Method and system for monitoring electromagnetic radiation of electric vehicle

Abstract

The invention belongs to the field of electric vehicle detection, and provides an electric vehicle electromagnetic radiation monitoring method and system, and the method comprises the steps: arranging electromagnetic radiation sensing nodes at a plurality of preset positions inside and outside an electric vehicle; establishing a multi-source input electromagnetic cognitive modeling system, and generating an atlas model for describing an electromagnetic radiation distribution evolution rule; constructing a coupling model among a plurality of electromagnetic source modules in the electric vehicle in the electromagnetic cognitive modeling system, and introducing an induced risk score as a correction factor into the atlas model; on the basis of the map model, predicting a high-radiation area possibly appearing under the current working condition of the electric vehicle, and determining the dynamic sampling priority of each sensing node; according to the sampling priority of each sensing node and a local real-time monitoring result, the sampling frequency is autonomously adjusted, and local enhanced sampling of a predicted high-risk area is realized through an inter-node cooperation mechanism.

Description

Technical Field

[0001] The present invention belongs to the field of electric vehicle detection, and particularly relates to an electromagnetic radiation monitoring method and system for electric vehicles. Background Art

[0002] With the rapid development of the electric vehicle industry, the degree of electrification of the whole vehicle has been significantly improved. A variety of electronic modules such as drive motors, high-frequency inverters, on-vehicle chargers, BMS systems, and on-vehicle communication terminals continuously work during vehicle operation, making the electromagnetic radiation environment of the whole vehicle increasingly complex. Electric vehicles may generate electromagnetic waves with different frequencies and intensities under various working conditions such as acceleration, braking, shifting, charging, and high-speed driving. These electromagnetic waves may not only interfere with the vehicle's electronic systems but also affect the electromagnetic exposure safety of the occupants in the vehicle.

[0003] Most of the existing electromagnetic radiation monitoring schemes are based on fixed layout points and equal-period sampling methods, which cannot dynamically adapt to the changing operating states of electric vehicles. Moreover, they usually ignore the electromagnetic coupling relationships between various electromagnetic source modules in the vehicle and cannot effectively identify non-linear radiation enhancement phenomena caused by module cooperation, frequency resonance, or structural coupling. In addition, although high-frequency continuous sampling improves the timeliness of monitoring, it also brings problems such as high power consumption, strong interference, and data redundancy, which are not conducive to the long-term stable operation of the system. Therefore, there is an urgent need for an electromagnetic radiation dynamic monitoring method for electric vehicles that can integrate multi-source input data, achieve spatio-temporal perception intelligent modeling, and have local enhanced response capabilities, so as to improve the recognition accuracy of high-risk areas and abnormal behaviors, reduce unnecessary monitoring overhead, and enhance the electromagnetic safety management level of the whole vehicle. Summary of the Invention

[0004] To solve the problems in the prior art, the present invention provides an electromagnetic radiation monitoring method for electric vehicles, including the following steps:

[0005] Deploy electromagnetic radiation sensing nodes at multiple preset positions inside and outside the electric vehicle, and each sensing node is used to collect the electromagnetic radiation time-domain signal and frequency-domain characteristics of the corresponding position;

[0006] Establish an electromagnetic cognitive modeling system with multi-source input, and the electromagnetic cognitive modeling system integrates multi-dimensional input variables including vehicle operation parameters, on-vehicle electronic module states, occupant distribution information, and external environment interference background to generate a map model for describing the evolution law of electromagnetic radiation distribution;

[0007] Construct a coupling model between multiple electromagnetic source modules inside the electric vehicle in the electromagnetic cognitive modeling system, establish an electromagnetic induction causal association graph to identify potential non-linear radiation enhancement events caused by module cooperation or resonance effects, and introduce the induced risk score as a correction factor into the map model;

[0008] Based on the atlas model, predict the high-radiation areas that the electric vehicle may encounter under the current working conditions, and determine the dynamic sampling priorities of each sensing node. The priorities are jointly determined by the electromagnetic change rate, spatial radiation gradient, and the risk level of inter-module coupling induction in the area where the node is located;

[0009] According to the sampling priorities of each sensing node and the local real-time monitoring results, autonomously adjust the sampling frequency, and achieve local enhanced sampling of the predicted high-risk areas through the collaborative mechanism between nodes.

[0010] Furthermore, the sensing nodes are arranged in the battery pack area, drive motor area, periphery of the power electronics module, location of the vehicle-mounted communication system, charging interface area, key positions in the passenger compartment, and external reference points of the vehicle body.

[0011] Furthermore, the nodes in the atlas model include state nodes, spatial nodes, module nodes, and risk nodes. The model is trained using a graph neural network and outputs the predicted value of the electromagnetic radiation intensity and the risk score based on the temporal evolution information and structural connection relationship of each node.

[0012] Furthermore, the inter-module coupling model includes: electrical direct coupling paths, spatial field coupling paths, and frequency modulation coupling paths. The coupling paths are represented in the form of a graph structure, where the weights of the edges are used to characterize the coupling strength or resonance probability between modules and participate in the construction of the electromagnetic induction causal association graph.

[0013] Furthermore, each sensing node includes a local trend detection module and a collaborative perception module. The local trend detection module is used to determine whether to increase the sampling frequency based on the fluctuation of the electromagnetic signal within a short time window, and the collaborative perception module is used to broadcast the abnormal state to neighboring nodes to achieve multi-node joint response sampling for high-risk areas.

[0014] On the other hand, the present invention also provides an electromagnetic radiation monitoring system for an electric vehicle, including the following modules:

[0015] Multiple electromagnetic radiation sensing nodes are arranged at multiple preset positions inside and outside the electric vehicle. Each sensing node is used to collect the electromagnetic radiation time-domain signal and frequency-domain characteristics of the corresponding position;

[0016] The electromagnetic cognitive modeling module is used to fuse multi-dimensional input variables including vehicle operation parameters, states of in-vehicle electronic modules, occupant distribution information, and external environmental interference backgrounds, and establish an atlas model describing the evolution law of electromagnetic radiation distribution;

[0017] A coupling analysis module, which is set in the electromagnetic cognition modeling system, is used to construct a coupling model between multiple electromagnetic source modules inside the electric vehicle, and establish an electromagnetic induction causal association graph, which is used to identify potential non-linear radiation enhancement events caused by module cooperation or resonance effects, and introduce the induced risk score as a correction factor into the atlas model;

[0018] A priority evaluation module is used to predict the high-radiation areas that the electric vehicle may appear in under the current working conditions based on the atlas model, and determine the dynamic sampling priorities of each sensing node. The priority is jointly determined by the electromagnetic change rate, spatial radiation gradient, and module coupling induced risk level in the area where the node is located;

[0019] A frequency adjustment and collaborative sampling module is used to autonomously adjust the sampling frequency according to the sampling priorities of each sensing node and the local real-time monitoring results, and achieve local enhanced sampling of the predicted high-risk areas through the node-to-node cooperation mechanism.

[0020] Furthermore, electromagnetic radiation sensing nodes are set in the battery pack area, drive motor area, periphery of the power electronics module, location of the vehicle-mounted communication system, charging interface area, key positions in the passenger compartment, and external reference points of the vehicle body.

[0021] Furthermore, the nodes in the atlas model include state nodes, spatial nodes, module nodes, and risk nodes. The atlas model is trained by a graph neural network, and based on the temporal evolution information and structural connection relationships of each node, it outputs the electromagnetic radiation intensity prediction value and risk score.

[0022] Furthermore, the module coupling model includes: an electrical direct coupling path, a spatial field coupling path, and a frequency modulation coupling path. The coupling paths are represented in the form of a graph structure, where the weight of the edge is used to characterize the coupling strength or resonance probability between modules, and participates in the construction of the electromagnetic induction causal association graph.

[0023] Furthermore, each sensing node includes a local trend detection module and a collaborative perception module. The local trend detection module is used to judge whether to increase the sampling frequency based on the fluctuation of electromagnetic signals within a short time window, and the collaborative perception module is used to broadcast the abnormal state to neighboring nodes to achieve multi-node joint response sampling for high-risk areas.

[0024] The present invention proposes an electromagnetic radiation monitoring method and system for electric vehicles. By arranging electromagnetic radiation sensing nodes at multiple key positions inside and outside the vehicle, combining multi-source state inputs to construct an electromagnetic cognition modeling system, and introducing an atlas model and a module coupling mechanism, it effectively overcomes the deficiencies of the prior art in aspects such as monitoring response, modeling ability, and energy consumption control, and has the following beneficial technical effects:

[0025] Adopt a graph neural network and a heterogeneous graph spectrum structure, integrate multi-source inputs such as vehicle operating status, module working information, occupant distribution, and external interference, establish a spatio-temporal joint modeling system, and can accurately predict the spatial distribution and intensity change trend of electromagnetic radiation under different working conditions.

[0026] By constructing an electromagnetic coupling model between modules and an induced causal graph, actively identify the non-linear radiation enhancement behavior caused by module collaboration or frequency resonance, can predict and intervene in potential high-risk events in advance, and improve the electromagnetic compatibility protection ability of the system.

[0027] Dynamically generate the sampling priorities of each sensing node according to the graph spectrum reasoning results and risk scores, combine local trend detection and inter-node collaboration mechanisms, realize local enhanced sampling in high-risk areas and intermittent or sleep control in low-risk areas, taking into account both monitoring accuracy and power consumption optimization.

[0028] Through a continuous feedback and online update mechanism, the system can continuously optimize the graph spectrum structure and parameter weights during long-term operation, adapt to the monitoring requirements of different vehicle models, different module layout structures, and different operating environments, and has strong versatility and scalability. Brief Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0030] Figure 1 is the flowchart of the method of the present invention;

[0031] Figure 2 is the system block diagram of the present invention. Detailed Embodiments

[0032] Next, in combination with the drawings and specific embodiments, a preferred description of the invention will be made.

[0033] This embodiment solves the above problems through the following steps:

[0034] In one embodiment, refer to Figure 1, the present invention provides an electromagnetic radiation monitoring method for electric vehicles, which is an electromagnetic radiation monitoring method for real-time monitoring, intelligent scheduling and dynamic optimization of electromagnetic radiation generated under different operating conditions. It is applicable to continuously sense and analyze the spatio-temporal characteristics of the electromagnetic radiation intensity that may appear in the internal key components of electric vehicles (including battery packs, drive motors, power electronic modules, charging systems and in-vehicle communication devices) and the vehicle interior and exterior occupant environments. Through the collaborative work of multiple sensing nodes and intelligent control strategies, key monitoring of high-risk areas and power consumption control of low-variation areas are achieved, thus ensuring the electromagnetic radiation safety and system stability of electric vehicles during market testing and actual operation.

[0035] The method specifically includes the following steps:

[0036] Step S1, deploy electromagnetic radiation sensing nodes at multiple preset positions inside and outside the electric vehicle, and each sensing node is used to collect the electromagnetic radiation time-domain signal and frequency-domain characteristics of the corresponding position.

[0037] Specifically, deploy electromagnetic radiation sensing nodes at multiple preset key positions inside and outside the electric vehicle for distributed and differential dynamic perception of the electromagnetic radiation signals generated by the whole vehicle under various operating conditions; each sensing node respectively collects the electromagnetic radiation time-domain signal (such as the variation curve of electric field strength with time) and frequency-domain characteristics (such as spectral energy distribution, main frequency point power density, harmonic amplitude, etc.) of its location, and uploads the collected data to the central processing module for subsequent modeling and scheduling.

[0038] The preset positions include but are not limited to the following areas:

[0039] Battery pack area: Deploy sensors to monitor the low-frequency radiation generated during high-current charging and discharging processes;

[0040] Near the drive motor: Used to capture the intermediate-frequency electromagnetic signals caused by motor commutation and rotating magnetic fields;

[0041] Periphery of power electronic modules (such as IGBT modules): Monitor high-frequency harmonic interference and transient electromagnetic pulses;

[0042] Near in-vehicle communication systems (such as Wi-Fi, Bluetooth, V2X antennas): Monitor the high-frequency continuous waves generated during communication;

[0043] Charging interface area: Evaluate the pulse electromagnetic interference during high-voltage connection and disconnection;

[0044] Key positions in the vehicle occupant compartment, such as the driver's seat and the child seat area: Used to monitor the radiation intensity and dose exposure level in the area close to the human body;

[0045] External reference points of the vehicle body: Set reference sensors for background noise calibration and external interference identification.

[0046] Preferably, each sensing node can adopt a broadband vector electromagnetic sensor module with the ability to sense both electric and magnetic fields, support multi-band induction in the range of 10 kHz to 1 GHz, and have the ability of local preliminary processing and data compression to adapt to the subsequent map model construction and sampling priority scheduling.

[0047] Step S2: Establish an electromagnetic cognitive modeling system with multi-source inputs. The electromagnetic cognitive modeling system integrates multi-dimensional input variables including vehicle operation parameters, in-vehicle electronic module status, occupant distribution information, and external environmental interference background, and generates a map model for describing the evolution law of electromagnetic radiation distribution.

[0048] As a highly electrified mobile system, the electromagnetic radiation of electric vehicles is not generated by a single source, but is excited by the combined operation of multiple electrical modules (such as drive motors, IGBT modules, charging systems, in-vehicle communication units, etc.) under different working conditions, and these radiation behaviors have significant dynamic and non-linear characteristics in terms of time, frequency, and space. Especially during the operation of electric vehicles, factors such as the mutual coupling between different modules, the switching of working states, and the change of modulation methods will complicate the radiation behavior, making it difficult for traditional monitoring methods based on static rules or empirical models to accurately predict high-risk areas.

[0049] Therefore, in order to achieve intelligent monitoring and active recognition of the electromagnetic radiation of electric vehicles, it is necessary to establish an electromagnetic cognitive modeling system that integrates multiple input data and has the ability of self-learning and reasoning. By introducing the graph structure modeling method, complex information such as vehicle working conditions, module status, and spatial structure can be organized into a learnable map form, so as to realize the dynamic modeling, reasoning prediction, and risk scoring of radiation distribution, and provide decision support for the subsequent optimization of sampling strategies.

[0050] The specific steps are as follows:

[0051] Step 21: Multi-source data acquisition and structured modeling

[0052] First, the system extracts multi-dimensional input variables from the in-vehicle network, including but not limited to:

[0053] Vehicle operation parameters: such as current speed, acceleration, motor current, battery voltage, motor temperature;

[0054] Module working status: such as IGBT duty cycle, PWM frequency, BMS charge and discharge status;

[0055] Occupant distribution information: Determine the seat usage status through pressure sensors, infrared / vision systems;

[0056] External interference information: Collect the electromagnetic background power spectral density, wireless communication interference level, etc. outside the vehicle.

[0057] Synchronize the above data in the time dimension and normalize it in a standardized form to construct an input tensor and a state description vector for subsequent modeling.

[0058] Step 22: Construct the topological structure of the electromagnetic map model

[0059] Based on the normalized data, construct an electromagnetic map model, specifically including:

[0060] Define node types: such as "battery status node", "current peak node", "occupant position node", "radiation risk node", etc.;

[0061] Define the semantic relationship of edges: such as "motor speed↑→PWM frequency↑→intermediate frequency radiation↑", and model it as a weighted directed edge;

[0062] Add spatial edges (such as the adjacency relationship of sensor layout points), module coupling edges (such as the feedback path between IGBT and BMS), and time edges (used to model state evolution) to the graph;

[0063] Construct a heterogeneous graph structure and initialize the graph neural network on it, setting the initial embedding vector of the nodes.

[0064] Step 23: Training and iterative learning

[0065] Use a graph neural network (such as GCN, GAT, or GraphSAGE) for training. The input is the constructed graph structure and the corresponding historical radiation data labels, and the output is:

[0066] The predicted radiation intensity values of each node;

[0067] The radiation risk scores of each area within the vehicle;

[0068] The spatial traceability path of potential abnormal radiation events.

[0069] During the training process, a supervised learning strategy (such as minimizing the mean square error between the predicted value and the measured value) or a semi-supervised strategy (combining the graph structure propagation mechanism when labels are missing) can be used.

[0070] Step 24: Real-time inference and scheduling decision support

[0071] Deploy the trained model to the in-vehicle central processing module, and in actual operation, input new vehicle state data in real time. Through the forward inference process, the following results are output:

[0072] The predicted map of the spatial distribution of electromagnetic radiation under the current working condition;

[0073] High-risk area identification results and confidence scores;

[0074] Sampling priority suggestions for each sampling node, which are used to guide subsequent sampling frequency adjustment and power consumption management.

[0075] In this step, by introducing multi-source state perception and graph modeling methods, the limitations of traditional monitoring methods based only on local sensor perception and static rule reasoning are broken. First, the system can perceive the potential coupling relationships and induction mechanisms among modules, and achieve high-risk identification at the module behavior level. Second, by introducing a structured graph representation and a graph neural network reasoning mechanism, the generation path and spatial propagation trend of electromagnetic anomalies can be identified in a high-dimensional non-linear state space, with good generalization ability. Third, the model has an adaptive learning ability and can continuously optimize its prediction performance with the accumulation of vehicle data, and is suitable for self-learning evolution under different vehicle models and working conditions.

[0076] Step S3: Construct a coupling model between multiple electromagnetic source modules inside the electric vehicle in the electromagnetic cognitive modeling system, establish an electromagnetic induction causal association graph, which is used to identify potential non-linear radiation enhancement events caused by module cooperation or resonance effects, and introduce the induced risk score as a correction factor into the graph model.

[0077] In step S2, an electromagnetic graph model integrating multi-source data such as vehicle operating status, module working conditions, and external interference has been constructed, which is used to describe the spatio-temporal distribution law of electromagnetic radiation under different operating conditions of the electric vehicle. However, in an actual electric vehicle system, some sudden electromagnetic radiation anomalies are not driven independently by a single module, but often result from non-linear amplification effects induced by the cooperative operation or frequency coupling among multiple electromagnetic source modules.

[0078] For example, if the control frequencies of the electric drive system and power devices are synchronized, it may trigger phenomena such as harmonic overlap and resonance excitation, resulting in a sudden increase in radiation power in certain areas. Due to the concealment, suddenness, and strong non-linearity of such behaviors, it is difficult to accurately capture them only relying on node state reasoning in traditional graphs. Therefore, it is necessary to further introduce an inter-module electromagnetic coupling model and an induced causal graph structure on the basis of graph modeling to identify such complex induction mechanisms, and introduce the corresponding risk score results into the graph reasoning process to improve the prediction ability for abnormal events.

[0079] The specific implementation process includes:

[0080] Step S31: Extract the main electromagnetic source modules and their electromagnetic interaction characteristics

[0081] Based on the system electrical schematic diagram and module control logic, extract the core modules with independent electromagnetic emission capabilities, such as the drive motor controller (MCU), IGBT power switch unit, BMS system, battery pack, radar, and V2X communication unit, etc. Establish the characteristic descriptions of each module, such as its typical electromagnetic emission frequency band, modulation strategy, harmonic behavior, etc.

[0082] Step S32: Build an electromagnetic coupling path model between modules

[0083] Combined with the actual circuit connections, electrical coupling conditions, and spatial arrangements, establish an electromagnetic coupling relationship network between modules, including:

[0084] Electrical direct coupling paths: shared power supply bus, grounding loop, control signal link;

[0085] Spatial field coupling paths: such as wire routing proximity, metal cavity reflection coupling, capacitive induction interference;

[0086] Frequency / modulation coupling paths: such as PWM frequency synchronization, modulation sideband coincidence, etc.

[0087] This coupling network is represented in a graph structure, with nodes being modules, edges being coupling relationships, and edge weights representing coupling strength or historical resonance probability.

[0088] Step S33: Establish an induced causal graph and define high-risk combination patterns

[0089] Based on the above coupling graph, through the analysis of historical operation data or simulation data:

[0090] Identify the statistical correlation between specific module combination states (such as A frequency ↑ & B duty cycle ↑) and electromagnetic anomaly events (sudden increase in E field strength) at the target sensing node;

[0091] Use causal reasoning methods (such as Bayesian networks, frequent itemset mining) to construct an induced causal graph;

[0092] Establish a logical mapping between each group of module combination states and potential anomalies, label them as high-risk combination patterns, and set corresponding induced probability values.

[0093] Step S34: Generate an induced risk score in real time

[0094] During the vehicle operation, the system continuously detects the working states of each module in real time and matches them with the high-risk combination patterns identified in the induced causal graph:

[0095] If the current module state is highly similar to a known combination, output a risk score between 0 and 1;

[0096] This score reflects the likelihood that the current system is on the verge of enhanced induced radiation;

[0097] Simultaneously record the areas on the coupling path where interference amplification may occur.

[0098] Step S35: Use the risk score as a map correction factor

[0099] Introduce the induced risk score into the map model constructed in step S2 as a correction factor:

[0100] If a high-risk score exists in a certain area, increase the predicted value of the radiation intensity in the map of that area;

[0101] In the sampling scheduling module, increase the priority of the corresponding node;

[0102] If there is a high-risk propagation path across nodes, adjacent area nodes can also be activated in advance for early defense.

[0103] This step enables the original map model to have the self-perception ability of induced risks within the system, breaking through the limitation of traditional graph modeling that only relies on static state input. After introducing the coupling model and causal inference structure, the system can not only judge the current risk according to the module state, but also infer the upcoming anomalies based on the coupling trend between modules, having a certain predictive foresight. In addition, as a dynamic correction quantity in map reasoning, the induced risk score significantly enhances the electromagnetic anomaly response ability of the model under the collaborative influence of multiple modules, improving the adaptability of the system to complex scenarios and the monitoring sensitivity at high-risk moments.

[0104] Exemplarily:

[0105] In a certain high-speed running test, when the vehicle enters the ramp acceleration stage, the following module states are detected:

[0106] The switching frequency of the IGBT module rises to 26 kHz and the duty cycle reaches 90%;

[0107] The rotational speed of the electric drive motor increases to 6000 rpm;

[0108] The radar system is in the active detection state and the V2X module periodically sends broadcast frames;

[0109] The instantaneous discharge current of the battery exceeds 220 A.

[0110] In the S2 stage, the map model of the system judges that the current overall radiation level is in the medium-risk range, but through the module coupling model and induced causal graph constructed in S3, it is further identified that the above state combination is a high-risk incentive combination for historical medium-frequency electromagnetic anomaly events, and the induced risk score is calculated to be 0.94.

[0111] Based on the atlas model, the radiation prediction values of the corresponding areas are corrected, and the front part of the passenger compartment and the top area of the vehicle motor compartment are marked as high-risk points, and the system schedules the sensors in this area to enter the high-frequency sampling mode.

[0112] Step S4: Based on the atlas model, predict the high-radiation areas that may occur in the electric vehicle under the current working conditions, and determine the dynamic sampling priorities of each sensing node. The priorities are jointly determined by the electromagnetic change rate, spatial radiation gradient, and module-coupling-induced risk level in the area where the node is located.

[0113] In step S2, the electromagnetic cognition modeling system constructed an atlas model to depict the overall electromagnetic radiation distribution trend of the electric vehicle by fusing data such as vehicle operating status, module working information, occupant position, and external environment; in step S3, an electromagnetic coupling mechanism between modules was introduced, and the possible non-linear enhanced radiation events were identified and quantified, and a risk score was output. The atlas model and the coupling risk score together constitute a systematic understanding of the electromagnetic radiation distribution and abnormal probability under the current working conditions.

[0114] However, the resources of the sensing nodes are limited, and high-frequency sampling will increase power consumption and background interference. In order to balance monitoring accuracy and system stability, a sampling priority scheduling mechanism needs to be constructed to dynamically adjust the working state of each node in a risk-driven manner, so that its sampling behavior adapts to the actual risk situation. That is to say, the prediction results output by the aforementioned model should be converted into sampling control quantities, giving priority to strengthening the perception coverage of high-risk areas and reducing the ineffective sampling in low-variation areas, so as to achieve the optimization of the intelligent perception of the electromagnetic state of the whole vehicle.

[0115] Specifically:

[0116] Step S41: Input the output data of the multi-source model

[0117] The system takes the following information output by the atlas model at the previous moment as input:

[0118] The predicted values of electromagnetic radiation in each spatial area of the whole vehicle at the current moment;

[0119] The electromagnetic change rate (i.e., the difference rate of the previous and subsequent predicted values) at the location of each sensing node;

[0120] Spatial gradient, that is, the first-order spatial derivative of the radiation intensity in adjacent areas;

[0121] The module-induced risk score value, which has been calculated in S3.

[0122] Step S42: Construct a sampling priority function

[0123] For each sensing node i, construct its sampling priority function P i, this function is a multi-factor weight model:

[0124]

[0125] Among them, represents the electromagnetic change rate in the area where the node is located;

[0126] represents the spatial radiation gradient (the difference in electromagnetic intensity from adjacent nodes);

[0127] R i represents the induced risk score obtained in step S3;

[0128] α, β, γ represent adjustable weighting coefficients for adapting to different task requirements (such as safety monitoring priority, power optimization priority, etc.).

[0129] Step S43: Priority normalization and level division

[0130] For the convenience of scheduling control, the system normalizes the P i

[0131] values of all nodes and divides them into multiple priority levels (such as 5 levels):

[0132] High priority (activation + high-frequency sampling);

[0133] Medium-high priority (activation + medium-frequency sampling);

[0134] Medium level (maintain the existing frequency);

[0135] Medium-low level (enter the intermittent sampling mode);

[0136] Low priority (enter the sleep / turn-off state).

[0137] Furthermore, the level division can be achieved by setting thresholds or based on clustering algorithms (such as K-means).

[0138] Step S44: Generate sampling scheduling instructions

[0139] According to the priority level, the following behavior parameters are assigned to each node:

[0140] Sampling frequency (such as 100Hz, 10Hz, 1Hz, 0.1Hz);

[0141] Sampling window duration (affects data granularity and compression ratio);

[0142] Activation period (such as periodic wake-up or event-driven wake-up);

[0143] Whether to perform relay transmission (affects data upload frequency and energy consumption);

[0144] The scheduling command is sent to each sensing node via the bus and executed by its local control module.

[0145] In this step, by converting the electromagnetic spectrum model and the module coupling risk scoring results into a sampling control strategy, a direct linkage from data cognition to hardware scheduling is achieved. On the one hand, it improves the monitoring density in high-risk areas, enabling rapid detection and response to sudden radiation anomalies; on the other hand, through the sleep control of low-risk areas, the overall power consumption of the system and the burden of redundant data processing are significantly reduced, the system operation life is extended, and the global monitoring efficiency is improved.

[0146] In addition, this mechanism has good self-adaptability and can dynamically adjust the sampling strategy according to the changes in the vehicle operation state. It does not rely on fixed rules but is based on risk perception for real-time regulation, making it suitable for flexible deployment under different platforms and different operating conditions.

[0147] Exemplarily:

[0148] In a road condition test in a certain city, when the vehicle enters the acceleration and merging state, the spectrum model predicts that the radiation levels near the front cabin of the vehicle and directly in front of the driver inside the vehicle rise rapidly, with a change rate 3.6 times the average of the previous 10 seconds and a significant enhancement of the spatial gradient; at the same time, the system determines that the IGBT and the motor controller frequencies are synchronized, and the induced risk score is as high as 0.88.

[0149] Based on the above data, the priority score of the sensing node in the middle of the vehicle head is calculated to be 0.91, and the system schedules it to a high-priority level, allocating a sampling frequency of 100 Hz to continuously record the EM interference waveform; the node located under the co-driver's feet, due to slow changes, gentle gradients, and no risk score, is only allocated a low-frequency sampling of 0.5 Hz and is set to enter the intermittent sleep mode.

[0150] Subsequent actual measurements show that the high-frequency sampling node successfully captured a 1.5-second electromagnetic anomaly pulse event with typical intermediate-frequency harmonic superposition characteristics, verifying the effectiveness of the scheduling mechanism; while the low-priority node maintained the continuity of the basic background data in the energy-saving state.

[0151] Step S5: Autonomously adjust the sampling frequency according to the sampling priorities of each sensing node and the local real-time monitoring results, and achieve local enhanced sampling of the predicted high-risk areas through the node cooperation mechanism.

[0152] In step S4, the system has calculated the sampling priority of each sensing node based on the atlas model and the induced risk score, and completed the global scheduling plan. However, the actual monitoring environment is dynamically changing. Some areas may enter a high-risk state within a very short time due to sudden interference, abnormal status, module failure, etc. Therefore, relying solely on the global priority setting may miss key anomalies within the data refresh interval.

[0153] To enhance the dynamic response ability of the system, each sensing node needs to have the intelligent ability of local perception + autonomous adjustment, that is, after receiving the initial priority, it combines the short-term trend of its own sampling data to dynamically fine-tune the sampling frequency. In addition, due to the spatial propagation of electromagnetic interference, a collaborative mechanism should be established between adjacent nodes. When detecting a local radiation enhancement trend, the sampling frequencies of multiple nodes are jointly increased to achieve local enhanced sampling of high-risk areas and ensure coverage density and data integrity.

[0154] Specifically, the implementation process includes the following steps:

[0155] Step S51: Initialize the sampling frequency

[0156] Each sensing node receives the priority parameter and the recommended sampling frequency f0 issued by the central system in step S4, and enters the sampling initialization stage at this frequency. The sampling module starts to regularly collect and cache local electromagnetic waveform data (electric field strength, electromagnetic spectrum, etc.).

[0157] Step S52: Local trend detection and adaptive sampling frequency adjustment

[0158] Each node is equipped with a lightweight trend analysis algorithm (such as moving average, variance detection, or short-time Fourier transform) for local judgment of electromagnetic signal feature changes. If any of the following situations occur in N consecutive samples:

[0159] The fluctuation amplitude of the electric field strength increases significantly (for example, exceeding 2 times the standard deviation of the mean);

[0160] The energy in the high-frequency band suddenly increases or a new frequency component appears;

[0161] The duration of the abnormal signal exceeds the preset threshold;

[0162] Then the node triggers the "sampling frequency adaptive adjustment mechanism" and increases the current frequency f to f′ = f0×(1 + δ), where

[0163] δ is the amplification factor (such as 0.5 - 2.0) to achieve enhanced local responsiveness.

[0164] Step S53: Inter-node collaborative perception mechanism

[0165] Each node regularly broadcasts its current monitoring status (such as average electric field strength, frequency characteristic change value, and whether anomalies are detected) to neighboring nodes to form a local state propagation subnetwork.

[0166] When a node A detects a rising trend in local risk, it sends a collaborative activation signal to its neighboring nodes B and C, prompting them to enter a collaborative warning state. Based on this, the neighboring nodes can determine whether they need to:

[0167] Upsampling in advance;

[0168] Activate standby low-power nodes;

[0169] Extend the sampling window time to obtain more stable spectral characteristics.

[0170] This mechanism realizes a rapid closed loop of local radiation events → multi-node joint response.

[0171] Step S54: Dynamic feedback and global strategy update

[0172] After each sampling cycle, each node uploads its frequency adjustment history, local exception capture records, and collaborative feedback information to the central system for correction of graph model input or recalculation of priority, thereby entering the scheduling closed loop for the next cycle.

[0173] This system has achieved a transition from "global static scheduling" to "local dynamic response," building a real-time sampling and control system based on risk perception. It can rapidly increase sampling density during periods of abnormalities and sudden incidents to capture subtle signal characteristics. It only increases sampling frequency in necessary areas, while maintaining low power consumption in other areas to avoid overall performance loss. Thanks to a node coordination mechanism, even if individual nodes misdetect or go dormant, they can leverage surrounding nodes for additional sensing, improving the system's fault tolerance. This mechanism is suitable for densely deployed sensor systems and can be extended to sparsely deployed structures through soft coordination.

[0174] For example:

[0175] During a complete vehicle immunity test on an electric vehicle, the vehicle entered charging mode and simultaneously activated its radar module. At this point, the sensor node deployed near the charging port operated at a 10Hz frequency. Within a short period of time, two strong intermediate-frequency pulse signals appeared in its electromagnetic spectrum, with amplitudes approaching the system's calibration threshold.

[0176] The node's internal analysis module detects abnormal fluctuations, automatically increasing the sampling frequency to 50Hz and activating a fast caching mechanism. Simultaneously, the node broadcasts an "electromagnetic disturbance coordination signal" to adjacent nodes under the vehicle, triggering the surrounding nodes to increase their frequency from 5Hz to 20Hz, forming a local enhanced perception array for the charging terminal area.

[0177] Under high-frequency sampling, the system successfully captured multiple repetitive pulse waveforms, and based on this, it was confirmed that there was improper modulation behavior in the radar module power controller. The tester intervened in a timely manner and adjusted the module parameters to prevent potential electromagnetic compatibility non-compliance risks.

[0178] See Figure 2 , in another embodiment, the present invention also provides an electromagnetic radiation monitoring system for electric vehicles, including:

[0179] Multiple electromagnetic radiation sensing nodes are arranged at multiple preset positions inside and outside the electric vehicle, and each sensing node is used to collect the electromagnetic radiation time-domain signal and frequency-domain characteristics of the corresponding position;

[0180] An electromagnetic cognition modeling module, which is used to fuse multi-dimensional input variables including vehicle operation parameters, in-vehicle electronic module status, occupant distribution information, and external environment interference background, and establish a map model describing the evolution law of electromagnetic radiation distribution;

[0181] A coupling analysis module is arranged in the electromagnetic cognition modeling system, which is used to construct a coupling model between multiple electromagnetic source modules inside the electric vehicle, and establish an electromagnetic induction causal association diagram, which is used to identify potential non-linear radiation enhancement events caused by module cooperation or resonance effects, and introduce the induced risk score as a correction factor into the map model;

[0182] A priority evaluation module is used to predict the high-radiation areas that the electric vehicle may appear in under the current working conditions based on the map model, and determine the dynamic sampling priorities of each sensing node. The priority is jointly determined by the electromagnetic change rate, spatial radiation gradient, and module-interference-induced risk level in the area where the node is located;

[0183] A frequency adjustment and collaborative sampling module is used to autonomously adjust the sampling frequency according to the sampling priorities of each sensing node and the local real-time monitoring results, and achieve local enhanced sampling of the predicted high-risk areas through the node-to-node collaborative mechanism.

[0184] In a further reality, the electromagnetic radiation sensing nodes are arranged in the battery pack area, drive motor area, periphery of the power electronics module, in-vehicle communication system location, charging interface area, key positions in the passenger compartment, and external reference points of the vehicle body.

[0185] In a further reality, the nodes in the map model include state nodes, space nodes, module nodes, and risk nodes. The map model is trained through a graph neural network, and based on the time-series evolution information and structural connection relationship of each node, the electromagnetic radiation intensity prediction value and risk score are output.

[0186] In a further scenario, the inter-module coupling model includes: an electrical direct coupling path, a spatial field coupling path, and a frequency modulation coupling path. The coupling paths are represented in the form of a graph structure, where the weights of the edges are used to characterize the coupling strength or resonance probability between modules and participate in the construction of the electromagnetic induction causal association graph.

[0187] In a further scenario, each sensing node includes a local trend detection module and a collaborative perception module. The local trend detection module is used to determine whether to perform up-sampling based on the fluctuations of electromagnetic signals within a short-time window, and the collaborative perception module is used to broadcast the abnormal state to neighboring nodes to achieve multi-node joint response sampling for high-risk areas.

[0188] It should be noted that the explanations of the foregoing embodiments of the electric vehicle electromagnetic radiation monitoring method also apply to the device of the embodiments of the present application, and will not be elaborated here.

[0189] Those of ordinary skill in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be implemented by a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0190] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0191] In several embodiments provided by the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (hereinafter referred to as ROM), random access memory (hereinafter referred to as RAM), magnetic disks, or optical discs that can store program codes.

[0192] The above are only specific embodiments of the present application. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. For the part of the module structure that is not specifically defined in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the foregoing background art part and specific embodiment part of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. An electromagnetic radiation monitoring method for an electric vehicle, characterized in that, It includes the following steps: Deploy electromagnetic radiation sensing nodes at multiple preset positions inside and outside the electric vehicle. Each sensing node is used to collect the electromagnetic radiation time-domain signal and frequency-domain characteristics at the corresponding position; Establish an electromagnetic cognitive modeling system with multi-source input. The electromagnetic cognitive modeling system integrates multi-dimensional input variables including vehicle operation parameters, in-vehicle electronic module status, occupant distribution information, and external environmental interference background, and generates a map model for describing the evolution law of electromagnetic radiation distribution; Construct a coupling model between multiple electromagnetic source modules inside the electric vehicle in the electromagnetic cognitive modeling system, and establish an electromagnetic induction causal association graph for identifying potential non-linear radiation enhancement events caused by module cooperation or resonance effects, and introduce the induced risk score as a correction factor into the map model; Based on the map model, predict the high-radiation areas that may occur under the current working conditions of the electric vehicle, and determine the dynamic sampling priority of each sensing node. The priority is jointly determined by the electromagnetic change rate, spatial radiation gradient, and module-interference induced risk level in the area where the node is located; According to the sampling priority of each sensing node and the local real-time monitoring results, autonomously adjust the sampling frequency, and achieve local enhanced sampling of the predicted high-risk areas through the node-to-node cooperation mechanism.

2. The electromagnetic radiation monitoring method for an electric vehicle according to claim 1, wherein The sensing nodes are arranged in the battery pack area, drive motor area, periphery of the power electronics module, in-vehicle communication system location, charging interface area, key positions in the passenger compartment, and external reference points of the vehicle body.

3. The electromagnetic radiation monitoring method for an electric vehicle according to claim 1, characterized in that, The nodes in the map model include state nodes, spatial nodes, module nodes, and risk nodes. The model is trained using a graph neural network, and based on the temporal evolution information and structural connection relationship of each node, output the electromagnetic radiation intensity prediction value and risk score.

4. The method for monitoring electromagnetic radiation of an electric vehicle according to claim 1, wherein The module-interference coupling model includes: electrical direct coupling path, spatial field coupling path, and frequency modulation coupling path. The coupling paths are represented in the form of a graph structure, where the weight of the edge is used to characterize the coupling strength or resonance probability between modules and participates in the construction of the electromagnetic induction causal association graph.

5. The method for monitoring electromagnetic radiation of an electric vehicle according to claim 1, characterized in that, Each sensing node includes a local trend detection module and a cooperative perception module. The local trend detection module is used to judge whether to increase the sampling frequency based on the fluctuation of the electromagnetic signal within a short time window, and the cooperative perception module is used to broadcast the abnormal state to adjacent nodes to achieve multi-node joint response sampling of high-risk areas.

6. An electromagnetic radiation monitoring system for an electric vehicle, characterized in that, The system includes the following modules: Multiple electromagnetic radiation sensing nodes, arranged at multiple preset positions inside and outside the electric vehicle. Each sensing node is used to collect the electromagnetic radiation time-domain signal and frequency-domain characteristics at the corresponding position; An electromagnetic cognitive modeling module, used to integrate multi-dimensional input variables including vehicle operation parameters, in-vehicle electronic module status, occupant distribution information, and external environmental interference background, and establish a map model for describing the evolution law of electromagnetic radiation distribution; The coupling analysis module is set in the electromagnetic cognition modeling system and is used to construct a coupling model among multiple electromagnetic source modules inside the electric vehicle, and establish an electromagnetic induction causal association graph, which is used to identify potential non-linear radiation enhancement events caused by module cooperation or resonance effects, and introduce the induced risk score as a correction factor into the atlas model; The priority evaluation module is used to predict the high-radiation areas that may occur in the electric vehicle under the current working conditions based on the atlas model, and determine the dynamic sampling priorities of each sensing node. The priorities are jointly determined by the electromagnetic change rate, spatial radiation gradient, and module coupling-induced risk level in the area where the node is located; The frequency adjustment and collaborative sampling module is used to autonomously adjust the sampling frequency according to the sampling priorities of each sensing node and the local real-time monitoring results, and achieve local enhanced sampling of the predicted high-risk areas through the node-to-node cooperation mechanism.

7. The electric vehicle electromagnetic radiation monitoring system according to claim 6, wherein The electromagnetic radiation sensing nodes are set in the battery pack area, drive motor area, periphery of the power electronics module, location of the vehicle-mounted communication system, charging interface area, key positions in the passenger compartment, and external reference points of the vehicle body.

8. The electromagnetic radiation monitoring system for electric vehicles according to claim 6, characterized in that The nodes in the atlas model include state nodes, spatial nodes, module nodes, and risk nodes. The atlas model is trained by a graph neural network, and based on the temporal evolution information and structural connection relationships of each node, it outputs the predicted value of the electromagnetic radiation intensity and the risk score.

9. The electric vehicle electromagnetic radiation monitoring system according to claim 6, characterized in that, The inter-module coupling model includes: an electrical direct coupling path, a spatial field coupling path, and a frequency modulation coupling path. The coupling paths are represented in the form of a graph structure, where the weights of the edges are used to characterize the coupling strength or resonance probability between modules, and participate in the construction of the electromagnetic induction causal association graph.

10. The electric vehicle electromagnetic radiation monitoring system according to claim 6, characterized in that, Each sensing node includes a local trend detection module and a collaborative perception module. The local trend detection module is used to judge whether to increase the sampling frequency based on the fluctuation of the electromagnetic signal within a short time window. The collaborative perception module is used to broadcast the abnormal state to adjacent nodes to achieve multi-node joint response sampling of high-risk areas.

Images