Method for suppressing coupling interference of communication line of high-voltage wiring harness of electric vehicle

Abstract

The invention discloses a method for suppressing coupling interference of a communication line of a high-voltage wiring harness of an electric vehicle, and the method comprises the steps: constructing an interference reference signal and a communication interference observation signal based on a current sampling signal of the high-voltage wiring harness and a receiving end voltage monitoring signal of the communication line, and generating an interference frequency band configuration parameter in combination with short-time Fourier transform, a cancellation signal is output under the constraint least mean square adaptive framework and is superposed and fed back through a cancellation injection network, so that the coupling interference voltage is constrained by a suppression threshold value when the interference level changes; the problem that it is difficult to carry out broadband and working condition self-adaptive suppression on a high-voltage wire harness communication line coupling channel in an existing electric vehicle noise elimination technology and an active electromagnetic interference filtering technology is effectively solved.

Description

Technical Field

[0001] This invention relates to the technical field of electromagnetic compatibility and anti-interference of vehicle communication for electric vehicles, and particularly to a method for suppressing coupling interference of high-voltage wiring harness communication lines in electric vehicles. Background Technology

[0002] As electric vehicle high-voltage platforms develop towards higher voltage levels, the high-frequency switching and modulation control of power electronic devices such as drive inverters, on-board chargers, and DC / DC converters cause strong pulsating currents and common-mode disturbances in high-voltage wiring harnesses. Simultaneously, the complex spatial layout of vehicle wiring harnesses, with high-voltage harnesses and communication lines such as CAN / LIN / vehicle Ethernet often arranged in parallel, intersecting, or shared corridors in local areas, easily creating conduction and near-field coupling channels. Existing suppression measures mostly rely on shielding, isolation, grounding optimization, and passive filtering. However, in multi-condition scenarios with broadband transients and harmonic superposition, fixed-parameter solutions have limited adaptability to interference frequency drift and amplitude fluctuations. Further expanding the suppression bandwidth by thickening the shielding layer, improving termination consistency, or increasing isolation levels would put pressure on wiring harness cost, weight, and assembly consistency, and could still lead to sudden increases in communication interference and decreased link stability during acceleration / deceleration, regenerative braking, and fast charging switching.

[0003] CN105577204A discloses a noise cancellation method for electric vehicles. Based on the control idea of ​​noise detection and cancellation signal output, it reflects the active suppression technical route to a certain extent. However, its focus is more on vehicle noise or generalized interference control. It does not establish a corresponding signal pair for the coupled link by the reference quantity on the current side of the high-voltage harness and the observed measurement on the voltage side of the communication receiver, nor does it provide a cross-operating condition interference frequency band identification and configuration mechanism.

[0004] CN108696117A discloses a compensated active electromagnetic interference filter, which proposes an active filter structure that senses and injects cancellation signals to detect noise in power conductors. It emphasizes high-frequency stability and low-frequency tolerance, but its suppression target is mainly EMI of the power supply side conductor. It lacks a feedforward adaptive parameter update framework with the communication link coupling interference voltage as the constraint target, and it is difficult to cover the condition-related coupling characteristics of high-voltage line harness disturbances on communication lines.

[0005] In summary, existing electric vehicle noise cancellation technologies and active electromagnetic interference filtering technologies still struggle to achieve broadband, condition-adaptive suppression of coupling interference in high-voltage harness communication lines. This invention provides a method for suppressing coupling interference in high-voltage harness communication lines of electric vehicles. It constructs an interference reference signal and a communication interference observation signal based on the high-voltage harness current sampling signal and the communication line receiver voltage monitoring signal. It then generates interference frequency band configuration parameters using short-time Fourier transform and outputs a cancellation signal under a constrained minimum mean square adaptive framework. This signal is then superimposed and fed back through a cancellation injection network, ensuring that the coupling interference voltage is constrained by a suppression threshold as the interference level changes. Summary of the Invention

[0006] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this section, the abstract and title of the invention. Such simplifications or omissions shall not be used to limit the scope of the present invention.

[0007] In view of the aforementioned existing problems, the present invention is proposed.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution: As a preferred embodiment of the method for suppressing coupling interference of high-voltage harness communication lines for electric vehicles described in this invention, the method involves: acquiring high-voltage harness current sampling signals and communication line receiving end voltage monitoring signals, and performing synchronous preprocessing at a unified sampling frequency to obtain interference reference signals and communication interference observation signals. Short-time Fourier transform analysis is performed on the interference reference signal to determine the broadband interference target frequency band, and interference frequency band configuration parameters corresponding to the broadband interference target frequency band are generated. During the communication data transmission process, the interference reference signal is input into the feedforward adaptive filter initialized with the interference frequency band configuration parameters. Based on the error signal between the communication interference observation signal and the output of the feedforward adaptive filter, the filter parameters of the feedforward adaptive filter are updated using the constrained minimum mean square algorithm to obtain the cancellation signal. The cancellation signal is superimposed onto the receiving end of the communication line by the cancellation injection network, and the residual interference error signal after cancellation is fed back to the feedforward adaptive filter, so as to keep the coupling interference voltage on the communication line below the preset suppression threshold when the interference level of the high voltage harness changes.

[0009] The beneficial effects of this invention are as follows: This invention collects high-voltage line harness current sampling signals and communication line receiving end voltage monitoring signals and performs synchronous preprocessing to form interference reference signals and communication interference observation signals, so that the high-voltage side disturbance and the communication side disturbance response have a consistent time reference; by performing short-time Fourier transform analysis on the interference reference signal and generating interference frequency band configuration parameters, the initialization and update range of the feedforward adaptive filter corresponds to the multi-condition interference spectrum changes; by using the constrained minimum mean square algorithm combined with the feedback of residual interference error signals to update the filter parameters and output a cancellation signal, the coupling interference voltage at the communication line receiving end is maintained within the preset suppression threshold constraint range when the high-voltage line harness interference level changes; compared with existing methods that mainly rely on shielding, isolation or fixed parameter filtering, the method of this invention requires less structural modification, the interference suppression basis is more in line with the operating condition changes, and the stability and consistency of the communication link are improved. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating the method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles, as shown in this invention. Figure 2 This is a wiring harness installation example diagram for the method of suppressing coupling interference of high-voltage wiring harness communication lines for electric vehicles, as shown in this invention. Detailed Implementation

[0011] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0012] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort should fall within the scope of protection of this invention.

[0013] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0014] According to an embodiment of the present invention, in combination Figure 1 , Figure 2The schematic diagram shown illustrates a method for suppressing coupling interference in communication lines of high-voltage harnesses in electric vehicles, specifically including the following steps: S1. Acquire the high-voltage line harness current sampling signal and the communication line receiver voltage monitoring signal, and perform synchronous preprocessing at a unified sampling frequency to obtain the interference reference signal and the communication interference observation signal. Note that the following should be noted in this step: S1.1. Collect the high-voltage harness current in the high-voltage circuit and record the sampling time at a uniform sampling frequency to obtain the high-voltage harness current sampling signal with time index; It should be noted that the high-voltage harness current can be acquired in the high-voltage circuit using at least one of the following methods: Hall effect current sensor, shunt resistor current sampling module, or flexible Rogowski coil. The current sensor is installed in the main power supply branch of the high-voltage harness or the input branch of the motor controller and connected to the on-board data acquisition unit. The unified sampling frequency is output from the master clock of the data acquisition unit to the current sensor sampling channel and the communication line voltage sampling channel. The unified sampling frequency can be selected as at least one of 200 kHz, 500 kHz, or 1 MHz to cover the switching frequency of the electric drive inverter and its harmonic energy distribution range. During the sampling process, the data acquisition unit records the sampling time and generates a time index each time sampling is triggered, so that the high-voltage harness current sampling signal forms a continuous sequence with a time index.

[0015] For example, when the electric drive switch frequency is in the range of 10 kHz to 20 kHz, a uniform sampling frequency of 500 kHz is selected, which can obtain distinguishable sampling sequences of 2nd, 3rd and higher order harmonics under time indexing, so as to locate the frequency band in the subsequent bandpass pre-filtering step.

[0016] S1.2. Collect the line voltage at the receiving end of the communication line and record the sampling time using the same sampling clock as the high-voltage line harness current sampling signal to obtain the communication line receiving end voltage monitoring signal with time index. It should be noted that the method of acquiring line voltage at the communication line receiving end is to use a combination of differential voltage probe and high-bandwidth isolated sampling front end. The sampling clock and the sampling clock of the high-voltage line harness current sampling signal are from the same source, and both are output by frequency division of the same master clock to avoid time drift caused by cross-clock domain. The frequency of the sampling clock is consistent with the unified sampling frequency, such as 500 kHz, and the sampling time is written into the time index field in each sampling period, so that the voltage monitoring signal at the communication line receiving end and the high-voltage line harness current sampling signal have a time reference that can be directly aligned.

[0017] For example, the communication line can be CAN, CAN FD, Ethernet twisted pair, or a high-speed diagnostic link inside a high-voltage component, and the receiving end voltage monitoring point can be set between the transceiver input and the common-mode reference node.

[0018] S1.3. Resample and time-align the high-voltage harness current sampling signal and the communication line receiving end voltage monitoring signal according to the time index to obtain synchronously aligned high-voltage harness current sampling signal and synchronously aligned communication line receiving end voltage monitoring signal. In a preferred embodiment, resampling and time alignment employ an interpolation alignment strategy based on time indexing. Specifically, a target time grid is first constructed with a unified sampling frequency, and the high-voltage harness current sampling signal and the communication line receiver voltage monitoring signal are mapped to the target time grid respectively. When a signal is missing a sampling value at the target time grid, linear interpolation is used to generate a compensation sampling value at the corresponding time point, thereby obtaining synchronously aligned high-voltage harness current sampling signal and synchronously aligned communication line receiver voltage monitoring signal.

[0019] For example, when the current sampling channel experiences a short-term sampling loss due to anti-interference protection, interpolation alignment can enable the synchronously aligned high-voltage harness current sampling signal to form a one-to-one corresponding sampling pair with the synchronously aligned communication line receiving voltage monitoring signal in terms of time index, which can then be used for subsequent pre-filtering and adaptive filtering processing with the same bandwidth.

[0020] S1.4 Apply a bandpass pre-filter centered on the electric drive switching frequency and its harmonics to the synchronously aligned high-voltage harness current sampling signal to filter out low-frequency operating condition variation components and unrelated high-frequency noise, and obtain the interference reference signal; In a preferred embodiment, the bandpass pre-filter employs a digital IIR bandpass filter or an FIR linear phase bandpass filter; the bandpass center frequency is set around the electric drive switching frequency and its harmonics, and the electric drive switching frequency can be read from the motor controller diagnostic interface.

[0021] For example, when the electric drive switching frequency is 12 kHz, the bandpass can be set to 8 kHz~40 kHz or 10 kHz~60 kHz to cover the main energy range of the 1st to 5th order harmonics. The order of the bandpass pre-filter is selected according to the unified sampling frequency and the target passband width. For example, under the sampling condition of 500 kHz, a 128th order FIR filter is selected to obtain a steeper transition band. After the bandpass pre-filter, the low-frequency operating condition variation component and uncorrelated high-frequency random noise in the synchronously aligned high-voltage harness current sampling signal are suppressed, and an interference reference signal characterizing the source of coupling interference is obtained.

[0022] S1.5 Apply a bandpass pre-filter with the same bandwidth as the interference reference signal to the voltage monitoring signal at the receiving end of the synchronously aligned communication line, and perform amplitude normalization processing to obtain the communication interference observation signal.

[0023] It should be noted that the bandpass pre-filter, which has the same bandwidth as the interference reference signal, uses the same filter type and passband boundary parameters as S1.4, so that the communication interference observation signal and the interference reference signal have a matching interference component expression in the frequency domain. Subsequently, the voltage monitoring signal at the receiving end of the synchronously aligned communication line is subjected to amplitude normalization processing, and the amplitude normalization adopts scale normalization based on the root mean square value.

[0024] For example, the root mean square amplitude of the voltage monitoring signal at the receiving end of the synchronously aligned communication line is first calculated within a preset time window, and the sampling sequence of the current window is divided by the corresponding root mean square amplitude to obtain a communication interference observation signal with consistent amplitude scale, which facilitates the migration and updating of parameters of the feedforward adaptive filter under different vehicle operating conditions.

[0025] Preferably, synchronous preprocessing can reduce the impact of cross-channel asynchronous errors and irrelevant frequency band noise on subsequent frequency band selection and filter weight learning, improve the correlation and distinguishability between interference reference and observation signal, and thus improve the stability and convergence consistency of subsequent cancellation processing.

[0026] S2. Perform short-time Fourier transform analysis on the interference reference signal to determine the broadband interference target frequency band and generate interference frequency band configuration parameters corresponding to the broadband interference target frequency band. Note that the following points should be noted in this step: S2.1 Perform a short-time Fourier transform on the interference reference signal according to the preset window length and step size to obtain the time-frequency distribution of the interference energy of the interference reference signal; It should be noted that the window function type, window length, and step size of the short-time Fourier transform are set according to the sampling frequency of the interference reference signal and the estimated interference main frequency range to obtain the short-time Fourier transform configuration parameters. The interference reference signal is segmented according to the window length and step size, and each segment of the interference reference signal is multiplied by the window function to obtain a windowed interference reference signal sequence. A fast Fourier transform is performed on each segment of the interference reference signal in the windowed interference reference signal sequence to obtain the amplitude spectrum of each segment of the interference reference signal on the frequency axis. The energy of each frequency point is calculated based on the amplitude spectrum, and the signals are arranged according to the corresponding time period and frequency point to obtain the time-frequency distribution of the interference energy of the interference reference signal.

[0027] For example, the short-time Fourier transform configuration parameters include: the sampling frequency of the interference reference signal, the estimated interference frequency range, the window function type, the window length, and the step size; the sampling frequency follows the unified sampling frequency of S1, 500 kHz; the estimated interference frequency range can be set to 8 kHz~80 kHz according to the electric drive switch frequency and its harmonic range; the window function type is selected as Hamming window; the window length is selected as 4096 points; and the step size is selected as 256 points.

[0028] Furthermore, the processing of segmentation and windowing of the interference reference signal can be carried out by sliding the signal according to the window length, and multiplying each segment of the signal with the window function point by point to form a windowed interference reference signal sequence. The fast Fourier transform adopts the FFT calculation process based on radix-2 splitting, and the complex spectrum is solved for each windowed sequence to obtain the amplitude spectrum.

[0029] As an example, the energy at each frequency point is calculated based on the amplitude spectrum in the following way: in, Let K be the interference energy at the k-th frequency point in the m-th time period. This is the complex spectrum value at the k-th frequency point obtained after performing an FFT on the m-th windowed interference reference signal; The amplitude of the complex spectrum value; Relating each time period to each frequency point Arranged in a time-frequency index manner, the time-frequency distribution of interference energy of the interference reference signal is obtained.

[0030] S2.2. In the time-frequency distribution of interference energy, the mean and variance of interference energy at each frequency point under multiple vehicle operating conditions are statistically analyzed. Based on the interference energy judgment threshold and the interference energy fluctuation judgment threshold, frequency intervals with high mean interference energy and large interference energy variance are selected to obtain a set of candidate interference frequency bands. S2.2.1 Collect the time-frequency distribution of interference energy under multiple vehicle operating conditions, and align the interference energy of all operating conditions according to the frequency points to form a set of interference energy samples indexed by the frequency points. It should be noted that for each frequency point in the interference energy sample set, the mean and variance of interference energy are calculated based on the interference energy under each vehicle operating condition to obtain the frequency point interference statistical characteristics. Based on the interference energy judgment threshold, frequency points with a mean interference energy greater than the interference energy judgment threshold are selected from the frequency point interference statistical characteristics to obtain the first frequency point set. Based on the interference energy fluctuation judgment threshold, frequency points with a variance of interference energy greater than the interference energy fluctuation judgment threshold are selected from the first frequency point set to obtain the second frequency point set. Adjacent frequency points in the second frequency point set are merged in order of frequency magnitude to form multiple continuous frequency intervals. The continuous frequency intervals are used as candidate interference frequency bands, and the lower limit, upper limit, and corresponding mean interference energy of each continuous frequency interval are recorded to obtain the candidate interference frequency band set.

[0031] In a preferred embodiment, multiple vehicle operating conditions include: different vehicle speed ranges, different torque output ranges, different regenerative braking intensity ranges, and different battery SOC ranges. Under each operating condition, a set of time-frequency distributions of interference energy is obtained according to S2.1.

[0032] When forming an interference energy sample set by aligning frequency points, the time-frequency energy matrix corresponding to each operating condition can be statistically aggregated along the time dimension to obtain an energy sample table of operating condition-frequency points, and then the sample set across operating conditions can be formed by superimposing the frequency point index.

[0033] For example, the mean and variance of interference energy are calculated using the following formulas: in, The average interference energy at the k-th frequency point under N vehicle operating conditions; Let Variance be the interference energy variance at the k-th frequency point under N vehicle operating conditions; is the statistical energy value at the k-th frequency point under the i-th vehicle operating condition; N is the number of selected vehicle operating conditions; Furthermore, the interference energy determination threshold is set to the entire frequency band. The 80th percentile value; the threshold for judging interference energy fluctuations is set to the entire frequency band. 70th percentile; When merging adjacent frequency points to form a continuous frequency range, the adjacent criterion is set to the frequency interval not exceeding 1 to 2 times the frequency resolution; for example, under the conditions of a 4096-point window length and 500 kHz sampling, the frequency resolution is 122 Hz, and the adjacent merging criterion is set to the frequency interval not exceeding 250 Hz. The lower and upper frequency limits of a continuous frequency range can be taken as the minimum and maximum frequency points within that range, respectively. For example, candidate ranges such as (11.8 kHz, 18.6 kHz) and (24.1 kHz, 37.5 kHz) can be formed, and the frequency values ​​within each range can be recorded. The weighted average is used as the corresponding mean of interference energy.

[0034] S2.3 Select a continuous frequency range covering the upper and lower parts of the communication frequency band from the candidate interference frequency band set, and determine the broadband interference target frequency band according to the lower frequency limit, the upper frequency limit and the corresponding interference energy weight; In a preferred embodiment, the communication frequency band can be determined according to the transceiver bandwidth specifications of the vehicle communication protocol. When selecting a continuous frequency range covering the upper and lower parts of the communication frequency band from the candidate interference frequency band set, the selection is carried out according to a combination of proximity priority and energy weight priority rules.

[0035] For example, when the communication frequency band is mainly concentrated in the differential signal spectrum below 1 MHz, and the main energy of electric drive interference is concentrated in 10 kHz to 60 kHz, the lower and upper intervals closest to the sensitive area of ​​the communication link can be selected from the candidate set, and the interference energy weight is defined according to the average interference energy ratio of the candidate intervals to form a broadband interference target frequency band covering the upper and lower sides.

[0036] S2.4 Generate interference band configuration parameters based on the frequency range and interference energy weight of the broadband interference target band. The interference band configuration parameters include the lower frequency limit, upper frequency limit, order of the feedforward adaptive filter, and adaptive update step size weight of the broadband interference target band.

[0037] In a preferred embodiment, the interference band configuration parameters can be generated by mapping parameters based on the bandwidth and energy weight of the broadband interference target frequency band. Specifically, when the broadband interference target frequency band is 10 kHz to 50 kHz, the lower and upper frequency limits can be directly written into the configuration parameters. The order of the feedforward adaptive filter can be selected according to the ratio of sampling frequency to target bandwidth. For example, under the conditions of 500 kHz sampling and 40 kHz target bandwidth, the order of the feedforward adaptive filter can be selected as 64, 96, or 128. The adaptive update step size weight can be set to discrete values ​​such as 0.2, 0.5, or 0.8 according to the interference energy weight classification to distinguish the convergence sensitivity of the high interference energy range and the secondary range.

[0038] Preferably, this step, through cross-condition statistics and weighted parameter mapping, can improve the adaptability of the subsequent adaptive canceller to changes in interference frequency bands and reduce the risk of overlearning non-target frequency bands.

[0039] S3. During communication data transmission, the interference reference signal is input into the feedforward adaptive filter initialized with interference frequency band configuration parameters. Based on the error signal between the observed communication interference signal and the output of the feedforward adaptive filter, the filter parameters of the feedforward adaptive filter are updated using the constrained least mean square algorithm to obtain the cancellation signal. It should be noted that the following points are important in this step: S3.1. Based on the interference frequency band configuration parameters, set the filter order, initial filter parameters, and step size of the constrained minimum mean square algorithm of the feedforward adaptive filter to obtain the initial state of the feedforward adaptive filter. Specifically, based on the lower and upper frequency limits in the interference band configuration parameters, the bandwidth of the broadband interference target band is calculated by subtracting the lower frequency limit from the upper frequency limit. The lower and upper frequency limits are then divided by the sampling frequency of the interference reference signal to obtain the normalized reference frequency range of the broadband interference target band within the range of [lower frequency limit / sampling frequency, upper frequency limit / sampling frequency]. Using the order of the feedforward adaptive filter in the interference band configuration parameters as the number of taps, a feedforward adaptive filter structure containing tap delay units, weight multiplication units, and addition units is constructed, and the filter parameters of the feedforward adaptive filter are initialized to zero. Based on the adaptive update step size weights in the interference band configuration parameters and the power estimate of the interference reference signal, combined with the normalized minimum mean square convergence condition, the upper and lower limits of the constrained minimum mean square algorithm are calculated to obtain the constrained step size interval. The feedforward adaptive filter structure, the filter parameter initialization results, and the constrained step size interval are combined to define the initial state of the feedforward adaptive filter.

[0040] In a preferred embodiment, the power estimate of the interfering reference signal can be obtained using a sliding window mean square estimation, for example, selecting a window with a length of... The power estimation window is used to calculate the square of the second norm of the vector formed by the current sampled interference reference signal and then normalize it. For example, the normalized least mean square convergence condition adopts the following constraint relationship: in, To constrain the step size of the minimum mean square algorithm; The power estimate of the interference reference signal; Combine adaptive step size weight update Define the upper and lower limits of the step size as follows: in, This is the upper limit of the step size; This is the lower limit of the step size; To adaptively update the step size weights; This is a power estimate.

[0041] For example, when the adaptive update step size weight is 0.5 and the power estimate is approximately 0.8, the upper limit of the step size is 1.25 and the lower limit of the step size is 0.625.

[0042] S3.2 During the communication data transmission process, the interference reference signal and the communication interference observation signal at the current moment are sampled according to a unified sampling period to obtain the current sampled interference reference signal and the current sampled communication interference observation signal. In a preferred embodiment, the uniform sampling period is determined by the data acquisition unit based on the uniform sampling frequency. For example, when the uniform sampling frequency is 500 kHz, the uniform sampling period is 2 μs. During the communication data transmission process, the current sampling interference reference signal and the current sampling communication interference observation signal are acquired and written into the uniform time index on the same sampling trigger edge to form a dual-channel synchronous sample pair for adaptive filtering.

[0043] S3.3 Input the current sampled interference reference signal into the feedforward adaptive filter, and calculate the current filtered output signal based on the filter parameters of the feedforward adaptive filter; Specifically, based on the filter order of the feedforward adaptive filter, a sampling sequence of the same length as the filter order is extracted from the current sampled interference reference signal and its preceding sampled values ​​to form the current interference reference vector. The current interference reference vector is input into the tap delay unit of the feedforward adaptive filter, multiplied element-by-element by the filter parameters, and the products are summed to obtain the current filtered output signal. At the same time, based on the difference between the power of the current interference reference vector and the smoothed power of the interference reference vector in the previous sampling period, the amplitude transition judgment threshold is compared. When the difference is greater than the amplitude transition judgment threshold, the feedforward adaptive filter is marked as the parameter frozen state of the current sampling period and output together with the current filtered output signal.

[0044] In a preferred embodiment, the current interference reference vector is represented as: in, This is the current interference reference vector for the nth sampling period; The current sampled value of the interference reference signal is denoted as ; M is the filter order of the feedforward adaptive filter. This is a transpose operation; The current filtered output signal is represented as: in, This is the current filtered output signal; This is the filter parameter vector for the nth sampling period; The smoothed power is estimated using an exponential moving average: in, The smoothed power of the current interference reference vector in the nth sampling period; This is a smoothing coefficient, with a value ranging from 0.9 to 0.99. The energy of the current disturbance reference vector; Further construct the power difference: in, This is the power difference; when Greater than the amplitude transition determination threshold When this happens, the feedforward adaptive filter is marked as being in a parameter-frozen state for the current sampling period, meaning it is not executed during that sampling period. The update operation only outputs... .

[0045] S3.4 Construct an error signal based on the difference between the current sampled communication interference observation signal and the current filtered output signal, and weight the step size of the constrained minimum mean square algorithm according to the weight range limited by the interference frequency band configuration parameters, and update the filter parameters of the feedforward adaptive filter. As an example, the error signal is represented as follows: in, This is an error signal; This is the current sampling signal used to observe communication interference. This is the current filtered output signal; The parameter update of the constrained least mean square algorithm is expressed as: in, This is the filter parameter vector for the next sampling period; The weighted adaptive step size; The weight range can be defined based on the interference frequency band configuration parameters. The selection is performed within the target frequency band and then discretely mapped based on the energy weights of the broadband interference target band.

[0046] S3.5 After completing the filter parameter update, output the current filtered output signal as the cancellation signal.

[0047] Preferably, in this step, the interference reference signal is input into the feedforward adaptive filter initialized based on the interference frequency band configuration parameters, and the constrained minimum mean square parameter update is performed according to the error signal between the communication interference observation signal and the current filtered output signal, thereby outputting the cancellation signal corresponding to the coupling interference component on the communication side; through the parameter freezing mechanism of step size interval constraint and amplitude transition discrimination, the risk of weight divergence caused by transient operating condition changes can be reduced, and the convergence stability of the broadband interference frequency band can be improved.

[0048] S4. The inverted signal of the cancellation signal is superimposed onto the receiving end of the communication line through the cancellation injection network, and the residual interference error signal after cancellation is fed back to the feedforward adaptive filter, thereby keeping the coupling interference voltage on the communication line below the preset suppression threshold when the interference level of the high-voltage harness changes. It should be noted that in this step: S4.1 Input the cancellation signal into the cancellation injection network, and obtain the inverse cancellation signal through inverted signal transformation and injection gain adjustment; Specifically, the cancellation signal is input into the cancellation injection network, and the injection gain coefficient corresponding to the cancellation signal is read from the cancellation injection network; the cancellation signal is inverted to obtain an intermediate inverted signal; the intermediate inverted signal is multiplied by the injection gain coefficient to obtain the gain-adjusted inverted cancellation signal; the gain-adjusted inverted cancellation signal is amplitude-limited to obtain the inverted cancellation signal and output to the superposition node at the receiving end of the communication line.

[0049] In a preferred embodiment, the cancellation injection network includes: a digital-to-analog converter, an inverting operation unit, a programmable gain unit, and a limiting protection unit; the injection gain coefficient can be configured by the vehicle calibration table according to different communication link lengths, shielding structures, and grounding topologies; the inverted cancellation signal after gain adjustment is represented as: in, This is the inverse cancellation signal after gain adjustment; This refers to the injection gain coefficient. To cancel the signal; Amplitude limiting is achieved using the following formula: in, This is the opposite of the cancellation signal; The upper limit of the amplitude can be set according to the safe voltage range of the communication transceiver input.

[0050] S4.2 At the superposition node of the communication line receiving end, the inverse cancellation signal is superimposed with the voltage monitoring signal of the communication line receiving end to obtain the canceled voltage signal of the communication line receiving end. In a preferred embodiment, the superposition node can be set at the analog coupling point or the front end of the common-mode rejection network at the front end of the communication transceiver; the method of superimposing the anti-cancellation signal with the voltage monitoring signal at the receiving end of the communication line can be achieved by using a resistor summing network or a differential injection coupling network.

[0051] For example, in a differential link, the inverse cancellation signal can be allocated according to the injection coefficients of the differential positive and differential negative terminals, so that the cancellation injection can cancel the coupling interference components without changing the protocol logic level decision.

[0052] S4.3. Perform bandpass pre-filtering on the voltage signal at the receiving end of the communication line after cancellation, according to the same bandpass pre-filtering configuration as the interference reference signal, to obtain the communication interference observation signal after cancellation. In a preferred embodiment, the voltage signal at the receiving end of the canceled communication line is processed using the same bandpass pre-filtering configuration as in S1.5 to obtain a canceled communication interference observation signal with the same frequency band expression vector as the interference reference signal.

[0053] For example, when the bandpass of S1.4 / S1.5 is set to 10 kHz to 60 kHz, the pre-filter of the canceled bandpass also uses the same passpass and filter order, which facilitates the subsequent threshold discrimination and feedback update of residual error.

[0054] S4.4 Construct a residual interference error signal based on the difference between the amplitude of the observed signal after cancellation of communication interference and the preset suppression threshold. When the residual interference error signal is greater than the preset suppression threshold, the residual interference error signal is used as the error signal for the next sampling period and input into the feedforward adaptive filter. When the residual interference error signal is less than or equal to the preset suppression threshold, the filter parameters of the feedforward adaptive filter follow the update results of the previous sampling period.

[0055] As an example, the residual interference error signal is defined as: in, This is a residual interference error signal; The current amplitude or root mean square amplitude of the observed signal after communication interference has been eliminated; The preset suppression threshold; Furthermore, the selection of the preset suppression threshold can be determined based on the immunity specifications of the communication transceiver and the EMC target limit of the whole vehicle.

[0056] For example, a preset suppression threshold can be set to a safety margin value for the common-mode / differential-mode disturbance voltage allowed at the transceiver input; when At that time, the residual interference error signal corresponding to The feedforward adaptive filter serves as the updated reference input to the error signal of the next sampling period; when At this time, the filter parameters of the feedforward adaptive filter use the update results of the previous sampling period.

[0057] Preferably, this step, through injection gain coefficient calibration, amplitude limiting protection, and residual error feedback rules based on a preset suppression threshold, can reduce the impact of cancellation injection on the safety voltage boundary of the communication front end and improve cancellation stability under broadband interference time-varying conditions.

[0058] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for suppressing coupling interference in communication lines of high-voltage harnesses for electric vehicles, characterized in that, include: The high-voltage line harness current sampling signal and the communication line receiver voltage monitoring signal are collected and synchronously preprocessed at a unified sampling frequency to obtain the interference reference signal and the communication interference observation signal. Short-time Fourier transform analysis is performed on the interference reference signal to determine the broadband interference target frequency band, and interference frequency band configuration parameters corresponding to the broadband interference target frequency band are generated. During the communication data transmission process, the interference reference signal is input into the feedforward adaptive filter initialized with the interference frequency band configuration parameters. Based on the error signal between the communication interference observation signal and the output of the feedforward adaptive filter, the filter parameters of the feedforward adaptive filter are updated using the constrained minimum mean square algorithm to obtain the cancellation signal. The cancellation signal is superimposed onto the receiving end of the communication line by the cancellation injection network, and the residual interference error signal after cancellation is fed back to the feedforward adaptive filter, so as to keep the coupling interference voltage on the communication line below the preset suppression threshold when the interference level of the high voltage harness changes.

2. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 1, characterized in that, Obtaining the interference reference signal and the communication interference observation signal includes: The high-voltage harness current is collected in the high-voltage circuit, and the sampling time is recorded at a uniform sampling frequency to obtain the high-voltage harness current sampling signal with time index; The line voltage is collected at the receiving end of the communication line, and the sampling time is recorded using the same sampling clock as the high-voltage line harness current sampling signal to obtain the communication line receiving end voltage monitoring signal with time index. The high-voltage harness current sampling signal and the communication line receiving end voltage monitoring signal are resampled and time-aligned according to the time index to obtain synchronously aligned high-voltage harness current sampling signal and synchronously aligned communication line receiving end voltage monitoring signal. A bandpass pre-filter centered on the electric drive switching frequency and its harmonics is applied to the synchronously aligned high-voltage harness current sampling signal to filter out low-frequency operating condition variation components and unrelated high-frequency noise, thereby obtaining an interference reference signal. The voltage monitoring signal at the receiving end of the synchronized communication line is subjected to a bandpass pre-filter with the same bandwidth as the interference reference signal and then subjected to amplitude normalization to obtain the communication interference observation signal.

3. The method for suppressing coupling interference in high-voltage wiring harness communication lines for electric vehicles according to claim 1 or 2, characterized in that, Generating the interference frequency band configuration parameters includes: A short-time Fourier transform is performed on the interference reference signal according to a preset window length and step size to obtain the time-frequency distribution of the interference energy of the interference reference signal; In the time-frequency distribution of interference energy, the mean and variance of interference energy at each frequency point under multiple vehicle operating conditions are statistically analyzed. Based on the interference energy judgment threshold and the interference energy fluctuation judgment threshold, frequency intervals with high mean interference energy and large interference energy variance are screened to obtain a set of candidate interference frequency bands. From the set of candidate interference frequency bands, select a continuous frequency range covering the area above and below the communication frequency band, and determine the broadband interference target frequency band based on the lower frequency limit, the upper frequency limit, and the corresponding interference energy weight; Interference band configuration parameters are generated based on the frequency range and interference energy weight of the broadband interference target band. The interference band configuration parameters include the lower frequency limit, upper frequency limit, order of the feedforward adaptive filter, and adaptive update step size weight of the broadband interference target band.

4. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 3, characterized in that, Obtaining the time-frequency distribution of the interference energy of the interference reference signal includes: Based on the sampling frequency of the interference reference signal and the estimated interference main frequency range, the window function type, window length, and step size of the short-time Fourier transform are set to obtain the short-time Fourier transform configuration parameters. The interference reference signal is segmented according to the window length and step size, and each segment of the interference reference signal is multiplied by the window function to obtain a windowed interference reference signal sequence. Perform a fast Fourier transform on each segment of the interference reference signal in the windowed interference reference signal sequence to obtain the amplitude spectrum of each segment of the interference reference signal on the frequency axis; The energy at each frequency point is calculated based on the amplitude spectrum, and the energy is arranged according to the corresponding time period and frequency point to obtain the time-frequency distribution of the interference energy of the interference reference signal.

5. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 4, characterized in that, The candidate interference frequency band set is obtained, including: The time-frequency distribution of interference energy under multiple vehicle operating conditions is collected, and the interference energy under all operating conditions is aligned according to the frequency point to form a set of interference energy samples indexed by the frequency point. For each frequency point in the interference energy sample set, the mean and variance of interference energy are calculated based on the interference energy under each vehicle operating condition to obtain the frequency point interference statistical characteristics. Based on the interference energy determination threshold, frequency points with an average interference energy greater than the interference energy determination threshold are selected from the frequency point interference statistical characteristics to obtain the first frequency point set. Based on the interference energy fluctuation judgment threshold, select frequency points from the first frequency point set whose interference energy variance is greater than the interference energy fluctuation judgment threshold to obtain a second frequency point set; In the second set of frequency points, adjacent frequency points are merged in order of frequency magnitude to form multiple consecutive frequency intervals; The continuous frequency intervals are used as candidate interference frequency bands. For each continuous frequency interval, the lower frequency limit, the upper frequency limit, and the corresponding average interference energy are recorded to obtain a set of candidate interference frequency bands.

6. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 3, characterized in that, Obtaining the cancellation signal includes: The filter order, initial filter parameters, and step size of the constrained minimum mean square algorithm of the feedforward adaptive filter are set according to the interference frequency band configuration parameters to obtain the initial state of the feedforward adaptive filter. During the communication data transmission process, the interference reference signal and the communication interference observation signal at the current moment are sampled according to a uniform sampling period to obtain the current sampled interference reference signal and the current sampled communication interference observation signal. The current sampled interference reference signal is input into the feedforward adaptive filter, and the current filtered output signal is calculated based on the filter parameters of the feedforward adaptive filter. An error signal is constructed based on the difference between the current sampled communication interference observation signal and the current filtered output signal. The step size of the constrained minimum mean square algorithm is weighted according to the weight range defined by the interference frequency band configuration parameters, and the filter parameters of the feedforward adaptive filter are updated. After the filter parameters are updated, the current filtered output signal is output as a cancellation signal.

7. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 6, characterized in that, Obtaining the initial state of the feedforward adaptive filter includes: Based on the lower and upper frequency limits in the interference band configuration parameters, the bandwidth of the broadband interference target band is calculated by subtracting the lower frequency limit from the upper frequency limit. The lower and upper frequency limits are then divided by the sampling frequency of the interference reference signal to obtain the normalized reference frequency range of the broadband interference target band within the range of [lower frequency limit / sampling frequency, upper frequency limit / sampling frequency]. Using the order of the feedforward adaptive filter in the interference frequency band configuration parameters as the number of taps, a feedforward adaptive filter structure containing tap delay units, weight multiplication units and addition units is constructed, and the filter parameters of the feedforward adaptive filter are initialized to zero. Based on the adaptive update step size weight in the interference frequency band configuration parameters and the power estimate of the interference reference signal, the upper limit and lower limit of the step size of the constrained minimum mean square algorithm are calculated in combination with the normalized minimum mean square convergence condition, and the constrained step size interval is obtained. The combination of the feedforward adaptive filter structure, the filter parameter initialization results, and the constraint step size interval is defined as the initial state of the feedforward adaptive filter.

8. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 6, characterized in that, The calculation of the current filtered output signal includes: Based on the filter order of the feedforward adaptive filter, a sampling sequence of the same length as the filter order is extracted from the current sampled interference reference signal and its preceding sampled values ​​to form the current interference reference vector; The current interference reference vector is input into the tap delay unit of the feedforward adaptive filter, multiplied element by element with the filter parameters, and the products are summed to obtain the current filtered output signal. Simultaneously, based on the difference between the power of the current interference reference vector and the smoothed power of the interference reference vector in the previous sampling period, an amplitude transition determination threshold is compared. When the difference is greater than the amplitude transition determination threshold, the feedforward adaptive filter is marked as the parameter frozen state of the current sampling period and output together with the current filtered output signal.

9. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 6, characterized in that, Also includes: The cancellation signal is input into the cancellation injection network, and an inverse cancellation signal is obtained through inverted signal transformation and injection gain adjustment; At the superposition node of the communication line receiving end, the inverse cancellation signal is superimposed with the voltage monitoring signal of the communication line receiving end to obtain the canceled voltage signal of the communication line receiving end. The canceled communication line receiving voltage signal is bandpass pre-filtered according to the same bandpass pre-filtering configuration as the interference reference signal to obtain the canceled communication interference observation signal. A residual interference error signal is constructed based on the difference between the amplitude of the canceled communication interference observation signal and a preset suppression threshold. When the residual interference error signal is greater than the preset suppression threshold, the residual interference error signal is used as the error signal for the next sampling period and input into the feedforward adaptive filter. When the residual interference error signal is less than or equal to the preset suppression threshold, the filter parameters of the feedforward adaptive filter follow the update results of the previous sampling period.

10. The method for suppressing coupling interference in high-voltage wiring harness communication lines of electric vehicles according to claim 9, characterized in that, The process of obtaining the inverse relative cancellation signal includes: The cancellation signal is input into the cancellation injection network, and the injection gain coefficient corresponding to the cancellation signal is read from the cancellation injection network; The cancellation signal is inverted to obtain an intermediate inverted signal; The intermediate inverted signal is multiplied by the injected gain coefficient to obtain the inverted cancellation signal after gain adjustment; The gain-adjusted inverse cancellation signal is subjected to amplitude limiting processing to obtain the inverse cancellation signal, which is then output to the superposition node at the receiving end of the communication line.

Images