A solid-state transformer multi-port routing control method and system based on digital twinning

Abstract

The application discloses a kind of solid-state transformer multi-port routing control method and system based on digital twinning, belong to transformer intelligent control technical field.The method and system are by collecting the real-time electrical waveform and temperature data of solid-state transformer power switch device, input to multi-physical field coupling digital twinning model, and the port stress state data reflecting the health state of each power port is calculated;Combined with real-time port power data, the instruction of dynamically distributing the transmission power of each port is generated by routing optimization controller and is executed;At the same time, based on the comparison deviation of new monitoring data and historical data obtained after instruction execution, the digital twinning model is modified on line, and the model self-evolution is realized.The application realizes the self-modification ability of digital twinning model, effectively improves the reliability, balance and life of solid-state transformer operation.

Description

Technical Field

[0001] This invention belongs to the field of intelligent control technology for transformers, and particularly relates to a multi-port routing control method and system for solid-state transformers based on digital twins. Background Technology

[0002] Solid-state transformers (SSTs), as key power electronic devices with multiple ports enabling hybrid AC / DC power distribution and flexible power control, hold significant application prospects in smart grids and renewable energy integration. Their core consists of numerous power switching devices, achieving high-frequency isolation and power conversion. During operation, different power ports of the SST typically undertake dynamically changing power transmission tasks. Traditional control methods optimize power allocation across ports primarily based on external grid demand and port voltage, aiming to achieve power balance, improve efficiency, or stabilize the DC bus voltage. However, these methods generally neglect the actual operating stress and health status of the power switching devices within the SST's critical components. Due to manufacturing tolerances, aging, and differences in operating conditions, the electrical stress (voltage, current) and thermal stress experienced by the power switching devices at each port are not uniform. Long-term adoption of fixed power allocation strategies or those optimized solely based on external electrical quantities may cause some ports to operate under high stress for extended periods, accelerating aging and even triggering failures, thus becoming a bottleneck restricting the long-term reliable operation and overall lifespan of the SST. Furthermore, the parameters of the digital twin model used for condition assessment and decision-making are prone to drift during long-term operation, leading to inaccurate model predictions and consequently affecting the accuracy and reliability of model-based optimization decisions. Therefore, there is an urgent need for an intelligent control method that can perceive the health status of key internal components in real time, dynamically adjust power allocation strategies, and ensure the accuracy of the model itself, in order to improve the operational reliability, lifespan, and overall performance of the SST. Summary of the Invention

[0003] To address the technical problem that existing solid-state transformers fail to consider the actual stress state of the power devices at each port during multi-port power distribution, leading to uneven stress between ports and affecting operational reliability, this invention provides a multi-port routing control method and system for solid-state transformers based on digital twins.

[0004] To achieve the above objectives, a first aspect of the present invention provides a multi-port routing control method for solid-state transformers based on digital twins, comprising the following steps: Acquire the first electrical waveform data and the first temperature monitoring data of the internal power switching devices of the target solid-state transformer during operation; The electrical waveform data and temperature monitoring data are input into a preset multi-physics coupled digital twin model. Based on the multi-physics coupled digital twin model, the port stress state data of the power port health status of the target solid-state transformer is calculated and output. The port stress state data and real-time port power data collected from the AC / DC ports of the solid-state transformer are input to the routing optimization controller. The routing optimization controller calculates and outputs a multi-port power allocation command for dynamically allocating the power transmission between the ports of the solid-state transformer. At the same time, the multi-port power allocation command is sent to the switching drive circuit of the solid-state transformer for execution, and the second electrical waveform data and the second temperature monitoring data are acquired in real time. The second electrical waveform data and the second temperature monitoring data are compared with the first electrical waveform data and the first temperature monitoring data to generate model prediction deviation data. When the deviation data indicates that the model is inaccurate, the parameter compensation amount of the model is calculated based on the deviation data, and the multiphysics coupled digital twin model is automatically corrected based on the parameter compensation amount.

[0005] Furthermore, it also includes: A voltage sensor, a current sensor, and a temperature sensor are provided on the power switching device of the target solid-state transformer; The port voltage waveform and branch current waveform of the power switching device under operating conditions are simultaneously collected as the first electrical waveform data, and the junction temperature or case temperature of the power switching device is simultaneously collected as the first temperature monitoring data.

[0006] Furthermore, the construction process of the preset multiphysics coupled digital twin model includes: Based on the physical parameters and circuit topology of the target solid-state transformer, the preset multi-physics coupled digital twin model is constructed. Within the digital twin model, electrical simulation calculations are performed using the first electrical waveform data to obtain the electrical stress parameters of each power port under the current operating conditions. Within the multiphysics coupled digital twin model, thermal simulation calculations are performed using the first temperature monitoring data and the power loss data obtained from the electrical simulation calculations to obtain the thermal stress parameters of the target solid-state transformer power port under the current operating conditions. The electrical stress parameters and the thermal stress parameters are processed based on preset mapping and fusion rules to output the port stress state data.

[0007] Furthermore, processing the electrical stress parameters and the thermal stress parameters based on preset mapping and fusion rules also includes the following steps: The electrical simulation calculations and thermal simulation calculations are performed interactively through a two-way coupling relationship, wherein: The power loss data calculated by the electrical simulation is used as a heat source input to the thermal simulation calculation; The temperature data output from the thermal simulation calculation is fed back to the electrical simulation calculation to correct the on-state voltage drop and switching time characteristic parameters of the power switching device online. The temperature data output from the thermal simulation calculation is further input into a preset life assessment model to calculate and update the cumulative damage value corresponding to each power port of the target solid-state transformer in real time.

[0008] Furthermore, it also includes, The multi-port power allocation command output by the routing optimization controller is sent to the switching drive circuit of the solid-state transformer to control the operation of the switching devices of each power port and adjust the power transmission path and allocation ratio of each port inside the solid-state transformer. After executing the multi-port power allocation command, the electrical waveform data and temperature monitoring data of the power switching devices inside the target solid-state transformer are re-acquired as the second electrical waveform data and the second temperature monitoring data.

[0009] Furthermore, the route optimization controller calculation also includes the following steps: The multi-port power allocation command is calculated by solving the minimum value of the following objective function, as follows: , in, Indicates from the first The port to the first The power transmitted by each port, The cost function representing power transfer between ports. Indicates the first A penalty function for each constraint. The penalty coefficient is... This represents the total number of ports on the solid-state transformer. The total number of constraints; the constraints include at least: port power capacity constraints. System power balance constraints .

[0010] Furthermore, the generated model prediction bias data includes: A multiphysics coupled digital twin model is constructed based on the first electrical waveform data and the first temperature monitoring data; the second electrical waveform data and the second temperature monitoring data are input as feedback quantities to the reference model; Calculate the dynamic difference between the feedback quantity and the predicted output quantity of the benchmark model at the corresponding time node; generate prediction deviation data based on the magnitude and duration of the dynamic difference.

[0011] Secondly, the present invention also provides a digital twin-based solid-state transformer multi-port routing control system, applied to any of the digital twin-based solid-state transformer multi-port routing control methods described above, comprising: The data acquisition module is used to acquire the first electrical waveform data and the first temperature monitoring data of the internal power switching devices of the target solid-state transformer during operation; A multi-physics coupled digital twin model module is used to receive the electrical waveform data and temperature monitoring data, and calculate and output the port stress state data of the health status of each power port of the target solid-state transformer. The routing optimization control module is used to receive the port stress state data and the real-time port power data collected from the AC / DC ports of the solid-state transformer, and calculate and output a multi-port power allocation command for dynamically allocating the power transmission between the ports of the solid-state transformer. The instruction execution module is used to send the multi-port power allocation instruction to the switch drive circuit of the solid-state transformer for execution, and to acquire the second electrical waveform data and the second temperature monitoring data in real time after execution; The model self-correction module is used to compare the second electrical waveform data and the second temperature monitoring data with the first electrical waveform data and the first temperature monitoring data to generate model prediction deviation data; when the deviation data indicates that the model is inaccurate, the module calculates the parameter compensation amount of the model based on the deviation data, and automatically corrects the multiphysics coupled digital twin model module based on the parameter compensation amount.

[0012] Furthermore, the multiphysics coupled digital twin model module is specifically used for: Receive first electrical waveform data and first temperature monitoring data from the data acquisition module; For each power port, electrical simulation is performed using the first electrical waveform data to calculate the electrical stress borne by the power switching device at that port. The electrical stress includes at least voltage stress and current stress. For each power port, thermal simulation is performed using the first temperature monitoring data and the loss data generated by the corresponding electrical simulation to calculate the thermal stress borne by the power switching device at that port. By integrating the electrical and thermal stresses of each power port and based on preset health assessment criteria, port stress state data is generated to quantitatively characterize the health status of each power port of the solid-state transformer, and then output to the routing optimization control module.

[0013] Furthermore, the model self-correction module is specifically used for: The second electrical waveform data and the second temperature monitoring data received from the instruction execution module are compared with the first electrical waveform data and the first temperature monitoring data obtained from the data acquisition module to calculate the model prediction deviation data that characterizes the simulation accuracy of the multi-physics coupled digital twin model module. When the model prediction deviation data exceeds a preset threshold, the parameter compensation amount of the key physical parameters in the multiphysics coupled digital twin model module is calculated based on the error characteristics reflected by the deviation data. Based on the parameter compensation amount, the internal model parameters of the multiphysics coupled digital twin model module are iteratively corrected online to reduce the model prediction bias and realize the adaptive tracking of the actual operating state of the solid-state transformer by the digital twin model.

[0014] The beneficial technical effects of the present invention are at least as follows: This invention provides a multi-port routing control method and system for solid-state transformers based on digital twin self-evolution, the advantages of which are mainly reflected in the following three aspects: Firstly, it significantly improves the operational reliability and lifespan of solid-state transformers. Traditional power allocation strategies do not consider the actual health status of internal components, which can easily lead to long-term overload of some ports. This invention uses a digital twin model to calculate and output "port stress state data" reflecting the health status of each power port in real time. This allows the routing optimization controller to make power allocation decisions based on this data, proactively allocating power to ports with lower stress and better health. This effectively balances the electrical and thermal stress of each port, avoids localized overload, and slows down the aging process of power switching devices, fundamentally improving the overall operational reliability and expected lifespan of solid-state transformers.

[0015] Secondly, it achieves continuous self-maintenance and optimization of the accuracy of the digital twin model. Existing digital twin models, after long-term operation, experience parameter drift due to factors such as device aging and environmental changes, leading to model inaccuracies. This invention automatically calculates the deviation and generates parameter compensation by comparing the model's predicted data with new monitoring data collected after actual command execution, thus correcting the twin model online. This closed-loop self-evolutionary mechanism of "monitoring-execution-comparison-correction" ensures that the digital twin model can continuously track the real physical state of the solid-state transformer, maintaining high accuracy over the long term, and providing a stable and reliable decision-making basis for model-based optimization control.

[0016] Thirdly, it optimizes the overall operating efficiency and performance of the system. This invention combines port health status with real-time power demand for optimization decisions, realizing a shift from "passively responding to external demands" to "actively coordinating internal and external states." This enables the solid-state transformer not only to protect itself while meeting power transmission requirements, but also to dynamically find the optimal or better operating point under multiple objectives such as system efficiency and stress balance, provided the model remains accurate. This improves the overall economy of energy transmission and the robustness of system performance, providing key technical support for building smarter and more robust power electronic conversion systems. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the working steps of a digital twin-based multi-port routing control method for solid-state transformers, as disclosed in one embodiment of the present invention. Figure 2 This is a diagram illustrating the steps involved in constructing the preset multiphysics coupled digital twin model, as disclosed in one embodiment of the present invention. Figure 3 This is a diagram illustrating the simulation calculation steps disclosed in one embodiment of the present invention; Figure 4 This is a schematic diagram of a digital twin-based solid-state transformer multi-port routing control system according to one embodiment of the present invention. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0020] Example 1 refer to Figure 1 This invention provides an embodiment of a multi-port routing control method for solid-state transformers based on digital twins, comprising the following steps: Acquire the first electrical waveform data and the first temperature monitoring data of the internal power switching devices of the target solid-state transformer during operation; The electrical waveform data and temperature monitoring data are input into a preset multi-physics coupled digital twin model. Based on the multi-physics coupled digital twin model, the port stress state data of the power port health status of the target solid-state transformer is calculated and output. The port stress state data and real-time port power data collected from the AC / DC ports of the solid-state transformer are input to the routing optimization controller. The routing optimization controller calculates and outputs a multi-port power allocation command for dynamically allocating the power transmission between the ports of the solid-state transformer. At the same time, the multi-port power allocation command is sent to the switching drive circuit of the solid-state transformer for execution, and the second electrical waveform data and the second temperature monitoring data are acquired in real time. The second electrical waveform data and the second temperature monitoring data are compared with the first electrical waveform data and the first temperature monitoring data to generate model prediction deviation data. When the deviation data indicates that the model is inaccurate, the parameter compensation amount of the model is calculated based on the deviation data, and the multiphysics coupled digital twin model is automatically corrected based on the parameter compensation amount.

[0021] In this embodiment, a sensor network deployed on the power switching devices (such as IGBTs) of the target solid-state transformer (e.g., a three-port AC-DC-AC solid-state transformer) collects first electrical waveform data (such as collector-emitter voltage Vce waveform and collector current Ic waveform) and first temperature monitoring data (such as case temperature measured by a thermistor or infrared sensor) in real time during its operation. The collected data is then input into a multiphysics coupled digital twin model pre-built and running in an industrial control computer or embedded processor. This model calculates port stress state data reflecting the health status of each power port (e.g., input AC port, high-voltage DC port, low-voltage DC port) through coupled electrical-thermal simulation. This data can be a vector containing voltage stress coefficient, current stress coefficient, and thermal stress coefficient. Then, the port stress state data, along with real-time port power (active power and reactive power) data collected from the solid-state transformer port measurement circuit, is input to a routing optimization controller (e.g., a controller implemented based on a model predictive control (MPC) algorithm). The controller performs optimization calculations with the goal of balancing port stress or extending overall lifespan, outputting multi-port power distribution commands (such as adjustments to the power setpoints of each port) for dynamically adjusting the power flow between ports. These commands are sent in real-time to the switching drive circuit of the solid-state transformer (e.g., a PWM wave generated by a DSP), controlling the corresponding switching devices to adjust the power distribution. Simultaneously, after command execution, second electrical waveform data and second temperature monitoring data from the power switching devices are collected again. Finally, the second set of data is compared with the first set (e.g., calculating the root mean square error of the waveform data and the absolute error of the temperature data) to generate model prediction deviation data. When the deviation exceeds a threshold, a parameter identification algorithm calculates the compensation amount for key model parameters based on the deviation data, and automatically corrects the internal parameters of the digital twin model online, completing the model's self-evolution update.

[0022] Furthermore, it also includes: A voltage sensor, a current sensor, and a temperature sensor are provided on the power switching device of the target solid-state transformer; The port voltage waveform and branch current waveform of the power switching device under operating conditions are simultaneously collected as the first electrical waveform data, and the junction temperature or case temperature of the power switching device is simultaneously collected as the first temperature monitoring data.

[0023] It should be noted that, in specific implementation, the steps of acquiring the first electrical waveform data and the first temperature monitoring data are achieved in the following way: A high-bandwidth differential voltage probe is connected to the power terminal of each power switching device of the target solid-state transformer to acquire the port voltage waveform, and a Rogowski coil or Hall current sensor is connected in series to acquire the branch current waveform. Simultaneously, a surface-mount PT100 temperature sensor or fiber optic temperature sensor is installed on the heat sink of each power switching device or at a location adjacent to the chip to monitor its junction temperature or directly measure the case temperature. The data acquisition card is triggered synchronously with all sensors to ensure that voltage, current, and temperature data are acquired synchronously at the same timestamp, collectively constituting the first electrical waveform data and the first temperature monitoring data.

[0024] Furthermore, the construction process of the preset multiphysics coupled digital twin model includes: Based on the physical parameters and circuit topology of the target solid-state transformer, the preset multi-physics coupled digital twin model is constructed. Within the digital twin model, electrical simulation calculations are performed using the first electrical waveform data to obtain the electrical stress parameters of each power port under the current operating conditions. Within the multiphysics coupled digital twin model, thermal simulation calculations are performed using the first temperature monitoring data and the power loss data obtained from the electrical simulation calculations to obtain the thermal stress parameters of the target solid-state transformer power port under the current operating conditions. The electrical stress parameters and the thermal stress parameters are processed based on preset mapping and fusion rules to output the port stress state data.

[0025] It should be noted that a high-fidelity simulation model of the dual active bridge circuit topology based on the target solid-state transformer and the device datasheet parameters is constructed in the simulation environment as the core of the digital twin model. During the model operation phase, the first electrical waveform data collected is input into the electrical simulation part of the model. By solving the circuit equations, the electrical stress parameters of the associated switching devices at each power port, including peak voltage and RMS current, are calculated in real time. Simultaneously, the first temperature monitoring data and the switching loss and conduction loss data calculated by the electrical simulation part are input into the thermal simulation part of the model. This part calculates the thermal stress parameters such as junction temperature fluctuation and average temperature of each power switching device based on the Foster thermal network model. Finally, the built-in processing unit of the model performs calculations according to preset rules, normalizes the electrical stress parameters, performs a weighted sum, and combines it with the thermal stress parameters through a stress fusion function to output a comprehensive port stress state data between 0 and 1. The higher the value, the worse the health status of the port and the higher the stress level.

[0026] Furthermore, constructing the preset multiphysics coupled digital twin model also includes processing the electrical stress parameters and the thermal stress parameters based on preset mapping and fusion rules, which further includes the following steps: The electrical simulation calculations and thermal simulation calculations are performed interactively through a two-way coupling relationship, wherein: The power loss data calculated by the electrical simulation is used as a heat source input to the thermal simulation calculation; The temperature data output from the thermal simulation calculation is fed back to the electrical simulation calculation to correct the on-state voltage drop and switching time characteristic parameters of the power switching device online. The temperature data output from the thermal simulation calculation is further input into a preset life assessment model to calculate and update the cumulative damage value corresponding to each power port of the target solid-state transformer in real time.

[0027] It should be noted that the more specific process of constructing the aforementioned pre-defined multiphysics coupled digital twin model includes the following steps. First, based on the actual physical parameters of the solid-state transformer, its electrical behavior model, equivalent thermal network model, and lifetime assessment model are established, with the lifetime assessment model constructed based on the Coffin-Manson equation. During model operation, the real-time power loss value calculated by the electrical behavior model is used as a heat source input to the equivalent thermal network model. The junction temperature data solved by the equivalent thermal network model is fed back to the electrical behavior model to dynamically adjust temperature-related electrical parameters such as the on-resistance and threshold voltage of the switching devices in the model to achieve bidirectional electrothermal coupling. On the other hand, this temperature data, along with time information, is input into the lifetime assessment model to calculate and update the cumulative damage value of each power switching device in real time. This lifetime information is also output as part of the port stress state data.

[0028] Furthermore, it also includes, The multi-port power allocation command output by the routing optimization controller is sent to the switching drive circuit of the solid-state transformer to control the operation of the switching devices of each power port and adjust the power transmission path and allocation ratio of each port inside the solid-state transformer. After executing the multi-port power allocation command, the electrical waveform data and temperature monitoring data of the power switching devices inside the target solid-state transformer are re-acquired as the second electrical waveform data and the second temperature monitoring data.

[0029] It should be noted that the specific steps of command issuance and re-acquisition in implementation are as follows: The multi-port power allocation command calculated by the routing optimization controller is issued to the central control unit of the solid-state transformer through the communication interface. Based on the new power command, the control unit generates a new pulse width modulation drive signal by adjusting the modulation signal of the phase shift control, and controls the on / off state of the switching devices corresponding to each power port via the drive circuit, thereby dynamically adjusting the internal power flow path and the actual power allocation ratio of each port. After the new power allocation state stabilizes, the system again triggers the sensor network and data acquisition card to re-acquire the voltage, current waveforms, and temperature data of the power switching devices in the same manner as acquiring the first set of data. This serves as the second electrical waveform data and the second temperature monitoring data for model verification and correction.

[0030] Furthermore, the route optimization controller calculation also includes the following steps: The multi-port power allocation command is calculated by solving the minimum value of the following objective function, as follows: , in, Indicates from the first The port to the first The power transmitted by each port, The cost function representing power transfer between ports. Indicates the first A penalty function for each constraint. The penalty coefficient is... This represents the total number of ports on the solid-state transformer. The total number of constraints; the constraints include at least: port power capacity constraints. System power balance constraints .

[0031] It should be noted that the controller's built-in optimization algorithm aims to minimize the total system cost, which consists of two parts: the cost of power transfer between ports and the penalty for power constraints imposed on the stress state of each port. The optimization variable is the power transferred between ports. The hard constraints that must be satisfied during the optimization solution include the constraint that the power transferred at each port must not exceed the upper and lower limits of its design capacity, and the power balance constraint that the sum of the net power of all ports must be zero for the entire solid-state transformer system. The controller uses the currently acquired port stress state data and real-time port power data to solve the above constrained optimization problem to obtain a set of optimal power allocation values ​​and converts them into specific control commands.

[0032] Furthermore, the generated model prediction bias data includes: A multiphysics coupled digital twin model is constructed based on the first electrical waveform data and the first temperature monitoring data; the second electrical waveform data and the second temperature monitoring data are input as feedback quantities to the reference model; Calculate the dynamic difference between the feedback quantity and the predicted output quantity of the benchmark model at the corresponding time node; generate prediction deviation data based on the magnitude and duration of the dynamic difference.

[0033] It should be noted that after the first correction of the multiphysics coupled digital twin model, the first electrical waveform data and the first temperature monitoring data collected at that moment are used as inputs to drive the multiphysics coupled digital twin model to simulate and predict the electrical waveform and temperature data of the next control cycle, and save them as the predicted output sequence. When the new control command is executed, the second electrical waveform data and the second temperature monitoring data collected are used as the actual feedback sequence. The feedback sequence is compared point by point with the predicted output sequence of the benchmark model at the same time node, and the dynamic differences such as voltage prediction error, current prediction error and temperature prediction error are calculated. The model self-correction module analyzes the magnitude and duration of these errors, and judges them by calculating the proportion of time when the mean square error and statistical error exceed the threshold. If the magnitude continuously exceeds the preset threshold and the duration reaches a certain length, it is determined that the model is inaccurate, and a model prediction deviation dataset containing the magnitude, direction and statistical characteristics of the error is generated for subsequent calculation of parameter compensation.

[0034] Secondly, the present invention also provides a digital twin-based solid-state transformer multi-port routing control system, applied to any of the digital twin-based solid-state transformer multi-port routing control methods described above, comprising: The data acquisition module is used to acquire the first electrical waveform data and the first temperature monitoring data of the internal power switching devices of the target solid-state transformer during operation; A multi-physics coupled digital twin model module is used to receive the electrical waveform data and temperature monitoring data, and calculate and output the port stress state data of the health status of each power port of the target solid-state transformer. The routing optimization control module is used to receive the port stress state data and the real-time port power data collected from the AC / DC ports of the solid-state transformer, and calculate and output a multi-port power allocation command for dynamically allocating the power transmission between the ports of the solid-state transformer. The instruction execution module is used to send the multi-port power allocation instruction to the switch drive circuit of the solid-state transformer for execution, and to acquire the second electrical waveform data and the second temperature monitoring data in real time after execution; The model self-correction module is used to compare the second electrical waveform data and the second temperature monitoring data with the first electrical waveform data and the first temperature monitoring data to generate model prediction deviation data; when the deviation data indicates that the model is inaccurate, the module calculates the parameter compensation amount of the model based on the deviation data, and automatically corrects the multiphysics coupled digital twin model module based on the parameter compensation amount.

[0035] Furthermore, the multiphysics coupled digital twin model module is specifically used for: Receive first electrical waveform data and first temperature monitoring data from the data acquisition module; For each power port, electrical simulation is performed using the first electrical waveform data to calculate the electrical stress borne by the power switching device at that port. The electrical stress includes at least voltage stress and current stress. For each power port, thermal simulation is performed using the first temperature monitoring data and the loss data generated by the corresponding electrical simulation to calculate the thermal stress borne by the power switching device at that port. By integrating the electrical and thermal stresses of each power port and based on preset health assessment criteria, port stress state data is generated to quantitatively characterize the health status of each power port of the solid-state transformer, and then output to the routing optimization control module.

[0036] Furthermore, the model self-correction module is specifically used for: The second electrical waveform data and the second temperature monitoring data received from the instruction execution module are compared with the first electrical waveform data and the first temperature monitoring data obtained from the data acquisition module to calculate the model prediction deviation data that characterizes the simulation accuracy of the multi-physics coupled digital twin model module. When the model prediction deviation data exceeds a preset threshold, the parameter compensation amount of the key physical parameters in the multiphysics coupled digital twin model module is calculated based on the error characteristics reflected by the deviation data. Based on the parameter compensation amount, the internal model parameters of the multiphysics coupled digital twin model module are iteratively corrected online to reduce the model prediction bias and realize the adaptive tracking of the actual operating state of the solid-state transformer by the digital twin model.

[0037] It should be noted that in this embodiment, the module continuously receives a second set of monitoring data from the instruction execution module and retrieves the first set of monitoring data that triggered the current control cycle from the cache. It first performs time alignment and preprocessing on the two sets of data, then calculates the deviation of key characteristic quantities, including comparing the difference between the predicted junction temperature and the actual estimated junction temperature of the same switching device at the same time, and comparing the difference between the predicted voltage overshoot and the actual overshoot. These deviation values ​​are summarized and statistically analyzed to form model prediction deviation data. When key indicators in the deviation data exceed a preset threshold for multiple consecutive cycles, the module activates the parameter identification algorithm. This algorithm, based on the characteristics of the deviation data, reverse-engineers which key physical parameters in the digital twin model module need compensation and calculates the specific compensation amount. Subsequently, the module iteratively updates the configuration file or internal variables of the digital twin model module online through the model interface, thereby reducing the deviation of subsequent predictions and achieving adaptive tracking of the actual degradation state of the transformer by the model. In the description of this specification, the references to terms such as "an embodiment," "example," "specific example," 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 the present invention. 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.

[0038] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes: USB flash drive, mobile hard drive, read-only memory (ROM). ROM (ROM), RAM (Random Access Memory), magnetic disks, optical disks, and other media that can store programs.

Claims

1. A multi-port routing control method for solid-state transformers based on digital twins, characterized in that, Includes the following steps: Acquire the first electrical waveform data and the first temperature monitoring data of the internal power switching devices of the target solid-state transformer during operation; The electrical waveform data and temperature monitoring data are input into a preset multi-physics coupled digital twin model. Based on the multi-physics coupled digital twin model, the port stress state data of the power port health status of the target solid-state transformer is calculated and output. The port stress state data and real-time port power data collected from the AC / DC ports of the solid-state transformer are input to the routing optimization controller. The routing optimization controller calculates and outputs a multi-port power allocation command for dynamically allocating the power transmission between the ports of the solid-state transformer. At the same time, the multi-port power allocation command is sent to the switching drive circuit of the solid-state transformer for execution, and the second electrical waveform data and the second temperature monitoring data are acquired in real time. The second electrical waveform data and the second temperature monitoring data are compared with the first electrical waveform data and the first temperature monitoring data to generate model prediction deviation data. When the deviation data indicates that the model is inaccurate, the parameter compensation amount of the model is calculated based on the deviation data, and the multiphysics coupled digital twin model is automatically corrected based on the parameter compensation amount.

2. The multi-port routing control method for solid-state transformers based on digital twins according to claim 1, characterized in that, include: A voltage sensor, a current sensor, and a temperature sensor are provided on the power switching device of the target solid-state transformer; The port voltage waveform and branch current waveform of the power switching device under operating conditions are simultaneously collected as the first electrical waveform data, and the junction temperature or case temperature of the power switching device is simultaneously collected as the first temperature monitoring data.

3. The multi-port routing control method for solid-state transformers based on digital twins according to claim 1, characterized in that, The construction process of the preset multiphysics coupled digital twin model includes: Based on the physical parameters and circuit topology of the target solid-state transformer, the preset multi-physics coupled digital twin model is constructed. Within the digital twin model, electrical simulation calculations are performed using the first electrical waveform data to obtain the electrical stress parameters of each power port under the current operating conditions. Within the multiphysics coupled digital twin model, thermal simulation calculations are performed using the first temperature monitoring data and the power loss data obtained from the electrical simulation calculations to obtain the thermal stress parameters of the target solid-state transformer power port under the current operating conditions. The electrical stress parameters and the thermal stress parameters are processed based on preset mapping and fusion rules to output the port stress state data.

4. The multi-port routing control method for solid-state transformers based on digital twins according to claim 3, characterized in that, The electrical simulation calculations and thermal simulation calculations are performed interactively through a two-way coupling relationship, wherein: The power loss data calculated by the electrical simulation is used as a heat source input to the thermal simulation calculation; The temperature data output from the thermal simulation calculation is fed back to the electrical simulation calculation to correct the on-state voltage drop and switching time characteristic parameters of the power switching device online. The temperature data output from the thermal simulation calculation is further input into a preset life assessment model to calculate and update the cumulative damage value corresponding to each power port of the target solid-state transformer in real time.

5. The multi-port routing control method for solid-state transformers based on digital twins according to claim 1, characterized in that, It also includes, The multi-port power allocation command output by the routing optimization controller is sent to the switching drive circuit of the solid-state transformer to control the operation of the switching devices of each power port and adjust the power transmission path and allocation ratio of each port inside the solid-state transformer. After executing the multi-port power allocation command, the electrical waveform data and temperature monitoring data of the power switching devices inside the target solid-state transformer are re-acquired as the second electrical waveform data and the second temperature monitoring data.

6. The multi-port routing control method for solid-state transformers based on digital twins according to claim 1, characterized in that, The route optimization controller calculation also includes the following steps: The multi-port power allocation command is calculated by solving the minimum value of the following objective function, as follows: ， in, Indicates from the first The port to the first The power transmitted by each port, The cost function representing power transfer between ports. Indicates the first A penalty function for each constraint. The penalty coefficient is... This represents the total number of ports on the solid-state transformer. This represents the total number of constraints. The constraints include at least: port power capacity constraints. System power balance constraints .

7. The multi-port routing control method for solid-state transformers based on digital twins according to claim 1, characterized in that, The generated model prediction bias data includes: A multiphysics coupled digital twin model is constructed based on the first electrical waveform data and the first temperature monitoring data; the second electrical waveform data and the second temperature monitoring data are input as feedback quantities to the reference model; Calculate the dynamic difference between the feedback quantity and the predicted output quantity of the benchmark model at the corresponding time node; generate prediction deviation data based on the magnitude and duration of the dynamic difference.

8. A digital twin-based solid-state transformer multi-port routing control system, applied to the digital twin-based solid-state transformer multi-port routing control method as described in any one of claims 1 to 7, characterized in that, include: The data acquisition module is used to acquire the first electrical waveform data and the first temperature monitoring data of the internal power switching devices of the target solid-state transformer during operation; A multi-physics coupled digital twin model module is used to receive the electrical waveform data and temperature monitoring data, and calculate and output the port stress state data of the health status of each power port of the target solid-state transformer. The routing optimization control module is used to receive the port stress state data and the real-time port power data collected from the AC / DC ports of the solid-state transformer, and calculate and output a multi-port power allocation command for dynamically allocating the power transmission between the ports of the solid-state transformer. The instruction execution module is used to send the multi-port power allocation instruction to the switch drive circuit of the solid-state transformer for execution, and to acquire the second electrical waveform data and the second temperature monitoring data in real time after execution; The model self-correction module is used to compare the second electrical waveform data and the second temperature monitoring data with the first electrical waveform data and the first temperature monitoring data to generate model prediction deviation data; when the deviation data indicates that the model is inaccurate, the module calculates the parameter compensation amount of the model based on the deviation data, and automatically corrects the multiphysics coupled digital twin model module based on the parameter compensation amount.

9. The solid-state transformer multi-port routing control system based on digital twin as described in claim 8, characterized in that, The multiphysics coupled digital twin model module is specifically used for: Receive first electrical waveform data and first temperature monitoring data from the data acquisition module; For the power port, electrical simulation is performed using the first electrical waveform data to calculate the electrical stress borne by the power switching device under the power port. The electrical stress includes at least voltage stress and current stress. For each power port, thermal simulation is performed using the first temperature monitoring data and the loss data generated by the corresponding electrical simulation to calculate the thermal stress borne by the power switching device at that port. By integrating the electrical and thermal stresses of each power port and based on preset health assessment criteria, port stress state data is generated to quantitatively characterize the health status of each power port of the solid-state transformer, and then output to the routing optimization control module.

10. The solid-state transformer multi-port routing control system based on digital twin according to claim 8, characterized in that, The model self-correction module is specifically used for: The second electrical waveform data and the second temperature monitoring data received from the instruction execution module are compared with the first electrical waveform data and the first temperature monitoring data obtained from the data acquisition module to calculate the model prediction deviation data that characterizes the simulation accuracy of the multi-physics coupled digital twin model module. When the model prediction deviation data exceeds a preset threshold, the parameter compensation amount of the key physical parameters in the multiphysics coupled digital twin model module is calculated based on the error characteristics reflected by the deviation data. Based on the parameter compensation amount, the internal model parameters of the multiphysics coupled digital twin model module are iteratively corrected online to reduce the model prediction bias and realize the adaptive tracking of the actual operating state of the solid-state transformer by the digital twin model.

Images