Method, apparatus, equipment, and medium for evaluating NVH performance attenuation of electric drive systems

Abstract

Conventional NVH performance evaluation methods rely on subjective evaluation and lack the ability to predict the NVH performance decay trend, thus having certain limitations. [Solution] The present application discloses a method, apparatus, equipment and medium for evaluating NVH performance attenuation of an electric drive system, relating to the field of NVH performance evaluation of an electric drive system, the method comprising the steps of: obtaining sequential values ​​of historical NVH performance; determining model parameters based on the sequential values; constructing a predictive model based on the model parameters, wherein the predictive model is used to predict the NVH performance attenuation status of the electric drive system at a predetermined point in time; and extracting feature values ​​based on the predictive model and establishing an evaluation model based on the feature values, wherein the evaluation model is used to evaluate the NVH performance attenuation status. The present application realizes the prediction of NVH performance attenuation trends.

Description

[Technical Field]

[0001] This application relates to the field of NVH performance evaluation, and more particularly to methods, apparatus, devices, and media for evaluating NVH performance attenuation of electric drive systems. [Background technology]

[0002] NVH in electric drive systems refers to the noise, vibration, and harshness generated by the electric drive system (including components such as motors, controllers, and gearboxes) applied to electric vehicles. With the rapid development of electric vehicle technology, NVH performance of electric drive systems has become an important indicator for measuring the overall quality and ride comfort of a vehicle. NVH performance decay in electric drive systems refers to the gradual deterioration of noise, vibration, and harshness in the electric drive system over time and with changes in usage conditions.

[0003] As the integration density of systems increases, electric drive systems in new energy vehicles may offer higher efficiency and performance, but at the same time, they may introduce more complex NVH (noise, vibration, and harshness) performance problems.

[0004] However, conventional NVH performance evaluation methods rely on subjective evaluation and lack the ability to predict the NVH performance decay trend, thus having certain limitations. [Overview of the project] [Problems that the invention aims to solve]

[0005] The object of this application is to provide a method, apparatus, equipment, and medium for evaluating NVH performance attenuation of an electric drive system, and to predict the trend of NVH performance attenuation. [Means for solving the problem]

[0006] To achieve the above objective, this application provides the following solutions.

[0007] According to the first aspect, the present application provides a method for evaluating the NVH performance attenuation of an electric drive system. Steps include obtaining sequential values ​​of historical NVH performance for the predicted target, The steps include determining model parameters based on the aforementioned series values, A step of constructing a predictive model based on the aforementioned model parameters, wherein the predictive model is used to predict the NVH performance decay status of the electric drive system at a predetermined time point. The steps include: extracting feature values ​​based on the prediction model and establishing an evaluation model based on the feature values, wherein the evaluation model is used to evaluate the NVH performance attenuation status.

[0008] According to a second aspect, the present application provides an evaluation device for NVH performance attenuation of an electric drive system. A module for obtaining sequential values ​​of historical NVH performance, A decision module for determining model parameters based on the aforementioned serial values, A construction module for constructing a predictive model based on the aforementioned model parameters, wherein the predictive model is used to predict the NVH performance decay status of the electric drive system at a predetermined time point. An evaluation module for extracting feature values ​​based on the prediction model and establishing an evaluation model based on the feature values, wherein the evaluation model includes an evaluation module used to evaluate the NVH performance attenuation status.

[0009] According to a third aspect, the present invention provides a computer device comprising memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program to realize the steps of the method for evaluating the NVH performance attenuation of an electric drive system described in any one of the above paragraphs.

[0010] According to the fourth aspect, the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, it realizes the steps of the method for evaluating the NVH performance attenuation of the electric drive system according to any one of the above items.

Advantages of the Invention

[0011] According to the specific embodiments provided in the present application, the present application discloses the following technical effects. The present application provides a method, device, equipment and medium for evaluating the NVH performance attenuation of an electric drive system, obtaining a series of values of historical NVH performance, determining model parameters based on the series of values, constructing a prediction model based on the model parameters, and the prediction model is used to predict the NVH performance attenuation situation of the electric drive system at a preset time point, realizing the prediction of the NVH performance attenuation trend. Extracting characteristic values based on the prediction model and establishing an evaluation model based on the characteristic values, and the evaluation model is used to evaluate the NVH performance attenuation situation, providing a scientific basis for the optimization and maintenance of the electric drive system.

Brief Description of the Drawings

[0012] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following briefly introduces the drawings that need to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative labor. [Figure 1] It is a diagram of the application environment of the method for evaluating the NVH performance attenuation of an electric drive system in an embodiment of the present application. [Figure 2] It is a schematic flowchart of the method for evaluating the NVH performance attenuation of an electric drive system provided in an embodiment of the present application. [Figure 3] It is a schematic flowchart of obtaining a series of values of the historical NVH performance of a prediction target provided in an embodiment of the present application. [Figure 4]This is a schematic diagram of a functional module of an evaluation device for NVH performance attenuation of an electric drive system provided in one embodiment of the present invention. [Figure 5] This is a schematic diagram of the structure of a computer device provided in one embodiment of the present invention. [Modes for carrying out the invention]

[0013] The following describes the technical solutions in the embodiments of the present application clearly and completely with reference to the drawings of the embodiments of the present application, and it is clear that the embodiments described are only some of the embodiments of the present application, not all of them. All other embodiments obtained based on the embodiments of the present application without creative work by a person skilled in the art are all within the scope of protection of the present application.

[0014] In cases where there are no conflicts, the embodiments and features described herein can be combined with each other. To make the above-mentioned objectives, features, and advantages of this application clearer and easier to understand, the present application will be described in more detail below with reference to the drawings and specific embodiments.

[0015] As can be understood, the NVH performance prediction method for electric drive systems provided in the embodiments of this application is used for electric drive systems of new energy vehicles, whose NVH performance is a future development trend.

[0016] NVH is an acronym for Noise, Vibration, and Harshness, a concept in engineering technology, particularly widely used in the automotive industry. NVH characteristics relate to a vehicle's ride comfort, quietness, and overall driving experience. Noise refers to various sounds generated during vehicle operation, including engine noise, tire noise, and wind noise, and these sounds can affect occupant comfort. Vibration refers to vibration phenomena of the vehicle's structure or components due to different factors, such as vibrations caused by engine operation or rattles caused by uneven road surfaces. Excessive vibration not only affects comfort but can also threaten the vehicle's durability and safety. Harshness is a more subjective evaluation index used to describe the degree of discomfort or unease experienced by occupants due to the synergistic effect of noise and vibration, and the impact of these elements on human perception relates to how "crude" or unacceptable they appear. Manufacturers of automobiles and other modes of transport analyze, simulate, and optimize these factors through NVH (Noise, Vibration, Harshness) engineering to improve the overall quality and market competitiveness of their products.

[0017] As shown in Figure 1, the method for evaluating the NVH performance attenuation of an electric drive system provided in the embodiment of the present invention can be applied to a hardware environment and / or software environment consisting of a terminal 102 and a server. Here, the terminal 102 communicates with the server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system may be installed independently, integrated with the server 104, or located in the cloud or on another server. The terminal 102 can transmit the acquired sequence values ​​of the historical NVH performance of the predicted target to the server 104. The server 104 receives the sequence values ​​of the historical NVH performance of the predicted target, determines model parameters based on the sequence values, constructs a prediction model based on the model parameters, the prediction model is used to predict the NVH performance attenuation status of the electric drive system at a preset point in time, extracts feature values ​​based on the prediction model, establishes an evaluation model based on the feature values, and the evaluation model is used to evaluate the NVH performance attenuation status. The server 104 can feed back the obtained prediction model and evaluation model to the terminal 102. Furthermore, in some embodiments, the NVH performance attenuation evaluation method may be implemented independently by the server 104 or terminal 102. For example, the terminal 102 may directly process the sequence values ​​of the historical NVH performance of the target prediction to be processed to obtain a prediction model and an evaluation model. Alternatively, the server 104 may obtain the sequence values ​​of the historical NVH performance of the target prediction to be processed from the data storage system, process these sequence values, and construct a prediction model and an evaluation model.

[0018] Here, terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smartphones, tablet computers, Internet of Things (IoT) devices, and portable wearable devices. Internet of Things devices may include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices may include smart watches, smart bracelets, head-mounted devices, etc. Server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or it may be a cloud server.

[0019] To facilitate understanding and explanation, the methods, apparatus, devices, and media for evaluating the NVH performance attenuation of an electric drive system provided in the embodiments of this application will be described in detail below with reference to Figures 2 to 5.

[0020] In one exemplary embodiment, as shown in Figure 2, a method for evaluating the NVH performance attenuation of an electric drive system is provided, the method being performed by computer equipment, specifically, by computer equipment such as a terminal or server alone, or by a terminal and a server together, and in the embodiment of this application, the method is described as being applied to server 104 in Figure 1 as an example, and includes the following steps S201 to S204. Here,

[0021] In step S201, the series values ​​of the historical NVH performance for the predicted target are obtained.

[0022] Specifically, NVH performance data refers to a series of numerical and analytical results obtained through a comprehensive quantization evaluation of three key indicators in automotive engineering: noise, vibration, and harshness. These data are important parameters for measuring the ride comfort, driving experience, and overall quality of a vehicle. To accurately monitor the operating state of an electric drive system during operation, a series of sensors are placed in key parts of the electric drive system. These sensors mainly include acceleration sensors and dedicated sensors for measuring sound pressure and acoustic power. Key parts include, but are not limited to, the motor, transmission belt, and bearings, as these are the main sources of vibration and noise. Exemplarily, multiple acceleration sensors and multiple sound pressure sensors are mounted on the rotor, stator, and housing of the electric drive system. These sensors allow for the real-time collection of vibration and noise signals during operation of the electric drive system at rated power.

[0023] In another exemplary embodiment of the present invention, as shown in Figure 3, step S201 above is replaced by the following steps S2011 to S2012.

[0024] In step S2011, NVH performance data is acquired based on durability tests of the same electric drive system, and a sensor is installed in the electric drive system. Here, the time period corresponding to the NVH performance data includes the initial time period and the decay time period, and the number of cycles predetermined for the durability test corresponds to the number of distances traveled under actual road conditions.

[0025] Specifically, NVH (Noise, Vibration, and Harshness) performance data can be collected by pre-configured sensors within the electric drive system, which includes collecting vibration and noise signals. In the NVH performance data collection process, one time period can be pre-configured, and an initial time period and a decay time period can be selected. The initial time period is the stage immediately after vibration and noise signals begin to be generated when the electric drive system starts operating, and the decay time period is the time period after the NVH performance has decayed over time.

[0026] Various sensors are placed in key parts of the electric drive system to enable the accurate acquisition of vibration and noise signals at will, for example, speed sensors, sound pressure sensors, and acoustic power sensors. These sensors can collect vibration and noise signals in real time while the electric drive system is in operation. Among these, acceleration sensors are mainly used to measure the acceleration of objects, and this can be used to evaluate the magnitude and frequency of vibrations, while sound pressure and acoustic power sensors measure the intensity and energy of sound and are used to understand the characteristics of the noise. By acquiring data from the above sensors in real time, the inventors can perform comprehensive monitoring and evaluation of the NVH performance of the electric drive system, thereby providing valuable data support for optimizing the performance of the electric drive system. At the same time, this data can be used to analyze and solve vibration and noise problems that may occur while the electric drive system is in operation, thereby improving the reliability and comfort of the electric drive system.

[0027] Specifically, the purpose of durability testing is to simulate wear and degradation of the electric drive system under actual use and predict its performance throughout its lifecycle. By repeatedly performing a series of standardized test cycles in a laboratory environment, the degradation of the electric drive system can be accelerated, and NVH performance data after long-term operation can be rapidly obtained.

[0028] To ensure accuracy and representativeness of the durability tests, an equivalence relationship between the number of cycles and the actual mileage is established. This relationship is based on research into the performance decay rate of electric drive systems under different operating conditions. Specifically, each test cycle includes, but is not limited to, several of the following types:

[0029] Urban Road Cycle: Simulates daily commuting, including frequent starting and parking processes, and low-speed driving situations commonly found in congested traffic.

[0030] High-speed cycle: The vehicle is traveling at a relatively high speed, simulating situations the driver might experience during a long-distance journey.

[0031] Mountain Road Cycling: This simulation includes driving on steep uphill and downhill sections, simulating driving on mountain roads. The vehicle must overcome gravity and provide sufficient power for uphill driving, while simultaneously managing braking force appropriately during downhill sections to prevent overheating of the braking system.

[0032] Each test cycle is equivalent to a certain actual driving distance, thereby ensuring that 1,000 cycles are equivalent to 300,000 kilometers of driving distance. Setting such equivalent driving distances not only saves time and resources but also provides a test environment similar to actual road conditions.

[0033] In step S2012, preprocessing is performed on the NVH performance data to obtain the series values ​​of the historical NVH performance.

[0034] Specifically, the original NVH performance data always contains a wide range of frequency components, such as noise signals from the surrounding environment and other interference sources. To improve noise quality, the data is preprocessed using a bandpass filter to remove irrelevant frequency components and retain important signal frequency bands. In this embodiment, the filter frequency range is set to 10 Hz to 1 kHz, which covers the main vibration and noise signals in the NVH characteristics of an electrically driven system. After filtering, the signal is usually represented in frequency domain form. An inverse Fourier transform (IFT) is used to convert the signal from a frequency domain signal to a time domain signal to facilitate subsequent feature extraction and analysis. The signal is normalized to adjust the signal amplitude to the same range and eliminate the influence of amplitude differences on subsequent analysis.

[0035] Time-domain features, such as the mean, standard deviation, maximum, minimum, and energy, are extracted from the signal. The signal mean reflects the signal's average level, the standard deviation measures the degree of signal intensity variation, the maximum represents the signal's maximum amplitude, the minimum represents the signal's minimum amplitude, and the energy represents the total energy of the signal within a specified time period. The mean, standard deviation, maximum, minimum, and energy of each signal collectively constitute the signal's features. For vibration and noise signals, the Short-Time Fourier Transform (STFT) is used to analyze the signal's local spectral characteristics, such as the principal frequency and frequency bandwidth, across the frequency bands. The Short-Time Fourier Transform is a time-frequency analysis tool that can reflect the frequency components of a signal as it changes over time. The wavelet transform method is used to extract local features in the time and frequency dimensions of the signal. All the extracted features are integrated into a single feature vector. Each element of the feature vector corresponds to a specific feature value. For example, if 10 time-domain features and 10 frequency-domain features are extracted, the length of the feature vector is 10.

[0036] In this embodiment, to balance temporal and frequency resolution, the signal is divided into multiple 1-second intervals that overlap by 0.5 seconds each, and boundary effects are reduced using a Hann window, which helps to smooth the edges of the signal and avoid artifacts caused by abrupt signal cutoffs. For each temporal window, the feature extraction step and the feature vector formation step must be repeated to generate one feature vector. A series of feature vectors are obtained as the temporal window moves. These feature vectors are arranged in temporal order to form a feature vector sequence, which can be considered a feature representation corresponding to the temporal change of the signal, i.e., a sequence value of the historical NVH performance.

[0037] In step S202, the model parameters are determined based on the series values.

[0038] Before step S202, it is necessary to perform a stationarity check on the series values ​​of the historical NVH performance. If it is stationary, the original series composed of the series values ​​of the historical NVH performance is a stationary series. If it is not stationary, a difference process is performed based on the series values ​​in the original series until a stationary series is obtained, and the number of difference processes is the difference order of the prediction model.

[0039] Specifically, stationarity primarily refers to the statistical stability of time series data. Stationarity means that the statistical properties of time series data, such as the mean, variance, and autocovariance function, do not change over the entire observation period and do not change significantly over time. Before performing a series analysis on the series values ​​of historical NVH performance, it is necessary to first determine whether these series values ​​satisfy the stationarity condition. If the series data is non-stationary, its statistical properties change over time, making the nature of the series data unstable, which in turn affects the predictive ability of the model.

[0040] As can be understood, when a series of values ​​is determined to be non-stationary, the series data is transformed using a difference processing method until a stationary time data series is obtained. The difference operation is used to calculate the difference between adjacent terms in the series, and this method removes trends and seasonal influences from the data, making the time series data more stable.

[0041] For example, a linear difference can be calculated using the following formula.

[0042] y t =x t -x t-1 That is the case.

[0043] Here, y t x is the series value after the difference, t x is the current value of the original series, t-1 This is the value of the previous term in the original series.

[0044] As can be understood, if the series still exhibits non-stationary characteristics after a first-order difference, it is usually necessary to perform a higher-order difference, such as a second-order difference, which can be understood as performing another difference on the first-order differenced series. By analogy, this means performing multiple differences on the series until the series exhibits stationarity. The number of times the series values ​​are differed affects the order of the prediction model. In time series analysis, the order of the difference can be the difference order in an autoregressive integrated moving average (ARIMA) model.

[0045] For example, if two differences are needed for a sequence to reach a steady state, the difference order of the prediction model is 2.

[0046] In one embodiment, the number of autoregressive terms p and moving average terms q of the model are determined based on the acquired stationary series using the AIC and BIC constant-order methods. A predictive model is constructed based on the number of autoregressive terms p, the number of moving average terms q, and the difference order d.

[0047] Specifically, the accuracy of prediction results can be further improved by determining the number of autoregressive terms p and the number of moving average terms q using two constant-order methods: AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion). First, it is necessary to obtain a set of already stationary series data. If the original data is not stationary, it is necessary to perform a difference process to make it stationary and record the difference order d. Then, a reasonable range is selected for the number of autoregressive terms p and the number of moving average terms q. For example, the range of possible values ​​for the number of autoregressive terms p and the number of moving average terms q is from 0 to 4, respectively. The ARIMA model is fitted to all possible combinations of (p,d,q), and the AIC and BIC values ​​of each model are calculated. Smaller AIC and BIC values ​​indicate a better model. By comparing the AIC and BIC values ​​of different models, the minimum AIC and BIC values ​​are selected to establish the optimal combination of (p,d,q), i.e., the optimal ARIMA(p,d,q) model.

[0048] The formula for AIC is expressed as follows:

[0049] AIC = 2k - 21n(L).

[0050] The formula for BIC is expressed as follows:

[0051] BIC = kln(n) - 21n(L).

[0052] Here, k represents the number of parameters in the model, n represents the number of samples, and L represents the maximum likelihood estimate of the model.

[0053] When performing time series analysis, it is possible to generate autocorrelation coefficients and partial autocorrelation coefficients corresponding to different values ​​of p and q, based on the range of possible values ​​of the autoregressive term p and the moving average term q. The time series is then analyzed based on the autocorrelation coefficients and partial autocorrelation coefficients. For the autocorrelation coefficients, it is first necessary to calculate the mean value of the time series, using any two observed values ​​Y separated by k time units in the time series data. t and Y t-kFor it, its covariance is calculated, and the covariance calculation formula is as follows.

[0054] JPEG2026085908000002.jpg1066

[0055] Here, Y t represents the observed value at time point t, and Y t-k is k hours before time t per JPEG2026085908000003.jpg12151

[0056] The autocorrelation coefficient is obtained by the covariance formula.

[0057] JPEG2026085908000004.jpg1133

[0058] Here, Var(Y t ) represents the variance of Y<x t , and for each lag period k, a single regression model is constructed, where Y t is the dependent variable, and Y t-1 , Y t-2 , Y t-3 ,..., Y t-(k-1) are used as arguments.

[0059] The partial autocorrelation coefficient can be expressed as follows.

[0060] π k =Corr(Y t ,Y t-k |Y t-1 ,Y t-k+1 )。

[0061] ] Here, Corr(Y t ,Y t-k |Y t-1 ,Y t-k+1 ) represents the magnitude of the correlation between Y t-1 and Y t-k+1 after removing the influence of Y t , Y t-k .

[0062] Specifically, the series values ​​in the sample set are substituted into the autocorrelation coefficient and partial autocorrelation coefficient to obtain a series of p and q values, where p and q represent the order parameters in the autoregressive model (AR) and moving average model (MA), respectively. Each specific p and q value corresponds to the functional values ​​of the autocorrelation coefficient and partial autocorrelation coefficient. After obtaining the autocorrelation coefficient and partial autocorrelation coefficient corresponding to different p and q values, the final p and q values ​​can be determined using an information criterion to construct an autoregressive model (AR), a moving average model (MA), or an ARIMA model.

[0063] In step S203, a predictive model is constructed based on the model parameters, and this predictive model is used to predict the NVH performance decay status of the electric drive system at a predetermined point in time.

[0064] Specifically, a predictive model is constructed based on the number of autoregressive terms p, the number of moving average terms q, and the difference order d. Feature values ​​characterizing NVH performance decay are extracted from the residuals of the predictive model. These feature values ​​include standard deviation, skewness, and kurtosis, and are used to predict the NVH performance decay status of an electrically driven system at a predetermined point in time.

[0065] For example, one time series dataset is pre-defined, and one appropriate ARIMA(p,d,q) model is determined. First, a linear difference is performed on the original time series to obtain a new series. The ARIMA model is then fitted to this new series. For each observation point, predictions are made using the ARIMA model. The difference between the observed and predicted values ​​is calculated as the residual. Feature values ​​characterizing NVH performance attenuation are extracted from the residuals, and these feature values ​​include standard deviation, skewness, and kurtosis.

[0066] The prediction model can be expressed as follows:

[0067] JPEG2026085908000005.jpg1576

[0068] JPEG2026085908000006.jpg12151 represents the coefficient of the Seth error term, B represents the backward operator, and ε t This represents the error term of the data at time t.

[0069] For example, the parameters of the ARIMA model are set to (1,1,2), meaning that the order of the prediction model, the number of autoregressive terms, and the number of moving average terms are d=1, p=1, and q=2, respectively. JPEG2026085908000007.jpg25151

[0070] JPEG2026085908000008.jpg559

[0071] The above prediction model can predict standard deviation, skewness, and kurtosis values ​​corresponding to different time points, and is used to characterize important features of NVH performance decay. For example, the standard deviation (σ_res=0.02g), skewness (S_res=0.05), and kurtosis (K_res=3.0) are extracted at time 0, and the standard deviation, skewness, and kurtosis values ​​for other time points are shown in Table 1. By analyzing these feature values, it is possible to more accurately predict the NVH performance decay status of electric drive systems at different time points.

[0072] In step S204, feature values ​​are extracted based on the prediction model, an evaluation model is established based on the feature values, and the evaluation model is used to evaluate the NVH performance attenuation status.

[0073] An evaluation model is established using the extracted feature values, and an evaluation model for NVH attenuation levels is established using the support vector machine method. This is used to evaluate the attenuation features, and the evaluation model is shown as follows.

[0074] JPEG2026085908000009.jpg1547

[0075] JPEG2026085908000010.jpg19151

[0076] As can be understood, based on the NVH attenuation level evaluation model, the root mean square of vibration acceleration and the sound pressure peak value can be obtained every hour between time 0 and time 1000, where the value of the root mean square of vibration acceleration is used to represent the vibration value, and the sound pressure peak value is used to represent the noise value. The root mean square of vibration acceleration and the sound pressure peak value at the start and end are obtained by direct measurement by the instrument, and the specific data are shown in Table 1.

[0077] [Table 1]

[0078] Observing the changes in the data in Table 1, we can see that the root mean square (RMS) of acceleration shows a gradual decrease over time. This trend clearly indicates that the force and intensity of the vibrations are weakening. Considering these changes, it is presumed that they are due to wear between mechanical components or other types of performance degradation. Furthermore, the decrease in the steady-state nature of the sound pressure peak values ​​may indicate that the sound insulation capacity of the electric drive system is gradually improving, or that the intensity of the noise source is gradually weakening. The residual standard deviation is also decreasing in stages, indicating an improvement in the predictive evaluation accuracy of the model. In addition, the residual distortion changes from slight rightward distortion to leftward distortion, suggesting that the operating conditions of the system may be changing to some extent. The residual kurtosis decreases from 3.00 to 1.80, which indicates that the distribution shape is gradually changing from a sharp peak to a flatter shape, and such a change may be a clear indication that the system performance is gradually becoming more stable and gradual from drastic changes.

[0079] Based on a similar inventive concept, embodiments of the present application further provide an evaluation apparatus for achieving the above-mentioned related NVH performance attenuation of electric drive systems. The means for solving the problems provided by the apparatus are similar to the means for solving the problems described in the above-described method, and therefore specific limitations in the apparatus embodiments of one or more electric drive system NVH performance attenuation evaluation methods provided below can refer to the limitations for the above-described electric drive system NVH performance attenuation evaluation method, and are omitted here.

[0080] In one exemplary embodiment, as shown in Figure 4, an evaluation apparatus for the NVH performance attenuation of an electric drive system is provided. Acquisition module 410 for obtaining sequential values ​​of historical NVH performance for the predicted target, A decision module 420 for determining model parameters based on the aforementioned serial values, A construction module 430 for constructing a predictive model based on the aforementioned model parameters, wherein the predictive model is used to predict the NVH performance attenuation status of the electric drive system at a predetermined time, An evaluation module 440 for extracting feature values ​​based on the prediction model and establishing an evaluation model based on the feature values, wherein the evaluation model includes an evaluation module 440 used to evaluate the NVH performance attenuation status.

[0081] As a selectable embodiment, the acquisition module 410 specifically includes: NVH performance data is acquired based on durability tests of the same electric drive system, with sensors installed within the electric drive system. Here, the time period corresponding to the NVH performance data includes the initial time period and the decay time period, and the number of cycles pre-set for the durability test corresponds to the number of distances traveled under actual road conditions. This is used to preprocess NVH performance data and obtain sequential values ​​of historical NVH performance.

[0082] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and whose internal structure diagram may be as shown in Figure 5. The computer device includes a processor, memory, an input / output interface (abbreviated as I / O), and a communication interface. Here, the processor, memory, and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. Here, the processor of the computer device is used to provide computation and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store a series of historical NVH performance values ​​of a predicted target. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to connect to and communicate with an external terminal via a network. The computer program implements a method for evaluating the NVH performance attenuation of an electrically driven system when executed by the processor.

[0083] As those skilled in the art will understand, the structure shown in Figure 5 is merely a block diagram of the structure of the part relating to the solution of the present invention, and does not limit the computer equipment to which the solution of the present invention applies. Specific computer equipment may include more or fewer components than shown in the figure, or may have a combination of several components or different component arrangements.

[0084] In one exemplary embodiment, a computer device is further provided, comprising memory and a processor, wherein a computer program is stored in the memory, and the processor performs the steps in each embodiment of the above method when executing the computer program.

[0085] In one exemplary embodiment, a computer-readable storage medium is provided on which a computer program is stored, and the computer program, when executed by a processor, realizes the steps in each embodiment of the above method.

[0086] Furthermore, the user information (including, but not limited to, user device information and user personal information) and data (including, but not limited to, data used for analysis, stored data, displayed data, etc.) relating to this application are all information and data permitted by the user or fully permitted by each party, and the collection, use, and processing of related data must comply with the relevant regulations.

[0087] Those skilled in the art will understand that implementing all or part of the flows in the above embodiments can be completed by instructing the relevant hardware with a computer program, which can be stored in a non-volatile computer-readable storage medium, and that when the computer program is executed, it can include the flows of each embodiment of the above embodiments. Any reference to memory, database, or other medium used in each embodiment provided herein can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, flexible disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) and external cache memory, etc. The description is not limited to RAM, which can take various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0088] The databases relating to each embodiment provided in this application may include at least one of relational databases and non-relational databases. Non-relational databases are not limited to these and may include blockchain-based distributed databases, etc. The processors relating to each embodiment provided in this application may be, but are not limited to, general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc.

[0089] Any combination of the technical features of the above embodiments is possible, and for the sake of brevity, not all possible combinations of the technical features in the above embodiments will be described; however, as long as there is no inconsistency in these combinations of technical features, they should be considered to fall within the scope described herein.

[0090] This specification describes the principles and embodiments of the present application by applying specific individual examples. The above description of the embodiments is useful for understanding the methods and core ideas of the present application. At the same time, those skilled in the art can modify the specific embodiments and scope of application based on the ideas of the present application. As described above, the contents of this specification should not be understood as limiting the present application.

Claims

1. A method for evaluating the NVH performance attenuation of an electric drive system, Steps include obtaining the series values ​​of the historical NVH performance for the predicted target, The steps include determining model parameters based on the aforementioned series values, A step of constructing a predictive model based on the aforementioned model parameters, and extracting feature values ​​that characterize NVH performance attenuation from the residuals of the predictive model, wherein the feature values ​​are used to predict the NVH performance attenuation status of the electric drive system at a predetermined time, and the feature values ​​include standard deviation, skewness, and kurtosis. A step of establishing an evaluation model based on the aforementioned feature values, wherein the evaluation model is used to evaluate the NVH performance decay status, The aforementioned evaluation model, It is represented as, The method includes the step of obtaining the root mean square of the vibration acceleration and the sound pressure peak value at one-hour intervals between time 0 and time 1000 based on a valence model, where the value of the root mean square of the vibration acceleration is used to represent the vibration value and the sound pressure peak value is used to represent the noise value, The step of obtaining a series of historical NVH performance values ​​for the predicted target specifically involves, A step of acquiring NVH performance data based on a durability test for the same electric drive system, wherein a sensor is provided in the electric drive system, and the time period corresponding to the NVH performance data includes an initial time period and a decay time period, and the number of cycles set in advance for the durability test corresponds to the number of distances traveled under actual road conditions. The step includes performing preprocessing on the NVH performance data to obtain a series of values ​​for the historical NVH performance, Before determining the model parameters based on the aforementioned serial values, the method for evaluating the NVH performance attenuation of the electric drive system further: The steps include: performing a steady-state determination on the series values ​​of the historical NVH performance; If the system is steady, the original series composed of the series values ​​of the historical NVH performance is a steady-state series; if it is not steady, a difference process is performed based on the series values ​​in the original series until a steady-state series is obtained, and the number of steps in the difference process is the difference order of the prediction model. The step of determining model parameters based on the aforementioned series values ​​specifically includes, A step in which, based on the acquired stationary series, the number of autoregressive terms and moving average terms of the prediction model are determined by the AIC and BIC constant order method, A method for evaluating the NVH performance decay of an electric drive system, characterized in that the model parameters include a step that includes the number of autoregressive terms, the number of moving average terms, and the difference order.

2. An evaluation device for the NVH performance attenuation of an electric drive system, An acquisition module for obtaining sequential values ​​of historical NVH performance for the predicted target, A decision module for determining model parameters based on the aforementioned serial values, A construction module for constructing a prediction model based on the aforementioned model parameters and extracting feature values ​​that characterize NVH performance attenuation from the residuals of the prediction model, wherein the feature values ​​are used to predict the NVH performance attenuation status of an electric drive system at a preset time, and the construction module includes standard deviation, skewness, and kurtosis. An evaluation module for establishing an evaluation model based on the aforementioned feature values, wherein the evaluation model is used to evaluate the NVH performance decay status, The aforementioned evaluation model, It is represented as, An evaluation module is included that can obtain the root mean square of vibration acceleration and the sound pressure peak value at one-hour intervals between time 0 and time 1000 based on a valuation model, where the value of the root mean square of vibration acceleration is used to represent the vibration value and the sound pressure peak value is used to represent the noise value. The acquisition module described above is, specifically, NVH performance data is acquired based on durability tests of the same electric drive system, and a sensor is installed within the electric drive system, where the time period corresponding to the NVH performance data includes the initial time period and the decay time period, and the number of cycles predetermined for the durability test corresponds to the number of distances traveled under actual road conditions. This is used to perform preprocessing on the aforementioned NVH performance data and to obtain the series values ​​of the historical NVH performance. Before determining the model parameters based on the aforementioned serial values, the evaluation device for the electric drive system NVH performance attenuation further: Used to perform a stationary determination on the series of values ​​of the aforementioned historical NVH performance, If the system is steady, the original series composed of the series values ​​of the historical NVH performance is a steady-state series; if it is not steady, a difference process is performed based on the series values ​​in the original series until a steady-state series is obtained, where the number of times the difference process is performed is the difference order of the prediction model. The aforementioned decision module specifically refers to: Based on the acquired stationary series, the number of autoregressive terms and moving average terms of the prediction model are determined by the AIC and BIC constant order method. An evaluation device for the NVH performance attenuation of an electric drive system, characterized in that the model parameters include the number of autoregressive terms, the number of moving average terms, and the difference order.

3. A computer device comprising memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program to realize the steps of the method for evaluating the NVH performance attenuation of an electric drive system described in claim 1.

4. A computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, it realizes the steps of the method for evaluating the NVH performance attenuation of an electric drive system described in claim 1.

Images