Real-time performance monitoring method and system for automobile electronic control system based on cloud computing

Abstract

The present invention relates to the field of data processing technology, and specifically to a real-time performance monitoring method and system for an automobile electronic control system based on cloud computing, including: screening out suspected abnormal moments from all moments through the difference between the neighborhood data at each moment in the speed difference sequence, and obtaining a number of suspected abnormal time periods, and obtaining the length factor of each suspected abnormal moment according to the length of each suspected abnormal time period; obtaining the fluctuation characteristics of each suspected abnormal time period through the fluctuation amplitude and periodicity of the data in the speed sequence of each suspected abnormal time period; and obtaining the speed credibility of each suspected abnormal moment, and correcting the speed data sequence through the credibility to obtain a new speed data sequence; and performing real-time performance monitoring of the automobile electronic control system through the new speed data sequence. The present invention improves the accuracy of speed data acquisition and improves the accuracy of real-time performance monitoring of the automobile electronic control system.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a real-time performance monitoring method and system for an automobile electronic control system based on cloud computing. Background Art

[0002] The automotive electronic control system is composed of highly integrated electronic hardware and software to achieve precise control of various functions of the car. In the automotive electronic control system, the engine is an important part of the car. By monitoring the abnormal engine speed, potential faults of the engine or related systems can be detected to improve the safety of personnel. Therefore, it is of great significance to monitor the abnormal engine speed.

[0003] When performing real-time performance monitoring of a car's engine speed, the engine speed data is collected through a crankshaft position sensor, and the automobile engine is analyzed to see if there is any abnormality based on the collected speed data, so as to complete real-time performance monitoring of the automobile's electronic control system. However, when collecting the engine speed through the crankshaft position sensor, the accuracy of the sensor is reduced due to the long-term use of the crankshaft position sensor, resulting in inaccurate collected speed data, which reduces the accuracy of real-time performance monitoring of the automobile's electronic control system. Summary of the invention

[0004] The present invention provides a real-time performance monitoring method and system for an automobile electronic control system based on cloud computing to solve the existing problems.

[0005] The cloud computing-based real-time performance monitoring method and system for automobile electronic control systems of the present invention adopt the following technical solutions:

[0006] An embodiment of the present invention provides a method for real-time performance monitoring of an automobile electronic control system based on cloud computing, the method comprising the following steps:

[0007] Obtaining the speed data sequence of the electronically controlled automobile engine;

[0008] A speed difference sequence is obtained according to the speed data sequence, and a number of suspected abnormal time periods within the time period corresponding to the speed data sequence are obtained through the differences between the local data in the speed difference sequence, wherein the moments within the suspected abnormal time period are suspected abnormal moments; the inverse of the length of each suspected abnormal time period is used as the length factor of each suspected abnormal moment within each suspected abnormal time period;

[0009] Obtain the speed sequence of each suspected abnormal time period, and obtain the fluctuation characteristics of each suspected abnormal time period through the fluctuation amplitude and periodicity of the data in the speed sequence of each suspected abnormal time period; obtain the speed credibility of each suspected abnormal moment through the length degree factor of each suspected abnormal moment in each suspected abnormal time period and the fluctuation characteristics, and correct the data in the speed data sequence through the credibility, and record the corrected sequence as a new speed data sequence;

[0010] Real-time performance monitoring of the automotive electronic control system is carried out through the new speed data sequence.

[0011] Furthermore, the step of obtaining the rotation speed difference sequence according to the rotation speed data sequence includes the following specific steps:

[0012] The data of each moment in the speed data sequence of the electronically controlled automobile engine is subtracted from the data of the adjacent previous moment to obtain a set of sequences, which are recorded as speed difference sequences.

[0013] Furthermore, the method of obtaining a plurality of suspected abnormal time periods within a time period corresponding to the speed data sequence by using the differences between the local data in the speed difference sequence includes the following specific steps:

[0014] According to the difference between the data of all moments in the local range of each moment in the speed difference sequence, the fluctuation degree of each moment is obtained;

[0015] Recording the moment when the fluctuation degree is greater than or equal to the preset threshold as a suspected abnormal moment;

[0016] Through the distribution of all suspected abnormal moments, several suspected abnormal time periods within the time period corresponding to the speed data sequence are obtained.

[0017] Furthermore, the step of obtaining the fluctuation degree at each moment according to the difference between the data at all moments within the local range at each moment in the rotation speed difference sequence includes the following specific steps:

[0018] Each moment in the speed difference sequence is taken as the center point of the local window, a local window is constructed according to a preset window size, and data within the local window at each moment is obtained;

[0019] The difference between any data in the local window at each moment in the speed difference sequence and the mean of all data is recorded as the first difference of any data in the local window at each moment, and the mean of the first differences of all data in the local window at each moment is recorded as the first fluctuation degree at each moment. The first fluctuation degree is linearly normalized to obtain the fluctuation degree at each moment.

[0020] Furthermore, the method of obtaining several suspected abnormal time periods within the time period corresponding to the speed data sequence through the distribution of all suspected abnormal moments includes the following specific steps:

[0021] The time period corresponding to the consecutive suspected abnormal moments is recorded as the suspected abnormal time period.

[0022] Furthermore, the fluctuation characteristics of each suspected abnormal time period are obtained by the fluctuation amplitude and periodicity of the data in the rotation speed sequence of each suspected abnormal time period, and the specific steps include the following:

[0023] The speed sequence of each suspected abnormal time period is curve-fitted by the least square method to obtain the speed curve of each suspected abnormal time period;

[0024] Obtaining extreme value points in the speed curve of each suspected abnormal time period, wherein the extreme value points include maximum value points and minimum value points;

[0025] The average of the absolute values ​​of the slopes between all adjacent extreme value points is recorded as the fluctuation amplitude characteristic of each suspected abnormal time period; the time intervals between all adjacent extreme value points are obtained, and the fluctuation period characteristics of each suspected abnormal time period are obtained by taking the average of the differences between all adjacent time intervals;

[0026] The fluctuation characteristics of each suspected abnormal time period are obtained through the fluctuation amplitude characteristics and the fluctuation period characteristics.

[0027] Furthermore, the reliability of the rotation speed at each suspected abnormal moment is obtained by using the length factor of each suspected abnormal moment in each suspected abnormal time period and the fluctuation characteristics, and the specific steps include the following:

[0028] The product of the length factor and the fluctuation feature is recorded as the first speed credibility at each suspected abnormal moment, and the first speed credibility is linearly normalized to obtain the speed credibility at each suspected abnormal moment.

[0029] Furthermore, the data in the rotation speed data sequence is corrected by using the credibility, and the corrected sequence is recorded as a new rotation speed data sequence, including the following specific steps:

[0030] All moments other than the suspected abnormal moments are recorded as non-suspected abnormal moments, and the speed credibility of the non-suspected abnormal moments is recorded as 1;

[0031] The calculation formula for correcting the speed data at each moment in the speed data sequence is:

[0032]

[0033] In the formula, Indicates The moment corresponds to the speed data in the speed data sequence, Indicates The reliability of the speed at a given moment, Indicates The reliability of the speed at a given moment, Indicates The moment and The time interval between moments, Indicates The corrected speed data at each moment, Indicates The corrected speed data at each moment;

[0034] The speed data at the first moment is not corrected, and the data at the first moment is used as the corrected speed data to obtain the corrected speed data at the second moment;

[0035] All the corrected speed data at all times are grouped into a sequence in chronological order, which is recorded as a new speed data sequence.

[0036] Furthermore, the real-time performance monitoring of the automobile electronic control system by using the new speed data sequence includes the following specific steps:

[0037] The ARIMA model is used to detect anomalies in the new speed data sequence, and the corresponding abnormal time and speed data are obtained to complete the real-time performance monitoring of the automotive electronic control system.

[0038] The present invention also provides a real-time performance monitoring system for an automobile electronic control system based on cloud computing, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the system implements any one of the steps of the above-mentioned method for real-time performance monitoring of an automobile electronic control system based on cloud computing.

[0039] The technical solution of the present invention has the following beneficial effects: the present invention selects suspected abnormal moments from all moments through the difference between the local data at each moment in the speed difference sequence, thereby improving the accuracy of the initial analysis of the data abnormal moments; through the distribution of all suspected abnormal moments, a number of suspected abnormal time periods in the time period corresponding to the speed data sequence are obtained; the reciprocal of the length of each suspected abnormal time period is used as the length factor of each suspected abnormal moment in each suspected abnormal time period, thereby improving the accuracy of the analysis of the length of the fluctuation of the suspected abnormal time period; through the fluctuation amplitude and periodicity of the data in the speed sequence of each suspected abnormal time period, the fluctuation characteristics of each suspected abnormal time period are obtained, thereby improving the accuracy of the analysis of the fluctuation amplitude and periodicity of the suspected abnormal time period; through the length factor and the fluctuation characteristics of each suspected abnormal moment in each suspected abnormal time period, the speed credibility of each suspected abnormal moment is obtained, thereby reducing the influence of the sensor on the data acquisition; the data in the speed data sequence is corrected through the credibility, and the corrected sequence is recorded as a new speed data sequence, thereby improving the accuracy of the speed data acquisition; the real-time performance monitoring of the automobile electronic control system is performed through the new speed data sequence, thereby improving the accuracy of the real-time performance monitoring of the automobile electronic control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 The present invention is a flowchart of the steps of the real-time performance monitoring method of the automobile electronic control system based on cloud computing;

[0042] Figure 2 This is a flow chart for real-time performance monitoring of automotive electronic control systems. DETAILED DESCRIPTION

[0043] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the real-time performance monitoring method and system of the automotive electronic control system based on cloud computing proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0045] The specific scheme of the real-time performance monitoring method and system of an automobile electronic control system based on cloud computing provided by the present invention is described in detail below with reference to the accompanying drawings.

[0046] See also Figure 1 , which shows a flowchart of a method for real-time performance monitoring of an automotive electronic control system based on cloud computing provided by an embodiment of the present invention, the method comprising the following steps:

[0047] Step S001: Collecting engine speed data in the automobile electronic control system.

[0048] It should be noted that in order to eliminate potential risks in automobiles, the engine in the automobile's electronic control system will be monitored for abnormalities in real time to discover potential risks in the automobile's electronic control system and reduce the possibility of danger.

[0049] Specifically, with a sampling time interval of 1 second, the engine speed data of the electronically controlled vehicle within two hours is continuously collected through the crankshaft position sensor; the engine speed data at all times are sorted in chronological order to form a set of sequences, which are recorded as speed data sequences.

[0050] At this point, the speed data sequence of the electronically controlled automobile engine is obtained.

[0051] Step S002: Obtain a speed difference sequence according to the speed data sequence, and obtain several suspected abnormal time periods within the time period corresponding to the speed data sequence through the differences between the local data in the speed difference sequence, wherein the moment within the suspected abnormal time period is the suspected abnormal moment; and use the inverse of the length of each suspected abnormal time period as the length factor of each suspected abnormal moment within each suspected abnormal time period.

[0052] It should be noted that when the car is operating normally, the car speed is stable. When accelerating or decelerating, the engine speed increases or decreases steadily. When the engine is abnormal, the engine speed fluctuates unstably. However, it is not clear whether the unstable fluctuation of the engine speed is caused by the engine abnormality or the accuracy of the crankshaft position sensor. Therefore, it is necessary to analyze the collected speed timing data.

[0053] It should be further explained that when the car is driving normally, the speed of the car is stable; when driving on a straight road, the accelerator may be stepped on to accelerate, and when turning, the accelerator may be appropriately released to decelerate, but whether it is deceleration or acceleration, the accelerator is stepped on slowly, so the change in speed is also stable to increase or decrease; so the fluctuation or abnormality of the data can be reflected by the change of local data at each moment; that is, the data changes of the local adjacent moments in the speed data sequence of the electronically controlled automobile engine are generally the same. Therefore, the speed difference sequence is obtained, and through the differences between the local data in the speed difference sequence, several suspected abnormal time periods in the time period corresponding to the speed data sequence are obtained.

[0054] Preferably, the data at each moment in the rotation speed data sequence of the electronically controlled automobile engine is subtracted from the data at the adjacent previous moment to obtain a set of sequences, which are recorded as rotation speed difference sequences.

[0055] Further, as an embodiment, the specific calculation method of the fluctuation degree at each moment is:

[0056] Preset a parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0057] Each moment in the speed difference sequence is regarded as the center point of the local window. is the local window size, and the data in the local window at each moment is obtained. Among them, the number of data on the left and right sides of the center point of the local window is the same, both are indivual.

[0058] The difference between any data and the mean of all data in the local window at each moment in the speed difference sequence is recorded as the first difference of any data in the local window at each moment, the mean of the first differences of all data in the local window at each moment is recorded as the first fluctuation degree at each moment, and the first fluctuation degree is linearly normalized to obtain the fluctuation degree at each moment;

[0059] The fluctuation degree is greater than or equal to the preset threshold The moment is recorded as the suspected abnormal moment.

[0060] In one embodiment of the present invention, it is specifically expressed by the formula:

[0061]

[0062] In the formula, Indicates the first The local window at the moment data, Indicates the first The mean of all data in the local window at time instant, Indicates the first The number of all data in the local window at a certain moment, is the absolute value symbol, Indicates The degree of fluctuation at a moment, represents the linear normalization function.

[0063] in, It indicates the difference of speed difference in the local range of the speed difference sequence. When the difference is larger, the fluctuation degree at that moment is larger, indicating that the possibility of abnormality at that moment is larger; when the difference is smaller, the fluctuation degree at that moment is smaller, indicating that the possibility of abnormality at that moment is smaller.

[0064] At this point, the degree of fluctuation at each moment is obtained.

[0065] In this embodiment, the preset threshold This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0066] Preferably, the time periods corresponding to the consecutive suspected abnormal moments are recorded as suspected abnormal time periods; thus, a number of suspected abnormal time periods are obtained.

[0067] It should be noted that during normal operation of the engine, the engine speed is determined by factors such as cylinder volume, number of cylinders, valves, throttle, humidity and temperature of the external environment. In reality, the cylinder volume and number of cylinders of each car are fixed, so the cylinder volume and number of cylinders cannot affect the speed; the valves and throttles are used to generate power through combustion to affect the engine speed. Therefore, by analyzing the abnormal speed, it is possible to analyze whether the valve and throttle system is normal. High temperature or high humidity will reduce air density, thereby affecting combustion efficiency and the maximum power output of the engine, which can also affect the speed. However, the temperature and humidity corresponding to cars in different areas are different, so the effects of temperature and humidity on the speed are temporary.

[0068] It should be further explained that valve abnormalities are generally caused by spring breakage, valve jamming or throat jamming, etc. Such problems may cause the valve to fail to open or close normally, resulting in sudden and drastic changes in speed. The throttle may be a temporary mechanical failure of the pedal, causing speed fluctuations; however, the abnormal data fluctuations caused by the valve and throttle are temporary. The crankshaft position sensor that collects speed data has reduced performance due to long-term use, but the abnormality of the sensor causes the abnormal fluctuations in the collected speed data to be long-term. Therefore, further analysis can be performed based on the length of time corresponding to each suspected fluctuation period.

[0069] Preferably, the length corresponding to each suspected abnormal time period is obtained, wherein the length corresponding to each suspected abnormal time period is the number of moments included. The reciprocal of the length of each suspected abnormal time period is used as the length factor of each suspected abnormal moment in each suspected abnormal time period. The length factors of all suspected abnormal moments in each suspected abnormal time period are the same.

[0070] At this point, the length factor of each suspected abnormal moment in each suspected abnormal time period is obtained.

[0071] Step S003: Obtain the speed sequence of each suspected abnormal time period, and obtain the fluctuation characteristics of each suspected abnormal time period through the fluctuation amplitude and periodicity of the data in the speed sequence of each suspected abnormal time period; obtain the speed credibility of each suspected abnormal moment through the length factor of each suspected abnormal moment in each suspected abnormal time period and the fluctuation characteristics, correct the data in the speed data sequence according to the credibility, and record the corrected sequence as a new speed data sequence.

[0072] It should be noted that if the speed data fluctuates periodically or with a stable frequency, it may be a problem with the crankshaft position sensor signal itself, which is usually caused by the instability or damage of the sensor signal; if the speed data fluctuates irregularly, the frequency is unstable, or is obviously related to the vehicle's accelerator pedal operation, it may be due to problems with the valve or throttle system that cause unstable engine output. Therefore, analysis can be performed based on the periodicity of the data.

[0073] It should be further explained that the amplitude of the sensor signal varies greatly, which may cause the speed data to fluctuate, but this fluctuation is usually within a small range and will not cause extreme speed deviation; the failure of the valve or throttle system may cause speed fluctuations within a large range, which may be accompanied by obvious power output changes or instability perceived by vehicle power. Therefore, the amplitude of the data fluctuation can be used for analysis. Therefore, the fluctuation characteristics of each suspected abnormal time period are obtained by the fluctuation amplitude and periodicity of the data in the speed sequence of each suspected abnormal time period.

[0074] Preferably, the speed data corresponding to each suspected abnormal time period in the speed data sequence is obtained, and a group of sequences is obtained in chronological order, which is recorded as the speed sequence of each suspected abnormal time period.

[0075] Further, as an embodiment, the specific calculation method of the fluctuation characteristics of each suspected abnormal time period is:

[0076] The speed sequence of each suspected abnormal time period is curve-fitted using a fifth-order polynomial by the least squares method to obtain a speed curve of each suspected abnormal time period. The least squares method is a well-known technique and will not be described in detail here. In this embodiment, a fifth-order polynomial is used for curve fitting, but it is not specifically limited and can be determined by the implementer according to the specific situation.

[0077] Obtain the extreme value points in the speed curve of each suspected abnormal time period; wherein the extreme value points include maximum value points and minimum value points.

[0078] The average of the absolute values ​​of the slopes between all adjacent extreme value points is recorded as the fluctuation amplitude characteristic of each suspected abnormal time period; the time intervals between all adjacent extreme value points are obtained, and the fluctuation period characteristics of each suspected abnormal time period are obtained by taking the average of the differences between all adjacent time intervals;

[0079] The fluctuation characteristics of each suspected abnormal time period are obtained through the fluctuation amplitude characteristics and the fluctuation period characteristics.

[0080] In one embodiment of the present invention, it is specifically expressed by the formula:

[0081]

[0082] In the formula, Indicates The speed curve of the suspected abnormal time period The extreme point and The slope between the extreme points, Indicates The speed curve of the suspected abnormal time period The extreme point and The time interval between extreme points, Indicates The speed curve of the suspected abnormal time period The extreme point and The time interval between extreme points, Indicates The number of all extreme value points on the speed curve of the suspected abnormal time period, is the absolute value symbol, Indicates The fluctuation characteristics of a suspected abnormal time period.

[0083] Among them, the larger the absolute value of the slope between adjacent extreme points on the speed curve of each suspected abnormal time period, the greater the fluctuation of the data in the suspected abnormal time period, that is, the greater the possibility of a real abnormal speed; the smaller the absolute value of the slope between adjacent extreme points on the speed curve of each suspected abnormal time period, the smaller the fluctuation of the data in the suspected abnormal time period, that is, the smaller the possibility of a real abnormal speed, and the greater the possibility that the fluctuation in the suspected abnormal time period is caused by the sensor. It represents the difference in time intervals between adjacent extreme value points. When the difference is smaller, it means that the data fluctuation in the suspected abnormal time period is more periodic, that is, the possibility of a real abnormal speed is smaller, and the possibility that the fluctuation in the suspected abnormal time period is caused by the sensor is greater; when the difference is larger, it means that the data fluctuation in the suspected abnormal time period is less periodic, that is, the possibility of a real abnormal speed is greater, and the possibility that the fluctuation in the suspected abnormal time period is caused by the sensor is smaller.

[0084] At this point, the fluctuation characteristics of each suspected abnormal time period are obtained.

[0085] The fluctuation characteristics of each suspected abnormal time period are used as the fluctuation characteristics of each suspected abnormal moment in the suspected abnormal time period; wherein the fluctuation characteristics of all suspected abnormal moments in each suspected abnormal time period are the same.

[0086] It should be noted that, when analyzing the abnormal speed of the engine, the normal speed data is identified as abnormal speed data due to the abnormality of the crankshaft position sensor. Therefore, the reliability of the speed data at each suspected abnormal moment is obtained by analyzing the real abnormality of the engine and the abnormality of the sensor. The abnormal speed data caused by the sensor is corrected by the reliability, and the real speed data is monitored by the corrected data. Therefore, the speed reliability of each suspected abnormal moment is obtained by the length factor and the fluctuation characteristics of each suspected abnormal moment in each suspected abnormal time period, and the data in the speed data sequence is corrected by the reliability, and the corrected sequence is recorded as the new speed data sequence.

[0087] Further, as an embodiment, the specific calculation method of the speed credibility at each suspected abnormal moment is:

[0088] The product of the length factor and the fluctuation feature is recorded as the first speed credibility at each suspected abnormal moment, and the first speed credibility is linearly normalized to obtain the speed credibility at each suspected abnormal moment.

[0089] In one embodiment of the present invention, it is specifically expressed by the formula:

[0090]

[0091] In the formula, Indicates The fluctuation characteristics of the suspected abnormal moment, Indicates The length factor of the suspected abnormal moment, Indicates The reliability of the speed at the suspected abnormal moment, Represents an exponential function with a natural constant as its base.

[0092] Among them, the smaller the fluctuation characteristics and length factor of each suspected abnormal moment, the more likely it is that the suspected abnormal moment is caused by the sensor, and the lower the credibility of the speed data at the suspected abnormal moment; conversely, the larger the abnormal degree and abnormal factor of each suspected abnormal moment, the less likely it is that the suspected abnormal moment is caused by the sensor, and the higher the credibility of the speed data at the suspected abnormal moment.

[0093] At this point, the reliability of the speed at each suspected abnormal moment is obtained.

[0094] All moments other than the suspected abnormal moments are recorded as non-suspected abnormal moments, and the speed credibility of the non-suspected abnormal moments is recorded as 1.

[0095] The calculation formula for correcting the speed data at each moment in the speed data sequence is:

[0096]

[0097] In the formula, Indicates The moment corresponds to the speed data in the speed data sequence, Indicates The reliability of the speed at a given moment, Indicates The reliability of the speed at a given moment, Indicates The moment and The time interval between moments, Indicates The corrected speed data at each moment, Indicates The speed data at the first moment is not corrected, and the data at the first moment is used as the corrected speed data to obtain the speed data at the second moment, and the speed data at each subsequent moment is corrected in turn.

[0098] All the corrected speed data at all times are grouped into a sequence in chronological order, which is recorded as a new speed data sequence.

[0099] Step S004: Real-time performance monitoring of the automobile electronic control system is performed through the new speed data sequence.

[0100] The ARIMA model is used to detect anomalies in the new speed data sequence, and the corresponding abnormal time and speed data are obtained, so as to complete the real-time performance monitoring of the automotive electronic control system; wherein, the ARIMA model is a well-known technology and will not be described in detail here.

[0101] At this point, the real-time performance monitoring of the automotive electronic control system is completed. The real-time performance monitoring flow chart of the automotive electronic control system is as follows: Figure 2 shown.

[0102] This embodiment provides a real-time performance monitoring system for an automobile electronic control system based on cloud computing, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the real-time performance monitoring method for an automobile electronic control system based on cloud computing in steps S001 to S004 is implemented.

[0103] At this point, this embodiment is completed.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A real-time performance monitoring method for an automotive electronic control system based on cloud computing, characterized in that: The method comprises the following steps: Obtaining the speed data sequence of the electronically controlled automobile engine; A speed difference sequence is obtained according to the speed data sequence, and a number of suspected abnormal time periods within the time period corresponding to the speed data sequence are obtained through the differences between the local data in the speed difference sequence, wherein the moments within the suspected abnormal time period are suspected abnormal moments; the inverse of the length of each suspected abnormal time period is used as the length factor of each suspected abnormal moment within each suspected abnormal time period; Obtain the speed sequence of each suspected abnormal time period, and obtain the fluctuation characteristics of each suspected abnormal time period through the fluctuation amplitude and periodicity of the data in the speed sequence of each suspected abnormal time period; obtain the speed credibility of each suspected abnormal moment through the length degree factor of each suspected abnormal moment in each suspected abnormal time period and the fluctuation characteristics, and correct the data in the speed data sequence through the credibility, and record the corrected sequence as a new speed data sequence; Real-time performance monitoring of automotive electronic control systems through new speed data sequences; The method for obtaining the speed credibility is: The product of the length factor and the fluctuation feature is recorded as the first speed credibility at each suspected abnormal moment, and the first speed credibility is linearly normalized to obtain the speed credibility at each suspected abnormal moment.

2. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 1 is characterized in that: The specific steps of obtaining the rotation speed difference sequence according to the rotation speed data sequence are as follows: The data of each moment in the speed data sequence of the electronically controlled automobile engine is subtracted from the data of the adjacent previous moment to obtain a set of sequences, which are recorded as speed difference sequences.

3. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 1 is characterized in that: The method of obtaining several suspected abnormal time periods within the time period corresponding to the speed data sequence by using the differences between the local data in the speed difference sequence includes the following specific steps: According to the difference between the data of all moments in the local range of each moment in the speed difference sequence, the fluctuation degree of each moment is obtained; Recording the moment when the fluctuation degree is greater than or equal to the preset threshold as a suspected abnormal moment; Through the distribution of all suspected abnormal moments, several suspected abnormal time periods within the time period corresponding to the speed data sequence are obtained.

4. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 3 is characterized in that: The method of obtaining the fluctuation degree at each moment according to the difference between the data at all moments within the local range of each moment in the rotation speed difference sequence includes the following specific steps: Each moment in the speed difference sequence is taken as the center point of the local window, a local window is constructed according to a preset window size, and data within the local window at each moment is obtained; The difference between any data in the local window at each moment in the speed difference sequence and the mean of all data is recorded as the first difference of any data in the local window at each moment, and the mean of the first differences of all data in the local window at each moment is recorded as the first fluctuation degree at each moment. The first fluctuation degree is linearly normalized to obtain the fluctuation degree at each moment.

5. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 3 is characterized in that: The method of obtaining a plurality of suspected abnormal time periods within the time period corresponding to the speed data sequence through the distribution of all suspected abnormal time periods includes the following specific steps: The time period corresponding to the consecutive suspected abnormal moments is recorded as the suspected abnormal time period.

6. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 1 is characterized in that: The method of obtaining the fluctuation characteristics of each suspected abnormal time period by using the fluctuation amplitude and periodicity of the data in the rotation speed sequence of each suspected abnormal time period includes the following specific steps: The speed sequence of each suspected abnormal time period is curve-fitted by the least square method to obtain the speed curve of each suspected abnormal time period; Obtaining extreme value points in the speed curve of each suspected abnormal time period, wherein the extreme value points include maximum value points and minimum value points; The average of the absolute values ​​of the slopes between all adjacent extreme value points is recorded as the fluctuation amplitude characteristic of each suspected abnormal time period; the time intervals between all adjacent extreme value points are obtained, and the fluctuation period characteristics of each suspected abnormal time period are obtained by taking the average of the differences between all adjacent time intervals; The fluctuation characteristics of each suspected abnormal time period are obtained through the fluctuation amplitude characteristics and the fluctuation period characteristics.

7. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 1 is characterized in that: The data in the rotation speed data sequence is corrected by using the credibility, and the corrected sequence is recorded as a new rotation speed data sequence, including the following specific steps: All moments other than the suspected abnormal moments are recorded as non-suspected abnormal moments, and the speed credibility of the non-suspected abnormal moments is recorded as 1; The calculation formula for correcting the speed data at each moment in the speed data sequence is: In the formula, Indicates The moment corresponds to the speed data in the speed data sequence, Indicates The reliability of the speed at a given moment, Indicates The reliability of the speed at a given moment, Indicates The moment and The time interval between moments, Indicates The corrected speed data at each moment, Indicates The corrected speed data at each moment; The speed data at the first moment is not corrected, and the data at the first moment is used as the corrected speed data to obtain the corrected speed data at the second moment; All the corrected speed data at all times are grouped into a sequence in chronological order, which is recorded as a new speed data sequence.

8. The real-time performance monitoring method of automobile electronic control system based on cloud computing according to claim 1 is characterized in that: The real-time performance monitoring of the automobile electronic control system by using the new speed data sequence includes the following specific steps: The ARIMA model is used to detect anomalies in the new speed data sequence, and the corresponding abnormal time and speed data are obtained to complete the real-time performance monitoring of the automotive electronic control system.

9. A real-time performance monitoring system for an automobile electronic control system based on cloud computing, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the real-time performance monitoring method of an automobile electronic control system based on cloud computing as described in any one of claims 1-8 are implemented.

Images