A method for locating faults in automotive engines based on data analysis

Abstract

The present invention discloses a method for locating faults in an automobile engine based on data analysis, belonging to the technical field of automobile fault detection, including Step 1: formulating the normal operation data range of each monitoring item of the engine under different environmental condition values; Step 2: judging whether the operation data of each monitoring item under this environmental condition value is normal according to the obtained operation data information; Step 3: when the operation data of the monitoring item is abnormal, generating a corresponding fault alarm and determining the danger level of the fault; Step 4: when the operation data of the monitoring item is normal, predicting whether there is a potential fault risk for each monitoring item. When the engine has not failed, the present invention predicts whether there is a potential fault risk for each monitoring item according to the historical accumulation of the operation data at the positions of each monitoring item, so as to timely remind the user to perform corresponding inspections and repairs on the corresponding positions, thereby avoiding the occurrence of faults.

Description

Technical Field

[0001] The present invention belongs to the technical field of automotive fault detection, and particularly relates to a method for locating automotive engine faults based on data analysis. Background Art

[0002] The automotive engine is an important part of the vehicle, providing power for the vehicle and can be said to be the heart of the vehicle. Therefore, its performance and condition have an important impact on the overall performance and driving safety of the vehicle.

[0003] Existing methods for detecting automotive engine faults mostly judge engine faults by whether the fault light is on. First, when the fault light is on, it can only be determined that the engine has a fault, but the specific location is not clear enough, and the severity of the fault cannot be determined. Second, when the fault light is on, the fault has already occurred, so it is impossible to judge potential fault risks, which is likely to cause additional property losses. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for locating automotive engine faults based on data analysis to solve the problems faced in the above background art.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for locating automotive engine faults based on data analysis, the method comprising:

[0007] Step 1: Install corresponding sensors at various places of the items to be monitored on the engine, collect multiple groups of operation data information of each monitored item of the engine in the normal working state, and thus draw up the normal operation data range of each monitored item under different environmental condition values;

[0008] Step 2: Collect the environmental information of the engine and the operation data information of each monitored item, and send the collected operation data information and environmental information to the fault judgment center for analysis and processing. Determine the environmental condition value of the current engine according to the environmental information, and judge whether the operation data of each monitored item under this environmental condition value is normal according to the operation data information;

[0009] Step 3: When the operation data of the monitored item is abnormal, generate a corresponding fault alarm, and determine the danger level of the fault according to the monitored operation data information;

[0010] Step 4: When the operation data of the monitored item is normal, predict whether there are potential fault risks for each monitored item according to the cumulative situation of the monitored operation data information.

[0011] Further, the method for determining the environmental condition value of the current engine according to the environmental information in Step 2 is:

[0012] Obtain environmental information such as the ambient temperature, ambient humidity, altitude, and air pressure intensity of the engine, and input it into a pre-trained artificial intelligence environmental prediction model to obtain the current environmental factors;

[0013] Compare the obtained environmental factors with the set environmental condition values to obtain the difference situation between the environmental factors and each environmental condition value, and select the environmental condition value with the smallest difference as the current environmental condition value of the engine.

[0014] Further, the method for judging whether the operation data of each monitoring item under this environmental condition value is normal according to the operation data information in step two is:

[0015] Obtain the operation data of each monitoring item , and thus compare it with the normal operation data range of each monitoring item under this environmental condition value When the operation data is not within the normal operation data range, it is judged that the operation data of this monitoring item is abnormal.

[0016] Further, the method for determining the danger level of the fault according to the monitored operation data information in step three is:

[0017] Through the formula Obtain the deviation values of each monitoring item ;

[0018] When , it is considered that the engine has a minor fault, and the fault danger level is determined to be primary;

[0019] When , it is considered that the engine has a general fault, and the fault danger level is determined to be intermediate;

[0020] When , it is considered that the engine has a major fault, and the fault danger level is determined to be high;

[0021] Among them, , and are the preset deviation thresholds for each monitoring item, and .

[0022] Further, the method for predicting whether there is a potential fault risk for each monitoring item according to the cumulative situation of the monitored operation data information in step four is:

[0023] When the operation data of the monitoring item is normal, at this time, obtain the curve of the operation data of the monitoring item changing with time within a period , while obtaining the deviation values of the monitoring item at different times within the cycle ;

[0024] Thus, through the formula the potential abnormal value of the monitoring item is obtained ;

[0025] When the obtained potential abnormal value exceeds the preset potential abnormal threshold , a corresponding warning signal is generated for warning and reminder;

[0026] Among them, , , and , m is the total number of deviation values obtained within the cycle, is the proportion of the time when the deviation value obtained within the cycle is higher than the average deviation value, is the start time of the cycle, is the end time of the cycle, is the preset curve of standard operating data changing with time.

[0027] Furthermore, when a minor fault occurs, at this time, the operating parameters are corrected accordingly through the formula ;

[0028] Among them, is the corresponding data change amount that needs to be corrected for the i-th monitoring item, is the maximum corresponding data change amount that the i-th monitoring item is allowed to correct, is the correction influence factor of the i-th monitoring item.

[0029] Furthermore, the sensors include a coolant temperature sensor, an injection sensor, a knock sensor, an oil pressure sensor, a rotational speed sensor, etc.

[0030] Advantages of the present invention:

[0031] The present invention directly sets corresponding sensors at the positions of the items that need to be monitored on the engine, so as to obtain the operating data information of the corresponding positions. According to the obtained operating data information, it is judged whether the operating data of each monitoring item position is normal. When it is abnormal, an alarm reminder is generated in time to remind the user. In this way, it is possible to directly determine whether a fault occurs at the corresponding position according to the operating data situation of the corresponding monitoring item position, so that the position point of the fault can be quickly determined when the engine fails, which is convenient for maintenance; and after the alarm reminder is generated, the danger level of the fault can be determined according to the monitored operating data information, which is convenient for the user to determine the severity of the fault occurrence, so as to perform corresponding processing to ensure driving safety.

[0032] When the engine is not malfunctioning, the present invention can also predict whether there is a potential failure risk for each monitoring item based on the historical accumulation of the operating data at the positions of each monitoring item. Once a potential failure risk is detected, the user can be promptly reminded to perform corresponding inspections and repairs on the corresponding positions, thereby avoiding the occurrence of failures, reducing property losses, and ensuring personal safety.

[0033] Of course, it is not necessary for any product implementing the present invention to simultaneously achieve all the above-mentioned advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for describing the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0035] Figure 1 It is a flowchart of the steps of the fault location method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0037] In one embodiment, a method for locating faults in an automotive engine based on data analysis is disclosed. As Figure 1 shown, the fault location method includes:

[0038] Step 1: Install corresponding sensors at various positions of the items to be monitored in the engine, collect multiple sets of operating data information of each monitoring item of the engine in the normal working state, and thus formulate the normal operating data intervals of each monitoring item under different environmental condition values.

[0039] Step 2: Collect the environmental information of the engine and the operating data information of each monitoring item, and send the collected operating data information and environmental information to the fault judgment center for analysis and processing. Determine the environmental condition value of the current engine according to the environmental information, and judge whether the operating data of each monitoring item under this environmental condition value is normal according to the operating data information.

[0040] Step 3: When the operating data of the monitoring item is abnormal, generate a corresponding fault alarm, and determine the danger level of the fault according to the monitored operating data information.

[0041] Step 4: When the operation data of the monitoring items is normal, predict whether there is a potential fault risk for each monitoring item according to the accumulation of the monitored operation data information.

[0042] Through the above technical solution, corresponding sensors are arranged at the positions of the items to be monitored of the engine in the present application. The sensors include a coolant temperature sensor, an injection sensor, a knock sensor, an oil pressure sensor, a rotational speed sensor, etc., so as to directly obtain the operation data information of the corresponding positions. Determine whether the operation data of each monitoring item position is normal according to the obtained operation data information. When it is abnormal, an alarm reminder is generated in time to remind the user. In this way, it is possible to directly determine whether a fault occurs at the corresponding position according to the operation data situation of the corresponding monitoring item position, so that the position point of the fault can be quickly determined when the engine fails, which is convenient for maintenance. And after generating the alarm reminder, the risk level of the fault can be determined according to the monitored operation data information, which is convenient for the user to determine the severity of the fault occurrence, so as to perform corresponding processing to ensure driving safety; at the same time, when the engine does not fail, it is also possible to predict whether there is a potential fault risk for each monitoring item according to the historical accumulation of the operation data of each monitoring item position. Once it is detected that there is a potential fault risk, the user can be reminded in time to perform corresponding inspections and repairs on the corresponding positions, thereby avoiding the occurrence of faults, reducing property losses, and ensuring personal safety.

[0043] As an implementation manner of the present invention, the method for determining the environmental condition value of the current engine according to the environmental information in step 2 is as follows:

[0044] Obtain environmental information such as the environmental temperature, environmental humidity, altitude, and air pressure intensity where the engine is located, and input it into a pre-trained artificial intelligence environmental prediction model to obtain the current environmental factor;

[0045] Compare the obtained environmental factor with each set environmental condition value to obtain the difference situation between the environmental factor and each environmental condition value, and select the environmental condition value with the smallest difference situation as the environmental condition value of the current engine.

[0046] Through the above technical solution, this embodiment provides a specific method for determining the environmental condition value of the current engine. Since the performance of the engine is different under different environments, in order to better judge whether the engine has a fault, first collect sufficient environmental information and multiple sets of operation data information of each monitoring item in the normal working state, and then use a machine learning algorithm to train a model library. The model library contains the normal operation data intervals under different environmental condition values. Then, obtain environmental information such as the environmental temperature, environmental humidity, altitude, and air pressure intensity of the engine, and bring it into the pre-trained artificial intelligence environmental prediction model to obtain the current environmental factor. Since each set environmental condition value is a specific value set, the obtained environmental factor will not match the set environmental condition value. Therefore, compare the obtained environmental factor with each set environmental condition value to obtain the difference between the environmental factor and each environmental condition value, and select the environmental condition value with the smallest difference as the environmental condition value of the current engine. In this way, the current environmental condition value can be obtained according to the size of the current environmental factor falling into the set environmental condition value.

[0047] As an implementation manner of the present invention, the method for judging whether the operation data of each monitoring item is normal according to the operation data information in step two is as follows:

[0048] Obtain the operation data of each monitoring item , so as to compare with the normal operation data interval of each monitoring item under this environmental condition value for comparison. When the operation data is not within the normal operation data interval, it is judged that the operation data of this monitoring item is abnormal.

[0049] Through the above technical solution, this embodiment provides a specific method for judging whether the operation data of each monitoring item is normal. Due to the differences between automobile engines, their operation data will not be exactly the same, but will be within a general interval. Therefore, according to the historical data of the same model of engines running in big data, a normal operation data interval is drawn up for each monitoring item , and then obtain the operation data of each monitoring item , so as to compare with the normal operation data interval of each monitoring item under this environmental condition value Compare. When the operating data is not within the normal operating data range, it is determined that the operating data of the monitoring item is abnormal. For example, in the monitoring item of the engine cooling system, the coolant temperature of the engine is obtained through the installed coolant temperature sensor, and then compared with the normal temperature range. If the obtained cooling temperature is not within the corresponding normal temperature range, it indicates that there is a fault problem with the engine cooling system. Through this operation, it is possible to accurately judge whether there is a fault at a certain position of the engine in a timely manner based on the operating data, thereby determining the fault problem and dealing with it in a timely manner to reduce the occurrence of safety accidents.

[0050] As an implementation manner of the present invention, the method for determining the risk level of a fault according to the monitored operating data information in step three is:

[0051] Through the formula Obtain the deviation value of each monitoring item ;

[0052] When It is considered that the engine has a minor fault, and the fault risk level is determined to be primary;

[0053] When It is considered that the engine has a general fault, and the fault risk level is determined to be intermediate;

[0054] When It is considered that the engine has a major fault, and the fault risk level is determined to be high;

[0055] Among them, 、 And Are the preset deviation thresholds for each monitoring item, and ;

[0056] When a minor fault occurs, at this time, through the formula Make corresponding corrections to the operating parameters;

[0057] Among them, Is the corresponding data change amount that needs to be corrected for the i-th monitoring item, Is the maximum corresponding data change amount allowed to be corrected for the i-th monitoring item, Is the correction influence factor for the i-th monitoring item.

[0058] Through the above technical solution, this embodiment provides a method for determining the risk level of a fault according to the monitored operating data information. When a fault occurs, only knowing that the fault has occurred but not knowing the risk level of the fault will affect the judgment of the user. For example, when the fault is very small, an inexperienced user may think that the engine has a major fault, causing panic; therefore, after obtaining the operating data of each monitoring item, through the formula Obtain the deviation values of each monitoring item , when , then set the deviation value to , when , then set the deviation value to , and then compare the deviation value with the preset deviation thresholds , and . When , it is considered that the engine has a minor fault, indicating that the fault is small at this time. After parameter adjustment, judgment can be made again, and the fault danger level is determined to be primary. When , it is considered that the engine has a general fault, indicating that the fault situation is general at this time. Then, it can be repaired after the operation ends, and the fault danger level is determined to be intermediate. When , it is considered that the engine has a major fault, and the operation needs to be stopped immediately at this time, and the fault danger level is determined to be high; in this way, the fault level can be determined in a timely manner according to the operation data, which is convenient for the user to understand the fault risk. When a minor fault occurs, the method of adjusting and modifying the operation parameters is: through the formula Perform corresponding corrections on the operation parameters is the corresponding data change amount that needs to be corrected for the i-th monitoring item is the maximum corresponding data change amount that the i-th monitoring item is allowed to correct is the correction influence factor of the i-th monitoring item; taking the monitoring of the engine cooling system item as an example, when it is detected that the coolant temperature is too high, the system improves the refrigeration efficiency of the cooling system. At this time, through the formula Increase the refrigeration efficiency of the cooling system on the original basis units to perform temperature correction to ensure the normal temperature of the coolant

[0059] In the above technical solution, the preset deviation thresholds , and , the corresponding data change amounts that each monitoring item needs to correct, the maximum corresponding data change amounts that each monitoring item is allowed to correct, and the correction influence factors of each monitoring item can all be determined according to historical data combined with empirical data, and will not be elaborated here

[0060] As an implementation manner of the present invention, the method for predicting whether there is a potential fault risk for each monitoring item according to the cumulative situation of the monitored operation data information in step four is

[0061] When the operation data of the monitoring item is normal, obtain the curve of the operation data of the monitoring item changing with time within a period , while obtaining the deviation values of the monitoring item at different times within the period ;

[0062] Thus, through the formula the potential abnormal value of the monitoring item is obtained ;

[0063] When the obtained potential abnormal value exceeds the preset potential abnormal threshold , a corresponding warning signal is generated for warning reminder;

[0064] Among them, , , and , m is the total number of deviation values obtained within the period, is the time proportion of the deviation values obtained within the period that are higher than the average deviation value, is the start time of the period, is the end time of the period, is the preset curve of standard operating data varying with time.

[0065] Through the above technical solution, this embodiment provides a method for predicting whether there is a potential failure risk for each monitoring item. First, when the operating data of the detection item is normal, the operating data of the monitoring item within a period is obtained, so as to formulate the curve of the operating data of the monitoring item varying with time within this period , the period duration can be determined artificially according to the actual situation. At the same time, the deviation values obtained at different times within this period for this monitoring item are obtained , thus through the formula the potential abnormal value of this monitoring item is obtained , and , , the formula represents the difference situation between the operating data obtained within this period and the preset standard operating data. The larger its value, the greater the potential risk may be. And the formula represents the fluctuation situation of the deviation values obtained within this period. When the fluctuation is greater, that is, when the value is larger, it indicates that the possibility of potential failure risk is greater. And represents the time proportion situation of the deviation values obtained within this period that are higher than the average deviation value. It can be seen that when the proportion is larger, it indicates that the possibility of potential failure risk is greater. Therefore, when the potential abnormal value obtained by the formula is larger, it indicates that the possibility of potential failure risk of the engine is greater. So when the obtained potential abnormal value exceeds the preset potential abnormal threshold When there is a potential fault risk in the engine, a corresponding warning signal is generated for warning and reminder. This can timely remind the user to process and repair the potential fault parts, reduce the actual occurrence of faults, thereby reducing property losses and ensuring personal safety.

[0066] In the above technical solution, the preset standard operation data change curve over time and the preset potential anomaly threshold can both be selected and determined according to the relevant data and historical data in big data.

[0067] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution. As long as they do not deviate from the concept of the invention or exceed the scope defined by this claim book, they should all belong to the protection scope of the present invention.

Claims

1. A method for locating automobile engine faults based on data analysis, characterized in that: The method comprises: Step 1: Install corresponding sensors at various locations of the engine that need to be monitored, collect multiple sets of operating data information of each monitoring item of the engine in normal working state, and thus formulate normal operating data intervals of each monitoring item under different environmental conditions; Step 2: Collect the engine's environment information and the operating data information of each monitoring item, and send the collected operating data information and environment information to the fault assessment center for analysis and processing, determine the current engine environment status value according to the environment information, and judge whether the operating data of each monitoring item under the environment status value is normal according to the operating data information; Step 3: When the operating data of the monitored item is abnormal, a corresponding fault alarm is generated, and the danger level of the fault is determined based on the monitored operating data information; Step 4: When the operating data of the monitored items are normal, predict whether there is a potential failure risk for each monitored item based on the accumulated monitored operating data information; The method for predicting whether each monitoring item has a potential failure risk based on the accumulated monitored operation data information in step 4 is as follows: when the operation data of the monitoring item is normal, a time-varying curve Q(t) of the operation data of the monitoring item within a period is obtained, and at the same time, the deviation value A of the monitoring item at different times within the period is obtained. j ; So through the formula The potential outlier value R of the monitoring item is obtained; When the acquired potential abnormal value R exceeds the preset potential abnormal threshold R th When the alarm occurs, a corresponding warning signal is generated for early warning reminder; in, And j∈(1,m), m is the total number of deviation values ​​obtained in the cycle, is the percentage of time that the deviation value obtained in the period is higher than the average deviation value, t a is the start time of the cycle, t b is the end time of the cycle, and Q0(t) is the preset standard operating data change curve over time.

2. The method for locating automobile engine faults based on data analysis according to claim 1, characterized in that: The method for determining the current engine environmental condition value according to the environmental information in step 2 is: Obtain the ambient temperature, humidity, altitude and air pressure information of the engine, and bring it into the pre-trained artificial intelligence environment prediction model to obtain the current environmental factors; The obtained environmental factor is compared with each set environmental condition value to obtain the difference between the environmental factor and each environmental condition value, and the environmental condition value with the smallest difference is selected as the environmental condition value of the current engine.

3. The method for locating automobile engine faults based on data analysis according to claim 2, characterized in that: The method for judging whether the operation data of each monitoring item under the environmental condition value is normal according to the operation data information in step 2 is: Get the operation data of each monitoring project Q i , thus the normal operating data interval of each monitoring item under the environmental condition value A comparison is performed. When the operating data is not within the normal operating data range, it is determined that the operating data of the monitoring item is abnormal.

4. The method for locating automobile engine faults based on data analysis according to claim 3, characterized in that: The method for determining the danger level of the fault based on the monitored operating data information in step 3 is: By formula Obtain the deviation value A of each monitoring item i ; when When the engine is considered to have a minor fault, the fault risk level is determined to be primary; when When the engine is considered to have a general fault, the fault risk level is determined to be medium; when When the engine is considered to have a major fault, the fault risk level is determined to be high; in, as well as Preset deviation thresholds for each monitoring item, and 5. The method for locating automobile engine faults based on data analysis according to claim 4, characterized in that: When a minor fault occurs, the formula Make corresponding corrections to the operating parameters; Among them, Rev i is the corresponding data change amount that needs to be corrected for the i-th monitoring item, is the maximum corresponding data change allowed to be corrected for the i-th monitoring item, δ i is the modified impact factor of the ith monitoring item.

6. The automobile engine fault location method based on data analysis according to claim 1, characterized in that: The sensors include a coolant temperature sensor, a fuel injection sensor, a knock sensor, an oil pressure sensor, and a rotation speed sensor.

Images