiple PCBA modules, ensuring that the carriers of multiple PCBA modules are in phase and at the same frequency.

[0023] It is understood that in this embodiment, data is collected collaboratively by multiple sensors in the data acquisition and processing module, combined with an adaptive filtering algorithm, to perform amplitude normalization and high-frequency interference suppression, effectively addressing grid harmonics and temperature drift interference, and improving parameter acquisition accuracy and data reliability under complex operating conditions. The deep reinforcement learning of the central control and decision module optimizes the voltage outer loop control algorithm, constructing a multi-objective reward function to optimize control parameters in real time under scenarios such as load mutations and grid anomalies, enhancing the adaptability and accuracy of control commands, improving system voltage stability and response speed, and reducing energy consumption. The repetitive control composite algorithm of the current inner loop drive control module, combined with command parsing and real-time current... Feedback data processing generates a high-precision pulse-width modulation (PWM) signal, enhancing current control accuracy and providing precise drive support for energy conversion. Through the wide-bandgap semiconductor device full-bridge topology and dynamic energy buffer circuit of the energy storage converter execution module, combined with high-precision PWM signal control, efficient bidirectional energy conversion between the energy storage battery and the grid is achieved, absorbing voltage and current fluctuations and reducing harmonic content, thus improving power quality and energy conversion efficiency. Utilizing the distributed synchronization algorithm of the multi-module carrier synchronization module and a high-speed fiber optic bus, efficient interaction and fusion correction of carrier phase information from each PCBA module are performed, eliminating carrier phase deviations between modules, enhancing multi-module parallel coordination, and improving operational stability. This solves the problems of insufficient data accuracy, poor algorithm adaptability, and difficulty in multi-module synchronization in existing technologies.

[0024] In this embodiment of the application, the data acquisition and processing module 100 further includes: Figure 2 As shown, there are parameter acquisition unit, signal preprocessing unit, and data distribution unit.

[0025] The parameter acquisition unit collects the terminal voltage and charging / discharging current of the energy storage battery pack, the input and output current and voltage of the energy storage converter, the operating temperature of each PCBA module, and the voltage, frequency, and phase signals of the grid side through voltage sensors, current sensors, temperature sensors, and grid monitoring units, respectively. The signal preprocessing unit performs amplitude normalization and high-frequency interference suppression on the collected data using an adaptive filtering algorithm. The data distribution unit synchronously transmits the processed data to the central control and decision-making module and the multi-module carrier synchronization module.

[0026] It should be noted that adaptive filtering algorithm refers to an intelligent signal processing method that can adjust filtering parameters in real time according to the characteristics of the input signal, dynamically suppress interference such as power grid harmonics and temperature drift, while retaining the effective signal characteristics.

[0027] It is understood that the parameter acquisition unit of the data acquisition and processing module in this application embodiment synchronously acquires battery status, converter operating conditions, module temperature, and grid parameters through voltage, current, temperature sensors, and grid monitoring unit. This allows for the comprehensive capture of core data on the operation of the energy storage system, providing full-dimensional information support for subsequent control strategy generation and module synchronous adjustment. The signal preprocessing unit uses an adaptive filtering algorithm to perform amplitude normalization and high-frequency interference suppression, which can eliminate differences in the dimensions of different parameters, filter out grid harmonics and sensor noise, effectively solve the problems of poor compatibility and significant interference signal impact of traditional acquisition data, and ensure data accuracy and effectiveness. The data distribution unit synchronously transmits the preprocessed data to the central control and decision module and the multi-module carrier synchronization module, ensuring that the two core modules obtain consistent real-time data, avoiding control deviations or synchronization failures caused by data asynchrony, significantly improving the comprehensiveness, reliability, and timeliness of system data acquisition, and laying a solid data foundation for the precise operation of the entire energy storage control system and the collaborative work of multiple modules.

[0028] For example, in a practical application of a 20MW / 40MWh grid-side energy storage power station, the data acquisition and processing module of this application, installed in the power station, uses 32 high-precision voltage sensors (measurement accuracy ±0.1%FS), 16 high-frequency current sensors (sampling frequency 5kHz), 48 temperature sensors (resolution 0.1℃), and a grid monitoring unit to synchronously acquire, every 5ms, the terminal voltage (350V-430V) and charging / discharging current (0-800A) of the energy storage battery pack, the input and output current (0-1200A) and voltage (380V / 35kV) of 12 energy storage converters, the operating temperature (25℃-70℃) of 48 PCBA modules, and the voltage (35kV±2%), frequency (50Hz±0.1Hz), and phase signal (accuracy ±0.2°) of the grid side. The system captures core data of the power plant operation. When the power grid experiences a sudden 10th harmonic (amplitude accounting for 20% of the fundamental frequency) or when the sensor experiences 8mV noise interference, the signal preprocessing unit uses an adaptive filtering algorithm to complete amplitude normalization within 50ms (mapping voltage, current, and temperature data to a standardized 0-1 range). At the same time, it improves the high-frequency interference suppression rate to over 95%, effectively solving the problems of poor data compatibility and significant interference signal impact in traditional data acquisition. The data distribution unit transmits the preprocessed data synchronously to the central control and decision-making module and the multi-module carrier synchronization module every 10ms via a gigabit fiber optic bus (transmission rate 1Gbps). This ensures that the data delay difference between the two modules is less than 2ms, avoiding the problem of current loop command error exceeding 2% due to data asynchrony. This provides stable data support for the subsequent accurate calculation of control algorithms and multi-module collaborative synchronization.

[0029] In this embodiment of the application, the central control and decision-making module 200 includes: Figure 3 As shown, there are a data receiving unit, a status analysis unit, and an algorithm calculation unit.

[0030] The data receiving unit receives and processes the data for verification and caching; the status analysis unit determines the system's operating status based on the received data, including normal operation, load surges, and grid anomalies; the algorithm operation unit constructs a multi-objective reward function with voltage stability, response speed, and energy consumption as objectives through a voltage outer loop control algorithm optimized by deep reinforcement learning, optimizes control parameters in real time, and generates carrier-level current loop control commands.

[0031] It should be noted that the voltage outer-loop control algorithm optimized by deep reinforcement learning refers to an adaptive control method that continuously interacts and learns the voltage regulation law under different operating conditions through trial and error between the agent and the system environment. Guided by a multi-objective reward function, it optimizes control parameters in real time to achieve coordinated optimization of voltage stability, response speed, and energy consumption. The formula is as follows: ; ; ; in, for The reward value at any given moment; As a weight for voltage stability; Voltage deviation rate; Weighted by response speed; For response latency; Energy consumption weighting; As a percentage of energy consumption; for The value of a moment; for The system state vector at time t; for Controlling actions at specific moments; Discount factor; To reinforce learning model parameters; The learning rate; For parameters Gradient calculation.

[0032] It is understood that the data receiving unit in this embodiment receives preprocessed data and performs verification and caching to avoid loss or errors in the original data transmission, ensuring the integrity and reliability of the data entering subsequent units. The caching mechanism balances the data transmission rate and the processing rate of subsequent units, preventing processing delays caused by sudden data influx. The status analysis unit judges the system operating status based on the verified reliable data, captures changes in system operating conditions in real time, breaks the limitation of traditional fixed control strategies being difficult to adapt to complex operating conditions, provides operating condition basis for the algorithm calculation unit, and avoids control inaccuracies caused by misjudgment of operating conditions. The algorithm calculation unit constructs a multi-objective reward function with voltage stability, response speed, and energy consumption as objectives through a voltage outer loop control algorithm optimized by deep reinforcement learning. This specifically solves the problem of response lag or excessive energy consumption caused by single-objective optimization of traditional control algorithms. By optimizing control parameters in real time and generating carrier-level current loop control commands, it improves the voltage stability, response timeliness, and operating economy of the system under complex operating conditions.

[0033] For example, in the daily operation scenario of a 20MW / 40MWh grid-side energy storage power station, when the power station receives a grid dispatch command to discharge 10MW of power, the data receiving unit first receives preprocessed data from the data acquisition and processing module—including the energy storage battery pack terminal voltage (405V), converter output current (24kA), PCBA module temperature (42℃), and grid voltage (10.2kV). After ensuring that there are no data transmission errors through CRC verification, the data is cached in the local register to avoid packet loss caused by the mismatch between the data burst transmission and the processing rate of subsequent units, thus ensuring data integrity and processing timeliness; subsequently, the status analysis unit, based on the buffer... The system continuously monitors real-time data. When a sudden load change occurs in the power grid (a factory suddenly connects to the grid, adding 8MW of load, and the grid voltage drops instantaneously to 9.5kV), the unit identifies the "power grid anomaly" condition within 50ms and immediately feeds back the condition indicator to the algorithm calculation unit. This breaks the limitation of traditional fixed control strategies that require more than 100ms to identify changes in the condition. After receiving the "power grid anomaly" signal, the algorithm calculation unit uses a voltage outer loop control algorithm optimized by deep reinforcement learning to construct a multi-objective reward function with voltage stability, response speed, and energy consumption as objectives. It calculates and optimizes the voltage outer loop PI parameters in real time and finally generates a carrier-level current loop control command within 80ms.

[0034] In this embodiment of the application, the current inner loop drive control module 300 includes: Figure 4 As shown, there are a parsing unit, a feedback acquisition unit, an algorithm processing unit, and a PWM generation unit.

[0035] The instruction parsing unit is used to parse the carrier-level current loop control instruction and extract the current control parameters; the feedback acquisition unit is used to acquire the real-time current feedback data of the energy storage converter execution module; the algorithm processing unit performs calculations on the instruction parameters and feedback data through a repetitive control composite algorithm to calculate the pulse width modulation signal parameters; the PWM generation unit generates a pulse width modulation signal based on the pulse width modulation signal parameters through the SVPWM algorithm, and transmits it to the energy storage converter execution module after power amplification.

[0036] It should be noted that the repetitive control composite algorithm refers to a composite control method that combines repetitive control with other control strategies. Through collaborative computation, it processes the deviation between the command parameters and feedback data, achieving dual suppression of periodic interference and aperiodic errors, thereby accurately calculating the pulse width modulation signal parameters. The formula is as follows: ; ; ; in, For real-time deviation; This is the command current; For feedback current; This is the error after repeated control compensation; For delay operators; The control quantity output by the algorithm processing unit; This is the proportionality coefficient; The integral coefficient; This is the periodic compensation amount for repetitive control; Let k be the duty cycle parameter of the PWM signal at time k. For the switching cycle, This is the DC side voltage.

[0037] It is understood that, in this embodiment, the instruction parsing unit parses the carrier-level current loop control instruction and extracts the current control parameters, converting the high-frequency pulse instruction into a computable parameter form, ensuring that subsequent algorithm processing has a clear target basis; the feedback acquisition unit collects real-time current feedback data from the energy storage converter execution module to capture the deviation between the actual output and the instruction target in real time, breaking the limitation of traditional open-loop control that makes it difficult to perceive the execution effect, and avoiding control inaccuracies caused by the accumulation of execution deviations; the algorithm processing unit performs calculations on the instruction parameters and feedback data through a repetitive control composite algorithm, specifically solving the complex control problem of the coexistence of periodic interference and aperiodic errors, accurately calculating the PWM signal parameters, and providing a quantitative basis for signal generation; the PWM generation unit generates a pulse width modulation signal through the SVPWM algorithm and amplifies the power before transmitting it to the execution module, significantly improving the current control accuracy, dynamic response speed, and operating efficiency of the converter.

[0038] For example, in a 2MW / 4MWh energy storage converter system on the 10kV distribution network side of an industrial park, when the power grid dispatch center issues a carrier-level current loop control command of "valley charging 300A", the command parsing unit first receives the high-frequency pulse command and parses the current control parameters such as "target charging current 300A, dynamic adjustment coefficient 1.2". Simultaneously, it verifies and eliminates parameter errors caused by transmission interference, ensuring that the target parameters are accurately transmitted to the algorithm processing unit. Subsequently, the feedback acquisition unit collects the current feedback data at the converter output terminal in real time through a Hall current sensor at a sampling frequency of 10kHz. At the moment of charging start, it captures the deviation of only 285A in the actual current and synchronizes the data to the algorithm processing unit. The algorithm processing unit immediately starts the repetitive control composite algorithm, first calculating the 15A deviation between the 300A command parameter and the 285A feedback data, and then using the repetitive control module to compensate for the periodic error caused by the 50Hz frequency fluctuation of the power grid (compensation amount 2). A) Simultaneously, relying on the PI control component, a 20V control quantity is quickly output, ultimately calculating the parameter of a PWM signal duty cycle of 42%. After receiving this parameter, the PWM generation unit generates a 10kHz pulse signal through the SVPWM algorithm. After being amplified by the IGBT drive circuit (increasing the signal amplitude to 15V), it is transmitted to the converter execution module to drive the power switch. During operation, the feedback acquisition unit continuously monitors. When the park suddenly adds an 800kW inductive load, causing the current feedback to drop to 290A, the algorithm processing unit recalculates within 2ms, adjusting the PWM duty cycle to 43.5%. The PWM generation unit synchronously updates the signal, causing the actual current to quickly rise back to 300A. Throughout the process, the current control accuracy is maintained within ±2A, and the harmonic distortion rate is less than 3%. This not only meets the grid's charging requirements for the energy storage system but also avoids current instability caused by load fluctuations, fully demonstrating the control value under complex operating conditions.

[0039] In this embodiment of the application, the energy storage converter execution module 400 includes, as follows: Figure 5 As shown, there are a power conversion unit, an energy buffer unit, a filtering unit, and a state feedback unit.

[0040] The power conversion unit adopts a wide bandgap semiconductor device full-bridge topology and performs energy conversion under the control of pulse width modulation signal; the energy buffer unit absorbs voltage and current fluctuations during the conversion process through a dynamic energy buffer circuit; the filtering unit filters the converted electrical energy to reduce harmonic content; and the status feedback unit collects the converter operating parameters and feeds them back to the data acquisition and processing module.

[0041] It should be noted that energy conversion refers to the conversion of the system's DC power into AC power that is compatible with the power grid or load, or vice versa, to achieve AC to DC power conversion, thus completing the core energy form conversion during the charging and discharging process of the energy storage system.

[0042] Harmonic content refers to the proportion of noise components with frequencies that are integer multiples of the fundamental frequency generated by factors such as the switching action of power electronic devices and load nonlinearity during the conversion or transmission of electrical energy. These noise components can affect the quality of power and may cause problems such as equipment overheating and decreased accuracy.

[0043] It is understood that the power conversion unit in this application uses a wide-bandgap semiconductor full-bridge topology to achieve the AC / DC energy conversion required for charging and discharging under PWM signal control, providing the system with core energy scheduling capabilities; the energy buffer unit absorbs voltage and current fluctuations during conversion through dynamic circuits to avoid damage to devices and affect conversion stability; the filtering unit filters the converted power to reduce harmonic content to meet the power grid's power quality requirements and reduce noise interference to the load; the status feedback unit collects and feeds back the converter's operating parameters to provide data support for system strategy adjustment and fault early warning, ultimately forming a complete energy processing link to improve the converter's safety, power quality, and controllability.

[0044] For example, in a peak-valley charging and discharging scenario of a 3MW / 6MWh energy storage power station on a 10kV distribution network, when the grid is in a valley and the energy storage system needs charging, the power conversion unit adopts a full-bridge topology composed of SiC wide-bandgap semiconductors. Under the control of a PWM signal (10kHz carrier, 52% duty cycle), it rectifies the 10kV AC power of the grid into 750V DC power to provide charging energy for the battery pack. Its wide-bandgap device characteristics reduce switching losses, improving the conversion efficiency by 3% compared to traditional silicon-based devices. During charging, due to instantaneous fluctuations in the grid voltage (dropping from 10.1kV to 9.8kV), a 20V voltage spike appears at the output of the power conversion unit. The energy buffer unit quickly absorbs this fluctuation through a built-in supercapacitor buffer circuit, preventing the voltage spike from impacting the battery pack. The converted DC power contains a small amount of high-frequency harmonics (mainly 20kHz harmonic components). The filtering unit filters out the harmonics through an LC low-pass filter circuit, reducing the total harmonic distortion (THD) of the output DC power from 8% to 2.5%, which meets the battery pack's requirements for charging power quality. At the same time, the status feedback unit collects the output voltage (750V), charging current (400A), and device temperature (58℃) of the power conversion unit in real time through voltage and current sensors, and transmits the data to the data acquisition and processing module. When the charging current is detected to rise to 420A (5% above the rated value), the module promptly adjusts the PWM signal duty cycle to 49% to reduce the charging current through the power conversion unit, ensuring the safety, efficiency, and power quality of the energy storage system during the entire charging process.

[0045] In this embodiment, the multi-module carrier synchronization module 500 includes, as follows: Figure 6As shown, there are a data interaction unit, an algorithm fusion unit, an instruction generation unit, and a synchronous execution unit.

[0046] The data interaction unit receives the operating data, carrier phase information, and system full-process data of each PCBA module through a high-speed fiber optic bus; the algorithm fusion unit uses a distributed synchronization algorithm to perform fusion calculations on the received data to determine the carrier synchronization deviation of each module; the instruction generation unit generates carrier phase adjustment instructions based on the synchronization deviation; and the synchronization execution unit distributes the adjustment instructions to each PCBA module to perform multi-module carrier phase-frequency control.

[0047] It should be noted that the distributed synchronization algorithm refers to a collaborative control algorithm that dynamically calculates and corrects the carrier frequency and phase deviation between modules by rapidly exchanging the operating data and carrier phase information of each module, and finally achieves the same frequency and phase of the carrier signals of all modules.

[0048] It is understood that, in this embodiment, the data interaction unit receives the operating data, carrier phase information, and full-process data of each PCBA module through a high-speed fiber optic bus, relying on the high bandwidth and low latency characteristics to ensure data real-time performance and provide complete data for deviation calculation; the algorithm fusion unit uses a distributed synchronization algorithm to fuse data and determine the carrier synchronization deviation, and reduces errors through multi-module data verification to ensure accurate deviation judgment; the instruction generation unit generates phase adjustment instructions based on the synchronization deviation, converting the deviation data into executable instructions, providing a clear basis for synchronization execution; the synchronization execution unit distributes the adjustment instructions to each module to achieve in-frequency and in-phase control, avoiding new deviations caused by adjustment timing differences, and improving the carrier synchronization accuracy of multiple modules and the stability of system operation.

[0049] For example, in a 5MW-level energy storage converter with 8 modules in parallel, the operation process of the multi-module carrier synchronization module is as follows: After system startup, the data interaction unit synchronously receives three types of data from each PCBA module via a 1.25Gbps high-speed fiber optic bus: operating data (e.g., module 2 output current 320A, module 5 IGBT temperature 65℃), carrier phase information (module 1 phase leading 3.2°, module 7 lagging 2.8°), and system full-process data. The algorithm fusion unit starts the distributed synchronization algorithm to fuse and calculate the three types of data: using the grid voltage phase as a reference to correct the carrier reference point, combining the harmonic distortion rate to determine the initial synchronization quality, referring to the IGBT temperature data to compensate for the temperature drift effect, and after 3 rounds of interactive verification to eliminate the 2.1° false deviation caused by the communication delay of module 4, the actual synchronization deviation of each module is finally determined (module 1 leading 1.8°, module 7 lagging 1.5°, etc.). The instruction generation unit generates quantized adjustment instructions (module 1 delayed by 50ns, module 7 advanced by 42ns) based on the deviation value and module hardware parameters (such as phase-locked loop response speed) in the full-process data. The synchronous execution unit distributes the instructions in parallel to each module via the fiber optic bus, triggering the internal carrier generator to adjust the timing. The entire process is completed within 100μs. During operation, module 3 experiences a phase drift of 0.6° due to a sudden load change. The data interaction unit captures the change within 3 sampling periods using full-process data (2A circulating current from the inner current loop feedback, and 0.5% fluctuation in the output voltage of the execution module). The algorithm fusion unit recalculates the deviation, and the instruction generation and execution units respond quickly, maintaining the system carrier synchronization accuracy within ±0.1°. This ensures that the inter-module circulating current is ≤2A when the converter is running at full power, meeting the grid connection standards.

[0050] The multi-module carrier synchronization control system for power storage waveborne PCBA proposed in this application utilizes a multi-sensor collaborative data acquisition and processing module. Combined with an adaptive filtering algorithm, it performs amplitude normalization and high-frequency interference suppression, effectively addressing grid harmonics and temperature drift interference, and improving parameter acquisition accuracy and data reliability under complex operating conditions. The system leverages deep reinforcement learning in the central control and decision-making module to optimize the voltage outer loop control algorithm, constructing a multi-objective reward function. This allows for real-time optimization of control parameters under scenarios such as load surges and grid anomalies, enhancing the adaptability and accuracy of control commands, improving system voltage stability and response speed, and reducing energy consumption. Finally, the system employs a repetitive control composite algorithm in the current inner loop drive control module, combined with... By analyzing and processing real-time current feedback data, a high-precision pulse-width modulation (PWM) signal is generated, enhancing current control accuracy and providing precise drive support for energy conversion. Through the wide-bandgap semiconductor device full-bridge topology and dynamic energy buffer circuit of the energy storage converter execution module, combined with high-precision PWM signal control, efficient bidirectional energy conversion between the energy storage battery and the grid is achieved, absorbing voltage and current fluctuations and reducing harmonic content, thus improving power quality and energy conversion efficiency. Utilizing the distributed synchronization algorithm of the multi-module carrier synchronization module and a high-speed fiber optic bus, efficient interaction and fusion correction of carrier phase information from each PCBA module are performed, eliminating carrier phase deviations between modules, enhancing multi-module parallel coordination, and improving operational stability. This solves the problems of insufficient data accuracy, poor algorithm adaptability, and difficulty in multi-module synchronization in existing technologies.

[0051] The following will illustrate the multi-module carrier synchronization control system for power storage waveborne PCBA through a specific embodiment, including: In a 5MW / 10MWh power storage system on the 10kV distribution network side of an industrial park, the data acquisition and processing module was the first to start working. At 6:00 AM, the system switched from standby mode to operating mode. The parameter acquisition unit immediately acquired the battery pack terminal voltage of 750V through the Hall voltage sensor installed at the output of the energy storage battery pack, and simultaneously obtained the initial charging and discharging current of 0A using the current sensor connected in series in the charging and discharging circuit. Temperature sensors distributed on the six PCBA modules monitored the operating temperature of each module in real time between 25-28℃, while the grid monitoring unit captured the grid-side voltage as 10.5kV, frequency 50Hz, and phase 0°. After these raw data were transmitted to the signal preprocessing unit, the adaptive filtering algorithm was quickly activated to suppress the 150Hz high-frequency noise in the voltage signal caused by cable interference, normalize the amplitude of the current signal to the 0-10V standard range, and the temperature data was also linearly corrected to eliminate sensor nonlinearity errors. After processing, the data distribution unit synchronously pushes data such as 750V battery voltage, 0A charging / discharging current, temperature values ​​of each module, 10.5kV grid voltage, 50Hz frequency, and 0° phase to the central control and decision-making module and the multi-module carrier synchronization module via the internal high-speed bus, providing basic data support for subsequent control decisions. As the system operates, the power load in the industrial park increases at 9:00 AM. The parameter acquisition unit detects a brief drop in grid voltage to 9.8kV and frequency fluctuations to 49.8Hz. These changes, after preprocessing, are also distributed in real time to ensure that each control module can promptly detect changes in the grid status.

[0052] After receiving the initial data from the data acquisition and processing module, the central control and decision-making module first verifies data such as the 750V battery voltage and 0A current. Data integrity is confirmed through CRC redundancy check, and then cached in a local data buffer to prevent data loss due to subsequent processing delays. The status analysis unit analyzes the cached data and, combined with parameters such as the grid voltage of 10.5kV and frequency of 50Hz, determines that the system is in normal operating condition, where the load is stable and there are no significant fluctuations. The algorithm calculation unit then initiates a voltage outer-loop control algorithm optimized by deep reinforcement learning, constructing a multi-objective reward function with voltage stability, response speed, and energy consumption as objectives. Considering the current low battery level, the algorithm prioritizes optimizing charging efficiency while ensuring that grid voltage fluctuations are within acceptable limits. After 10ms of iterative calculation, a carrier-level current loop control command is generated, requiring the energy storage converter execution module to absorb power from the grid at a current of 300A to charge the battery pack. At noon, a 2MW inductive load suddenly connected to the industrial park, causing the grid voltage to drop to 9.2kV and the frequency to fluctuate to 49.5Hz. The data receiving unit quickly verified and buffered this change data, the state analysis unit immediately identified the sudden load change, and the algorithm calculation unit quickly adjusted the weight of the multi-objective reward function to prioritize voltage stability and response speed. It then regenerated the adjusted carrier-level current loop control command to temporarily reduce the charging current to 200A to avoid increasing the burden on the grid.

[0053] Upon receiving the 300A charging current command from the central control and decision-making module, the inner loop drive control module immediately parses the carrier-level current loop control command, extracting current control parameters, including key information such as the target current of 300A, dynamic adjustment coefficient of 1.2, and response time threshold of 5ms. The command is then validated to ensure its accuracy. Simultaneously, the feedback acquisition unit starts, acquiring real-time current feedback data from the energy storage converter's execution module output. At the moment of charging startup, the actual current is captured as 285A, deviating from the target current by 15A. This data is continuously transmitted to the algorithm processing unit. The algorithm processing unit uses a repetitive control composite algorithm to calculate the 300A command parameters and the 285A feedback data. It first calculates the basic deviation of 15A, then compensates for the periodic error of 2A caused by the 50Hz frequency fluctuation of the power grid through the repetitive control module. Simultaneously, it rapidly outputs a 20V control quantity using a PI control component, ultimately calculating the duty cycle of the pulse width modulation signal to be 42%. After receiving the parameter, the PWM generation unit generates a 10kHz pulse signal using the SVPWM algorithm. This signal is then amplified to 15V by the IGBT drive circuit and transmitted to the power conversion unit of the energy storage converter execution module. When a sudden load change at noon causes the central control and decision module to adjust the charging current to 200A, the instruction parsing unit quickly parses the new instruction, extracting parameters such as the target current of 200A. The feedback acquisition unit monitors in real time that the actual current fluctuates with the grid to 190A. The algorithm processing unit then recalculates, adjusting the PWM signal duty cycle to 28% to ensure the actual current quickly stabilizes at 200A.

[0054] Upon receiving the 10kHz pulse signal from the inner current loop drive control module, the full-bridge topology composed of wide-bandgap semiconductor devices (SiCMOSFETs) in the power conversion unit immediately responds. Under the control of the pulse signal, it rectifies the 10kV AC input from the grid into 750V DC to charge the energy storage battery pack. During the conversion process, due to grid voltage fluctuations and semiconductor device switching, a 20V voltage spike and a 5A current fluctuation appear at the output of the power conversion unit. The energy buffer unit quickly absorbs these fluctuations through its built-in supercapacitor dynamic energy buffer circuit, preventing the voltage and current spikes from impacting the battery pack and other components. The converted DC contains a small amount of high-frequency harmonics, mainly 20kHz harmonic components. The LC low-pass filter circuit in the filtering unit then filters these components, reducing the total harmonic distortion rate of the output DC from 8% to 2.5%, meeting the battery pack's requirements for charging power quality. The status feedback unit collects real-time operating parameters such as the power conversion unit's output voltage of 750V, charging current of 300A, and SiC MOSFET junction temperature of 58℃, and feeds this data back to the data acquisition and processing module, forming a data closed loop. When the system adjusts to 200A charging, the power conversion unit adjusts its conversion efficiency accordingly, the energy buffer unit continues to absorb any fluctuations, the filtering unit maintains low harmonic output, and the status feedback unit continuously updates and feeds back the operating parameters.

[0055] The multi-module carrier synchronization module begins operation at system startup. The data interaction unit synchronously receives operational data from six PCBA modules via a 1.25Gbps high-speed fiber optic bus (e.g., module 2 output current 300A, module 5 SiCMOSFET temperature 58℃), carrier phase information (initially detected as module 1 leading 3.2°, module 6 lagging 2.8°), and full-process system data (including grid voltage phase of 0° from the data acquisition and processing module, 10kHz carrier reference command from the central control and decision-making module, switching timing parameters from the current inner loop drive control module, and harmonic distortion rate of 2.5% from the energy storage converter execution module). The algorithm fusion unit initiates a distributed synchronization algorithm to fuse and calculate this data. Using the grid voltage phase as a reference, the carrier reference point is corrected. The initial synchronization quality is determined by combining the harmonic distortion rate with the data from each module, and the temperature drift effect is compensated by referring to the temperature data of each module. After three rounds of interactive verification, a 2.1° false deviation caused by sensor error in module 3 is eliminated, ultimately determining the actual synchronization deviation of each module (module 1 leading 1.8°, module 6 lagging 1.5°, etc.). The instruction generation unit generates quantized adjustment instructions (module 1 delayed by 50ns, module 6 advanced by 42ns) based on the deviation value and module hardware parameters (such as phase-locked loop response speed) in the full-process data. The synchronous execution unit distributes the instructions in parallel to each module via the fiber optic bus, triggering the internal carrier generator to adjust the timing. The entire process is completed within 100μs. During operation, at 3 PM, module 4 experienced a temperature rise to 65℃ due to a cooling fan failure, causing a phase drift of 0.6°. The data interaction unit detected this change within 3 sampling cycles (30μs) through the full-process data (2A circulating current increment from the inner current loop feedback, and the harmonic content of the execution module rising to 3.0%). The algorithm fusion unit recalculates the deviation, and the instruction generation and execution units respond quickly, ensuring that the system carrier synchronization accuracy remains within ±0.1°, guaranteeing that the circulating current between modules is ≤2A, and ensuring the stable and efficient operation of the entire energy storage system under complex operating conditions. At 6 PM, the power load in the industrial park decreased, the power grid returned to stability, and the central control and decision-making module adjusted the command to continue charging the battery pack with a current of 250A. All modules worked together, the data acquisition and processing module updated the data in real time, the current inner loop drive control module precisely adjusted the PWM signal, the energy storage converter execution module efficiently converted energy, and the multi-module carrier synchronization module kept the carriers of all modules synchronized, working together to complete the energy storage task.

[0056] In summary, this embodiment of the application forms a complete energy storage system control link through the coordinated operation of a data acquisition and processing module, a central control and decision-making module, a current inner loop drive control module, an energy storage converter execution module, and a multi-module carrier synchronization module. Each module has a clear division of labor and deep linkage, sensing real-time changes in system operating status and grid conditions. Based on algorithm optimization, it generates precise control commands to ensure high energy conversion efficiency and power quality compliance. Simultaneously, multi-module carrier synchronization eliminates phase deviations and suppresses circulating currents, preventing fluctuations in operating conditions from affecting system stability. Ultimately, this achieves safe, stable, and economical operation of the energy storage system under complex operating conditions, enhancing its core value in terms of reliability, adaptability, and operational efficiency.

[0057] Next, referring to the accompanying drawings, a method for multi-module carrier synchronization control of power storage waveborne PCBA proposed according to an embodiment of this application is described.

[0058] like Figure 7 As shown, the multi-module carrier synchronization control method for power storage waveborne PCBA includes the following steps: In step S101, the voltage, current, carrier frequency, phase difference, and operating temperature parameters of each module are acquired.

[0059] It is understood that the embodiments of this application obtain voltage, current, carrier frequency, phase difference and operating temperature parameter data of each module to provide comprehensive original basis for subsequent filtering processing, algorithm optimization and instruction generation, avoid one-sided control decisions due to data loss, and provide support for the safe and stable operation of the system.

[0060] In step S102, noise is filtered from the voltage, current, carrier frequency, phase difference, and operating temperature parameter data of each module using an adaptive filtering algorithm. Combined with a voltage outer loop control algorithm optimized by deep reinforcement learning, a multi-objective reward function is constructed to optimize the control parameters in real time, generating a carrier-level current loop control command.

[0061] Among them, the carrier-level current loop control command is a high-frequency, precise command that matches the carrier frequency and is based on the inner current loop control of the power storage converter. It is used to clarify the target value of current regulation, dynamic response parameters, and timing requirements for synchronization with the carrier, providing the core basis for generating pulse width modulation signals.

[0062] It is understood that the embodiments of this application transform the top-level optimized control objective into specific parameters that match the carrier frequency, ensuring that the pulse width modulation signal generated by the current inner loop drive control module accurately adapts to the power conversion requirements, improving the real-time performance of current regulation, providing a unified current control benchmark for multi-module carrier synchronization, avoiding current fluctuations or synchronization deviations caused by ambiguous commands, and ensuring the stability of energy conversion and power quality.

[0063] For example, in a 5MW energy storage converter system, when the grid voltage fluctuates by 10%, the data acquisition module obtains the voltage (750±30V), current (300±15A), 10kHz carrier frequency, 2° phase difference, and temperature data (55-62℃) of each PCBA module. This raw data is processed by an adaptive filtering algorithm to remove 150Hz high-frequency noise, narrowing the voltage fluctuation range to 750±5V and stabilizing the current at 300±3A, providing a clean data source for subsequent calculations. The central control module calls a voltage outer-loop algorithm optimized by deep reinforcement learning to construct a multi-objective reward function based on voltage stability (weight 40%), response speed (30%), and energy consumption (30%). The algorithm iterates in real time using filtered data, completing 100 parameter optimizations within 20ms, and ultimately generating a carrier-level current loop control command: a target current of 302A, a dynamic adjustment coefficient of 1.1, a trigger timing synchronized with a 10kHz carrier (updated at 10μs per cycle), and a derating threshold (current drops to 280A) when the temperature exceeds 65℃. This command precisely connects top-level decision-making with bottom-level execution, providing a clear benchmark for generating PWM signals in the inner current loop.

[0064] In step S103, based on the carrier-level current loop control command, the current control parameters are extracted, and combined with the real-time current feedback data, the pulse width modulation signal parameters are calculated through the repetitive control composite algorithm, and the pulse width modulation signal is generated using the SVPWM algorithm.

[0065] Among them, extracting current control parameters based on carrier-level current loop control commands means first performing format verification and protocol parsing on the commands, separating the data segments from the command frames, and then extracting the target current value, dynamic adjustment coefficient, carrier synchronization timing, and overcurrent or overheat protection thresholds according to the preset parameter mapping relationship. After data verification, a standardized set of current control parameters is output.

[0066] It should be noted that the SVPWM algorithm is a control method that generates pulse width modulation signals in a three-phase inverter circuit by synthesizing a space voltage vector, thereby precisely controlling the amplitude and phase of the output voltage.

[0067] It is understood that the embodiments of this application convert the carrier-level current loop control command into standardized, calculable current control parameters to ensure that the control target is not deviated or lost. Through space voltage vector synthesis, the parameters are converted into PWM signals adapted to power devices to accurately regulate the amplitude and phase of the output voltage. The two work together to ensure the real-time performance and accuracy of the current regulation of the energy storage converter, avoid power quality problems caused by signal conversion deviations, adapt to the high-frequency switching characteristics of wide bandgap semiconductor devices, improve energy conversion efficiency, and lay a stable driving signal foundation for multi-module carrier synchronization, thus helping the entire energy storage system to operate safely and efficiently under complex conditions.

[0068] In step S104, bidirectional energy conversion between the energy storage battery and the power grid is performed based on the pulse width modulation signal to obtain converter operating status feedback data. Combined with the full process data, the carrier synchronization deviation of each module is calculated using a distributed synchronization algorithm to generate carrier phase adjustment commands and perform multi-module carrier frequency and phase control.

[0069] It is understood that the embodiments of this application adapt to changes in grid supply and demand through bidirectional energy conversion, improve the energy dispatch flexibility and economy of the energy storage system, eliminate carrier phase deviation between modules through real-time feedback and synchronous control, avoid circulating current and harmonic exceedance problems caused by asynchrony, reduce power device losses, improve the operational reliability and power quality of the energy storage converter, and provide support for the safe and efficient grid connection of high-power energy storage systems.

[0070] For example, in a 5MW energy storage system on the 10kV distribution network side of an industrial park, when the grid load surges during the midday peak electricity demand and the voltage drops to 9.2kV, the pulse width modulation signal (10kHz, duty cycle 52%) generated by the current inner loop drive control module drives the SiCMOSFET full-bridge topology operation of the energy storage converter execution module. This inverts the 750V DC power from the energy storage battery pack into 10kV AC power and injects it into the grid, realizing energy conversion from the battery to the grid and alleviating the power supply pressure on the grid. Simultaneously, the converter status feedback unit collects real-time operating data: output current 480A, AC side voltage 9.8kV, IGBT junction temperature 62℃, and uploads this data to the data acquisition and processing module. The multi-module carrier synchronization module, combining the full-process data (including grid frequency 49.8Hz and initial phase difference of 1.5° between each module's carriers), calculates through a distributed synchronization algorithm that module 2 is 0.9° ahead and module 5 is 0.7° behind. Subsequently, adjustment instructions are generated (module 2 delayed by 25ns, module 5 advanced by 20ns). The synchronous execution unit issues the instructions and completes the carrier timing adjustment of each module within 100μs, ensuring that the carriers of the six PCBA modules are in phase and frequency. At this time, the circulating current between modules drops from 8A to below 2A, ensuring stable and efficient energy conversion and helping the power grid to smoothly pass through peak electricity demand.

[0071] The multi-module carrier synchronization control method for power storage carrier PCBA proposed in this application utilizes a multi-sensor collaborative data acquisition and processing module. Combined with an adaptive filtering algorithm, amplitude normalization and high-frequency interference suppression are performed to effectively address grid harmonics and temperature drift interference, improving parameter acquisition accuracy and data reliability under complex operating conditions. The deep reinforcement learning optimization algorithm of the central control and decision module optimizes the voltage outer loop control algorithm, constructing a multi-objective reward function to optimize control parameters in real time under scenarios such as load mutations and grid anomalies, enhancing the adaptability and accuracy of control commands, improving system voltage stability and response speed, and reducing energy consumption. Furthermore, the repetitive control composite algorithm of the current inner loop drive control module, combined with… Command parsing and real-time current feedback data processing generate high-precision pulse-width modulation (PWM) signals, enhancing current control accuracy and providing precise drive support for energy conversion. Through the wide-bandgap semiconductor device full-bridge topology and dynamic energy buffer circuit of the energy storage converter execution module, combined with high-precision PWM signal control, efficient bidirectional energy conversion between the energy storage battery and the grid is achieved, absorbing voltage and current fluctuations and reducing harmonic content, thus improving power quality and energy conversion efficiency. Utilizing the distributed synchronization algorithm of the multi-module carrier synchronization module and a high-speed fiber optic bus, efficient interaction and fusion correction of carrier phase information from each PCBA module are performed, eliminating carrier phase deviations between modules, enhancing multi-module parallel coordination, and improving operational stability. This solves the problems of insufficient data accuracy, poor algorithm adaptability, and difficulty in multi-module synchronization in existing technologies.

[0072] The following will illustrate the multi-module carrier synchronization control method for power storage waveborne PCBA through a specific embodiment, including: In an industrial park energy storage system, the primary task upon switching from standby mode to operation is to acquire key parameter data. At this time, the voltage detection element installed at the output of the energy storage battery pack begins operation, capturing the battery pack's terminal voltage in real time. The initial detection value is 750V, and as the system gradually activates, the voltage stabilizes within the range of 750±2V. The current detection element connected in series in the charging and discharging circuit starts simultaneously. Since the system has just started, the initial charging and discharging current is 0A, subsequently rising slowly to 50A during the pre-charging phase. The device used to monitor carrier characteristics continuously acquires the carrier frequency, consistently maintaining a stable 10kHz, while simultaneously recording the carrier phase difference between modules. Initially, a slight deviation of ±1.2° was detected in some phases. Temperature detection elements distributed in key parts of the energy storage converter provide real-time feedback on the operating temperature. Influenced by ambient temperature, the initial temperature is between 25-28℃, and as the elements gradually heat up, the temperature slowly rises to around 30℃. These parameters cover the core status information of the energy storage system, including the voltage and current of the energy storage end, the carrier characteristics of the control signal, and the operating temperature of the equipment. They provide comprehensive and real-time raw data support for subsequent control decisions, ensuring that the system can carry out subsequent operations based on the actual operating status.

[0073] After acquiring complete parameter data, the system initiates the data processing and control command generation process. First, an adaptive filtering algorithm is used to filter noise from the collected voltage, current, carrier frequency, phase difference, and operating temperature parameters. For voltage data, the algorithm effectively filters out 150Hz high-frequency noise caused by line interference, narrowing the voltage fluctuation range from 750±2V to 750±0.5V. Instantaneous pulse interference in the current data is successfully eliminated, resulting in a smoother current change curve with no abnormal fluctuations during the rise from 0A to 50A. After filtering, the stability of the 10kHz carrier frequency value is further improved, with frequency fluctuations controlled within ±0.1Hz. Random interference signals in the phase difference data are filtered out, improving the detection accuracy of phase deviation and more accurately reflecting the actual phase relationship. Filtering of the operating temperature data eliminates interference from instantaneous changes in ambient temperature, making the temperature feedback more consistent with the actual heat generation of the equipment. After noise filtering, the system combines a voltage outer-loop control algorithm optimized by deep reinforcement learning to construct a multi-objective reward function. The reward function focuses on voltage stability, response speed, and energy consumption as its core objectives. Voltage stability is weighted at 40% to ensure that battery and grid voltages remain within reasonable ranges; response speed is weighted at 30% to ensure the system can react quickly to changes in operating conditions; and energy consumption is weighted at 30% to achieve economical system operation. By analyzing filtered parameter data in real time, the algorithm iteratively optimizes control parameters. After a 20ms calculation process, it finally generates a carrier-level current loop control command. This command specifies key control information such as a target current of 300A, a dynamic adjustment coefficient of 1.1, carrier synchronization timing requirements, and a current derating threshold when the temperature exceeds 65℃. This provides precise control information for subsequent pulse width modulation signal generation.

[0074] Based on the generated carrier-level current loop control command, the system enters the pulse width modulation (PWM) signal generation stage. First, key current control parameters are extracted from the carrier-level current loop control command, including the target current of 300A, dynamic adjustment coefficient of 1.1, carrier synchronization timing parameters, and overcurrent protection threshold. During extraction, the system rigorously verifies the validity of the parameters to ensure each parameter is within a reasonable range, avoiding subsequent control errors due to parameter anomalies. Subsequently, the system performs calculations based on real-time current feedback data. At this point, the real-time current feedback value is 280A, deviating from the target current of 300A by 20A. To address this deviation, the system uses a repetitive control composite algorithm. This algorithm first calculates the basic deviation compensation amount and, considering the impact of grid frequency fluctuations and load changes on the current, introduces a periodic error compensation term, effectively offsetting the ±2A current fluctuation caused by the 50Hz grid frequency fluctuation. Through algorithm calculation, the duty cycle parameter of the PWM signal is finally determined to be 45%. After obtaining the duty cycle parameter, the system uses the SVPWM algorithm to generate the PWM signal. The SVPWM algorithm synthesizes a space voltage vector, determines the sector of the voltage vector based on current current control requirements and grid conditions, calculates the duration of adjacent effective vectors and the allocation of zero vectors, and ultimately generates a pulse width modulation signal synchronized with a 10kHz carrier frequency. This signal has precise pulse timing and duty cycle, enabling accurate switching of power devices and providing a reliable drive signal for energy conversion in the energy storage converter, ensuring stable current output according to the target value.

[0075] After generating a pulse width modulation (PWM) signal, the system performs bidirectional energy conversion between the energy storage battery and the grid, and implements multi-module carrier synchronous frequency and phase control. During off-peak hours, when grid load is low and electricity prices are relatively cheap, the system, following the PWM signal's instructions, controls the energy storage converter to rectify the grid's AC power into DC power to charge the energy storage battery pack. During charging, the PWM signal precisely controls the switching on and off of power switching devices, stabilizing the rectified DC voltage at 750V and gradually increasing the current from 50A to 300A, achieving efficient charging. During peak hours, when the industrial park's electricity load surges and the grid voltage declines, the system switches to discharge mode. The PWM signal controls the energy storage converter to invert the DC power from the energy storage battery pack into AC power synchronized with the grid, injecting it into the grid to alleviate power supply pressure. During bidirectional energy conversion, the system acquires real-time feedback data on the converter's operating status, including output voltage, current, power factor, and device temperature. For example, during the discharge phase, the output voltage stabilizes at 10kV, the current dynamically adjusts between 200-400A according to grid demand, the power factor remains above 0.98, and the device temperature is controlled below 60℃. Simultaneously, the system combines data from the entire process, including previously acquired parameter data, filtered and processed data, and control command information, and uses a distributed synchronization algorithm to calculate the carrier synchronization deviation of each module. Algorithm analysis reveals a phase deviation of ±0.8° in some modules. Based on the calculation results, the system generates carrier phase adjustment commands to correct the phase of modules with deviations. The adjustment commands are rapidly transmitted via a high-speed communication link, and each module adjusts its carrier timing in real-time according to the commands, completing phase correction within 100μs, achieving multi-module carrier synchronous frequency and phase control.

[0076] In summary, this application's embodiments provide a high-quality data foundation for subsequent control decisions by accurately collecting multi-dimensional operating parameters and optimizing them through adaptive filtering. Combined with a multi-objective reward function constructed using deep reinforcement learning, it can generate carrier-level commands adapted to the operating conditions, ensuring that current regulation simultaneously meets the requirements of voltage stability, energy consumption optimization, and response speed. Relying on the pulse signals generated by the repetitive control composite algorithm and the SVPWM algorithm, efficient bidirectional energy conversion between the energy storage battery and the grid can be achieved. Furthermore, a distributed synchronization algorithm eliminates module carrier deviation, effectively suppressing circulating current and reducing device losses. The overall process not only ensures the stable operation of the energy storage system during off-peak charging and peak discharging, but also improves power quality and energy conversion efficiency, while providing strong support for system reliability under complex operating conditions, accurately adapting to the energy dispatching needs of the industrial park's distribution network.

[0077] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 801, the processor 802, and the computer program stored on the memory 801 and capable of running on the processor 802.

[0078] When the processor 802 executes the program, it implements the multi-module carrier synchronization control method for power storage waveborne PCBA provided in the above embodiments.

[0079] Furthermore, electronic devices also include: Communication interface 803 is used for communication between memory 801 and processor 802.

[0080] The memory 801 is used to store computer programs that can run on the processor 802.

[0081] The memory 801 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0082] If the memory 801, processor 802, and communication interface 803 are implemented independently, then the communication interface 803, memory 801, and processor 802 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 8 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0083] Optionally, in a specific implementation, if the memory 801, processor 802, and communication interface 803 are integrated on a single chip, then the memory 801, processor 802, and communication interface 803 can communicate with each other through an internal interface.

[0084] The processor 802 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0085] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described power storage waveborne PCBA multi-module carrier synchronization control method.

[0086] Furthermore, this application also provides a computer program product, including a computer program or instructions, which, when executed, implement the above-mentioned power storage wavecarrier PCBA multi-module carrier synchronization control method.

[0087] In the description of this specification, the references to "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0088] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0089] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0090] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0091] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0092] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A multi-module carrier synchronization control system for power storage waveborne PCBA, characterized in that, include: The module comprises a data acquisition and processing module, a central control and decision-making module, a current inner loop drive control module, an energy storage converter execution module, and a multi-module carrier synchronization module; among which, The data acquisition and processing module is used to collect voltage, current, temperature and grid parameter data during the operation of the power energy storage system; The central control and decision-making module is used to receive the data, construct a multi-objective reward function to optimize control parameters in real time through a voltage outer loop control algorithm optimized by deep reinforcement learning, and generate carrier-level current loop control commands. The current inner loop drive control module receives the carrier-level current loop control command, combines it with real-time current feedback data, and generates a pulse width modulation signal through a repetitive control composite algorithm and an SVPWM algorithm. The energy storage converter execution module performs bidirectional energy conversion between the energy storage battery and the grid based on the pulse width modulation signal, using a wide bandgap semiconductor device full-bridge topology and a dynamic energy buffer circuit. The multi-module carrier synchronization module is based on full-process data and utilizes a distributed synchronization algorithm to perform carrier phase information interaction and fusion correction between multiple PCBA modules through a high-speed optical fiber bus, so as to achieve carrier synchronization of multiple PCBA modules.

2. The power storage carrier PCBA multi-module carrier synchronization control system according to claim 1, characterized in that, The data acquisition and processing module includes a parameter acquisition unit, a signal preprocessing unit, and a data distribution unit. The parameter acquisition unit acquires the terminal voltage and charging / discharging current of the energy storage battery pack, the input and output current and voltage of the energy storage converter, the operating temperature of each PCBA module, and the voltage, frequency, and phase signals of the grid side through voltage sensors, current sensors, temperature sensors, and a grid monitoring unit. The signal preprocessing unit performs amplitude normalization and high-frequency interference suppression on the acquired data using an adaptive filtering algorithm. The data distribution unit synchronously transmits the processed data to the central control and decision-making module and the multi-module carrier synchronization module.

3. The power storage carrier PCBA multi-module carrier synchronization control system according to claim 1, characterized in that, The central control and decision-making module includes a data receiving unit, a status analysis unit, and an algorithm calculation unit. The data receiving unit receives the processed data for verification and buffering. The status analysis unit determines the system's operating status based on the received data, including normal operation, load fluctuations, and grid anomalies. The algorithm calculation unit uses a voltage outer loop control algorithm optimized through deep reinforcement learning to construct a multi-objective reward function targeting voltage stability, response speed, and energy consumption, optimizes control parameters in real time, and generates carrier-level current loop control commands.

4. The power storage carrier PCBA multi-module carrier synchronization control system according to claim 1, characterized in that, The current inner loop drive control module includes an instruction parsing unit, a feedback acquisition unit, an algorithm processing unit, and a PWM generation unit. The instruction parsing unit parses the carrier-level current loop control instruction and extracts the current control parameters. The feedback acquisition unit acquires real-time current feedback data from the energy storage converter execution module. The algorithm processing unit performs calculations on the instruction parameters and feedback data using a repetitive control composite algorithm to calculate the pulse width modulation signal parameters. The PWM generation unit generates a pulse width modulation signal based on the pulse width modulation signal parameters using the SVPWM algorithm, amplifies the signal, and transmits it to the energy storage converter execution module.

5. The power storage carrier PCBA multi-module carrier synchronization control system according to claim 1, characterized in that, The energy storage converter execution module includes a power conversion unit, an energy buffer unit, a filtering unit, and a status feedback unit. The power conversion unit employs a wide-bandgap semiconductor device full-bridge topology and performs energy conversion under pulse width modulation signal control. The energy buffer unit absorbs voltage and current fluctuations during the conversion process through a dynamic energy buffer circuit. The filtering unit filters the converted electrical energy to reduce harmonic content. The status feedback unit collects converter operating parameters and feeds them back to the data acquisition and processing module.

6. The power storage carrier PCBA multi-module carrier synchronization control system according to claim 1, characterized in that, The multi-module carrier synchronization module includes a data interaction unit, an algorithm fusion unit, an instruction generation unit, and a synchronization execution unit. The data interaction unit receives operating data, carrier phase information, and system-wide process data from each PCBA module via a high-speed fiber optic bus. The algorithm fusion unit uses a distributed synchronization algorithm to perform fusion calculations on the received data to determine the carrier synchronization deviation of each module. The instruction generation unit generates carrier phase adjustment instructions based on the synchronization deviation. The synchronization execution unit distributes the adjustment instructions to each PCBA module for multi-module carrier phase-frequency control.

7. A method for multi-module carrier synchronization control of a power storage waveborne PCBA applied to any one of claims 1-6, characterized in that, include: Acquire the voltage, current, carrier frequency, phase difference, and operating temperature parameters of each module; The voltage, current, carrier frequency, phase difference, and operating temperature parameter data of each module are filtered for noise using an adaptive filtering algorithm. Combined with a voltage outer loop control algorithm optimized by deep reinforcement learning, a multi-objective reward function is constructed to optimize the control parameters in real time and generate carrier-level current loop control commands. Based on the carrier-level current loop control command, the current control parameters are extracted, and combined with real-time current feedback data, the pulse width modulation signal parameters are calculated through a repetitive control composite algorithm, and the pulse width modulation signal is generated using the SVPWM algorithm. Based on the pulse width modulation signal, bidirectional energy conversion is performed between the energy storage battery and the power grid to obtain converter operating status feedback data. Combined with the full-process data, the carrier synchronization deviation of each module is calculated using a distributed synchronization algorithm to generate carrier phase adjustment commands and perform multi-module carrier frequency and phase control.

8. An electronic device, characterized in that, include: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the power storage waveborne PCBA multi-module carrier synchronization control method of claim 7.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed, they implement the power storage waveborne PCBA multi-module carrier synchronization control method of claim 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed, they implement the power storage waveborne PCBA multi-module carrier synchronization control method of claim 7.

Images