Conduction disturbance and immunity integrated test system in complex electromagnetic environment

Abstract

The invention relates to the technical field of electromagnetic environment simulation, in particular to a conducted disturbance and immunity comprehensive test system in a complex electromagnetic environment, which comprises a signal generation module, a holographic electromagnetic wave processing module, a conduction path simulation module, a monitoring and analysis module and a control and management module. Through AI dynamic optimization and a programmable metasurface technology, real-time regulation and control of phased array excitation weight and electromagnetic wave front are realized, through combination of digital twin pre-calculation and multi-physics field coupling modeling, spatial characteristics and interference propagation paths of a complex electromagnetic environment can be efficiently and accurately simulated, authenticity and reliability of testing are improved, and the method is suitable for large-scale popularization and application. A high-precision conduction path model can be constructed by utilizing a programmable impedance network and a modular configuration architecture in combination with a non-intrusive interference injection and data driving optimization technology, an electromagnetic conduction environment is vividly simulated, and a reliable environment basis is provided for testing.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic environment simulation, and in particular to a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment. Background Art

[0002] The electromagnetic environment refers to the comprehensive environment within a specific space, composed of all electromagnetic phenomena (including natural and man-made electromagnetic radiation, electromagnetic fields, and electromagnetic signals). With the rapid development of modern electronic technology, electronic equipment and systems are becoming increasingly dense and complex, and electromagnetic interactions between different devices are becoming more frequent. This results in the actual electromagnetic environment being characterized by multi-band interweaving, multi-source interference superposition, and rapid dynamic changes. This complex electromagnetic environment places extremely high demands on the electromagnetic compatibility of electronic equipment, requiring precise test systems to evaluate the device's immunity and interference characteristics in real-world scenarios.

[0003] However, in the existing technology, electromagnetic compatibility test systems have many problems when simulating complex electromagnetic environments: on the one hand, the interference signal generation method is relatively simple, and it is difficult to dynamically reproduce complex interference scenarios with multiple frequency bands and multiple modulation methods, and cannot meet diversified testing needs; on the other hand, in terms of simulating electromagnetic space characteristics, traditional technologies rely on static models and lack the ability to real-time control the propagation path, phase and polarization state of electromagnetic waves, resulting in low simulation efficiency; in addition, the simulation accuracy of the conduction path is insufficient, and there is a general lack of real-time dynamic optimization mechanisms based on measured data, which makes the test results deviate greatly from the actual environment, making it difficult to meet the increasingly stringent testing standards requirements of multiple industries such as aerospace, automotive electronics, and industrial automation.

[0004] Based on this, the present invention provides a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a comprehensive test system for conducted interference and immunity in a complex electromagnetic environment. The software-defined radio architecture and intelligent interference generation algorithm of the present invention can dynamically generate composite interference signals with multiple frequency bands and multiple modulation modes, support the import of custom scenarios, and accurately simulate the interference conditions in various complex electromagnetic environments to meet different testing requirements. Based on AI dynamic optimization and programmable metasurface technology, real-time regulation of phased array excitation weights and electromagnetic wavefronts is achieved. Combined with digital twin pre-calculation and multi-physics field coupling modeling, the spatial characteristics and interference propagation paths of complex electromagnetic environments can be simulated efficiently and accurately, thereby improving the authenticity and reliability of the test. By utilizing programmable impedance networks and modular configuration architectures, combined with non-invasive interference injection and data-driven optimization technology, a high-precision conduction path model can be constructed to realistically simulate the electromagnetic conduction environment, providing a reliable environmental foundation for testing.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment, including a signal generation module, a holographic electromagnetic wave processing module, a conduction path simulation module, a monitoring and analysis module, and a control and management module, wherein:

[0008] The signal generation module is used to dynamically generate electromagnetic interference signals with multiple frequency bands and multiple modulation modes through a software-defined radio architecture combined with an intelligent interference generation algorithm;

[0009] The holographic electromagnetic wave processing module: Based on AI dynamic optimization and programmable metasurface technology, it uses deep learning to control the phased array excitation weights and electromagnetic wavefront in real time, and combines digital twin pre-calculation and multi-physics field coupling modeling;

[0010] The conduction path simulation module: Based on a programmable impedance network and modular configuration architecture, it builds a conduction path model that supports multiple industry standards and performs high-precision simulation of the electromagnetic conduction environment through non-intrusive interference injection and data-driven optimization technology;

[0011] The monitoring and analysis module is used to monitor and record the impact of electromagnetic disturbances and the response of the equipment during the test in real time, and provide data analysis functions;

[0012] The control and management module is used to execute the test according to the predetermined plan and provide the operator with a user interface to set parameters, start / stop the test and view the results.

[0013] The signal generation module includes a software-defined radio unit, an intelligent interference generation algorithm unit, and a radio frequency up-conversion and power amplification unit, wherein:

[0014] The software-defined radio unit: implements digital signal processing based on FPGA / ASIC chips to generate multi-band, multi-modulation baseband signals;

[0015] The intelligent interference generation algorithm unit is used to dynamically generate a composite interference signal using a machine learning algorithm and supports the import of custom scenarios;

[0016] The radio frequency up-conversion and power amplification unit is used to up-convert the baseband signal to the target frequency band and output an electromagnetic interference signal with adjustable amplitude after power amplification.

[0017] The intelligent interference generation algorithm unit uses a machine learning algorithm to dynamically generate a composite interference signal and supports the import of custom scenarios. The specific operations are as follows:

[0018] A1: Scenario parameter analysis: Receive interference characteristic parameters imported from custom scenarios, such as frequency range f min -f max , modulation type, pulse repetition frequency PRF, and construct the interference parameter probability distribution function P(x) based on historical test data, where x is the variable of interference amplitude and phase;

[0019] A2: Signal pre-generation based on generative adversarial networks:

[0020] ①Build a GAN model, whose loss function is:

[0021]

[0022] ② Use the generator network G to generate the initial interference signal S init (t), optimize the signal authenticity through the discriminator network D, and iterate the training until convergence;

[0023] A3: Reinforcement learning dynamic adjustment:

[0024] ① Define the state space S as the real-time response data of the device under test, such as bit error rate and voltage fluctuation, and the action space A as the interference signal adjustment parameters, such as amplitude adjustment step ΔA and frequency offset Δf;

[0025] ② Based on the deep Q-network algorithm, the action value is evaluated through the reward function R(s,a):

[0026]

[0027] A4: Composite signal generation: The pre-generated signal S init (t) and the signal S after RL adjustment adj (t) Perform weighted fusion to generate the final composite interference signal:

[0028] S final (t):S final (t) = α·S init (t)+(1-α)·S adj (t)

[0029] Among them, the weight coefficient α is adaptively adjusted according to the complexity of the scene.

[0030] The holographic electromagnetic wave processing module includes a programmable metasurface array unit, an AI dynamic optimization and deep learning unit, and a digital twin pre-calculation unit, wherein:

[0031] The programmable metasurface array unit is composed of thousands of electromagnetic units, which are used to control the phase, polarization and amplitude distribution of electromagnetic waves in real time to simulate the spatial characteristics of complex electromagnetic environments;

[0032] The AI dynamic optimization and deep learning unit is used to analyze monitoring data through the CNN network, generate a phased array excitation weight matrix, dynamically adjust the electromagnetic wavefront, and achieve holographic reconstruction of multi-source interference scenarios;

[0033] The digital twin pre-calculation unit is used to store a holographic database of typical electromagnetic scenes and pre-calculate interference propagation paths based on a multi-physics field coupling model.

[0034] The AI dynamic optimization and deep learning unit uses a CNN network to analyze monitoring data, generate a phased array excitation weight matrix, and dynamically adjust the electromagnetic wavefront to achieve holographic reconstruction of multi-source interference scenarios. The specific operations are as follows:

[0035] B1: Data preprocessing: Receive the time-domain voltage waveform V(t) and frequency-domain electric field strength E(f) output by the monitoring and analysis module, and eliminate the dimension effect through normalization:

[0036]

[0037] Where, μv, σv are the mean and standard deviation of voltage, μE, σE are the mean and standard deviation of electric field intensity;

[0038] B2: Feature extraction: Use the 3D convolution layer of CNN to process the frequency domain electric field data E(f) and extract the space-frequency feature map FCNN. The convolution operation is defined as:

[0039]

[0040] Where K is the convolution kernel, and its size is (2a+1)×(2b+1)×(2c+1);

[0041] Using LSTM network to process time domain voltage series The hidden layer state update formula is:

[0042] h t =tanh(W in x t +W hh h t-1 +b h )

[0043] Among them, W ih , W hh is the weight matrix, b h is the bias vector;

[0044] B3: Weight matrix generation: concatenate the spatial features extracted by CNN and the temporal features extracted by LSTM into a joint feature vector F joint , mapped to the phased array excitation weight matrix through the fully connected layer: W:W=σ(W fc ·Fjoint +b fc , where σ is the Sigmoid activation function, W fc , b fc is the fully connected layer parameter;

[0045] B4: Dynamic adjustment feedback: Update the network parameters θ based on the mean square error between the measured and simulated electric field distributions. The loss function is:

[0046]

[0047] The Adam optimizer is used for iterative optimization with a learning rate of η = 1e-4.

[0048] The conduction path simulation module includes a programmable impedance network unit, a modular interface component unit, a non-intrusive interference injection unit, and a data-driven optimization unit, wherein:

[0049] The programmable impedance network unit is used to dynamically simulate the impedance characteristics of the conduction path by being composed of MEMS adjustable elements, with a switching speed of ≤50μs;

[0050] The modular interface component unit is used to provide standardized interfaces for power ports and signal ports, supports quick replacement within 5 minutes, and is compatible with multiple industry testing standards;

[0051] The non-intrusive interference injection unit is used to indirectly inject interference signals into the conduction path through the electromagnetic coupling clamp and the current probe to avoid damaging the original circuit structure of the device under test;

[0052] The data-driven optimization unit automatically optimizes impedance network parameters based on a particle swarm optimization algorithm.

[0053] The data-driven optimization unit uses a particle swarm optimization algorithm to automatically optimize impedance network parameters. The specific operations are as follows:

[0054] C1: Initialize the population parameters of the particle swarm optimization algorithm, including the number of particles, inertia weight and learning factor;

[0055] C2: Establish a fitness function based on the error between the current impedance network output and the target electromagnetic environment simulation;

[0056] C3: Iteratively update the position and velocity of each particle to find the optimal impedance parameter combination;

[0057] C4: Feeding back the optimized impedance parameters to the programmable impedance network unit (301) to achieve dynamic adjustment of the conduction path model.

[0058] In C3, the speed and position are updated by the particle swarm optimization algorithm. The specific formula is:

[0059]

[0060] Among them, v id and x id They represent the velocity and position of the i-th particle in the d-th dimension, w is the inertia weight, ranging from 0.4 to 0.9, c1 and c2 are learning factors, ranging from 1.4 to 2.0, r1 and r2 are random numbers, and p id is the individual optimal solution, g d is the global optimal solution.

[0061] The monitoring and analysis module includes a multi-parameter synchronous acquisition unit, a holographic data analysis unit, and an immunity evaluation unit, wherein:

[0062] The multi-parameter synchronous acquisition unit is used to collect the voltage, current, frequency, digital signal error rate and functional status of the device under test in real time;

[0063] The holographic data analysis unit is used to use electromagnetic topology theory and wavelet packet decomposition algorithm to construct a three-dimensional field distribution model of the conduction path, locate disturbance sensitive points, and generate visualization results of spectrum diagrams and waterfall diagrams;

[0064] The immunity evaluation unit is used to automatically determine the equipment immunity level according to industry standards and output a failure mode analysis report.

[0065] The control and management module includes an automation control unit, a graphical user interface unit, and a data storage and tracing unit, wherein:

[0066] The automated control unit executes a predetermined test process based on state machine logic, supports sequential testing, cyclic testing, and conditional trigger testing, and is compatible with remote control;

[0067] The graphical user interface unit is used to provide a parameter setting interface, a real-time data monitoring dashboard and a test report generation tool, and preset industry test templates for automotive electronics and industrial automation;

[0068] The data storage and traceability unit is used to encrypt and store test raw data, analysis results and operation logs, supports fast retrieval by time / project dimensions, and complies with ISO17025 laboratory management requirements.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The present invention uses software-defined radio architecture and intelligent interference generation algorithm to dynamically generate composite interference signals with multiple frequency bands and multiple modulation modes, supports the import of custom scenarios, and can accurately simulate interference conditions in various complex electromagnetic environments to meet different testing requirements. Based on AI dynamic optimization and programmable metasurface technology, it realizes real-time regulation of phased array excitation weights and electromagnetic wavefronts. Combined with digital twin pre-calculation and multi-physics field coupling modeling, it can efficiently and accurately simulate the spatial characteristics and interference propagation paths of complex electromagnetic environments, thereby improving the authenticity and reliability of the test. By utilizing programmable impedance networks and modular configuration architectures, combined with non-invasive interference injection and data-driven optimization technology, it can construct a high-precision conduction path model, realistically simulate the electromagnetic conduction environment, and provide a reliable environmental foundation for testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a system diagram of a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to the present invention.

[0072] Figure 2 This is a system architecture topology diagram of a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to the present invention.

[0073] Figure 3 This is a workflow timing diagram of a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to the present invention.

[0074] Description of Figure Numbers:

[0075] 100. Signal generation module; 101. Software-defined radio unit; 102. Intelligent interference generation algorithm unit; 103. RF up-conversion and power amplification unit; 200. Holographic electromagnetic wave processing module; 201. Programmable metasurface array unit; 202. AI dynamic optimization and deep learning unit; 203. Digital twin pre-calculation unit; 300. Conduction path simulation module; 301. Programmable impedance network unit; 302. Modular interface component unit; 303. Non-intrusive interference injection unit; 304. Data-driven optimization unit; 400. Monitoring and analysis module; 401. Multi-parameter synchronous acquisition unit; 402. Holographic data analysis unit; 403. Immunity assessment unit; 500. Control and management module; 501. Automation control unit; 502. Graphical user interface unit; 503. Data storage and traceability unit. DETAILED DESCRIPTION

[0076] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0077] Example:

[0078] like Figure 1-Figure 3 As shown, this embodiment provides a comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment, including a signal generation module 100, a holographic electromagnetic wave processing module 200, a conduction path simulation module 300, a monitoring and analysis module 400, and a control and management module 500, wherein: the signal generation module 100 is used to dynamically generate electromagnetic interference signals in multiple frequency bands and multiple modulation modes through a software-defined radio architecture and in combination with an intelligent interference generation algorithm; the holographic electromagnetic wave processing module 200 is based on AI dynamic optimization and programmable metasurface technology, and controls the phased array excitation weight and electromagnetic wavefront in real time through deep learning, and combines Digital twin pre-calculation and multi-physics field coupling modeling; conduction path simulation module 300: based on programmable impedance network and modular configuration architecture, builds a conduction path model that supports multiple industry standards, and uses non-intrusive interference injection and data-driven optimization technology to perform high-precision simulation of the electromagnetic conduction environment; monitoring and analysis module 400: used to monitor and record the impact of electromagnetic interference and the response of the equipment during the test in real time, and provide data analysis functions; control and management module 500: used to execute the test according to the predetermined plan, and provide the operator with a user interface to set parameters, start / stop the test and view the results.

[0079] In this embodiment, it should be noted that: the signal generation module 100 generates a multimodal electromagnetic interference signal, which is precisely controlled by the holographic electromagnetic wave processing module 200 and simulated by the conduction path simulation module 300 for high-fidelity conduction environment simulation. The monitoring and analysis module 400 then collects device response data in real time, and finally realizes intelligent closed-loop control of the test process through the control and management module 500.

[0080] In the present invention, the signal generation module 100 includes a software-defined radio unit 101, an intelligent interference generation algorithm unit 102, and a radio frequency up-conversion and power amplification unit 103, wherein: the software-defined radio unit 101: implements digital signal processing based on an FPGA / ASIC chip to generate a baseband signal with multiple frequency bands and multiple modulation modes; the intelligent interference generation algorithm unit 102: is used to dynamically generate a composite interference signal using a machine learning algorithm and supports the import of custom scenarios; the specific operations are as follows: A1: Scenario parameter parsing: receiving interference feature parameters imported from a custom scenario, such as the frequency range fmin -f max , modulation type, pulse repetition frequency PRF, and construct the interference parameter probability distribution function P(x) based on historical test data, where x is the variable of interference amplitude and phase; A2: Signal pre-generation based on generative adversarial network: ① Construct a GAN model, and its loss function is:

[0081]

[0082] ② Use the generator network G to generate the initial interference signal S init (t), optimize the signal authenticity through the discriminator network D, and iterate the training until convergence; A3: Reinforcement learning dynamic adjustment: ① Define the state space S as the real-time response data of the device under test, such as bit error rate and voltage fluctuation, and the action space A as the interference signal adjustment parameters, such as amplitude adjustment step ΔA and frequency offset Δf; ② Based on the deep Q network algorithm, the reward function R(s,a) is used to evaluate the action value:

[0083]

[0084] A4: Composite signal generation: The pre-generated signal S init (t) and the signal S after RL adjustment adj (t) Perform weighted fusion to generate the final composite interference signal:

[0085] S final (t):S final (t) = α·S init (t)+(1-α)·S adj (t)

[0086] The weight coefficient α is adaptively adjusted according to the complexity of the scene. RF up-conversion and power amplification unit 103 is used to up-convert the baseband signal to the target frequency band and output an electromagnetic interference signal with adjustable amplitude after power amplification.

[0087] In this embodiment, it should be noted that: the software-defined radio unit 101 generates a baseband signal, which is pre-generated by the GAN network of the intelligent interference generation algorithm unit 102 and dynamically optimized by reinforcement learning to form a composite interference signal, which is finally frequency-converted and amplified and output by the RF up-conversion and power amplification unit 103.

[0088] In addition, it is important to note that the multi-channel RF signal generator supports the simultaneous output of at least four independently adjustable RF signals, each with a frequency range of 10kHz to 6GHz and an amplitude accuracy of ≤±0.5dB. The intelligent interference generation algorithm, based on a generative adversarial network, dynamically generates composite interference signals of pulse group interference and RF field-induced conducted interference that comply with the CISPR 22 standard. Multiple frequency bands, such as VHF / UHF / LTE / WiFi / millimeter wave, and various modulation methods, such as AM / FM / PSK / QAM, are supported.

[0089] In the present invention, the holographic electromagnetic wave processing module 200 includes a programmable metasurface array unit 201, an AI dynamic optimization and deep learning unit 202, and a digital twin pre-calculation unit 203, wherein: the programmable metasurface array unit 201: is composed of thousands of electromagnetic units, and is used to control the phase, polarization and amplitude distribution of electromagnetic waves in real time to simulate the spatial characteristics of complex electromagnetic environments; the AI dynamic optimization and deep learning unit 202: is used to analyze monitoring data through a CNN network, generate a phased array excitation weight matrix, dynamically adjust the electromagnetic wave front, and realize holographic reconstruction of multi-source interference scenarios; the specific operations are as follows: B1: Data preprocessing: Receive the time domain voltage waveform V(t) and frequency domain electric field strength E(f) output by the monitoring and analysis module 400, and eliminate the dimensionality effect through normalization processing:

[0090]

[0091] Where μv and σv are the mean and standard deviation of voltage, and μE and σE are the mean and standard deviation of electric field intensity. B2: Feature extraction: The 3D convolutional layer of CNN is used to process the frequency domain electric field data E(f) and extract the space-frequency feature map FCNN. The convolution operation is defined as:

[0092]

[0093] Where K is the convolution kernel, and its size is (2a+1)×(2b+1)×(2c+1);

[0094] Using LSTM network to process time domain voltage series The hidden layer state update formula is:

[0095] h t =tanh(W ih x t +W hh h t-1 +b h )

[0096] Among them, W ih , W hh is the weight matrix, b his the bias vector;

[0097] B3: Weight matrix generation: concatenate the spatial features extracted by CNN and the temporal features extracted by LSTM into a joint feature vector F joint , mapped to the phased array excitation weight matrix through the fully connected layer: W:W=σ(W fc ·F joint +b fc , where σ is the Sigmoid activation function, W fc , b fc is the fully connected layer parameter;

[0098] B4: Dynamic adjustment feedback: Update the network parameters θ based on the mean square error between the measured and simulated electric field distributions. The loss function is:

[0099]

[0100] The Adam optimizer is used for iterative optimization, with a learning rate of η = 1e-4. The digital twin pre-calculation unit 203 is used to store a holographic database of typical electromagnetic scenes and pre-calculate interference propagation paths based on a multi-physics field coupling model.

[0101] In this embodiment, it should be noted that: the scene database and propagation path pre-calculation provided by the digital twin pre-calculation unit 203, the CNN-LSTM hybrid network of the AI dynamic optimization and deep learning unit 202 analyzes the monitoring data in real time and generates an optimized weight matrix, and finally realizes precise control of electromagnetic wave parameters through thousands of electromagnetic units of the programmable metasurface array unit 201.

[0102] In addition, it should be noted that the multi-physics coupled model electromagnetic-thermal-mechanical programmable metasurface array consists of ≥1024 electromagnetic units, each unit supports 0-360° phase control and 0-1 linear polarization control, with a control resolution of ≤1° / 0.01.

[0103] In the present invention, the conduction path simulation module 300 includes a programmable impedance network unit 301, a modular interface component unit 302, a non-invasive interference injection unit 303, and a data-driven optimization unit 304, wherein: the programmable impedance network unit 301 is used to dynamically simulate the impedance characteristics of the conduction path by being composed of MEMS adjustable elements, with a switching speed of ≤50μs; the modular interface component unit 302 is used to provide standardized interfaces for power ports and signal ports, supporting rapid replacement within 5 minutes, and adapting to multi-industry test standards; the non-invasive interference injection unit 303 is used to indirectly inject interference signals into the conduction path through the device of electromagnetic coupling clamps and current probes to avoid damaging the original circuit structure of the device under test; the data-driven optimization unit 304 automatically optimizes the impedance network parameters based on the particle swarm optimization algorithm. The specific operations are as follows: C1: Initialize the population parameters of the particle swarm optimization algorithm, including the number of particles, inertia weight, and learning factor; C2: Establish a fitness function based on the current impedance network output and the target electromagnetic environment simulation error; C3: Iteratively update the position and velocity of each particle to find the optimal impedance parameter combination; In C3, the velocity and position are updated using the particle swarm optimization algorithm. The specific formula is:

[0104]

[0105] Among them, v id and x id They represent the velocity and position of the i-th particle in the d-th dimension, w is the inertia weight, ranging from 0.4 to 0.9, c1 and c2 are learning factors, ranging from 1.4 to 2.0, r1 and r2 are random numbers, and p id is the individual optimal solution, g d C4: Feedback the optimized impedance parameters to the programmable impedance network unit 301 to achieve dynamic adjustment of the conduction path model.

[0106] In this embodiment, it should be noted that: the modular interface component unit 302 adapts to the device under test, the basic impedance characteristics are constructed by the programmable impedance network unit 301, and after the interference signal is coupled by the non-intrusive interference injection unit 303, the impedance parameters are optimized in real time by the particle swarm algorithm of the data-driven optimization unit 304.

[0107] In addition, it should be noted that: impedance characteristics such as cable distribution parameters and connector contact resistance. Adjust the capacitance / resistance / inductance of the components. The programmable impedance network unit 301 is composed of MEMS adjustable capacitors (0.1pF~10nF), thin film resistors (1Ω~1MΩ) and chip inductors (1nH~100μH). Dynamic impedance switching is achieved through FPGA control, and the switching speed is ≤50μs. C2: Establish a fitness function based on the current impedance network output and the target electromagnetic environment simulation error. The specific formula is: f(θ)=||Ε sim (θ)-Ε target || 2 .

[0108] In the present invention, the monitoring and analysis module 400 includes a multi-parameter synchronous acquisition unit 401, a holographic data analysis unit 402, and an immunity evaluation unit 403, wherein: the multi-parameter synchronous acquisition unit 401 is used to collect the voltage, current, frequency, digital signal bit error rate and functional status of the device under test in real time; the holographic data analysis unit 402 is used to use electromagnetic topology theory and wavelet packet decomposition algorithm to construct a three-dimensional field distribution model of the conduction path, locate interference sensitive points, and generate visual results of spectrum diagrams and waterfall diagrams; the immunity evaluation unit 403 is used to automatically determine the device immunity level according to industry standards and output a failure mode analysis report.

[0109] In this embodiment, it should be noted that: the multi-parameter synchronous acquisition unit 401 obtains the multi-dimensional response data of the device under test in real time, and after the electromagnetic topology and wavelet packet algorithm of the holographic data analysis unit 402 perform three-dimensional field reconstruction and sensitive point positioning, the anti-interference evaluation unit 403 automatically generates a quantitative evaluation report according to industry standards.

[0110] In addition, it should be noted that: functional status such as indicator lights, data transmission, voltage (accuracy ±0.1% FS), current (accuracy ±0.2% FS), frequency (accuracy ±1ppm) and digital signal error rate (resolution ≤10 -9 ).

[0111] In the present invention, the control and management module 500 includes an automation control unit 501, a graphical user interface unit 502, and a data storage and traceability unit 503, wherein: the automation control unit 501: executes a predetermined test process based on state machine logic, supports sequential testing, cyclic testing and conditional triggering testing, and is compatible with remote control; the graphical user interface unit 502: is used to provide a parameter setting interface, a real-time data monitoring dashboard and a test report generation tool, and presets industry test templates for automotive electronics and industrial automation; the data storage and traceability unit 503: is used to encrypt and store test raw data, analysis results and operation logs, supports rapid retrieval by time / project dimensions, and complies with ISO17025 laboratory management requirements.

[0112] In this embodiment, it should be noted that: the automation control unit 501 intelligently schedules the test process, realizes human-computer interaction and visual monitoring through the graphical user interface unit 502, and finally completes the encrypted archiving and traceable management of the test data through the data storage and traceability unit 503.

[0113] In addition, it should be noted that the remote control interface is compatible with TCP / IP and GPIB communication protocols, supports multi-site collaborative testing through the cloud platform, and the data synchronization delay is ≤200ms.

[0114] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these 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 any one or more embodiments or examples.

[0115] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A comprehensive test system for conducted disturbance and immunity in complex electromagnetic environment, characterized by: The system comprises a signal generation module (100), a holographic electromagnetic wave processing module (200), a conduction path simulation module (300), a monitoring and analysis module (400), and a control and management module (500), wherein: The signal generation module (100) is used to dynamically generate electromagnetic interference signals with multiple frequency bands and multiple modulation modes by using a software-defined radio architecture and combining an intelligent interference generation algorithm; The holographic electromagnetic wave processing module (200) is based on AI dynamic optimization and programmable metasurface technology, and controls the phased array excitation weight and electromagnetic wavefront in real time through deep learning, and combines digital twin pre-calculation and multi-physics field coupling modeling; The conduction path simulation module (300) is based on a programmable impedance network and a modular configuration architecture to construct a conduction path model that supports multiple industry standards, and to perform high-precision simulation of the electromagnetic conduction environment through non-intrusive interference injection and data-driven optimization technology; The monitoring and analysis module (400) is used to monitor and record the impact of electromagnetic disturbances and the response of the equipment during the test in real time, and provide data analysis functions; The control and management module (500) is used to execute the test process according to a predetermined plan and provide a user interface for operators to set parameters, start / stop the test and view the results.

2. The comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to claim 1, characterized in that: The signal generation module (100) comprises a software defined radio unit (101), an intelligent interference generation algorithm unit (102), and a radio frequency up-conversion and power amplification unit (103), wherein: The software defined radio unit (101) implements digital signal processing based on an FPGA / ASIC chip to generate baseband signals with multiple frequency bands and multiple modulation modes; The intelligent interference generation algorithm unit (102) is used to dynamically generate a composite interference signal using a machine learning algorithm and supports the import of custom scenarios; The radio frequency up-conversion and power amplification unit (103) is used to up-convert the baseband signal to a target frequency band and output an electromagnetic interference signal with adjustable amplitude after power amplification.

3. The comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to claim 2, characterized in that: The intelligent interference generation algorithm unit (102) uses a machine learning algorithm to dynamically generate a composite interference signal and supports the import of custom scenarios. The specific operations are as follows: A1: Scenario parameter analysis: Receive interference characteristic parameters imported from custom scenarios, such as frequency range f min -f max , modulation type, pulse repetition frequency PRF, and construct the interference parameter probability distribution function P(x) based on historical test data, where x is the variable of interference amplitude and phase; A2: Signal pre-generation based on generative adversarial networks: ①Build a GAN model, whose loss function is: ② Use the generator network G to generate the initial interference signal S init (t), optimize the signal authenticity through the discriminator network D, and iterate the training until convergence; A3: Reinforcement learning dynamic adjustment: ① Define the state space S as the real-time response data of the device under test, such as bit error rate and voltage fluctuation, and the action space A as the interference signal adjustment parameters, such as amplitude adjustment step ΔA and frequency offset Δf; ② Based on the deep Q-network algorithm, the action value is evaluated through the reward function R(s,a): A4: Composite signal generation: The pre-generated signal S init (t) and the signal S after RL adjustment adj (t) Perform weighted fusion to generate the final composite interference signal: S final (t):S final (t)=α·S init (t)+(1-α)·S adj (t) Among them, the weight coefficient α is adaptively adjusted according to the complexity of the scene.

4. The comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to claim 1, characterized in that: The holographic electromagnetic wave processing module (200) comprises a programmable metasurface array unit (201), an AI dynamic optimization and deep learning unit (202), and a digital twin pre-calculation unit (203), wherein: The programmable metasurface array unit (201) is composed of thousands of electromagnetic units and is used to control the phase, polarization and amplitude distribution of electromagnetic waves in real time to simulate the spatial characteristics of a complex electromagnetic environment; The AI dynamic optimization and deep learning unit (202) is used to analyze monitoring data through a CNN network, generate a phased array excitation weight matrix, dynamically adjust the electromagnetic wavefront, and realize holographic reconstruction of multi-source interference scenes; The digital twin pre-calculation unit (203) is used to store a holographic database of typical electromagnetic scenes and pre-calculate interference propagation paths based on a multi-physics field coupling model.

5. The comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to claim 4 is characterized in that: The AI dynamic optimization and deep learning unit (202) analyzes monitoring data through a CNN network, generates a phased array excitation weight matrix, dynamically adjusts the electromagnetic wavefront, and realizes holographic reconstruction of a multi-source interference scene. The specific operations are as follows: B1: Data preprocessing: receiving the time domain voltage waveform V(t) and frequency domain electric field intensity E(f) output by the monitoring and analysis module (400), and eliminating the dimension effect through normalization processing: Where, μv, σv are the mean and standard deviation of voltage, μE, σE are the mean and standard deviation of electric field intensity; B2: Feature extraction: Use the 3D convolution layer of CNN to process the frequency domain electric field data E(f) and extract the space-frequency feature map FCNN. The convolution operation is defined as: Among them, K is the convolution kernel, the size is (2a+1)×(2b+1)×(2c+1); Using LSTM network to process time domain voltage series The hidden layer state update formula is: h t =tanh(W ih x t +W hh h t-1 +b h ) Among them, W ih , W hh is the weight matrix, b h is the bias vector; B3: Weight matrix generation: concatenate the spatial features extracted by CNN and the temporal features extracted by LSTM into a joint feature vector F joint , mapped to the phased array excitation weight matrix through the fully connected layer: W:W=σ(W fc ·F joint +b fc , where σ is the Sigmoid activation function, W fc , b fc is the fully connected layer parameter; B4: Dynamic adjustment feedback: Update the network parameters θ based on the mean square error between the measured and simulated electric field distributions. The loss function is: The Adam optimizer is used for iterative optimization with a learning rate of η = 1e-4.

6. The comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to claim 1, characterized in that: The conduction path simulation module (300) comprises a programmable impedance network unit (301), a modular interface component unit (302), a non-intrusive interference injection unit (303), and a data-driven optimization unit (304), wherein: The programmable impedance network unit (301) is used to dynamically simulate the impedance characteristics of the conduction path by being composed of MEMS adjustable elements, with a switching speed of ≤50μs; The modular interface component unit (302) is used to provide a standardized interface for a power port and a signal port, supports rapid replacement within 5 minutes, and is compatible with multiple industry test standards; The non-intrusive interference injection unit (303) is used to indirectly inject interference signals into the conduction path through an electromagnetic coupling clamp and a current probe, thereby avoiding damage to the original circuit structure of the device under test; The data driven optimization unit (304) automatically optimizes the impedance network parameters based on a particle swarm optimization algorithm.

7. The comprehensive test system for conducted disturbance and immunity in complex electromagnetic environment according to claim 6, characterized in that: The data driven optimization unit (304) automatically optimizes the impedance network parameters based on the particle swarm optimization algorithm, and the specific operations are as follows: C1: Initialize the population parameters of the particle swarm optimization algorithm, including the number of particles, inertia weight and learning factor; C2: Establish a fitness function based on the error between the current impedance network output and the target electromagnetic environment simulation; C3: Iteratively update the position and velocity of each particle to find the optimal impedance parameter combination; C4: Feedback the optimized impedance parameters to the programmable impedance network unit to achieve dynamic adjustment of the conduction path model.

8. The comprehensive test system for conducted disturbance and immunity in complex electromagnetic environment according to claim 7, characterized in that: In C3, the speed and position are updated by the particle swarm optimization algorithm. The specific formula is: Among them, v id and x id They represent the velocity and position of the i-th particle in the d-th dimension, w is the inertia weight, ranging from 0.4 to 0.9, c1 and c2 are learning factors, ranging from 1.4 to 2.0, r1 and r2 are random numbers, and p id is the individual optimal solution, g d is the global optimal solution.

9. The comprehensive test system for conducted disturbance and immunity in a complex electromagnetic environment according to claim 1, characterized in that: The monitoring and analysis module (400) comprises a multi-parameter synchronous acquisition unit (401), a holographic data analysis unit (402), and an immunity evaluation unit (403), wherein: The multi-parameter synchronous acquisition unit (401) is used for real-time acquisition of the voltage, current, frequency, digital signal error rate and functional status of the device under test; The holographic data analysis unit (402) is used to use electromagnetic topology theory and wavelet packet decomposition algorithm to construct a three-dimensional field distribution model of the conduction path, locate the disturbance sensitive point, and generate visual results of the spectrum diagram and waterfall diagram; The anti-interference evaluation unit (403) is used to automatically determine the equipment anti-interference level according to industry standards and output a failure mode analysis report.

10. The comprehensive test system for conducted disturbance and immunity in complex electromagnetic environment according to claim 1, characterized in that: The control and management module (500) includes an automation control unit (501), a graphical user interface unit (502), and a data storage and tracing unit (503), wherein: The automation control unit (501) executes a predetermined test process based on state machine logic, supports sequential testing, cyclic testing and conditional triggering testing, and is compatible with remote control; The graphical user interface unit (502) is used to provide a parameter setting interface, a real-time data monitoring dashboard and a test report generation tool, and preset industry test templates for automotive electronics and industrial automation; The data storage and traceability unit (503) is used for encrypting and storing test raw data, analysis results and operation logs, supporting fast retrieval by time / project dimension, and complying with ISO17025 laboratory management requirements.

Images