Electric vehicle remote fault analysis method and system based on Internet of Things

Abstract

The present invention relates to the field of data processing technology, and in particular to a method and system for remote fault analysis of electric vehicles based on the Internet of Things. The method comprises the following steps: obtaining a window segment and a current data segment of each data point in a time series of motor speed data; obtaining the degree of numerical difference in the data point window segment by the difference between adjacent data in the data point window segment; obtaining the difference between the median and mean of the data point window segment and the corresponding current data segment, respectively, to obtain the degree of correlation between the speed and current in the window segment; calculating the degree of fluctuation of the window segment; calculating the length of a subwindow after iteration in the data point window segment; and using the length of the subwindow after iteration in the data point window segment in a fuzzy entropy algorithm to calculate the fuzzy entropy value of the data point window segment to implement remote fault analysis of the electric vehicle, thereby effectively improving the accuracy of the remote fault analysis results of the electric vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for remote fault analysis of electric vehicles based on the Internet of Things. Background Art

[0002] Electric vehicles are powered by high-energy-density batteries. The stored energy in these batteries is then transferred to the electric motor, enabling environmentally friendly and energy-efficient operation. The electric motor is the core power unit of an electric vehicle, and a failure or overload of the motor can lead to a loss of power or even serious safety hazards such as fire.

[0003] To improve the safety of electric vehicles during use, existing technologies offer a variety of fault identification methods. For example, patent application publication number CN118194241A discloses a method and apparatus for diagnosing electric vehicle motor faults. This application performs fault diagnosis by inputting real-time motor operating data from the electric vehicle to a multimodal data fusion motor fault diagnosis model. The multimodal data fusion motor fault diagnosis model is constructed based on historical motor operating data.

[0004] However, when an electric motor fails, a quick response and processing are required. However, neural network models are easily restricted by known failure modes during training and have low sensitivity to unknown or emerging failure modes. Especially when electric vehicle motor failure data is scarce, it is difficult to accurately obtain motor fault identification results, posing a major safety hazard.

[0005] Based on this, how to accurately implement remote fault analysis of electric vehicles based on the Internet of Things is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0006] In order to solve the technical problem of how to accurately implement remote fault analysis of electric vehicles based on the Internet of Things, the present invention provides a remote fault analysis method and system for electric vehicles based on the Internet of Things.

[0007] In a first aspect, the present invention provides an electric vehicle remote fault analysis method based on the Internet of Things, which adopts the following technical solutions:

[0008] The electric vehicle remote fault analysis method based on the Internet of Things includes the following steps:

[0009] The invention relates to a method for obtaining a window segment of each data point in a time series of motor speed data and a current data segment corresponding to the window segment; obtaining the degree of numerical difference in the window segment of the data point by multiplying the cumulative sum of the speed differences between the adjacent data in the window segment of the data point by the cumulative sum of the absolute values of the speed differences; obtaining the difference between the median and the mean of the window segment of the data point and the corresponding current data segment as a first difference and a second difference respectively, and obtaining the degree of correlation between the speed and the current in the window segment by the degree of proximity between the first difference and the second difference; calculating the degree of fluctuation of the window segment, which is positively correlated with the degree of numerical difference in the window segment and negatively correlated with the degree of correlation; calculating the length of the subwindow after iteration in the data point window segment according to the degree of fluctuation of the data point window segment, a preset subwindow length and an iteration step; using the length of the subwindow after iteration in the data point window segment in a fuzzy entropy algorithm, obtaining the fuzzy entropy value of the data point window segment, and realizing remote fault analysis of electric vehicles based on the comparison result of the fuzzy entropy value and a preset threshold.

[0010] The present invention uses a fuzzy entropy algorithm to calculate the fuzzy entropy value of each data point in the speed data, which can identify small changes in speed data and thus accurately obtain the results of remote fault analysis of electric vehicles. In this process, the present invention considers that factors such as sensor wear can affect the fluctuation trend of speed data caused by motor faults, which may affect the accuracy of fault identification. Based on this, the present invention accurately identifies the fluctuating data caused by motor faults by analyzing the data differences in the speed data itself and the current data that is correlated with the speed data, effectively improving the accuracy of remote fault analysis results of electric vehicles.

[0011] According to the Internet of Things-based electric vehicle remote fault analysis method provided by the present invention, the window segments of each data point in the motor speed data time series sequence and the current data segments corresponding to the window segments are obtained, which also includes: obtaining the motor speed data and current data at the same acquisition time, and obtaining the speed data time series sequence and the current data time series sequence after preprocessing; determining the window segments of each data point in the speed data time series sequence based on the length of a preset window segment; and obtaining the current data segment in the current data time series sequence that is in the same acquisition time period as the window segment.

[0012] The present invention takes into account that the originally collected speed data and current data may have data missing, and therefore improves the overall quality of the data through preprocessing to facilitate subsequent data processing.

[0013] According to the electric vehicle remote fault analysis method based on the Internet of Things provided by the present invention, the method for obtaining the degree of correlation between the speed and the current in the window segment includes:

[0014] ;

[0015] is the correlation between the speed and current in the window segment of the jth data point, 、 are the median and mean of the j-th data point window segment, 、 are the median and mean of the current data segment corresponding to the j-th data point window segment, is an exponential function with base e, is a linear normalization function.

[0016] The present invention provides a method for calculating the degree of correlation between the rotational speed and the current in a window segment of an accurate data point. By respectively obtaining the difference between the median and the mean of the window segment and the corresponding current data segment, the degree of negative correlation between the rotational speed data and the current data is obtained, thereby accurately obtaining the degree of correlation between the rotational speed and the current in the window segment of the data point.

[0017] According to the electric vehicle remote fault analysis method based on the Internet of Things provided by the present invention, the fluctuation degree of the window segment satisfies the relationship:

[0018] ;

[0019] is the degree of fluctuation of the j-th data point window segment, 、 are the numerical difference degree of the j-th data point window segment, the correlation degree between the speed and current, is a hyperparameter.

[0020] According to the electric vehicle remote fault analysis method based on the Internet of Things provided by the present invention, the fuzzy entropy algorithm uses the length of the subwindow after iteration in the data point window segment to obtain the fuzzy entropy value of the data point window segment, including: using the preset subwindow length and the subwindow after iteration of the data point window segment in the fuzzy entropy algorithm respectively to obtain the fuzzy membership of the window segment before and after the subwindow length iteration; normalizing the difference between the fuzzy membership of the window segment before and after the subwindow length iteration in the data point window segment to obtain the fuzzy entropy value of the data point window segment;

[0021] in, ; is the length of the sub-window after iteration in the j-th data point window segment, To preset the sub-window length, is the iteration step length, is the degree of fluctuation of the j-th data point window segment, The symbol for rounding up.

[0022] The present invention calculates the fuzzy entropy value of each data point window segment through a fuzzy entropy algorithm, which can accurately capture small changes in the speed data, so that abnormal data can be identified in a timely and accurate manner when an abnormality occurs in the motor.

[0023] According to the Internet of Things-based electric vehicle remote fault analysis method provided by the present invention, the electric vehicle remote fault analysis is implemented based on the comparison result of the fuzzy entropy value and the preset threshold, including: if the fuzzy entropy value of the data point window segment is greater than the preset threshold, then the electric vehicle's motor has a fault; otherwise, the electric vehicle's motor has not a fault.

[0024] According to the Internet of Things-based electric vehicle remote fault analysis method provided by the present invention, the electric vehicle remote fault analysis is realized based on the comparison result of the fuzzy entropy value and the preset threshold value, and then also includes: in response to a fault in the electric vehicle's motor, reporting the fault occurrence time of the electric vehicle and issuing an abnormal prompt.

[0025] The present invention takes into account that wear and tear may occur when an abnormality occurs in the motor, affecting the user experience and safety of the electric vehicle user. Therefore, the abnormal fault is reported to facilitate timely processing by the staff and reduce safety hazards.

[0026] In a second aspect, the present invention provides an electric vehicle remote fault analysis system based on the Internet of Things, which adopts the following technical solutions:

[0027] The electric vehicle remote fault analysis system based on the Internet of Things includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the electric vehicle remote fault analysis method based on the Internet of Things is implemented.

[0028] By adopting the above technical solution, the above-mentioned electric vehicle remote fault analysis method based on the Internet of Things is generated into a computer program and stored in a memory to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0029] The present invention has the following technical effects:

[0030] Based on the above technical solution, the present invention, when implementing remote fault analysis for electric vehicles, calculates the fuzzy entropy value of each data point in the speed data using a fuzzy entropy algorithm. This allows for the identification of slightly varying speed data, thereby accurately obtaining remote fault analysis results for the electric vehicle. In this process, the present invention considers that factors such as sensor wear can affect the apparent downward trend in speed data volatility caused by motor faults, thereby affecting the accuracy of fault identification. Based on this, the present invention accurately identifies fluctuating data caused by motor faults by analyzing the data differences in the speed data itself and the current data that is correlated with the speed data, effectively improving the accuracy of remote fault analysis results for electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flowchart of a remote fault analysis method for electric vehicles based on the Internet of Things is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0033] It should be understood that when the terms "first," "second," and the like are used in the claims, description, and drawings of the present invention, they are merely used to distinguish between different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the description and claims of the present invention indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0034] Electric vehicles are powered by high-energy-density batteries. The stored energy in these batteries is then transferred to the electric motor, enabling environmentally friendly and energy-efficient operation. The electric motor, the core power unit of an electric vehicle, maintains a stable speed under normal conditions. However, a malfunction or overload can cause a loss of power, resulting in a fluctuating downward speed.

[0035] The fuzzy entropy method is a method for calculating information entropy based on fuzzy set theory. It can accurately capture subtle changes in data while retaining the advantages of anti-noise and anti-interference.

[0036] Based on this, an embodiment of the present invention discloses a remote fault analysis method for electric vehicles based on the Internet of Things. This method identifies abnormal data in the motor speed data through a fuzzy entropy method, and can accurately capture slight changes in the motor speed data, thereby more effectively identifying anomalies and reducing false alarms and missed alarms.

[0037] For details, please refer to Figure 1 As shown, Figure 1 A flow chart of a method for remote fault analysis of an electric vehicle based on the Internet of Things is provided in an embodiment of the present invention. The method specifically includes the following steps.

[0038] S1: Get the window segment of each data point in the motor speed data time series.

[0039] For example, in an embodiment of the present invention, obtaining the window segment of each data point in the motor speed data time series sequence further includes: preprocessing the motor speed data obtained at each acquisition time to obtain the speed data time series sequence.

[0040] Among them, preprocessing can be missing data interpolation, data format conversion, standardization processing, etc., which can be set according to actual needs.

[0041] Specifically, the speed data of the electric vehicle motor is collected through a speed sensor. The collection time can be 2 hours and the collection frequency can be 10 times per second. The speed data collected each time is used as a data point to form a speed data time series sequence.

[0042] The acquisition duration and acquisition frequency can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0043] It should be noted that when the fuzzy entropy algorithm identifies abnormal data in the speed data, it usually calculates the fuzzy membership value between each subwindow and other subwindows in the window segment of the data point, and takes the average fuzzy membership value between all subwindows and other subwindows in the window segment as the fuzzy membership value before the iteration of the subwindow length of the current window segment; adds 1 to the length of the subwindow, and then calculates the average fuzzy membership value between all subwindows and other subwindows as the fuzzy membership value after the iteration of the subwindow length of the current window segment; the fuzzy entropy value of the window segment of the data point is obtained by taking the difference between the fuzzy membership values before and after the iteration of the subwindow length in the window segment of the data point.

[0044] However, in the process of collecting motor speed data through the speed sensor, the sensor may be distorted due to long-term use, wear and tear, etc., resulting in the collected speed data being unable to accurately reflect the downward trend of fluctuation when the motor fails. If the iteration step size of the sub-window uses a fixed value of 1, it will cause false alarms and missed alarms of electric vehicle faults, reducing the accuracy of the electric vehicle remote fault analysis results.

[0045] Based on this, the embodiment of the present invention analyzes the data fluctuations in the data point window segment and adjusts the length of the sub-window based on the data fluctuations, so that when there are abnormal fluctuations in the data, the potential overload fault of the motor can be captured more sensitively. When the sensor is distorted and the numerical reflection of the actual downward trend of the volatility of the speed data is weak, a larger fuzzy entropy value can be calculated by increasing the length of the sub-window after iteration, thereby effectively reducing the impact of the speed data sampling error, and the accuracy of the fuzzy entropy value in the speed data window segment is higher.

[0046] For example, in an embodiment of the present invention, the window segment of each data point in the speed data time series is determined based on the length of the preset window segment, specifically including the following two possible implementation methods:

[0047] Among them, the length of the window segment It can be preset to 300. The length of the window segment can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0048] In one possible implementation, the historical data to the left of the current data point can be obtained. data points as the window segment of the current data point.

[0049] In another possible implementation, the two sides of the current data point can be obtained separately. data points as the window segment of the current data point.

[0050] It can be understood that the length of the window segment obtained based on any of the above implementations is the number of data points included in the window segment, and the window segment of the current data point does not include the current data point itself.

[0051] After obtaining the window segments of each data point based on the above steps, continue to perform the following steps.

[0052] S2: The degree of numerical difference in the data point window segment is obtained by multiplying the cumulative sum of the speed differences between the adjacent data in the data point window segment by the cumulative sum of the absolute values of the speed differences.

[0053] It should be noted that when the motor is overloaded, the motor speed data may show a fluctuating downward trend, and the speed difference between adjacent data points can reflect the changing trend of the window segment. If the speed difference between adjacent data points in the window segment is positive, it means that the speed value of the previous data point is greater than the speed value of the next data point, and the speed data value has dropped significantly.

[0054] Based on this, the embodiment of the present invention obtains the data fluctuation in the window segment by analyzing the speed data difference between adjacent data points in the data point window segment, and then obtains the speed difference between adjacent data to characterize the data fluctuation trend in the window segment, thereby accurately obtaining the degree of data difference in the window segment.

[0055] For example, in an embodiment of the present invention, the degree of difference in the values of the data point window segment is determined, and the details can be seen in the following relationship:

[0056] ;

[0057] is the numerical difference degree of the j-th data point window segment, is the cumulative sum of the speed differences between the adjacent data before and after the j-th data point window segment, It is the cumulative sum of the absolute values of the speed differences between the adjacent data before and after the j-th data point window segment.

[0058] In the above formula, Indicates the degree of data change within the window segment. The larger the value, the higher the degree of data difference within the window segment, the more drastic the change, the higher the possibility of a clear upward or downward trend in the window segment, and the greater the degree of numerical difference in the window segment.

[0059] It indicates that the overall change trend of the data points in the window segment is a downward trend. The larger the value, the more obvious the downward trend in the window segment, and the greater the corresponding numerical difference.

[0060] After obtaining the numerical difference degree of each data point window segment based on the above steps, continue to perform the following steps.

[0061] S3: Obtain the current data segment corresponding to each data point window segment in the motor speed data time series, and obtain the degree of correlation between the speed and current in the window segment by the degree of proximity between the median and mean difference between the window segment of the data point and the corresponding current data segment.

[0062] It should be noted that by analyzing the fluctuations and trends of data within a window segment using the above steps, the degree of numerical variation within the window segment can be determined. However, if the engine is overloaded, the downward trend in fluctuation within the window segment obtained by analyzing the speed data itself using the above steps will not be obvious, resulting in the calculated fuzzy entropy value being considered normal, leading to missed or incorrect motor anomaly reports.

[0063] It's important to note that, under normal circumstances, the current in an electric vehicle's motor is primarily affected by the load, while the motor's speed is inversely proportional to the load. That is, the greater the load, the lower the speed. To maintain a stable speed, the motor increases the current to provide more torque. Therefore, when the motor load is high, its current increases and its speed decreases. Conversely, when the load is low, the motor current decreases and the speed increases. Therefore, the motor current and speed have a good negative correlation.

[0064] Based on this, the embodiment of the present invention corrects the numerical difference degree of the speed data point window segment through the degree of connection between the change characteristics of the motor current data and the speed data, so as to accurately obtain the fluctuation degree of the speed data point window segment.

[0065] For example, in an embodiment of the present invention, the current data segments corresponding to each data point window segment in the motor speed data timing sequence are obtained, which also includes: obtaining the current data of the motor at the same acquisition time of the speed data, and obtaining the current data timing sequence after preprocessing; obtaining the current data segment in the current data timing sequence that is in the same acquisition time period as the data point window segment.

[0066] Specifically, the current data of the motor can be collected through a current sensor.

[0067] In order to facilitate data processing, the acquisition time, acquisition frequency and pre-processing method of the current data may refer to the acquisition time, acquisition frequency and pre-processing method of the speed data.

[0068] After obtaining the current data segment corresponding to the data point window segment based on the above method, the following steps can be continued to obtain the degree of correlation between the data point window segment and the corresponding current data segment. The higher the degree of correlation, the higher the possibility that the speed data is true data.

[0069] For example, in an embodiment of the present invention, when calculating the degree of correlation between the speed and the current in a window segment, the difference between the median and the mean of the window segment of the data point and the corresponding current data segment can be obtained respectively as the first difference and the second difference, and the degree of correlation between the speed and the current in the window segment can be obtained by the degree of proximity between the first difference and the second difference.

[0070] For example, in an embodiment of the present invention, a method for obtaining the degree of correlation between the rotational speed and the current in the window segment includes:

[0071] ;

[0072] is the correlation between the speed and current in the window segment of the jth data point, 、 are the median and mean of the j-th data point window segment, 、 are the median and mean of the current data segment corresponding to the j-th data point window segment, is an exponential function with base e, is a linear normalization function.

[0073] In the above formula, represents the first difference of the j-th data point, The second difference value of the j-th data point is represented. The larger the difference between the median and the mean, the greater the existence of extreme values in the speed data and the current data.

[0074] Indicates the degree of negative correlation between speed and current in the window segment of the jth data point. The smaller the value, the and The closer they are, the stronger the negative correlation between the speed and current is, the higher the correlation between the corresponding speed and current is, and the higher the possibility that the speed data can reflect the true state of the motor.

[0075] After obtaining the correlation degree between the rotational speed and the current in the window segment of the data point based on the above steps, the fluctuation degree of the data point window segment can be obtained through the correlation degree between the rotational speed and the current in the window segment of the data point and the degree of numerical difference in the window segment.

[0076] S4: Calculate the degree of fluctuation of the window segment. The degree of fluctuation is positively correlated with the degree of difference in the values in the window segment and negatively correlated with the degree of association. Use the length of the iterated subwindow in the data point window segment in the fuzzy entropy algorithm to obtain the fuzzy entropy value of the data point window segment.

[0077] For example, in an embodiment of the present invention, the degree of fluctuation of a data point window segment is calculated, and the details can be found in the following relationship:

[0078] ;

[0079] is the degree of fluctuation of the j-th data point window segment, is the numerical difference degree of the j-th data point window segment, is the correlation between the speed and current in the jth data point window segment, is a hyperparameter, is a linear normalization function.

[0080] The hyperparameter can be set to 10. The purpose of the hyperparameter is to prevent the length of the final sub-window after iteration from being too large or too small. The hyperparameter can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0081] It should be noted that the greater the degree of difference in the numerical values in the window segment, the higher the degree of fluctuation of the data point window segment, and the smaller the correlation between the speed and current in the window segment, which means that there is a greater possibility of distorted data caused by sensor failure in the speed data of the window segment. The actual downward trend of the speed data in the window segment corresponding to the acquisition time period should be greater, the possibility of abnormal speed data will be greater, and the corresponding degree of fluctuation will also be greater.

[0082] After obtaining the true degree of fluctuation in the data point window segment based on the above steps, the iterative step size of the sub-window in the window segment when calculating the fuzzy entropy value can be adjusted based on the degree of fluctuation in the data point window segment, thereby accurately obtaining the fuzzy entropy value of the window segment.

[0083] Specifically, the length of the sub-window in the data point window segment after iteration can be calculated according to the fluctuation degree of the data point window segment, the preset sub-window length and the iteration step.

[0084] For example, in an embodiment of the present invention, the length of the sub-window after iteration in the data point window segment is calculated as follows:

[0085] ;

[0086] is the length of the sub-window after iteration in the j-th data point window segment, To preset the sub-window length, is the iteration step length, is the degree of fluctuation of the j-th data point window segment, The symbol for rounding up.

[0087] Among them, the iteration step size is the fixed value initially set in the fuzzy entropy algorithm, that is, The preset sub-window length can be set to 10, and can be set according to actual needs.

[0088] For example, the number of sub-windows is the window segment length minus the sub-window length plus 1.

[0089] In the above formula, the greater the fluctuation within the data point window segment, the greater the likelihood of an abnormal downward trend in the speed data fluctuation within the window segment, and the greater the possibility of a potential overload failure. To accurately identify abnormal data in the speed data, the length of the subwindow after iteration should also be larger to ensure that the calculated fuzzy entropy value of the data point window segment is larger.

[0090] For example, in an embodiment of the present invention, the length of the sub-window after iteration in the data point window segment is used in the fuzzy entropy algorithm to obtain the fuzzy entropy value of the data point window segment, including: using the preset sub-window length and the length of the sub-window after iteration of the data point window segment in the fuzzy entropy algorithm respectively to obtain the fuzzy membership of the window segment before and after the sub-window length iteration; normalizing the difference between the fuzzy membership of the window segment before and after the sub-window length iteration in the data point window segment to obtain the fuzzy entropy value of the data point window segment.

[0091] Specifically, when calculating the fuzzy membership of a data point window segment, a similarity tolerance can be preset; the similarity tolerance and the preset sub-window length are substituted into the fuzzy entropy calculation formula to obtain the fuzzy membership of the window segment before the sub-window length iteration; the similarity tolerance and the length of the sub-window in the window segment after iteration are substituted into the fuzzy entropy calculation formula to obtain the fuzzy membership of the window segment after the sub-window length iteration.

[0092] For example, the similarity tolerance may be preset to 0.25. The similarity tolerance may be specifically set according to actual needs, and the embodiment of the present invention does not impose any excessive restrictions thereon.

[0093] Among them, the specific steps of using the length of the sub-window in the data point window segment in the fuzzy entropy algorithm to obtain the fuzzy entropy value of the data point window segment can be implemented by the existing formula in the fuzzy entropy algorithm, and the embodiments of the present invention will not be repeated here.

[0094] After obtaining the fuzzy entropy value of each data point window segment in the speed data time series based on the above steps, abnormal data in the speed data can be accurately identified.

[0095] S5: Realize remote fault analysis of electric vehicles based on the comparison results of fuzzy entropy value and preset threshold.

[0096] The preset threshold value may be set to 0.9. The preset threshold value may be set according to actual needs, and the embodiment of the present invention does not impose any excessive restrictions on this.

[0097] For example, in an embodiment of the present invention, remote fault analysis of an electric vehicle is implemented based on the comparison result of the fuzzy entropy value and the preset threshold, including: if the fuzzy entropy value of the data point window segment is greater than the preset threshold, the electric motor of the electric vehicle has a fault; otherwise, the electric motor of the electric vehicle has not a fault.

[0098] It is understandable that the fuzzy entropy value of the data point window segment is greater than the preset threshold, indicating that a potential overload fault may have occurred in the motor during the time period corresponding to the window segment, and the staff needs to be reminded in time to handle it.

[0099] For example, in an embodiment of the present invention, remote fault analysis of an electric vehicle is implemented based on the comparison result of the fuzzy entropy value and the preset threshold value, and then further includes: in response to a fault in the electric vehicle's motor, reporting the fault occurrence time of the electric vehicle and issuing an abnormal prompt.

[0100] In this way, the embodiments of the present invention can ensure that the vehicle resumes normal operation in a short time by quickly identifying and responding to motor faults, reducing excessive wear and damage to the motor, thereby extending its service life and reducing safety risks.

[0101] It can be seen that in an embodiment of the present invention, when analyzing the remote fault of an electric vehicle, a window segment of each data point in the motor speed data time series and a current data segment corresponding to the window segment can be obtained; the degree of numerical difference in the data point window segment is obtained by multiplying the cumulative sum of the speed differences between the adjacent data in the data point window segment by the cumulative sum of the absolute values of the speed differences; the difference between the median and the mean of the window segment of the data point and the corresponding current data segment is respectively obtained as a first difference and a second difference, and the degree of correlation between the speed and the current in the window segment is obtained by the degree of proximity between the first difference and the second difference; the degree of fluctuation of the window segment is calculated, and the degree of fluctuation is positively correlated with the degree of numerical difference in the window segment and negatively correlated with the degree of correlation; the length of the sub-window after iteration in the data point window segment is calculated according to the degree of fluctuation of the data point window segment, the preset sub-window length and the iteration step; the length of the sub-window after iteration in the data point window segment is used in the fuzzy entropy algorithm to obtain the fuzzy entropy value of the data point window segment, and the remote fault analysis of the electric vehicle is implemented based on the comparison result of the fuzzy entropy value and the preset threshold, thereby effectively improving the accuracy of the remote fault analysis result of the electric vehicle.

[0102] An embodiment of the present invention also discloses an electric vehicle remote fault analysis system based on the Internet of Things, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, the electric vehicle remote fault analysis method based on the Internet of Things provided by the present invention is implemented.

[0103] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0104] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. The electric vehicle remote fault analysis method based on the Internet of Things is characterized by: include: Obtaining the window segment of each data point in the motor speed data time series and the current data segment corresponding to the window segment; The degree of numerical difference in the data point window segment is obtained by multiplying the cumulative sum of the speed differences between the adjacent data in the data point window segment by the cumulative sum of the absolute values of the speed differences; The difference between the median and the mean of the window segment of the data point and the corresponding current data segment is recorded as a first difference and a second difference, and the degree of correlation between the speed and the current in the window segment is obtained according to the proximity between the first difference and the second difference; Calculate the degree of fluctuation of the window segment. The degree of fluctuation is positively correlated with the degree of difference of the values in the window segment and negatively correlated with the degree of correlation. Calculate the length of the sub-window after iteration in the data point window segment based on the degree of fluctuation of the data point window segment, the preset sub-window length, and the iteration step size. The fuzzy entropy algorithm uses the iterative length of the sub-window in the data point window segment to obtain the fuzzy entropy value of the data point window segment, and realizes remote fault analysis of electric vehicles based on the comparison result of the fuzzy entropy value and the preset threshold. The method for obtaining the degree of correlation between the rotational speed and the current in the window segment includes: ； is the correlation between the speed and current in the window segment of the jth data point, 、 are the median and mean of the j-th data point window segment, 、 are the median and mean of the current data segment corresponding to the j-th data point window segment, is an exponential function with base e, is a linear normalization function.

2. The electric vehicle remote fault analysis method based on the Internet of Things according to claim 1 is characterized in that: The method of obtaining the window segments of each data point in the motor speed data time series and the current data segments corresponding to the window segments also includes: The speed data and current data of the motor are acquired at the same acquisition time, and a speed data time series sequence and a current data time series sequence are obtained after preprocessing; the window segment of each data point in the speed data time series sequence is determined based on the length of a preset window segment; and the current data segment that is in the same acquisition time period as the window segment is acquired in the current data time series sequence.

3. The electric vehicle remote fault analysis method based on the Internet of Things according to claim 1 is characterized in that: The degree of fluctuation of the window segment satisfies the relationship: ； is the degree of fluctuation of the j-th data point window segment, 、 are the numerical difference degree of the j-th data point window segment, the correlation degree between the speed and current, is a hyperparameter.

4. The electric vehicle remote fault analysis method based on the Internet of Things according to claim 1 is characterized in that: The method of using the iterated length of the sub-window in the data point window segment in the fuzzy entropy algorithm to obtain the fuzzy entropy value of the data point window segment includes: In the fuzzy entropy algorithm, the preset sub-window length and the length after sub-window iteration of the data point window segment are used respectively to obtain the fuzzy membership of the window segment before and after the sub-window length iteration; Normalize the difference between the fuzzy membership of the window segment before and after the sub-window length iteration in the data point window segment to obtain the fuzzy entropy value of the data point window segment; in, ; is the length of the sub-window after iteration in the j-th data point window segment, To preset the sub-window length, is the iteration step length, is the degree of fluctuation of the j-th data point window segment, The symbol for rounding up.

5. The electric vehicle remote fault analysis method based on the Internet of Things according to claim 1 is characterized in that: The method for realizing remote fault analysis of electric vehicles based on the comparison result of the fuzzy entropy value and the preset threshold value includes: If the fuzzy entropy value of the data point window segment is greater than a preset threshold, the electric motor of the electric vehicle is faulty; otherwise, the electric motor of the electric vehicle is not faulty.

6. The electric vehicle remote fault analysis method based on the Internet of Things according to claim 5 is characterized in that: The method realizes remote fault analysis of electric vehicles based on the comparison result of the fuzzy entropy value and the preset threshold, and further includes: In response to a fault in the electric motor of the electric vehicle, the fault occurrence time of the electric vehicle is reported and an abnormal prompt is issued.

7. The electric vehicle remote fault analysis system based on the Internet of Things is characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the electric vehicle remote fault analysis method based on the Internet of Things according to any one of claims 1 to 6 is implemented.

Images