Software-defined smart electromagnetic equipment digital twin device, method and apparatus

Abstract

The application relates to the technical field of digital twin design equipment, and provides a software-defined intelligent electromagnetic equipment digital twin device, method and equipment, wherein a software radio is used as a signal processing base, and equipment simulation modules, equipment decomposition modules, equipment diagnosis modules, equipment drill modules and AI intelligent agent modules are arranged, which are used for constructing a full-link real-time sensing, multi-physical field high-fidelity simulation, software dynamic reconstruction and knowledge-driven intelligent decision system, and deeply integrating physical equipment characteristics and software definition capabilities; with the aid of an AI digital equipment intelligent agent, a user can be intelligently assisted in completing full-cycle management from design, training, operation and maintenance to reconstruction in each link of the digital twin device, the engineering problems of weak adaptability, high maintenance cost and poor drill effect of existing electromagnetic equipment in a complex environment are solved, and the continuous optimization of equipment performance and the fine management of the whole life cycle are realized.

Description

Technical Field

[0001] This invention belongs to the field of equipment digital twin design technology, and relates to a software-defined intelligent electromagnetic equipment digital twin device, method and equipment. Background Technology

[0002] With the increasing complexity of modern electromagnetic environments and the growing informatization of equipment, the management and maintenance of traditional electromagnetic equipment, such as radar and communication radios, face significant challenges. Traditional hardware systems struggle to quickly adjust waveforms and algorithms via software, resulting in poor equipment adaptability and long upgrade cycles. Furthermore, when the electromagnetic environment changes, the equipment cannot flexibly switch operating modes and signal systems, severely restricting its environmental adaptability. Existing digital twin systems primarily focus on static 3D displays or simple fault diagnosis, lacking real-time dynamic mapping of physical characteristics such as electromagnetic signals. More critically, existing systems cannot reverse-engineer the physical equipment to reconstruct parameters, leaving digital twins merely at the visualization level and failing to establish a closed-loop feedback mechanism.

[0003] Fault diagnosis is often based on post-incident alarms, meaning it's only discovered after a fault has occurred and had a real impact. It lacks data-driven diagnostics of equipment physical characteristics and preventative maintenance strategies, making it impossible to predict potential faults in the early stages of performance degradation, let alone achieve an automated closed loop from observation to diagnosis and correction. There are discrepancies between drills and real-world operational scenarios. Existing training systems mostly rely on pre-set scripts and cannot adaptively adjust strategies based on real-time situations; the lack of skill profiling based on actual data and personalized assessment mechanisms makes it difficult to directly translate drill results into improved actual mission capabilities. Therefore, how to deeply integrate physical equipment characteristics with software-defined capabilities to achieve a digital twin device that enables closed-loop management of the entire process from design, training, operation, and reconfiguration has become one of the most pressing technical challenges. Summary of the Invention

[0004] To address the problems existing in the aforementioned traditional technologies, this invention proposes a software-defined intelligent electromagnetic equipment digital twin device, method, and equipment, which can deeply integrate the characteristics of physical equipment with software-defined capabilities to realize a digital twin device with closed-loop management of the entire process from design, training, operation and maintenance to reconfiguration.

[0005] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions: On the one hand, a software-defined intelligent electromagnetic equipment digital twin device is provided, which uses software radio as the signal processing base and includes: The equipment simulation module is configured to be based on a software radio architecture, and performs waveform signal design and generation and operation interface construction according to the waveform signal design requirements of electromagnetic equipment. The equipment decomposition module is configured to decouple the core functions of electromagnetic equipment in layers according to the signal processing link and connect the graphical display component to the bypass branch of each functional module; the graphical display component is used to build a real-time data perspective window; The equipment diagnostic module is configured to build a fault knowledge graph and expert rule base for electromagnetic equipment, and perform source tracing and intelligent diagnosis of full-link data. The equipment drill module is configured to build a distributed simulation environment and execute multi-mode drills and task action strategy verification. The AI ​​intelligent agent module is communicatively connected to the equipment simulation module, equipment disassembly module, equipment diagnosis module, and equipment drill module, and is configured to provide intelligent assistance during the operation of each module.

[0006] In one embodiment, the equipment simulation module includes: The waveform signal design function unit is configured to be based on a software radio architecture. It calls the waveform component library and the visual waveform design engine, selects waveform components according to the waveform signal design requirements of electromagnetic equipment, establishes data links to execute the transmission of data flow between waveform components, and sets the parameters of waveform components. The equipment operation interface construction unit is configured to call the corresponding control from the control library according to the signal parameter type and configure the control interface to build the electromagnetic equipment operation interface; the control library includes basic controls, display controls and pane controls.

[0007] In one embodiment, the AI ​​agent module is configured to perform semantic parsing on the input equipment waveform signal design requirements during waveform signal design, automatically call the corresponding waveform component from the waveform component library, and then complete the connection establishment and parameter preset between waveform components. The AI ​​intelligent agent module is also configured to automatically call controls that support the corresponding data type based on the parameter data type of the equipment waveform signal when the operation interface is built, and to automatically adjust the layout and attributes of the controls according to preset operation habits.

[0008] In one embodiment, the equipment disassembly module is layered and decoupled and includes: The logic mapping unit is configured to automatically identify and split core sub-tasks based on the business logic and signal flow of the electromagnetic equipment. The interface standardization unit is configured to define standard input and output data formats for each functional module; the data format includes data structure type, sampling rate, bit width, and protocol format; The metadata encapsulation unit is configured to automatically associate context tags with each functional module and supports parallel output of metadata streams and real-time waveform streams in a set format. The context tags include timestamp synchronization information, sampling frequency, quantization bits, center frequency, and module running status code, and can be set in JSON or XML format.

[0009] In one embodiment, the graphical display component accessed in the bypass branch is configured to construct a real-time data pivot window; the real-time data pivot window includes: The time-domain view is used to plot the waveform amplitude change curve over time in real time, presenting pulse characteristics, noise floor, and dynamic range. Frequency domain view, used to generate spectrum and waterfall plots, showing the signal bandwidth occupied, location of interference sources, and frequency stability; Advanced analysis views are used to generate constellation diagrams, power spectral density curves, and eye diagrams, showing modulation quality and bit error rate trends.

[0010] In one embodiment, the equipment diagnostic module includes: The end-to-end data traceability unit is configured to encapsulate each processing module within the equipment as an independent data node. It collects the original in-phase and quadrature I / Q waveform streams and associated metadata of the input and output in real time through a bypass interface, visualizes and compares the input and output data side by side, and automatically stores historical data snapshots of key nodes. The associated metadata includes timestamps and sampling rates. The intelligent diagnostic unit is configured to automatically identify and mark abnormal modules based on preset threshold rules and real-time indicators, perform root cause localization calculations in conjunction with data flow logic, and generate parameter adjustment schemes based on the diagnostic results and the linkage waveform design engine.

[0011] In one embodiment, the intelligent diagnostic unit is further configured as follows: When the signal-to-noise ratio drop exceeds a preset threshold, an abnormal event is triggered. The fault knowledge graph and current data stream features are invoked to perform reverse reasoning operations, infer the physical fault source, generate a confidence score, and then generate a quantitative assessment report including symptoms, causes, and solutions.

[0012] In one embodiment, the equipment training module includes: The distributed simulation architecture unit is configured to support multiple digital twin devices to autonomously construct dynamic network topology in virtual space. It calculates the link status, hop count routing and bandwidth fluctuations between nodes in real time through the built-in mobile ad hoc network protocol stack model, and connects the single-device digital twin to the external joint training system through standard interfaces. The multi-dimensional exercise mode unit is configured to perform single-unit operation exercises, formation coordination exercises, and game simulations, respectively.

[0013] On the other hand, a method for creating a digital twin of intelligent electromagnetic equipment based on the aforementioned software-defined intelligent electromagnetic equipment digital twin device is also provided, comprising the following steps: Using software-defined radio as the signal processing foundation, the design and generation of waveform signals and the construction of the operation interface are performed according to the waveform signal design requirements of electromagnetic equipment. The core functions of electromagnetic equipment are decoupled in layers according to the signal processing link, and graphical display components are connected to the bypass branches of each functional module. Construct a fault knowledge graph and expert rule base for electromagnetic equipment, and perform source tracing, comparison and intelligent diagnosis of full-link data; Build a distributed simulation environment to perform multi-mode drills and verify task action strategies; The AI ​​intelligent agent module provides intelligent assistance in waveform signal design and generation, operation interface construction, hierarchical decoupling, data traceability and comparison, intelligent diagnosis, and collaborative training processes.

[0014] In another aspect, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described intelligent electromagnetic equipment digital twin method.

[0015] One of the above technical solutions has the following advantages and beneficial effects: The aforementioned software-defined intelligent electromagnetic equipment digital twin device, method, and equipment, by using software-defined radio as the signal processing foundation and deploying equipment simulation, equipment decomposition, equipment diagnosis, equipment training, and AI intelligent agent modules, are used to construct an intelligent decision-making system with full-link real-time perception, high-fidelity simulation of multi-physics fields, dynamic software reconstruction, and knowledge-driven capabilities. It deeply integrates the characteristics of physical equipment with software-defined capabilities, and with the help of AI digital equipment intelligent agents, it intelligently assists users in completing the entire lifecycle management from design, training, operation and maintenance to reconstruction at each stage of the digital twin device. It solves the engineering problems of existing electromagnetic equipment having weak adaptability in complex environments, high maintenance costs, and poor training effects, and realizes continuous optimization of equipment performance and refined management of the entire lifecycle.

[0016] By employing a technical approach that integrates software-defined radio infrastructure, a layered decoupling architecture, and deep AI-enabled intelligence, we have successfully addressed the real-world pain points in traditional electromagnetic equipment development, such as long design cycles, high debugging thresholds, difficulty in fault location, and rigid simulation environments. This approach transforms the traditional reliance on physical hardware and human experience into a new engineering model driven by data, intelligent decision-making, and closed-loop optimization, significantly improving the efficiency and intelligence level of the entire lifecycle management of electromagnetic equipment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the modular architecture of a software-defined intelligent electromagnetic equipment digital twin device in one embodiment; Figure 2 This is a schematic diagram of equipment disassembly in one embodiment; Figure 3 This is a flowchart of a digital twin method for intelligent electromagnetic equipment in one embodiment. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0020] It should be noted that, in this document, the reference to "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The presentation of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will understand that the embodiments described herein can be combined with other embodiments. The term "and / or" as used herein refers to any combination of one or more of the associated listed items, and all possible combinations, including such combinations.

[0021] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0022] In one embodiment, such as Figure 1 As shown, a software-defined intelligent electromagnetic equipment digital twin device is provided. With software radio as the signal processing base, it unifies the signal processing paradigm at the underlying architecture level, enabling each module to work collaboratively under the same signal processing framework. This fundamentally solves the problems of upgrade difficulties and data interoperability barriers between modules caused by the hardware fixation of traditional electromagnetic equipment.

[0023] The software-defined intelligent electromagnetic equipment digital twin device may include an equipment simulation module 11, an equipment disassembly module 13, an equipment diagnostic module 15, an equipment drill module 17, and an AI agent module 19. The equipment simulation module 11 is configured based on a software-defined radio architecture to design and generate waveform signals and construct the user interface according to the waveform signal design requirements of the electromagnetic equipment. The equipment disassembly module 13 is configured to decouple the core functions of the electromagnetic equipment in layers according to the signal processing link and connect graphical display components to the bypass branches of each functional module; the graphical display components are used to construct real-time data perspective windows. The equipment diagnostic module 15 is configured to construct a fault knowledge graph and expert rule base for the electromagnetic equipment, and perform source tracing and intelligent diagnosis of the entire link data. The equipment drill module 17 is configured to construct a distributed simulation environment to perform multi-mode drills and task action strategy verification. The AI ​​agent module 19 is communicatively connected to the equipment simulation module, equipment disassembly module, equipment diagnostic module, and equipment drill module, and is configured to provide intelligent assistance during the operation of each module.

[0024] It is understandable that the five modules of the aforementioned software-defined intelligent electromagnetic equipment digital twin device achieve deep task collaboration through the AI ​​intelligent agent module as the core scheduler: that is, after the equipment simulation module 11 generates waveform signals and an operating interface, the equipment decomposition module 13 transparently displays its internal signal processing links; the equipment diagnosis module 15 performs real-time monitoring and fault diagnosis based on the transparent data links; the equipment exercise module 17 uses the digital twin to conduct multi-dimensional exercise verification; and the AI ​​intelligent agent module 19 is present throughout, providing intelligent assistance at every stage. This architectural design enables the device to form a complete closed loop from design, observation, diagnosis to verification.

[0025] Existing digital twin systems typically focus on a single aspect, such as pure signal simulation or static 3D visualization, lacking cross-aspect collaborative mechanisms. This embodiment, through the orderly integration of five major modules and the overall scheduling of the AI ​​agent module 19, forms a complete digital twin system with real-time perception, intelligent diagnosis, and closed-loop optimization capabilities.

[0026] Within this overall architecture, the software-defined radio (SDR) platform provides a unified signal processing support platform for the entire device, enabling waveform signal design and generation to be flexibly adjusted via software configuration, rather than relying on dedicated hardware. The equipment simulation module 11 utilizes this feature to achieve rapid iterative waveform design; the equipment decomposition module 13, based on the standardized data interface under the SDR architecture, performs bypass observation of each processing node; the equipment diagnostic module 15 uses the raw data acquired through the bypass interface for end-to-end tracing; and the equipment drill module 17 realistically maps the signal environment of the digital twin's operation onto the virtual space. The operational data from each of the aforementioned basic modules is input into the AI ​​agent module 19 for analysis in real time, and the analysis results from the AI ​​agent module 19 are then fed back to each basic module in real time to guide its next operation, forming a real-time interactive intelligent closed loop.

[0027] The aforementioned software-defined intelligent electromagnetic equipment digital twin device, using software-defined radio as its signal processing foundation, deploys equipment simulation, equipment decomposition, equipment diagnosis, equipment training, and AI agent modules. This enables the construction of an intelligent decision-making system with full-link real-time perception, high-fidelity simulation of multi-physics fields, dynamic software reconfiguration, and knowledge-driven capabilities. It deeply integrates physical equipment characteristics with software-defined capabilities. Leveraging the AI ​​digital equipment agent, it intelligently assists users in completing the entire lifecycle management from design, training, operation and maintenance to reconfiguration at every stage of the digital twin device. This solves the engineering problems of existing electromagnetic equipment's weak adaptability in complex environments, high maintenance costs, and poor training effectiveness, achieving continuous optimization of equipment performance and refined management throughout its entire lifecycle.

[0028] In one embodiment, the equipment simulation module 11 may include two functional units: a waveform signal design function unit and an equipment operation interface construction unit. This is used to solve the problem that waveform design and operation interface construction are independent and iterative, resulting in a long development cycle in traditional electromagnetic equipment R&D. After the waveform signal design function unit completes the signal design, the output signal parameter data type is automatically passed to the equipment operation interface construction unit. The operation interface construction module automatically matches the controls accordingly to realize one-click generation from signal design to operation interface.

[0029] Furthermore, the AI ​​agent module 19 is configured to perform semantic parsing on the input equipment waveform signal design requirements during waveform signal design, automatically call the corresponding waveform component from the waveform component library, and complete the connection establishment and parameter preset between waveform components. The AI ​​agent module 19 is also configured to automatically call controls that support the corresponding data type according to the parameter data type of the equipment waveform signal during the construction of the operation interface, and automatically perform layout and attribute adjustment of the controls according to preset operation habits.

[0030] Specifically, the waveform signal design function unit is configured based on a software radio architecture, calling a waveform component library and a visual waveform design engine, selecting waveform components according to the waveform signal design requirements of electromagnetic equipment, establishing data links to execute the transmission of data flow between waveform components, and setting the parameters of waveform components.

[0031] The waveform signal design function unit utilizes the system's built-in waveform component library and visual waveform design engine. The waveform component library stores existing basic components such as carrier generators, pulse shaping filters, modulation mappers, and power amplifier models, each with standardized input / output interfaces. The visual waveform design engine provides a drag-and-drop component arrangement interface, allowing users to intuitively select and connect components within a graphical environment.

[0032] In practical operation, users can first input the design requirements for the waveform signal corresponding to the electromagnetic equipment being targeted, such as designing a communication waveform that operates in a specified frequency band, with a specific bandwidth, and using a specified modulation method. The AI ​​agent module 19 automatically performs semantic parsing on this requirement, calling an existing natural language processing model to perform entity recognition and intent decomposition, extracting key parameters such as frequency band, bandwidth, and modulation method. Subsequently, the AI ​​agent module 19 connects to the existing waveform component library of the software radio architecture, retrieves matching basic components based on the extracted parameters, automatically calculates the sampling rate conversion requirements between each component, and establishes a data link. For example, if the upstream output sampling rate does not match the downstream required sampling rate, the AI ​​agent module 19 automatically inserts a decimation filter component into the link to meet the sampling rate conversion requirements.

[0033] During the parameter preset phase, the AI ​​agent module 19 can automatically fill in default parameters based on its built-in knowledge base. For example, the filter cutoff frequency can be set to 1.2 times the signal bandwidth to ensure adequate transition band protection, and the pulse shaping roll-off factor can be set to 0.35 to balance bandwidth efficiency with the suppression of inter-symbol interference. These preset values ​​are derived from the AI ​​agent module 19's learning of extensive electromagnetic equipment waveform design experience, enabling even non-professional operators to obtain initial design parameters approaching expert levels.

[0034] After the system generates a preliminary waveform flow diagram, the user can fine-tune it based on the recommendations of the AI ​​agent module 19. Once confirmed, the settings are saved and the system is started. User actions such as rejecting AI-recommended parameters and manually adjusting them to other values ​​will be recorded by the system and stored in the training set as training data for subsequent model fine-tuning. This allows the AI ​​agent module 19 to continuously optimize its understanding of user habits and specific equipment characteristics.

[0035] The equipment operation interface construction unit is configured to call the corresponding control from the control library according to the signal parameter type and configure the control interface to build the electromagnetic equipment operation interface; the control library includes basic controls, display controls and pane controls.

[0036] It is understandable that the equipment operation interface construction unit starts after the waveform signal design is completed. The AI ​​agent module 19 analyzes the parameter data types of the signal flow, including complex I / Q in-phase quadrature data, real power values, and integer status codes. For different data types, the AI ​​agent module 19 can automatically select appropriate controls. For example, for complex I / Q in-phase quadrature data, it automatically selects constellation diagram and I / Q trajectory diagram controls; for time-varying signals, it automatically selects real-time line graph controls; for control commands, it automatically generates start / stop buttons and parameter slider controls. Control sources can include basic controls, display controls, and other related controls. Basic controls can include, but are not limited to, input boxes, dropdown lists, labels, dates, buttons, menus, knobs, rich text, sliders, and switches. Display controls can include, but are not limited to, dividers, time-domain graphs, spectrum graphs, waterfall graphs, constellation graphs, maps, and console controls. Pane controls can include, but are not limited to, dividers, layout blocks, and tables.

[0037] The AI ​​agent module 19 can automatically arrange control positions and adjust window sizes based on user-preset operating habits, such as prioritizing the display of spectrum diagrams (refer to the layout example provided to the AI ​​agent). The system automatically generates backend control interface code, mapping the control actions on the interface to commands issued by the waveform engine, thus completing the WYSIWYG interface design. After the operation interface is built, it can be saved and published to complete the operation interface design of the electromagnetic equipment.

[0038] The parameter types output by the waveform signal design function unit determine the control categories that the operation interface construction module needs to select; and the real-time adjustments of signal parameters by the user in the operation interface construction module can be transmitted back to the waveform signal design function unit through the control interface to achieve dynamic fine-tuning of parameters. The AI ​​intelligent agent module 19 plays a bridging role between the two, automatically completing the interface construction based on the signal design results, and continuously optimizing the recommendation strategy based on the user's operating habits, thereby realizing an integrated intelligent construction process from waveform design to operation interface.

[0039] In one embodiment, the equipment decomposition module 13 is layered and decoupled and may include three parts: a logic mapping unit, an interface standardization unit, and a metadata encapsulation unit, in order to solve the problem of opaque signal processing links in traditional electromagnetic equipment and the inability to independently observe each processing node; after the logic mapping unit completes the functional decomposition, the interface standardization unit defines a standard data format for the decomposed modules, and the metadata encapsulation unit automatically associates context labels with each module based on this.

[0040] Furthermore, the graphical display components connected in the bypass branch are configured to build real-time data pivot windows, such as... Figure 2 As shown, the real-time data pivot window can include: a time-domain view, used to plot waveform amplitude changes over time in real time, presenting pulse characteristics, noise floor, and dynamic range; a frequency-domain view, used to generate spectrum diagrams and waterfall plots, presenting signal bandwidth, interference source location, and frequency stability; and a high-order analysis view, used to generate constellation diagrams, power spectral density curves, and eye diagrams, displaying modulation quality and bit error rate trends.

[0041] Specifically, the logic mapping unit is configured to automatically identify and break down core sub-tasks based on the operational logic and signal flow of the electromagnetic equipment. When the user inputs a top-level logic description of the equipment, such as a radar signal processing link, the AI ​​agent module 19 analyzes its operational logic and automatically breaks it down into independent functional components such as radio frequency front-end reception, digital down-conversion (DDC), and pulse compression. Each functional component corresponds to a specific link in the signal processing link, and the connection relationships between the components fully reflect the original equipment's signal processing flow.

[0042] The standardized interface unit configuration defines standard input and output data formats for each functional module. These formats include parameters such as data structure type, sampling rate, bit width, and protocol format. For example, for the digital down-conversion functional component, its input is defined as a 16-bit I / Q in-phase quadrature complex stream from the RF front-end, with a sampling rate of 40MHz; its output is defined as a baseband I / Q in-phase quadrature data stream after down-conversion and decimation processing, with the sampling rate reduced to match the rate of the subsequent pulse compression functional component. This standardized interface definition ensures data interoperability between different functional modules, enabling seamless integration of modules from different designers.

[0043] The metadata encapsulation unit is configured to automatically associate context tags with each functional module and supports parallel output of metadata streams and real-time waveform streams in a defined format. Context tags include timestamp synchronization information, sampling frequency, quantization bits, center frequency, and module running status code, and are set in JSON or XML format. When each functional module starts, the AI ​​agent module 19 assists in generating a metadata configuration file in JSON or XML format containing the above context tags, ensuring that the waveform stream and metadata stream are output in parallel. The parallel output design allows the backend log system to simultaneously record the signal data itself and its running context, providing a complete structured data foundation for subsequent end-to-end data traceability.

[0044] In the bypass branches of key processing nodes of each functional module, graphical display components can be integrated to construct real-time data pivot windows. The AI ​​agent module 19 automatically recommends and inserts the most suitable graphical analysis components based on the signal processing characteristics of the currently decomposed module. For example, it automatically selects a waterfall plot for broadband signals and automatically configures a time-domain waveform view for pulse signals. The AI ​​agent module 19 automatically completes the connection mapping between each bypass channel and the main data stream, as well as the adaptive setting of coordinate axis parameters, achieving plug-and-play deployment of observation windows.

[0045] Real-time data pivot windows can include three view types. The time-domain view plots waveform amplitude changes over time, visually presenting pulse characteristics, noise floor, and dynamic range. The frequency-domain view generates high-resolution spectrograms and waterfall plots, clearly showing signal bandwidth usage, interference source locations, and frequency stability. The high-order analysis view supports constellation diagram analysis, power spectral density analysis, and eye diagram analysis, directly reflecting modulation quality and bit error rate trends. These three views are effectively complementary; for example, the time-domain view reflects the transient characteristics of the signal, the frequency-domain view reflects the signal's spectral distribution, and the high-order analysis view evaluates signal performance from a modulation quality perspective. Together, they constitute a complete evaluation system for signal processing quality.

[0046] It's important to note that by decoupling the signal processing chain into independent, accessible nodes, bypass branches can access and acquire valid raw data at the precise node locations. Furthermore, the multi-dimensional graphical display provides an intuitive visual verification method for each functional module after decoupling, helping users determine the accuracy of the module decomposition and the rationality of the data interface definitions. This structured decomposition and visual feedback mutually reinforce each other, jointly achieving the transformation from black-box processing to white-box observation.

[0047] In one embodiment, the equipment diagnostic module 15 may include two parts: a full-link data tracing unit and an intelligent diagnostic unit. This addresses the problems of traditional electromagnetic equipment fault diagnosis, which relies on manual point-by-point troubleshooting, is inefficient, and makes it difficult to identify related faults. The two units work closely together to achieve data-driven diagnostic reasoning. Specifically, the full-link data tracing unit provides structured and comparable input and output data for diagnosis, while the intelligent diagnostic unit performs automated anomaly detection and root cause localization based on this data.

[0048] Specifically, the end-to-end data traceability unit is configured to encapsulate each processing module within the equipment as an independent data node. It collects the original in-phase and quadrature I / Q waveform streams and associated metadata of the input and output in real time through a bypass interface, visualizes and compares the input and output data side by side, and automatically stores historical data snapshots of key nodes. The associated metadata may include timestamps and sampling rates.

[0049] The processing module may include, but is not limited to, existing signal processing basic units such as filtering modules, mixing modules, and modulation modules. The system acquires the raw I / Q in-phase and quadrature waveforms and associated metadata of each data node in real time via a bypass interface. Each node is assigned a unique data node identifier, enabling precise tracking of the signal flow throughout the entire link.

[0050] The core capability of the end-to-end data traceability unit is to support the visual side-by-side comparison of input and output data. After the user selects a node (such as a mixer), the AI ​​agent module 19 automatically retrieves the signal data before input and the signal data after output for that node, displaying them side-by-side in the same view. The system automatically calculates and highlights the spectral component changes, amplitude gain curve deviations, and phase drift trajectories of the two data points, intuitively presenting the specific changes that occur to the signal after passing through the processing module. For example, if the output spectrum of a filter node contains spurious components that are not present at the input, the side-by-side comparison view will automatically mark the frequency position and amplitude of the spurious components, helping the user quickly locate the source of signal distortion.

[0051] The end-to-end data traceability unit can also automatically store historical data snapshots of key nodes. The system can maintain a circular buffer to continuously record the data status of each node over a recent period. Users can review the signal status at any historical moment at any time, facilitating the reproduction of fault scenarios or long-term performance trend analysis. Comparison between historical data snapshots and real-time data further provides a comprehensive view of how signal performance changes over time.

[0052] The intelligent diagnostic unit is configured to automatically identify and mark abnormal modules based on preset threshold rules and real-time indicators, perform root cause localization calculations in conjunction with data flow logic, and generate parameter adjustment schemes based on the diagnostic results and the linkage waveform design engine.

[0053] Furthermore, the intelligent diagnostic unit is also configured to: trigger an abnormal event when the signal-to-noise ratio decreases by more than a preset threshold, call the fault knowledge graph and the current data stream features to perform reverse reasoning operations, infer the physical fault source and generate a confidence score, and then generate a quantitative assessment report including symptoms, causes and solutions.

[0054] It is understandable that real-time metrics may include, for example, signal-to-noise ratio (SNR), frequency offset, and error vector magnitude (EVM). When any metric exceeds the preset threshold range for its corresponding metric, the intelligent diagnostic unit automatically marks the module as abnormal. Anomaly marking is not only targeted at single points but can also be combined with data flow logic analysis to determine whether the anomaly is caused by propagation from an upstream anomaly, thereby distinguishing between primary and secondary faults.

[0055] The intelligent diagnostic unit further combines data flow logic to perform root cause localization calculations. When high-frequency spurious interference occurs at the output, the intelligent diagnostic unit does not simply mark the output as abnormal, but traces the signal link in reverse to analyze the relationship between the cutoff frequency setting of the front-end filter (e.g., 40MHz) and the signal bandwidth (e.g., 50MHz). If it is found that the filter's cutoff frequency is lower than the signal bandwidth, causing a mismatch in filtering characteristics, the root cause is determined to be improper front-end filter parameter settings rather than saturation of the subsequent amplifier. This reverse reasoning mechanism based on link topology allows fault localization to go from the symptoms to the root cause.

[0056] When a decrease in the signal-to-noise ratio (SNR) exceeds a preset threshold, such as -3dB, the intelligent diagnostic unit immediately triggers an abnormal event, invoking the fault knowledge graph and performing reverse reasoning operations based on the current data stream characteristics. The fault knowledge graph stores the correlations between symptoms, causes, and solutions in the electromagnetic equipment field. For example, it records the causal relationship between high-frequency spurious emissions and the mismatch of the front-end filter cutoff frequency, as well as the causal relationship between a specific frequency offset and the frequency drift of the local oscillator. Through graph reasoning, the intelligent diagnostic unit can infer the physical fault source and provide a confidence score; for example, a 92% confidence score indicates that the current anomaly is most likely caused by a specific reason. After diagnosis, the system directly generates a quantitative evaluation report containing symptom descriptions, root cause analysis, and recommended solutions. Based on the diagnostic results, it also triggers a parameter adjustment scheme through the waveform design engine, such as suggesting increasing the filter order from 4th to 8th. After user confirmation, the parameter adjustment can be applied with a single click, and the system automatically simulates and verifies the expected effect of the parameter adjustment, thus forming a complete closed loop from observation to diagnosis to correction.

[0057] The aforementioned end-to-end data tracing provides a precise data foundation for intelligent diagnosis, enabling diagnostic reasoning to be based on measured signal characteristics rather than indirect inference. Conversely, the analysis results of intelligent diagnosis guide the focus of end-to-end data tracing, allowing the system to increase data collection density and snapshot frequency for nodes with detected anomalies. Data tracing ensures that diagnosis is based on evidence, while intelligent diagnosis ensures that tracing is directional; the two support each other, jointly achieving a technological upgrade from manual investigation to automated intelligent diagnosis.

[0058] In one embodiment, the equipment training module 17 may include a distributed simulation architecture unit and a multi-dimensional training mode unit to address the problems of traditional electromagnetic equipment training systems having fixed scripts, being unable to adapt to dynamically changing environments, and being disconnected from actual applications. The distributed simulation architecture unit provides the basic operating environment for training, on which the multi-dimensional training mode unit executes specific training tasks.

[0059] The distributed simulation architecture unit is configured to support multiple digital twin devices in autonomously constructing dynamic network topologies in virtual space. Each digital twin device can correspond to a real electromagnetic equipment, and the devices can communicate and interact with each other through virtual wireless channels. The built-in Mobile Ad Hoc Network (MANET) protocol stack model calculates link status, hop count routing, and bandwidth fluctuations between nodes in real time, and connects individual digital twins to external joint training systems through standard interfaces.

[0060] In the MANET protocol stack model, changes in node location and movement speed lead to dynamic changes in link quality. The protocol stack model updates the routing table in real time to adapt to the continuous evolution of the network topology, simulating collaborative communication scenarios in a decentralized architecture. Standard interfaces follow distributed simulation standards such as High-Level Architecture (HLA) or Data Distribution Service (DDS) to achieve state synchronization and interaction between heterogeneous systems. Through standard interface adapters, the state data of external physical systems and the virtual equipment data within the digital twin are spatiotemporally aligned, including time synchronization to Network Time Protocol (NTP / PTP time synchronization) with microsecond-level accuracy and pose coordinate transformation, ensuring state consistency in hybrid exercises.

[0061] The multi-dimensional exercise mode unit is configured to perform single-equipment operation exercises, formation coordination exercises, and game theory simulations. In essence, the multi-dimensional exercise mode unit supports three levels of exercise modes: single-equipment operation exercises focus on the logical verification of the equipment's internal systems, supporting parameter fine-tuning and fault injection testing of subsystems such as radar and communication terminals, and recording operational response latency and execution success rate. Formation coordination exercises, based on a multi-agent reinforcement learning architecture, simulate formation maintenance, dynamic reorganization, and collaborative task execution of squads. The system automatically calculates collaborative quality indicators such as relative distance error, angle deviation, and collaborative communication latency. Game theory simulations construct a task environment based on game theory, which can introduce AI adversaries to support adaptive task action strategy generation and real-time evaluation of both sides' situational awareness capabilities and decision-making efficiency.

[0062] In the game simulation mode, the player's digital twin, based on a reinforcement learning model, dynamically adjusts its formation spacing and communication frequency according to the environmental situation. The opposing AI agent, acting as an intelligent adversary, does not have a pre-set fixed script but generates adaptive interference strategies in real time based on the radiation characteristics of the opponent's digital twin. These strategies include noise suppression that varies with power and deceptive interference targeting specific waveforms.

[0063] After the simulation, the AI ​​agent module 19 automatically calculates key performance indicators such as task completion rate and communication connectivity, generating a debriefing report. The report identifies the root causes of failures, such as communication interruptions caused by the opponent's frequency-agile interference dropping the signal-to-noise ratio (SNR) below -15dB at a specific point in the simulation, and recommends adjustments to task action strategies, such as enabling frequency hopping anti-jamming mode or changing the formation structure. All status frames, signaling interaction logs, and environmental interference parameter sequences from the entire simulation process are saved as structured data files (e.g., .log or .h5 format) for post-simulation analysis and algorithm training.

[0064] The dynamic topology construction capability provided by the aforementioned distributed simulation architecture unit enables formation collaborative exercises and game simulations to be conducted under near-realistic network conditions. Meanwhile, the task data generated in the multi-dimensional exercise mode is fed back to the distributed simulation architecture unit to optimize network topology generation and channel model parameters. The architecture provides the runtime environment for the modes, while the modes provide optimization feedback to the architecture; the two mutually promote each other, jointly achieving the technological upgrade from static script simulation to dynamic intelligent game simulation.

[0065] Compared with existing technologies, the above-mentioned design of this invention significantly lowers the development threshold, achieving zero-code or low-code intelligent equipment design. Through the semantic parsing and automatic component matching capabilities of the AI ​​agent module 19, even operators without in-depth signal processing expertise can complete complex electromagnetic equipment waveform design and interface construction. It achieves end-to-end white-box visualization of equipment operation status, completely solving the black-box problem of debugging. Through the layered decoupling design of the equipment decomposition modules and the bypass visualization observation mechanism, the opaque signal processing process inside traditional electromagnetic equipment is transformed into a transparent link that can be visualized and observed node by node. Users can view the specific changes in the signal before and after each processing node in real time.

[0066] A knowledge graph-based closed-loop intelligent diagnostic system was constructed, significantly shortening fault diagnosis time. Through the collaborative work of the end-to-end data traceability unit and intelligent diagnostic unit in the equipment diagnostic module, a complete diagnostic chain is achieved, from automatic anomaly detection and intelligent root cause localization to closed-loop optimization suggestions, transforming the traditional manual troubleshooting mode into an automated intelligent diagnostic mode. This effectively enhances the realism of task action simulation and the intelligence level of strategy iteration. Through the distributed simulation architecture and AI opponent game mechanism of the equipment exercise module, adaptive strategy generation and real-time performance evaluation are supported, forming a data-driven task action optimization closed loop. It possesses high standardization and interoperability capabilities, supporting complex system collaboration. Through standardized data format definitions and standard interface support, collaborative training between different manufacturers and equipment at different levels is achieved under a unified spatiotemporal benchmark, breaking down data barriers between traditional heterogeneous systems.

[0067] Each module in the aforementioned software-defined intelligent electromagnetic equipment digital twin device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of a device with data processing capabilities, or stored in software within the memory of the aforementioned device, so that the processor can call and execute the operations corresponding to each module. The aforementioned device can be, but is not limited to, various types of computer equipment already existing in the field.

[0068] In one embodiment, such as Figure 3 As shown, a method for creating a digital twin of intelligent electromagnetic equipment based on the above-described software-defined intelligent electromagnetic equipment digital twin device is also provided, which may include the following steps S10 to S18: S10 uses software radio as its signal processing foundation to design and generate waveform signals and construct the operation interface according to the waveform signal design requirements of electromagnetic equipment.

[0069] The process begins by receiving the user's input of equipment waveform signal design requirements. Semantic parsing is then performed to extract key parameters, which may include the operating frequency band, signal bandwidth, and modulation method. Based on the extracted parameters, the corresponding waveform components are retrieved from the waveform component library. Data links are established between the components to execute data flow transmission, and the parameters of the waveform components are set. After the waveform signal design is completed, the corresponding controls are retrieved from the control library based on the signal parameter data type, and the control interface is configured to build an operation interface that matches the waveform signal.

[0070] S12 decouples the core functions of electromagnetic equipment by performing layered decoupling according to the signal processing link and connects graphical display components to the bypass branches of each functional module to build a real-time data perspective window.

[0071] Understandably, the system can automatically identify and break down core sub-tasks based on the business logic and signal flow of electromagnetic equipment. Standard input / output data formats are defined for each sub-module, including data structure type, sampling rate, bit width, and protocol format. Context tags are automatically associated with each module; these tags can include timestamp synchronization information, sampling frequency, quantization bits, center frequency, and module running status code. A metadata stream in JSON or XML format is generated and output in parallel with the real-time waveform stream. A graphical display component integrated in the bypass branch is used to generate time-domain views, frequency-domain views, and high-order analysis views.

[0072] S14: Construct a fault knowledge graph and expert rule base for electromagnetic equipment, and perform source tracing, comparison and intelligent diagnosis of full-link data.

[0073] It is understandable that each processing module within the equipment can be encapsulated as an independent data node. Through a bypass interface, the raw I / Q in-phase quadrature waveform streams and associated metadata of the input and output can be acquired in real time. Visualized side-by-side comparison of input and output data is performed, and historical data snapshots of key nodes are automatically stored. Based on preset threshold rules and real-time indicators, abnormal modules are automatically identified and marked. Root cause localization calculations are performed in conjunction with data flow logic, and parameter adjustment schemes are generated based on the diagnostic results. When a decrease in the signal-to-noise ratio (SNR) exceeds a preset threshold, an abnormal event is triggered. A fault knowledge graph is invoked, combined with the current data flow characteristics, to perform reverse reasoning calculations, inferring the physical fault source and generating a confidence score.

[0074] S16: Construct a distributed simulation environment to execute multi-mode drills and verify task action strategies.

[0075] It is understandable that it can support multiple digital twin devices to autonomously construct dynamic network topologies in virtual space, calculate link status, hop count routing, and bandwidth fluctuations between nodes in real time, and support single-device digital twins to access external joint training systems through standard interfaces. It can execute three levels of training modes: single-device operation drills, formation coordination drills, and game theory simulations, to verify the effectiveness of task action strategies.

[0076] S18 uses the AI ​​intelligent agent module to provide intelligent assistance in waveform signal design and generation, operation interface construction, hierarchical decoupling, data traceability and comparison, intelligent diagnosis, and collaborative training processes.

[0077] It is understandable that the AI ​​agent module can monitor the running status of each process and the system resource load in real time, and automatically plan the task execution sequence based on the current status. When a user initiates a new task, the AI ​​agent module automatically breaks down the task chain and coordinates the execution order of each process.

[0078] The aforementioned intelligent electromagnetic equipment digital twin method, by constructing an intelligent decision-making system that integrates real-time perception across the entire chain, high-fidelity simulation of multi-physics fields, dynamic software reconstruction, and knowledge-driven approaches, deeply integrates the characteristics of physical equipment with software-defined capabilities. With the help of AI digital equipment intelligent agents, it intelligently assists users in completing the entire lifecycle management from design, training, operation and maintenance to reconstruction at every stage of the digital twin device. This solves the engineering problems of existing electromagnetic equipment, such as weak adaptability in complex environments, high maintenance costs, and poor training results, and achieves continuous optimization of equipment performance and refined management throughout the entire lifecycle.

[0079] By employing a technical approach that integrates software-defined radio infrastructure, a layered decoupling architecture, and deep AI-enabled intelligence, we have successfully addressed the real-world pain points in traditional electromagnetic equipment development, such as long design cycles, high debugging thresholds, difficulty in fault location, and rigid simulation environments. This approach transforms the traditional reliance on physical hardware and human experience into a new engineering model driven by data, intelligent decision-making, and closed-loop optimization, significantly improving the efficiency and intelligence level of the entire lifecycle management of electromagnetic equipment.

[0080] It is understood that the specific limitations of the features in the various embodiments of the above-mentioned intelligent electromagnetic equipment digital twin method can be found in the explanation of the corresponding features in the various embodiments of the software-defined intelligent electromagnetic equipment digital twin device above, and will not be repeated here.

[0081] It should be understood that, although Figure 3 The steps are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order in which these steps are executed; they can be performed in other orders. Figure 3 At least some of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0082] In one embodiment, a computer device is also provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the intelligent electromagnetic equipment digital twin method in the above embodiments.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of protection of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and all such modifications and improvements fall within the scope of protection of the present invention.

Claims

1. A software-defined intelligent electromagnetic equipment digital twin device, characterized in that, Software-defined radio as the signal processing foundation and including: The equipment simulation module is configured to be based on a software radio architecture and to perform waveform signal design and generation and operation interface construction according to the waveform signal design requirements of electromagnetic equipment. The equipment decomposition module is configured to decouple the core functions of electromagnetic equipment in layers according to the signal processing link and connect the graphical display component to the bypass branch of each functional module; the graphical display component is used to build a real-time data perspective window; The equipment diagnosis module is configured to build a fault knowledge graph and expert rule base for electromagnetic equipment, and perform source tracing and intelligent diagnosis of full-link data. The equipment exercise module is configured to build a distributed simulation environment and perform multi-mode exercises and mission action strategy verification. Among them, the equipment simulation module generates waveform signals and operation interface according to the waveform signal design requirements of electromagnetic equipment. The equipment decomposition module transparently displays the internal signal processing link of electromagnetic equipment. The equipment diagnosis module performs real-time monitoring and fault diagnosis based on the transparent data link. The equipment exercise module uses digital twins to carry out multi-dimensional exercise verification. The AI ​​intelligent agent module is communicatively connected to the equipment simulation module, equipment disassembly module, equipment diagnosis module, and equipment exercise module. It is configured to provide intelligent assistance in waveform signal design and generation, operation interface construction, hierarchical decoupling, data traceability and comparison, intelligent diagnosis, and exercise process coordination. The operating data of the equipment simulation module, equipment disassembly module, equipment diagnosis module, and equipment exercise module are input into the AI ​​intelligent agent module for analysis in real time. The analysis results of the AI ​​intelligent agent module are fed back to each module in real time to guide the next operation of each module, forming a real-time interactive intelligent closed loop.

2. The software-defined intelligent electromagnetic equipment digital twin device according to claim 1, characterized in that, The equipment simulation module includes: The waveform signal design function unit is configured to be based on a software radio architecture. It calls the waveform component library and the visual waveform design engine, selects waveform components according to the waveform signal design requirements of electromagnetic equipment, establishes data links to execute the transmission of data flow between waveform components, and sets the parameters of waveform components. The equipment operation interface construction unit is configured to call the corresponding control from the control library according to the signal parameter type and configure the control interface to build the electromagnetic equipment operation interface; the control library includes basic controls, display controls and pane controls.

3. The software-defined intelligent electromagnetic equipment digital twin device according to claim 2, characterized in that, The AI ​​intelligent agent module is configured to perform semantic parsing on the input equipment waveform signal design requirements during waveform signal design, automatically call the corresponding waveform component from the waveform component library, and complete the connection establishment and parameter preset between waveform components. The AI ​​intelligent agent module is also configured to automatically call controls that support the corresponding data type based on the parameter data type of the equipment waveform signal when the operation interface is built, and to automatically adjust the layout and attributes of the controls according to preset operation habits.

4. The software-defined intelligent electromagnetic equipment digital twin device according to any one of claims 1 to 3, characterized in that, The equipment disassembly module is layered and decoupled and includes: The logic mapping unit is configured to automatically identify and split core sub-tasks based on the business logic and signal flow of the electromagnetic equipment. The interface standardization unit is configured to define standard input and output data formats for each functional module; the data format includes data structure type, sampling rate, bit width, and protocol format; The metadata encapsulation unit is configured to automatically associate context tags with each functional module and supports parallel output of metadata streams and real-time waveform streams in a set format. The context tags include timestamp synchronization information, sampling frequency, quantization bits, center frequency, and module running status code, and can be set in JSON or XML format.

5. The software-defined intelligent electromagnetic equipment digital twin device according to claim 4, characterized in that, The graphical display component accessed in the bypass branch is configured to build a real-time data pivot window; the real-time data pivot window includes: The time-domain view is used to plot the waveform amplitude change curve over time in real time, presenting pulse characteristics, noise floor, and dynamic range. Frequency domain view, used to generate spectrum and waterfall plots, showing the signal bandwidth occupied, location of interference sources, and frequency stability; Advanced analysis views are used to generate constellation diagrams, power spectral density curves, and eye diagrams, showing modulation quality and bit error rate trends.

6. The software-defined intelligent electromagnetic equipment digital twin device according to claim 1, characterized in that, The equipment diagnostic module includes: The end-to-end data traceability unit is configured to encapsulate each processing module within the equipment as an independent data node. It collects the original in-phase and quadrature I / Q waveform streams and associated metadata of the input and output in real time through a bypass interface, visualizes and compares the input and output data side by side, and automatically stores historical data snapshots of key nodes. The associated metadata includes timestamps and sampling rates. The intelligent diagnostic unit is configured to automatically identify and mark abnormal modules based on preset threshold rules and real-time indicators, perform root cause localization calculations in conjunction with data flow logic, and generate parameter adjustment schemes based on the diagnostic results and the linkage waveform design engine.

7. The software-defined intelligent electromagnetic equipment digital twin device according to claim 6, characterized in that, The intelligent diagnostic unit is also configured as follows: When the signal-to-noise ratio drop exceeds a preset threshold, an abnormal event is triggered. The fault knowledge graph and current data stream features are invoked to perform reverse reasoning operations, infer the physical fault source, generate a confidence score, and then generate a quantitative assessment report including symptoms, causes, and solutions.

8. The software-defined intelligent electromagnetic equipment digital twin device according to claim 1, characterized in that, The equipment training module includes: The distributed simulation architecture unit is configured to support multiple digital twin devices to autonomously construct dynamic network topology in virtual space. It calculates the link status, hop count routing and bandwidth fluctuations between nodes in real time through the built-in mobile ad hoc network protocol stack model, and connects the single-device digital twin to the external joint training system through standard interfaces. The multi-dimensional exercise mode unit is configured to perform single-unit operation exercises, formation coordination exercises, and game simulations, respectively.

9. A method for creating a digital twin of intelligent electromagnetic equipment based on a software-defined intelligent electromagnetic equipment digital twin device according to any one of claims 1 to 8, characterized in that, Including the following steps: Using software-defined radio as the signal processing foundation, the design and generation of waveform signals and the construction of the operation interface are performed according to the waveform signal design requirements of electromagnetic equipment. The core functions of electromagnetic equipment are decoupled in layers according to the signal processing link, and graphical display components are connected to the bypass branches of each functional module. Construct a fault knowledge graph and expert rule base for electromagnetic equipment, and perform source tracing, comparison and intelligent diagnosis of full-link data; Build a distributed simulation environment to perform multi-mode drills and verify task action strategies; The AI ​​intelligent agent module provides intelligent assistance in waveform signal design and generation, operation interface construction, hierarchical decoupling, data traceability and comparison, intelligent diagnosis, and collaborative training processes.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the digital twin method for intelligent electromagnetic equipment as described in claim 9.

Images