Active noise reduction method and device for automobile electromagnetic compatibility test

Abstract

The invention discloses an active noise reduction method and device for an automobile electromagnetic compatibility test. The method comprises the following steps: acquiring an original interference noise signal through a pre-arranged reference sensor array; processing the original interference noise signal through a pre-constructed intelligent noise reduction model to obtain a noise reduction control signal; a noise reduction signal is emitted according to the noise reduction control signal, and an electromagnetic field counteracting the original interference noise signal is formed; detecting the offset resultant field intensity signal in real time through a pre-arranged error sensor; and performing parameter optimization on the intelligent noise reduction model according to the synthesized field intensity signal. According to the method, the problem that the test accuracy is seriously influenced by environmental background noise and electromagnetic interference generated by test equipment in an automobile electromagnetic compatibility test can be solved.

Description

Technical Field

[0001] This application relates to the field of automotive testing technology, specifically to an active noise reduction method and apparatus for automotive electromagnetic compatibility testing. Background Technology

[0002] In the field of automotive electromagnetic compatibility testing, ensuring measurement accuracy is paramount. Testing is typically conducted in anechoic chambers, but in real-world environments, background interference such as external broadcast signals and power grid noise can still penetrate the chamber. Furthermore, the testing equipment itself and non-tested vehicle components also generate electromagnetic noise, severely contaminating the test data. Currently, the industry primarily relies on physical shielding, filtering, or conducting tests during low-interference periods such as nighttime. These methods are not only costly and inefficient, but more importantly, they cannot dynamically adapt to real-time changes in the interference environment, resulting in poor testing flexibility and difficulty in guaranteeing accuracy. Summary of the Invention

[0003] In view of the above problems, this application provides an active noise reduction method and apparatus for automotive electromagnetic compatibility testing, which can solve the problem that environmental background noise and electromagnetic interference generated by the testing equipment itself seriously affect the accuracy of testing in automotive electromagnetic compatibility testing.

[0004] In a first aspect, this application provides an active noise reduction method for automotive electromagnetic compatibility testing, comprising: The original interference noise signal is acquired by a pre-arranged array of reference sensors; The original interference noise signal is processed by a pre-built intelligent noise reduction model to obtain a noise reduction control signal; A noise reduction signal is emitted according to the noise reduction control signal to form an electromagnetic field that cancels out the original interference noise signal; The synthesized field strength signal after cancellation is detected in real time by pre-positioned error sensors; The parameters of the intelligent noise reduction model are optimized based on the synthesized field strength signal.

[0005] In the above technical solution, the method can comprehensively collect the original interference signal through a reference sensor array, providing accurate basis for noise reduction; generate targeted noise reduction control signal through intelligent noise reduction model processing to adapt to dynamic interference; emit noise reduction signal based on the signal to construct a canceling electromagnetic field, directly weakening the interference and replacing high-cost shielding and inefficient time period selection; error sensor detects the composite field strength after cancellation in real time and provides feedback on the noise reduction effect; and optimize the model parameters accordingly to continuously improve the noise reduction accuracy, ensure test accuracy and flexibility, and thus achieve active noise reduction accurately and efficiently.

[0006] In some embodiments, before acquiring the original interference noise signal through a pre-arranged array of reference sensors, the method further includes: Pre-deploy a reference sensor array and an error sensor; The noise signal is emitted by the interference source, and the original interference noise signal is acquired by a pre-arranged array of reference sensors.

[0007] In the above technical solution, the method can lay a solid foundation for subsequent accurate noise reduction and model optimization by deploying sensors in advance and using interference sources to assist in signal acquisition.

[0008] In some embodiments, the reference sensor array includes at least an environmental monitoring antenna, an equipment monitoring probe, and a vehicle monitoring probe; The error sensor is a standard measuring antenna used in automotive electromagnetic compatibility testing.

[0009] In the above technical solution, the method can collect interference signals from different sources such as environment, equipment and vehicles by classification, so as to ensure that the original interference is captured comprehensively and accurately, and at the same time, the cancellation effect is detected by using a standard measurement antenna.

[0010] In some embodiments, transmitting the noise-reduced signal according to the noise reduction control signal includes: The noise reduction control signal is converted from digital to analog to obtain an analog signal; The analog signal is amplified to obtain a noise-reduced signal; The noise-reduced signal is transmitted by a cancellation antenna positioned along the interference propagation path.

[0011] In the above technical solution, the method can convert the noise reduction control signal into an analog signal that can be actually transmitted and amplify it, and then accurately transmit the noise reduction signal through the cancellation antenna on the interference propagation path to ensure the effective construction of the cancellation electromagnetic field.

[0012] In some implementations, optimizing the parameters of the intelligent noise reduction model based on the synthesized field strength signal includes: The parameters of the intelligent noise reduction model are optimized with the minimum mean square value of the synthesized field strength signal as the objective.

[0013] In the above technical solution, the method can use the minimum mean square value of the synthesized field strength signal as the optimization target to accurately adjust the parameters of the intelligent noise reduction model, so as to continuously reduce residual interference.

[0014] In some embodiments, the method further includes: Construct an original noise reduction model; wherein the original noise reduction model includes at least an input layer, a long short-term memory network layer, a fully connected layer, and an output layer; Obtain the training dataset; The original denoising model is trained using backpropagation, gradient descent algorithm, and the training dataset to obtain a trained intelligent denoising model.

[0015] In the above technical solution, the method can construct an original denoising model (including an input layer, an LSTM layer, a fully connected layer, and an output layer), and train it using a training dataset through backpropagation and gradient descent algorithms to obtain an intelligent denoising model that is adapted to the actual test scenario. This provides reliable model support for subsequent real-time and accurate denoising and improves the generalization ability and interference suppression performance of the active denoising system.

[0016] In some implementations, processing the original interference noise signal using a pre-built intelligent noise reduction model to obtain a noise reduction control signal includes: Obtain a pre-built intelligent noise reduction model; Multiple time-series signals are obtained based on the noise reduction control signal; The multi-channel time series signals are input into the input layer of the intelligent noise reduction model; The long short-term memory network layer of the intelligent noise reduction model extracts features from the data transmitted by the input layer to obtain signal features; The signal features are transformed using the fully connected layer of the intelligent noise reduction model to obtain transformed data. The output layer of the intelligent noise reduction model outputs a noise reduction control signal based on the transformed data.

[0017] In the above technical solution, the method can use a pre-built intelligent noise reduction model to input multiple time series signals, extract features through an LSTM layer, transform them through a fully connected layer, and finally output a noise reduction control signal, thereby achieving accurate prediction and control of dynamic interference and improving the real-time performance and targeting of noise reduction.

[0018] Secondly, this application provides an active noise reduction device for automotive electromagnetic compatibility testing, comprising: The acquisition unit is used to acquire raw interference noise signals through a pre-arranged array of reference sensors; The processing unit is used to process the original interference noise signal through a pre-built intelligent noise reduction model to obtain a noise reduction control signal; The noise-reducing transmission unit is used to transmit a noise-reducing signal according to the noise reduction control signal to form an electromagnetic field that cancels out the original interference noise signal; The detection unit is used to detect the synthesized field strength signal after cancellation in real time through pre-arranged error sensors; An optimization unit is used to optimize the parameters of the intelligent noise reduction model based on the synthesized field strength signal.

[0019] In the above technical solution, the device can comprehensively collect the original interference signal through a reference sensor array, providing accurate basis for noise reduction; generate targeted noise reduction control signal through intelligent noise reduction model processing to adapt to dynamic interference; emit noise reduction signal based on the signal to construct a canceling electromagnetic field, directly weakening the interference and replacing high-cost shielding and inefficient time period selection; error sensor detects the composite field strength after cancellation in real time and provides feedback on the noise reduction effect; and optimize the model parameters accordingly to continuously improve the noise reduction accuracy, ensure test accuracy and flexibility, and thus achieve active noise reduction accurately and efficiently.

[0020] In some embodiments, the active noise cancellation device for automotive electromagnetic compatibility testing further includes: An arrangement unit is used to pre-arrange the reference sensor array and the error sensor before acquiring the original interference noise signal through the pre-arranged reference sensor array; The noise emission unit is used to emit noise signals through the interference source and trigger the acquisition unit to acquire the original interference noise signals through a pre-arranged reference sensor array.

[0021] In the above technical solution, the device can lay a solid foundation for subsequent accurate noise reduction and model optimization by deploying sensors in advance and using interference sources to assist in signal acquisition.

[0022] Thirdly, this application provides an electronic device including a memory and a processor, the memory storing a computer program, and the processor running the computer program to cause the electronic device to perform the active noise reduction method for automotive electromagnetic compatibility testing as described in any one of the first aspects.

[0023] Fourthly, this application provides a readable storage medium storing a computer program, which, when executed by a processor, performs the active noise reduction method for automotive electromagnetic compatibility testing as described in any one of the first aspects.

[0024] Fifthly, this application provides a computer program product comprising a computer program that, when executed by a processor, performs the active noise reduction method for automotive electromagnetic compatibility testing as described in any one of the first aspects.

[0025] The beneficial effects of this application are as follows: it can effectively suppress background noise and significantly improve the signal-to-noise ratio, thereby making weak signal measurement possible and enabling more accurate testing; it can also reduce the dependence on the "clean" test time window, enabling 24 / 7 all-weather testing; it can also appropriately reduce the requirements for the shielding effectiveness of the anechoic chamber, thereby reducing construction and maintenance costs; and it can also learn and adapt to different vehicle platforms and test configurations through AI algorithms, thereby improving versatility. Attached Figure Description

[0026] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart illustrating an active noise reduction method for automotive electromagnetic compatibility testing in some embodiments of this application. Figure 2 This application presents an AI-based active noise cancellation system framework for automotive EMC testing in some embodiments. Figure 3 This is a flowchart of an AI noise processing algorithm in some embodiments of this application; Figure 4 This is a typical layout diagram of an anechoic chamber in some embodiments of this application; Figure 5 This is a schematic diagram of the structure of an active noise cancellation device for automotive electromagnetic compatibility testing in some embodiments of this application; Figure 6 This is a schematic diagram of the structure of an electronic device in some embodiments of this application. Detailed Implementation

[0028] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0030] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more (including two), similarly, "multiple sets" refers to two or more sets (including two sets), and "multiple pieces" refers to two or more pieces (including two pieces) unless otherwise explicitly defined.

[0031] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0032] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0033] Existing technologies in automotive EMC testing primarily rely on semi-anechoic or fully anechoic chambers for testing. They also employ shielding, filtering, and selecting periods with clean electromagnetic environments, such as nighttime, to combat interference. However, these methods struggle to identify and distinguish between non-target interference noise (including ambient background noise and the device's own noise) and the actual transmitted noise of the device under test in real time. They cannot dynamically generate and transmit precise "anti-noise" signals to effectively cancel out non-target interference noise at the test antenna or sensor. Furthermore, they fail to effectively improve the test signal-to-noise ratio, lower the test threshold to enhance the accuracy and reliability of the results, and cannot support testing in more relaxed environments such as daytime.

[0034] To address the aforementioned technical issues, this application provides an active noise reduction method for automotive electromagnetic compatibility testing. This method involves collecting interference signals (such as 150MHz narrowband noise generated by a cooling fan) from a reference sensor after system startup, identifying signal characteristics using an AI model, predicting its morphology at the test antenna, and generating an anti-noise signal. This signal is then transmitted by a cancellation antenna to significantly suppress interference (e.g., reducing the peak value at the 150MHz frequency point by 25dB). Ultimately, the true transmission level of the target ECU is obtained, thereby significantly improving the test signal-to-noise ratio, increasing test efficiency (supporting all-weather testing), reducing test costs (reducing anechoic chamber shielding requirements), and possessing intelligent adaptive capabilities to adapt to different vehicle platforms and test configurations.

[0035] like Figure 1 As shown, some embodiments of this application provide an active noise reduction method for automotive electromagnetic compatibility testing. This active noise reduction method for automotive electromagnetic compatibility testing includes: S101. Acquire the original interference noise signal through a pre-arranged reference sensor array; S102. The original interference noise signal is processed by a pre-built intelligent noise reduction model to obtain a noise reduction control signal; S103. Based on the noise reduction control signal, transmit the noise reduction signal to form an electromagnetic field that cancels out the original interference noise signal; S104. Real-time detection of the synthesized field strength signal after cancellation using pre-arranged error sensors; S105. Optimize the parameters of the intelligent noise reduction model based on the synthesized field strength signal.

[0036] In some embodiments, a reference sensor array refers to a collection of multiple sensors arranged in the dark room (such as a corner or doorway), near power lines, auxiliary equipment (such as a chassis dynamometer or cooling system), and non-test components of the vehicle under test (such as a wiper motor or air conditioning compressor). It may include environmental monitoring antennas, current clamps, voltage probes, near-field probes, etc., for collecting raw interference noise signals upstream of the interference noise propagation path or at the source.

[0037] In some embodiments, the original interference noise signal refers to conducted interference noise generated by external environment noise (such as broadcast signals, cellular signals), auxiliary equipment (such as chassis dynamometers, battery simulators, cooling systems), and electromagnetic noise generated when non-test components of the vehicle under test (such as wiper motors, air conditioning compressors) are working.

[0038] In some embodiments, the intelligent noise reduction model refers to a deep learning model deployed in the AI ​​control and processing core, which is trained offline and adaptively optimized online. It preferably consists of an input layer, 2-3 layers of LSTM network, a fully connected layer and an output layer. It can receive time series signals from multiple reference sensors and realize interference noise feature identification, test point noise pattern prediction and anti-noise control signal generation.

[0039] In some embodiments, the noise reduction control signal refers to the digital signal generated by the output layer of the intelligent noise reduction model, which can drive the noise-resistant transmission system after digital-to-analog conversion. Its signal characteristics need to satisfy the requirement that it forms a cancellation effect with the original interference noise signal having opposite phase and matching amplitude at the test point.

[0040] In some embodiments, the noise reduction signal refers to the analog signal amplified by the power amplifier in the noise-resistant transmission system and transmitted through the cancellation antenna. Essentially, it is an "anti-noise" electromagnetic wave that complements the vector characteristics of the original interference noise signal and is used to superimpose and cancel the interference noise at the error sensor.

[0041] In some embodiments, the cancelling electromagnetic field refers to the electromagnetic field formed in space by the noise reduction signal emitted by the cancelling antenna (arranged on the path of interference propagation to the error sensor, with the polarization direction consistent with the interference noise), which can be vector-superimposed with the electromagnetic field of the original interference noise at the error sensor location to achieve interference suppression.

[0042] In some embodiments, the error sensor refers to an EMC test standard measurement antenna (such as a biconical antenna or a log-periodic antenna) placed at a standard test distance (such as 1 meter or 3 meters) directly in front of the vehicle under test. It is used to collect the composite field strength signal of "actual transmitted noise of the device under test + interference noise - anti-noise" as a feedback basis for the online adaptive optimization of the AI ​​model.

[0043] In some embodiments, the synthetic field strength signal refers to the electromagnetic field strength signal obtained by superimposing the actual emitted noise of the device under test, the original interference noise, and the noise reduction signal at the location of the error sensor. Its data is used to evaluate the noise reduction effect and is the core basis for the intelligent noise reduction model to adjust parameters to minimize residual interference.

[0044] In some embodiments, parameter optimization refers to the intelligent noise reduction model using the mean square value of the synthesized field strength signal as the loss function, and adjusting the model parameters through two methods: offline training (backpropagation, gradient descent algorithm) and online adaptation (fine-tuning network weights with a low learning rate), to achieve dynamic tracking and optimization of the cancellation effect of time-varying interference noise.

[0045] In the above embodiments, the method can comprehensively collect the original interference signal through a reference sensor array, providing accurate basis for noise reduction; generate targeted noise reduction control signal through intelligent noise reduction model processing to adapt to dynamic interference; emit noise reduction signal based on the signal to construct a canceling electromagnetic field, directly weakening the interference and replacing high-cost shielding and inefficient time period selection; the error sensor detects the composite field strength after cancellation in real time and provides feedback on the noise reduction effect; and optimize the model parameters accordingly to continuously improve the noise reduction accuracy, ensure test accuracy and flexibility, and thus achieve active noise reduction accurately and efficiently.

[0046] In some embodiments, the method further includes, before acquiring the original interference noise signal via a pre-arranged array of reference sensors: Pre-deploy a reference sensor array and an error sensor; The noise signal is emitted by the interference source and the original interference noise signal is acquired by a pre-arranged array of reference sensors.

[0047] In some embodiments, the method involves placing a reference sensor (near-field probe) near the cooling fan and a device monitoring probe on the power line.

[0048] In some embodiments, a 150MHz narrowband noise is generated by a cooling fan (interference source), and the original interference noise signal is acquired by a reference sensor.

[0049] In the above embodiments, the method can lay a solid foundation for subsequent accurate noise reduction and model optimization by deploying sensors in advance and using interference sources to assist in signal acquisition.

[0050] In some embodiments, the reference sensor array includes at least an environmental monitoring antenna, an equipment monitoring probe, and a vehicle monitoring probe; The error sensor is a standard measuring antenna used in automotive electromagnetic compatibility testing.

[0051] In some embodiments, the reference sensor includes a near-field probe near the cooling fan and a device monitoring probe on the power line.

[0052] In some embodiments, the test antenna (i.e., error sensor) used in effect verification and accuracy testing is a standard measurement antenna for automotive EMC testing.

[0053] In the above embodiments, the method can collect interference signals from different sources such as environment, equipment, and vehicles by classification, ensuring comprehensive and accurate capture of the original interference, while using a standard measurement antenna to detect the cancellation effect.

[0054] In some embodiments, transmitting a noise-reduced signal according to a noise reduction control signal includes: The noise reduction control signal is converted from digital to analog to obtain an analog signal; The analog signal is amplified to obtain a noise-reduced signal; Noise-reduced signals are transmitted by canceling antennas positioned along the interference propagation path.

[0055] In some embodiments, after the AI ​​processes and cancels the generated anti-noise signal, the method performs digital-to-analog conversion and power amplification on the noise reduction control signal (i.e., the anti-noise signal is amplified by a power amplifier), and then transmits the noise reduction signal through a cancellation antenna arranged on the interference propagation path.

[0056] In the above embodiments, the method can convert the noise reduction control signal into an analog signal that can be actually transmitted and amplify it, and then accurately transmit the noise reduction signal through the cancellation antenna on the interference propagation path to ensure the effective construction of the cancellation electromagnetic field.

[0057] In some embodiments, parameter optimization of the intelligent noise reduction model is performed based on the synthesized field strength signal, including: The parameters of the intelligent noise reduction model are optimized with the goal of finding the minimum mean square value of the synthesized field strength signal.

[0058] In some embodiments, the AI ​​model optimizes the parameters of the intelligent noise reduction model based on the 150MHz peak change (i.e., the feedback of the synthetic field strength signal) displayed by the spectrum analyzer during effect verification, with the minimum mean square value of the synthetic field strength signal as the target, to ensure that the interference is effectively canceled.

[0059] In the above embodiments, the method can accurately adjust the parameters of the intelligent noise reduction model with the minimum mean square value of the synthesized field strength signal as the optimization target, so that the residual interference is continuously reduced.

[0060] In some embodiments, the method further includes: Construct the original noise reduction model; the original noise reduction model includes at least an input layer, a long short-term memory network layer, a fully connected layer, and an output layer; Obtain the training dataset; The original denoising model is trained using backpropagation, gradient descent algorithm, and training dataset to obtain a trained intelligent denoising model.

[0061] In some embodiments, the prerequisite for the initial operation of the AI ​​model after system startup is: to pre-build an original denoising model containing an input layer, a long short-term memory network layer, a fully connected layer, and an output layer, to obtain a training dataset, and to train the original denoising model using backpropagation, gradient descent algorithm, and the training dataset to obtain a trained intelligent denoising model.

[0062] In the above embodiments, the method can construct an original denoising model (including an input layer, an LSTM layer, a fully connected layer, and an output layer), and train it using a training dataset through backpropagation and gradient descent algorithms to obtain an intelligent denoising model adapted to the actual test scenario. This provides reliable model support for subsequent real-time and accurate denoising, and improves the generalization ability and interference suppression performance of the active denoising system.

[0063] In some embodiments, the original interference noise signal is processed by a pre-built intelligent noise reduction model to obtain a noise reduction control signal, including: Obtain a pre-built intelligent noise reduction model; Multiple time-series signals are acquired based on the noise reduction control signal; Multiple time-series signals are input into the input layer of the intelligent noise reduction model; The long short-term memory network layer of the intelligent noise reduction model extracts features from the data transmitted from the input layer to obtain signal features; The signal features are transformed by the fully connected layer of the intelligent noise reduction model to obtain transformed data. The output layer of the intelligent noise reduction model outputs noise reduction control signals based on the transformed data.

[0064] In some embodiments, the method can acquire a pre-built intelligent noise reduction model, acquire multiple time-series signals (150MHz sine wave signals) collected by near-field probes and device monitoring probes according to the noise reduction control signal, input the multiple time-series signals into the input layer of the intelligent noise reduction model, extract signal features with a frequency of 150MHz, specific amplitude and phase through a long short-term memory network layer, perform data transformation processing on the signal features through a fully connected layer, and output the output layer outputs an anti-phase noise control signal according to the transformed data.

[0065] In the above embodiments, the method can utilize a pre-built intelligent noise reduction model to input multiple time series signals, extract features through an LSTM layer, transform them through a fully connected layer, and finally output a noise reduction control signal, thereby achieving accurate prediction and control of dynamic interference and improving the real-time performance and targeting of noise reduction.

[0066] To make the purpose, technical solution and advantages of this application clearer, the technical solution in this application will be clearly and completely described below.

[0067] In some embodiments, such as Figure 2 As shown, the AI-based active noise cancellation system for automotive EMC testing, which applies this active noise cancellation method, includes a reference sensor array, an error sensor, an AI control and processing core, and an anti-noise emission system. (1) The reference sensor array consists of multiple sensors, including: Environmental monitoring antenna: One or more broadband antennas are placed in a corner or near the door of the anechoic chamber to collect environmental background noise (such as broadcast signals and cellular signals) that seeps into the anechoic chamber from the outside. Equipment monitoring probes: current clamps and voltage probes, installed on the power lines supplying auxiliary equipment (such as chassis dynamometers, battery simulators, and cooling systems), are used to collect conducted interference noise generated by these devices.

[0068] Vehicle monitoring probe: A near-field magnetic field or electric field probe, temporarily placed near non-test components of the vehicle under test (such as wiper motors and air conditioning compressors) to collect the noise generated during their operation.

[0069] The core function of the aforementioned reference sensor is to acquire the original waveform of the interference noise upstream or near its source.

[0070] (2) The error sensor is a standard measurement antenna used in EMC testing (such as a biconical antenna or a log-periodic antenna).

[0071] It is placed at a standard test distance (e.g., 1 meter, 3 meters), directly facing the vehicle under test. Its function is to collect the final composite field strength, which is the actual emitted noise of the device under test (EUT) + all interfering noise - the noise immunity generated by the system.

[0072] This signal serves as a feedback signal for the entire system and is the basis for the AI ​​algorithm to perform adaptive optimization.

[0073] (3) The AI ​​control and processing core is the "brain" of the system, typically a control cabinet with a built-in high-speed data acquisition card and GPU computing unit. The AI ​​control and processing core may include: Data acquisition and preprocessing module: Synchronously acquires signals from all reference sensors and error sensors, performs analog-to-digital conversion (ADC), anti-aliasing filtering, and gain adjustment.

[0074] The AI ​​noise processing module receives preprocessed multi-channel reference signals and inputs them into a pre-trained deep learning model. This model outputs a complex control signal in real time.

[0075] Digital-to-analog conversion module: Used to convert control signals generated by AI models into analog signals.

[0076] (4) Noise-resistant emission system: Power amplifier: Receives analog control signals from the AI ​​core and amplifies them to a sufficient power level.

[0077] Cancellation antennas: One or more transmitting antennas connected to the output of the power amplifier. The placement of these antennas requires careful design and optimization; they are typically located along the path of interference noise propagating to the error sensor, and their polarization must align with the interference noise. The cancellation antennas transmit "noise-resistant" electromagnetic waves generated by the system.

[0078] In some embodiments, the input of the AI-based active noise reduction system for automotive EMC testing is the real-time signal of a reference sensor array; the output is the characteristics of the identified interference noise (such as frequency, amplitude, phase, and modulation method), the predicted form that the interference noise will present when it propagates to the error sensor (test point) (this requires the model to learn the propagation path characteristics of noise in a complex environment of a dark room), and the calculated "anti-noise" signal (opposite phase and appropriate amplitude) that can cancel the predicted noise.

[0079] In some embodiments, the system employs an adaptive algorithm. Specifically, this algorithm can learn online through an AI model, continuously fine-tuning the model parameters based on feedback from the error sensor to achieve dynamic tracking and compensation, thereby addressing the time-varying characteristics of the noise source.

[0080] In some embodiments, the working principle of this system is briefly described as follows: After system startup, the reference sensor array continuously captures various types of interference noise. Based on these noise signals, the AI ​​core predicts their combined effect at the error sensor and immediately generates an anti-noise signal with opposite phase and matching amplitude. This signal is then transmitted via a power amplifier to drive the cancellation antenna. The transmitted anti-noise field and the original interference field are vector-superimposed at the error sensor, ideally canceling each other out. The residual error signal detected by the error sensor is fed back to the AI ​​core in real time, allowing the AI ​​model to fine-tune its parameters, achieving closed-loop control and dynamically tracking and suppressing changing interference.

[0081] In some embodiments, Figure 3 A flowchart of an AI noise processing algorithm is shown. This method preferably uses a deep learning model combining a Long Short-Term Memory (LSTM) network with a fully connected (DenseLayer) layer. LSTM is particularly adept at processing time-series data, capable of capturing the temporal dependencies and dynamic characteristics of noise signals.

[0082] In some embodiments, the input is preprocessed synchronous time series data of multiple reference sensor signals, denoted as X(t)=[x1(t),x2(t),...,xn(t)], where n is the number of reference sensors.

[0083] In some embodiments, the input data first enters an LSTM network consisting of 2-3 layers. This layer is responsible for learning the dynamic temporal characteristics of interference noise and the correlation between signals from different sensors.

[0084] In some embodiments, the output of the LSTM layer is then fed into a fully connected layer containing multiple neurons. This layer performs nonlinear combinations and transformations on the learned features.

[0085] In some embodiments, the output layer (another fully connected layer) generates the final control signal Y(t).

[0086] In some embodiments, the output layer outputs Y(t), which is the noise immunity signal required to drive the cancellation antenna.

[0087] In some embodiments, the training process (offline) is as follows: (1) Data collection: In a darkroom, without activating the cancellation function, various interference sources (such as auxiliary equipment, non-tested vehicle components) are operated, and the signals X(t) of the reference sensor array (101) and e(t) of the error sensor (104) are recorded simultaneously. A large number of data pairs (X(t), e(t)) are collected.

[0088] (2) Define the loss function: The loss function L is directly defined as the mean square error (MSE) of the error sensor signal: L = mean(e(t)^2). The goal of training is to minimize this loss function, that is, to minimize the signal energy received by the error sensor.

[0089] (3) Model training: Using the collected dataset, the LSTM model is trained through backpropagation and gradient descent algorithms (such as the Adam optimizer). The essence of training is to teach the model a mapping function from X(t) to Y(t) so that when the system executes Y(t), e(t) is minimized.

[0090] (4) Online Adaptation (Online Learning): In actual testing, the model is not fixed. The system will continuously learn online at a low learning rate, fine-tuning the network weights based on small data fragments collected in real time, in order to adapt to the slow changes in interference noise and achieve the best cancellation effect.

[0091] In some embodiments, Figure 4 A typical layout diagram of an anechoic chamber is shown. The vehicle under test is placed on a turntable in the center of the chamber. The standard test antenna (i.e., error sensor) is located 3 or 10 meters away from the front bumper of the vehicle.

[0092] In some embodiments, the reference sensor arrangement is as follows: An environmental monitoring antenna is placed near the door of the anechoic chamber to monitor infiltrating noise; The current clamp is attached to the power supply cable of the chassis dynamometer; The near-field probe is placed near the air vent on the passenger side of the vehicle to monitor the noise of the air conditioning fan; In some embodiments, the cancelling antenna is placed on the path between the interference source and the test antenna, but off the main axis, to avoid obstructing or affecting the actual radiation of the vehicle under test.

[0093] For example, the cancellation antenna can be placed at a certain distance to the side or rear of the test antenna. Its height and polarization direction must be consistent with those of the test antenna.

[0094] Figure 5 A schematic diagram of an active noise cancellation device for automotive electromagnetic compatibility testing is shown. It should be understood that this device is related to... Figure 1 The method executed in the middle corresponds to the steps involved in the aforementioned method. The specific functions and effects of the device can be found in the description above. To avoid repetition, detailed descriptions are omitted here.

[0095] The active noise cancellation device for automotive electromagnetic compatibility testing includes: Acquisition unit 210 is used to acquire raw interference noise signals through a pre-arranged array of reference sensors; The processing unit 220 is used to process the original interference noise signal through a pre-built intelligent noise reduction model to obtain a noise reduction control signal; The noise-reducing transmitting unit 230 is used to transmit a noise-reducing signal according to the noise reduction control signal to form an electromagnetic field that cancels out the original interference noise signal; The detection unit 240 is used to detect the canceled composite field strength signal in real time through pre-arranged error sensors; The optimization unit 250 is used to optimize the parameters of the intelligent noise reduction model based on the synthesized field strength signal.

[0096] In some embodiments, the active noise cancellation device for automotive electromagnetic compatibility testing further includes: Arrangement unit 260 is used to pre-arrange the reference sensor array and error sensor before acquiring the original interference noise signal through the pre-arranged reference sensor array; The noise emission unit 270 is used to emit noise signals through the interference source and trigger the acquisition unit 210 to acquire the original interference noise signals through a pre-arranged reference sensor array.

[0097] In some embodiments, the reference sensor array includes at least an environmental monitoring antenna, an equipment monitoring probe, and a vehicle monitoring probe; The error sensor is a standard measuring antenna used in automotive electromagnetic compatibility testing.

[0098] In some embodiments, the noise-resistant transmitting unit 230 includes: The digital-to-analog conversion subunit 231 is used to convert the noise reduction control signal from digital to analog to obtain an analog signal; Amplification subunit 232 is used to amplify the power of the analog signal to obtain a noise-reduced signal; Transmitting subunit 233 is used to transmit noise-reduced signals through a cancellation antenna arranged in the interference propagation path.

[0099] In some embodiments, the optimization unit 250 is specifically used to optimize the parameters of the intelligent noise reduction model with the goal of minimizing the mean square value of the synthesized field strength signal.

[0100] In some embodiments, the active noise cancellation device for automotive electromagnetic compatibility testing further includes: Building unit 280 is used to build the original denoising model; wherein the original denoising model includes at least an input layer, a long short-term memory network layer, a fully connected layer, and an output layer; Unit 290 is used to acquire the training dataset; Training unit 300 is used to train the original denoising model using backpropagation, gradient descent algorithm and training dataset to obtain a trained intelligent denoising model.

[0101] In some embodiments, the processing unit 220 includes: Acquire subunit 221, used to acquire a pre-built intelligent noise reduction model; The acquisition subunit 221 is also used to acquire multiple time-series signals based on the noise reduction control signal; Input subunit 222 is used to input multiple time series signals into the input layer of the intelligent noise reduction model; Extraction subunit 223 is used to extract features from the data transmitted from the input layer through the long short-term memory network layer of the intelligent noise reduction model to obtain signal features; Transform subunit 224 is used to perform data transformation processing on signal features through the fully connected layer of the intelligent noise reduction model to obtain transformed data; Output subunit 225 is used to output noise reduction control signals based on the transformed data through the output layer of the intelligent noise reduction model.

[0102] like Figure 6 As shown, this application provides an electronic device 400, which includes a processor 401 and a memory 402. The processor 401 and the memory 402 are interconnected and communicate with each other through a communication bus 403 and / or other forms of connection mechanism (not shown). The memory 402 stores a computer program that can be executed by the processor 401. When the computing device is running, the processor 401 executes the computer program to perform the method in any of the aforementioned optional implementations.

[0103] This application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method in any of the aforementioned optional implementations.

[0104] The computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0105] This application provides a computer program product, which includes a computer program that, when run by a processor, executes the method in any of the aforementioned optional implementations.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. An active noise reduction method for automotive electromagnetic compatibility testing, characterized in that, include: The original interference noise signal is acquired by a pre-arranged array of reference sensors; The original interference noise signal is processed by a pre-built intelligent noise reduction model to obtain a noise reduction control signal; A noise reduction signal is emitted according to the noise reduction control signal to form an electromagnetic field that cancels out the original interference noise signal; The synthesized field strength signal after cancellation is detected in real time by pre-positioned error sensors; The parameters of the intelligent noise reduction model are optimized based on the synthesized field strength signal.

2. The active noise reduction method for automotive electromagnetic compatibility testing according to claim 1, characterized in that, Before acquiring the original interference noise signal through a pre-arranged reference sensor array, the method further includes: Pre-deploy a reference sensor array and an error sensor; The noise signal is emitted by the interference source, and the original interference noise signal is acquired by a pre-arranged array of reference sensors.

3. The active noise reduction method for automotive electromagnetic compatibility testing according to claim 1, characterized in that, The reference sensor array includes at least an environmental monitoring antenna, an equipment monitoring probe, and a vehicle monitoring probe; The error sensor is a standard measuring antenna used in automotive electromagnetic compatibility testing.

4. The active noise reduction method for automotive electromagnetic compatibility testing according to claim 1, characterized in that, The step of transmitting a noise reduction signal according to the noise reduction control signal includes: The noise reduction control signal is converted from digital to analog to obtain an analog signal; The analog signal is amplified to obtain a noise-reduced signal; The noise-reduced signal is transmitted by a cancellation antenna positioned along the interference propagation path.

5. The active noise reduction method for automotive electromagnetic compatibility testing according to claim 1, characterized in that, The step of optimizing the parameters of the intelligent noise reduction model based on the synthesized field strength signal includes: The parameters of the intelligent noise reduction model are optimized with the minimum mean square value of the synthesized field strength signal as the objective.

6. The active noise reduction method for automotive electromagnetic compatibility testing according to claim 1, characterized in that, The method further includes: Construct an original noise reduction model; wherein the original noise reduction model includes at least an input layer, a long short-term memory network layer, a fully connected layer, and an output layer; Obtain the training dataset; The original denoising model is trained using backpropagation, gradient descent algorithm, and the training dataset to obtain a trained intelligent denoising model.

7. The active noise reduction method for automotive electromagnetic compatibility testing according to claim 1, characterized in that, The process of processing the original interference noise signal using a pre-built intelligent noise reduction model to obtain a noise reduction control signal includes: Obtain a pre-built intelligent noise reduction model; Multiple time-series signals are obtained based on the noise reduction control signal; The multi-channel time series signals are input into the input layer of the intelligent noise reduction model; The long short-term memory network layer of the intelligent noise reduction model extracts features from the data transmitted by the input layer to obtain signal features; The signal features are transformed using the fully connected layer of the intelligent noise reduction model to obtain transformed data. The output layer of the intelligent noise reduction model outputs a noise reduction control signal based on the transformed data.

8. An active noise reduction device for automotive electromagnetic compatibility testing, characterized in that, The active noise cancellation device for automotive electromagnetic compatibility testing includes: The acquisition unit is used to acquire raw interference noise signals through a pre-arranged array of reference sensors; The processing unit is used to process the original interference noise signal through a pre-built intelligent noise reduction model to obtain a noise reduction control signal; The noise-reducing transmission unit is used to transmit a noise-reducing signal according to the noise reduction control signal to form an electromagnetic field that cancels out the original interference noise signal; The detection unit is used to detect the synthesized field strength signal after cancellation in real time through pre-arranged error sensors; An optimization unit is used to optimize the parameters of the intelligent noise reduction model based on the synthesized field strength signal.

9. The active noise reduction device for automotive electromagnetic compatibility testing according to claim 8, characterized in that, The active noise cancellation device for automotive electromagnetic compatibility testing also includes: An arrangement unit is used to pre-arrange the reference sensor array and the error sensor before acquiring the original interference noise signal through the pre-arranged reference sensor array; The noise emission unit is used to emit noise signals through the interference source and trigger the acquisition unit to acquire the original interference noise signals through a pre-arranged reference sensor array.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to cause the electronic device to perform the active noise reduction method for automotive electromagnetic compatibility testing as described in any one of claims 1 to 7.

11. A readable storage medium, characterized in that, The readable storage medium stores a computer program, which, when executed by a processor, performs the active noise reduction method for automotive electromagnetic compatibility testing as described in any one of claims 1 to 7.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, performs the active noise reduction method for automotive electromagnetic compatibility testing as described in any one of claims 1 to 7.

Images