Engine characteristic data acquisition method and system based on same timestamp storage
By adopting a method of acquiring characteristic data saved with the same timestamp in high-power engines and combining it with engine parameters for comprehensive fault judgment, the problem of fault judgment with strong correlation between multiple parameters is solved, the accuracy and real-time nature of the data are achieved, the workload of manual processing is reduced, and the accuracy and efficiency of the artificial intelligence model are improved.
Patent Information
- Application Number
- CN202510462944.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-14
- Publication Date
- 2025-09-12
Smart Images
Figure CN120636015A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of engine artificial intelligence technology, and in particular relates to a method and system for acquiring engine characteristic data based on storage with the same timestamp. Background Art
[0002] Real-time monitoring and data logging are crucial for the operation and maintenance of high-power engines. Current engine alarms rely on threshold alarms, which trigger an alarm when a parameter exceeds a certain limit. However, when faced with a fault involving multiple parameters that are highly correlated and do not trigger a threshold alarm, existing alarm methods are unable to identify and issue an alarm. These multi-parameter faults are difficult to diagnose directly and, over time, can lead to more serious failures and damage the machine. This often requires manual review of historical data to identify and analyze the cause after a fault occurs.
[0003] In recent years, artificial intelligence (AI) technology has rapidly developed and is being applied to the engine sector, offering an effective approach to addressing multi-parameter, multi-feature early warning issues. However, AI models require extensive training with feature data, which typically requires manual processing. This data often requires manual analysis, cleaning, conversion, and construction, making the data processing process cumbersome, time-consuming, and labor-intensive. The real-time, diverse, and frequently updated nature of engine data poses significant challenges to its collection and utilization. This inability to accurately and effectively identify and diagnose faults such as single-cylinder anomalies, power surges, turbocharger anomalies, intercooler anomalies, and rail pressure anomalies is a major obstacle to the application of AI technology. Summary of the Invention
[0004] In order to solve the above problems, the present invention proposes a method and system for acquiring engine feature data based on the same timestamp storage. The present invention performs average aggregation calculations according to preset time intervals, reduces the pressure of saving high-frequency data, and saves feature tags and engine operation data according to the same timestamp, ensuring data accuracy and real-time performance. By applying a time series database, operation data can be stored for a long time, and required feature data can be downloaded at any time, reducing the work of manual cleaning and data screening.
[0005] In order to achieve the above object, the present invention is implemented through the following technical solutions:
[0006] In a first aspect, the present invention provides a method for acquiring engine characteristic data based on storage with a timestamp, comprising:
[0007] Obtain relevant engine operating parameters and characteristic marker data corresponding to different engine fault types; wherein the fault types include single cylinder abnormality, power sudden change, supercharger abnormality, intercooler abnormality, and rail pressure abnormality;
[0008] Performing an average aggregation operation on the relevant operating parameters and the characteristic marking data at preset time intervals to obtain fault characteristics and parameters to be marked;
[0009] The fault characteristics and parameters to be marked are saved in a preset time series database according to the same timestamp and marked to obtain characteristic data.
[0010] Furthermore, the feature tags are replaced with numerical tags; the relevant operating parameters, the feature tag data and the feature data are all stored in the influxDB database, and the feature data is set in the edge device.
[0011] Furthermore, whether a single cylinder is working abnormally is comprehensively judged based on the engine single-cylinder exhaust temperature, engine speed, and engine load rate; specifically, when the engine is operating normally, at a preset speed and load, the single-cylinder exhaust temperature is fixed within a preset value range. When the value is higher or lower than the preset value range and exceeds a threshold, it is judged that the single cylinder is working abnormally.
[0012] Furthermore, it is determined whether the external load has a power mutation based on the engine speed and the engine load rate; specifically, when the engine load rate increases or decreases, if the engine speed does not change suddenly, it is determined to be normal loading or unloading; if the engine speed changes suddenly, it is determined to be a power mutation.
[0013] Furthermore, whether the supercharger is operating abnormally is indirectly inferred through the engine load rate, the exhaust temperature before the turbine, the exhaust temperature after the turbine, the supercharger speed and the boost pressure. Specifically, when the supercharger is operating normally, the relationship between the engine load rate, the exhaust temperature before the turbine, the exhaust temperature after the turbine, the supercharger speed and the boost pressure is that at a preset engine load rate, the exhaust temperature before the turbine is stable within a preset value range, the difference between the exhaust temperature after the turbine and the exhaust temperature before the turbine is fixed, and the supercharger speed and the boost pressure value are proportional; when the supercharger is operating abnormally, the exhaust temperature before the turbine exceeds the preset value, the difference between the exhaust temperature after the turbine and the exhaust temperature before the turbine becomes smaller, and / or the ratio of the supercharger speed to the boost pressure becomes larger.
[0014] Furthermore, whether the intercooler is abnormal is comprehensively judged based on the air temperature before intercooling, the air temperature after intercooling, the cooling water temperature and the cooling water pressure; specifically, when the cooling water pressure and the cooling water temperature are stable at preset values, the difference between the air temperature before intercooling and the air temperature after intercooling becomes smaller, and / or when the cooling water is stable, the cooling water pressure decreases, and it is judged that the intercooler is abnormal.
[0015] Furthermore, the engine speed, engine load rate and rail pressure are combined to determine whether the rail pressure is abnormal; specifically, when the engine speed and engine load rate are stable at preset values, the average rail pressure value is stable within a preset range, and when the average value of the rail pressure value deviation is above the preset average value, the rail pressure is considered abnormal.
[0016] In a second aspect, the present invention further provides a system for acquiring engine characteristic data based on storage with the same timestamp, comprising:
[0017] The data acquisition module is configured to obtain relevant engine operating parameters and characteristic marker data corresponding to different engine fault types, wherein the fault types include single-cylinder abnormality, power surge, supercharger abnormality, intercooler abnormality, and rail pressure abnormality;
[0018] The average value aggregation module is configured to: perform average value aggregation operation on the relevant operating parameters and the characteristic marking data at preset time intervals to obtain fault characteristics and parameters to be marked;
[0019] The characteristic data determination module is configured to save the fault characteristics and the parameters to be marked in a preset time series database according to the same timestamp for marking to obtain characteristic data.
[0020] In a third aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for acquiring engine characteristic data based on storage with the same timestamp as described in the first aspect.
[0021] In a fourth aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, the steps of the method for acquiring engine characteristic data based on the same timestamp storage described in the first aspect are implemented.
[0022] In a fifth aspect, the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the method for acquiring engine characteristic data based on the same timestamp storage described in the first aspect are implemented.
[0023] Compared with the prior art, the present invention has the following beneficial effects:
[0024] 1. In the present invention, first, the obtained relevant operating parameters and feature mark data are aggregated according to the preset time interval to obtain the fault characteristics and parameters to be marked; then, the fault characteristics and parameters to be marked are saved in a preset time series database according to the same timestamp; the average value aggregation operation is performed according to the preset time interval to reduce the pressure of saving high-frequency data, and the feature marks and engine operating data are saved according to the same timestamp, ensuring data accuracy and real-time performance; by applying the time series database, the operating data can be stored for a long time, and the required feature data can be downloaded at any time, reducing the work of manual cleaning and screening of data.
[0025] 2. In the present invention, whether a single cylinder is operating abnormally is comprehensively judged based on the engine single-cylinder exhaust temperature, engine speed, and engine load rate; whether the external load has a power mutation is judged based on the engine speed and engine load rate; whether the supercharger is operating abnormally is indirectly inferred through the engine load rate, exhaust temperature before the turbine, exhaust temperature after the turbine, supercharger speed, and boost pressure; whether the intercooler is abnormal is comprehensively judged based on the air temperature before the intercooler, air temperature after the intercooler, cooling water temperature, and cooling water pressure; and whether the rail pressure is abnormal is jointly judged in combination with the engine speed, engine load rate, and rail pressure. The present invention can accurately and effectively judge faults such as single-cylinder abnormality, power mutation, supercharger abnormality, intercooler abnormality, and rail pressure abnormality, solving the problem that the traditional threshold alarm method cannot judge the fault when facing a fault with strong correlation between multiple parameters and the threshold alarm is not triggered. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] The drawings constituting a part of the specification of this embodiment are used to provide a further understanding of this embodiment. The schematic embodiments and descriptions of this embodiment are used to explain this embodiment and do not constitute an improper limitation on this embodiment.
[0027] Figure 1 This is a block diagram of the method and system structure of Example 1 of the present invention;
[0028] Figure 2 Graph showing the relationship between engine characteristic importance parameters and characteristic marks in Example 1 of the present invention. DETAILED DESCRIPTION
[0029] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0030] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.
[0031] Example 1:
[0032] This embodiment provides a method for acquiring engine feature data based on timestamp storage, as well as a corresponding system. This method can store feature tags and engine operating data using the same timestamp, ensuring data accuracy and real-time performance. By utilizing a time-series database, operating data can be stored long-term and required feature data can be downloaded at any time, reducing the need for manual data cleaning and screening. By standardizing data formats, the accuracy of artificial intelligence training and prediction can be improved.
[0033] Optional, such as Figure 1As shown, the engine operation data and feature tags are saved in the InfluxDB database at the same time, and the two parts of data together constitute the feature data; the InfluxDB database is installed on the edge device, and the database is manually connected through a laptop to write the feature tags into the database; the data cleaning work is completed through the InfluxDB database, and after the feature data is manually downloaded, the feature data is improved and optimized according to a specific feature tagging method. The optimized data can be used for artificial intelligence model training.
[0034] Specifically, the system provides RS485, CAN, and RJ45 interfaces for receiving various engine-related data. All data is stored uniformly in the InfluxDB database. However, this data is written at different intervals, hindering data analysis and unsuitable for machine learning. Therefore, the database performs an average aggregation operation every one second, organizing all data into one-second intervals and saving the organized data into bucket 1. Table 1 shows the raw data written to the database, which contains a large number of null values, hindering data analysis. Table 2 shows the data after the average aggregation operation.
[0035] Table 1 Raw data
[0036]
[0037]
[0038]
[0039]
[0040] Table 2 Data after average value aggregation operation
[0041]
[0042]
[0043] like Figure 1 As shown, a laptop connects to the edge device via WiFi and uses Python scripts to read and write data to the InfluxDB database. During engine operation, the fault type is manually determined and the corresponding feature tags are written to the database. Simultaneously, the InfluxDB database automatically saves the 1-second mean data with the feature tags to bucket 2. Table 3 shows some of the data in bucket 2, with the feature tags added in the last column.
[0044] Table 3 1-second mean data with feature markers
[0045]
[0046]
[0047] Using character strings to represent feature tags is not conducive to data processing, so feature tags are replaced by numerical labels, as shown in Table 4.
[0048] Table 4 Feature markers replaced by numerical labels
[0049]
[0050]
[0051] When feature data is needed, the laptop connects to the edge device via WiFi and directly downloads the data in bucket 2. Manual verification and optimization of the feature data content are performed, and the feature tags of each piece of data are checked to obtain accurate prediction targets during training and improve the accuracy of the training data.
[0052] like Figure 2 As shown, in the feature marking of this embodiment, feature importance parameters are selected from the engine operating parameters, and the relationship between the feature importance parameters and the feature marking is determined. The basis and method for each feature marking are:
[0053] Based on the engine's single-cylinder exhaust temperature, engine speed, and engine load factor, a comprehensive assessment can be made of whether a single cylinder is operating abnormally. During normal engine operation, at a certain speed and load, the single-cylinder exhaust temperature remains within a fixed range. A significant increase or decrease above or below this fixed value indicates a single-cylinder abnormality. When a single-cylinder abnormality is manually identified, the corresponding numerical tag is written to the database, and the data is then downloaded for verification. The verified data is shown in Table 5. The data in Table 5 is merely an example to demonstrate the marking method and does not represent actual engine operating data. The bolded data indicates abnormal data. The marking method is to mark any abnormal single-cylinder exhaust temperature as a single-cylinder abnormality.
[0054] Table 5 Verified data
[0055]
[0056]
[0057] Engine speed and load factor can be used to determine whether the external load is experiencing a sudden power change, which can easily shorten engine life. Engine speed is measured, while load factor is calculated by the engine ECU. When the engine load factor increases or decreases, if the engine speed does not change suddenly, it can be considered normal loading or unloading. If the engine speed changes suddenly, it can be considered a sudden power change. A sudden change in engine speed can be understood as exceeding a threshold within a preset period of time, or the rate of change exceeds a preset rate. This method is used to identify sudden power changes. Example data is shown in Table 6.
[0058] Table 6 Power mutation characteristic markers
[0059]
[0060] The health of the supercharger directly affects engine reliability. While it's difficult to directly determine whether a supercharger is functioning properly, the engine load factor, pre-turbine exhaust temperature, post-turbine exhaust temperature, supercharger speed, and boost pressure can be used to indirectly infer whether the supercharger is operating abnormally. When the supercharger is operating normally, the relationship between these parameters is that at a certain engine load factor, the pre-turbine exhaust temperature is essentially stable within a range of values, the difference between the post-turbine exhaust temperature and the pre-turbine exhaust temperature is essentially fixed, and the supercharger speed and boost pressure are proportional. When the supercharger is operating abnormally, several situations can be identified: the pre-turbine exhaust temperature is excessively high, the difference between the post-turbine exhaust temperature and the pre-turbine exhaust temperature decreases, and the ratio of the supercharger speed to the boost pressure increases. This method is used to identify abnormal supercharger characteristics, with example data shown in Table 7.
[0061] Table 7 Abnormal characteristic mark of supercharger
[0062]
[0063] The intercooler uses cooling water to cool the pressurized air. After prolonged operation, it can become clogged, resulting in low cooling efficiency and impacting engine performance. Intercooler anomalies can be comprehensively identified based on the pre-intercooler air temperature, post-intercooler air temperature, cooling water temperature, and cooling water pressure. When the cooling water pressure and temperature are stable, the difference between the pre-intercooler air temperature and post-intercooler air temperature decreases. When the cooling water level is stable, the cooling water pressure decreases. This method can be used to identify intercooler anomalies. Example data is shown in Table 8.
[0064] Table 8 Intercooler abnormality characteristic mark
[0065]
[0066] Rail pressure refers to the high-pressure fuel pressure of a high-pressure common rail diesel engine. This parameter varies at different speeds and loads, making it difficult to determine an abnormality using a single threshold. A combination of engine speed, engine load factor, and rail pressure is required to determine whether a rail pressure abnormality exists. The determination method is to ensure that the engine speed and engine load factor are stable at the same value, and the average rail pressure remains within a stable range. If the rail pressure deviates from the average by more than 20%, the rail pressure is considered abnormal. An example is shown in Table 9.
[0067] Table 9 Abnormal rail pressure characteristic mark
[0068]
[0069]
[0070] In this embodiment, the engine data is written into the InfluxDB database, and the InfluxDB client program is installed in the engine edge device. The edge device has a built-in database writing program that can write the engine operation data and feature tags into the database at the same time. The data is preliminarily cleaned inside the database, and the original data is divided into 1s time intervals to ensure a unified data format. The database saves the data containing feature tags separately to avoid manual screening of data from a large amount of data and reduce manual labor intensity. When feature data needs to be obtained, the feature data is downloaded to the local computer through the built-in program, and then the data is manually improved and optimized to finally generate accurate feature data for model training. The engine operation data can be efficiently recorded, and the feature tags can be saved at the same time, and the feature data can be automatically processed through the database. The feature tags are divided into 5 types, and the accuracy of the feature tags and the operation data is manually checked to finally generate accurate feature data for model training.
[0071] Example 2:
[0072] This embodiment provides a system for acquiring engine characteristic data based on time stamp storage, including:
[0073] The data acquisition module is configured to obtain relevant engine operating parameters and characteristic marker data corresponding to different engine fault types, wherein the fault types include single-cylinder abnormality, power surge, supercharger abnormality, intercooler abnormality, and rail pressure abnormality;
[0074] The average value aggregation module is configured to: perform average value aggregation operation on the relevant operating parameters and the characteristic marking data at preset time intervals to obtain fault characteristics and parameters to be marked;
[0075] The characteristic data determination module is configured to save the fault characteristics and the parameters to be marked in a preset time series database according to the same timestamp for marking to obtain characteristic data.
[0076] The working method of the system is the same as the method for obtaining engine characteristic data based on the same timestamp storage in Example 1, and will not be repeated here.
[0077] Example 3:
[0078] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the method for acquiring engine characteristic data based on storage with the same timestamp as described in Example 1 are implemented.
[0079] Example 4:
[0080] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, the steps of the method for obtaining engine characteristic data based on the same timestamp storage described in Example 1 are implemented.
[0081] Example 5:
[0082] This embodiment provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the method for acquiring engine characteristic data based on storage with the same timestamp as described in Example 1 are implemented.
[0083] The above description is merely a preferred embodiment of this embodiment and is not intended to limit this embodiment. Those skilled in the art will readily appreciate that this embodiment may be modified and varied in various ways. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this embodiment shall be within the scope of protection of this embodiment.
Claims
1. A method for acquiring engine characteristic data based on storage with the same timestamp, characterized in that: include: Obtain relevant engine operating parameters and characteristic marker data corresponding to different engine fault types; wherein the fault types include single cylinder abnormality, power sudden change, supercharger abnormality, intercooler abnormality, and rail pressure abnormality; Performing an average aggregation operation on the relevant operating parameters and the characteristic marking data at preset time intervals to obtain fault characteristics and parameters to be marked; The fault characteristics and parameters to be marked are saved in a preset time series database according to the same timestamp and marked to obtain characteristic data.
2. The method for acquiring engine characteristic data based on storage with the same timestamp as claimed in claim 1, characterized in that: The feature tags are replaced with numerical tags; the relevant operating parameters, the feature tag data and the feature data are all stored in the influxDB database, and the feature data is set in the edge device.
3. The method for acquiring engine characteristic data based on storage with the same timestamp as claimed in claim 1, characterized in that: Whether a single cylinder is working abnormally is determined based on the engine's single-cylinder exhaust temperature, engine speed, and engine load rate. Specifically, when the engine is working normally, at a preset speed and load, the single-cylinder exhaust temperature is fixed within a preset value range. When the value is higher or lower than the preset value range and exceeds a threshold, it is determined that the single cylinder is working abnormally.
4. The method for acquiring engine characteristic data based on storage with the same timestamp as claimed in claim 1, characterized in that: Determine whether the external load has a power mutation based on the engine speed and engine load rate; specifically, when the engine load rate increases or decreases, if the engine speed does not change suddenly, it is determined to be normal loading or unloading; if the engine speed changes suddenly, it is determined to be a power mutation.
5. The method for acquiring engine characteristic data based on storage with the same timestamp as claimed in claim 1, characterized in that: Whether the supercharger is operating abnormally is indirectly inferred through the engine load rate, the exhaust temperature before the turbine, the exhaust temperature after the turbine, the supercharger speed and the boost pressure. Specifically, when the supercharger is operating normally, the relationship between the engine load rate, the exhaust temperature before the turbine, the exhaust temperature after the turbine, the supercharger speed and the boost pressure is that at a preset engine load rate, the exhaust temperature before the turbine is stable within a preset value range, the difference between the exhaust temperature after the turbine and the exhaust temperature before the turbine is fixed, and the supercharger speed and the boost pressure value are proportional; when the supercharger is operating abnormally, the exhaust temperature before the turbine exceeds the preset value, the difference between the exhaust temperature after the turbine and the exhaust temperature before the turbine becomes smaller, and / or the ratio of the supercharger speed to the boost pressure becomes larger.
6. The method for acquiring engine characteristic data based on storage with the same timestamp as claimed in claim 1, characterized in that: Whether the intercooler is abnormal is comprehensively judged based on the air temperature before intercooling, the air temperature after intercooling, the cooling water temperature and the cooling water pressure; specifically, when the cooling water pressure and the cooling water temperature are stable at preset values, the difference between the air temperature before intercooling and the air temperature after intercooling becomes smaller, and / or when the cooling water is stable, the cooling water pressure decreases, and it is judged that the intercooler is abnormal.
7. The method for acquiring engine characteristic data based on storage with the same timestamp as claimed in claim 1, characterized in that: The engine speed, engine load rate and rail pressure are combined to determine whether the rail pressure is abnormal. Specifically, when the engine speed and engine load rate are stable at preset values, the average rail pressure is stable within a preset range. When the average value of the rail pressure deviation is above the preset average value, the rail pressure is considered abnormal.
8. An engine characteristic data acquisition system based on time stamp storage, characterized in that: include: The data acquisition module is configured to obtain relevant engine operating parameters and characteristic marker data corresponding to different engine fault types, wherein the fault types include single-cylinder abnormality, power surge, supercharger abnormality, intercooler abnormality, and rail pressure abnormality; The average value aggregation module is configured to: perform average value aggregation operation on the relevant operating parameters and the characteristic marking data at preset time intervals to obtain fault characteristics and parameters to be marked; The characteristic data determination module is configured to save the fault characteristics and the parameters to be marked in a preset time series database according to the same timestamp for marking to obtain characteristic data.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor executes the program, the steps of the method for acquiring engine characteristic data based on storage with the same timestamp as described in any one of claims 1 to 6 are implemented.
10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the method for acquiring engine characteristic data based on storage with the same timestamp as described in any one of claims 1 to 6 are implemented.