Transient data processing method for long-period data recording

Abstract

In order to solve the problem of long-time storage record under limited storage capacity, the application provides a transient data processing method for long-period data record, which is used for processing armored vehicle electrical system data, and can not only meet the use requirement of long-time record, but also record the change information of vehicle state without influence or within the allowable range of influence, for data analysis in later period. The method takes a standard frame data segment of 8 bytes as a standard storage unit; when storing data, the storage module contains a time field and a data field; the time field occupies a standard storage unit and is used for recording data acquisition time; the data field occupies a standard storage unit and is used for recording the data segment extracted from the CAN standard data frame.

Description

Technical Field

[0001] This invention relates to a data processing method, specifically a method for processing transient data recorded over a long period in an in-vehicle embedded device, belonging to the field of data processing technology. Background Technology

[0002] With the increasing complexity, integration, and intelligence of modern weaponry, in order to meet the demands of modern joint operations and network-centric warfare for sustained, agile, accurate, and economical support capabilities in a more economical and effective manner, there is an urgent need to transform from the current extensive and large-scale maintenance support based on reactive and periodic maintenance to an intensive, precise, and agile maintenance support based on proactive maintenance and condition-based predictive maintenance. This involves utilizing sensors installed throughout the vehicle's information system to collect technical status data of key components. Computers then use artificial intelligence to analyze and reason about this data, predicting the failures and remaining lifespan of key components in real time. This data is then managed and scheduled through a decision support system to generate support plans.

[0003] With the changing landscape of land warfare and the development of information technology, the functions of tank and armored vehicle information systems are becoming increasingly complex, their structures more integrated, and their signal types more diverse. Taking the electrical system—a crucial support system for armored vehicles—as an example, it involves hundreds of parameters, such as short circuits and overcurrents in hundreds of power distribution channels, the acquisition of dozens of analog and switching signals, fault reporting information, diagnostic information, and electrical system faults caused by abnormal vehicle control and various support system failures. During testing and armored vehicle development, it is essential to readily access relevant electrical system parameters. As a large open-loop system in armored vehicles, the electrical system is closely related to all electrical loads, information networks, vehicle control, and logical functions, making fault location far more complex than in other subsystems. The new high-voltage power distribution system involves actuators of various power levels, creating greater complexity and placing higher demands on insulation and other safety aspects. Just as traditional Chinese medicine emphasizes "treating illness before it manifests," future equipment support will inevitably shift from "treating existing illnesses" to "preventing illnesses," focusing on early fault detection, prevention before disease occurs, and mitigation of existing problems.

[0004] The CAN network is the primary channel for acquiring data from tanks and armored vehicles and controlling them. By acquiring data from the CAN bus, data storage technology is used to design storage devices or components within the system to achieve real-time information storage. During maintenance, data export technology is used to export the stored information for processing and analysis. At the component level, embedded processors are typically microcontrollers, DSPs, ARM chips, etc. Data storage is usually accomplished using onboard data storage chips. Depending on the storage format, storage time, and frequency, the embedded processor selects the storage address of the storage chip via the address bus and writes data to the selected storage address via the data bus.

[0005] In the design of convenient embedded data storage for armored vehicle electrical systems, a pair of contradictory problems arise: On the one hand, armored vehicles are used in a wide range of scenarios, and the harsh service environment is not only a major cause of equipment performance degradation and even functional failure, but also results in equipment malfunctions exhibiting great complexity and uncertainty, such as transient and intermittent failures. Especially when the fault mechanism is unclear and there is a lack of sufficient fault samples, it is often necessary to sample and store real vehicle data for a long time in order to analyze the entire performance degradation process, thus requiring long-term data recording and storage. Taking a certain armored vehicle as an example, the data storage time requirement reaches 30 days. On the other hand, due to the limitations of the system's microcontroller's internal storage space and control mode, the maximum recording capacity of the selected onboard Flash is only 512 MPa, thus posing a problem of limited storage capacity.

[0006] By classifying and summarizing the bus information, hundreds of frames contain digital and analog signals, heartbeat data, version information, fault diagnosis information, etc. The signal transmission cycle baud rates are 1s, 500ms, 200ms, and 100ms, with some data, such as starting current, at 20ms. Furthermore, the data format is scattered, and the data length varies from 2 to 8 bytes. Recording according to the CAN bus standard 8-bit format would result in a significant waste of storage space. Without changing the storage format, calculations show that a 512MPa storage capacity can only store all the information data for 24 hours, far from meeting the 30-day storage requirement. Summary of the Invention

[0007] To address the problem of long-term (data storage time requirement of up to 30 days) storage recording under limited storage capacity, this invention proposes a transient data processing method for long-cycle data recording. This method processes the electrical system data of armored vehicles, which not only meets the requirements for long-term recording, but also records changes in vehicle status without affecting or within permissible limits, for later data analysis.

[0008] The transient data processing method for long-period data recording uses an 8-byte standard frame data segment as the standard storage unit.

[0009] When storing data, the storage module includes a time field and a data field;

[0010] The time field occupies one standard storage unit and is used to record the data acquisition time;

[0011] The data field occupies a standard storage unit and is used to record data segments extracted from CAN standard data frames.

[0012] As a preferred embodiment of the present invention: the 8 bytes of the time field, from low to high, are: year low, year high, month, day, hour, minute, second and empty space, and the year low and year high bytes together form the year.

[0013] As a preferred embodiment of the present invention: for data frames that are less than 4 bytes in the data field, a merging process is adopted to merge multiple data frames of the same type to fill an 8-byte standard storage unit.

[0014] As a preferred embodiment of the present invention: in the standard storage unit used to store the data field, Data0 to Data7 are the actual data stored, and the rest are annotation content to help explain the data frame name to which the data stored in the corresponding standard storage unit belongs.

[0015] As a preferred embodiment of the present invention: when the stored data is switch-type data:

[0016] Step 1: The CAN data receiving and processing unit determines whether the current data is the first data at the specified ID position. If it is, it records the first switch status data at the specified ID position and then proceeds to Step 2; otherwise, it proceeds directly to Step 2.

[0017] Step 2: Determine if the switch data at the specified ID position has changed. If it has changed, proceed to Step 3; otherwise, proceed to Step 4.

[0018] Step 3: Set the state change flag to "1", and then invert the state value at the specified ID position;

[0019] Step 4: Determine if 1 second has elapsed. If not, return to Step 1; if so, write a constant string for 1 second.

[0020] As a preferred embodiment of the present invention: when the stored data is analog data:

[0021] In the analog data, for data frames that are less than 4 bytes, a merging process is adopted to combine two data frames into a standard 8-byte storage unit.

[0022] As a preferred embodiment of the present invention: when the stored data is heartbeat data:

[0023] The heartbeat data is an independent data frame, occupying only 1 byte; 8 heartbeat data frames are merged into 8-byte data, occupying a standard 8-byte storage unit.

[0024] As a preferred embodiment of the present invention: when the stored data is working time:

[0025] Four working time data frames are combined into one 8-byte data, which occupies one 8-byte standard storage unit.

[0026] As a preferred embodiment of the present invention: when the stored data is fault code data:

[0027] The fault code data uses an independent storage format and is only briefly recorded when a fault occurs.

[0028] The fault code data includes a time field, a fault code field, and a data field.

[0029] Beneficial effects:

[0030] (1) The data processing method of the present invention is designed for CAN network data. It adopts a compact data format and an optimized storage recording method, which greatly reduces the data storage space. This method has universal properties. Taking an armored vehicle as an example, the optimized data storage capacity is 1 / 20 of the original bus data storage capacity, which can greatly extend the storage time of bus monitoring data in a microcontroller system. It is very suitable for the application of embedded diagnostic systems for small microcontroller systems.

[0031] (2) In the data field, for data frames that are less than 4 bytes, a merging process is adopted to merge multiple data frames of the same type to fill an 8-byte standard storage unit, which can greatly save storage space.

[0032] (3) In the data processing method of the present invention, a time field is set to facilitate real-time playback and retrieval of the collected and stored data.

[0033] (4) In the data processing method of the present invention, the fault code data adopts an independent storage format and is only recorded for a short period of time when the fault occurs, thus saving storage space.

[0034] (5) An armored vehicle adopts an optimized storage and recording method. Based on the recorded data, it can accurately determine the abnormal state of the vehicle's electrical system, trace and locate transient faults, abnormal operations and abnormal system states, and update and optimize the internal diagnostic library of the electrical system, providing a basis for the later maintenance and support of the equipment. Attached Figure Description

[0035] Figure 1 Taking the electrical system data of a certain armored vehicle as an example, here is a sequence diagram of transient data ID records;

[0036] Figure 2 Example of padding with zeros for storage units that do not meet the 8-byte standard;

[0037] Figure 3 This is an example of a data frame that occupies only 2 bytes;

[0038] Figure 4 Store the position order for the time field;

[0039] Figure 5 This is a schematic diagram of the vehicle data storage mode;

[0040] Figure 6 Example of data storage for the electrical system of an armored vehicle;

[0041] Figure 7 Example of switch data for a portion of the electrical system of an armored vehicle;

[0042] Figure 8 This is a data storage method for switch-type data;

[0043] Figure 9 Example of merging heartbeat data frames;

[0044] Figure 10 This is a schematic diagram of the working time arrangement in this embodiment;

[0045] Figure 11 This is a schematic diagram of the fault code data storage format in this embodiment. Detailed Implementation

[0046] The present invention will now be described in detail with reference to the accompanying drawings and preferred embodiments, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more explicit definition of the scope of protection of the present invention.

[0047] This embodiment proposes a transient data processing method for long-cycle data recording, which enables long-cycle recording of transient data in vehicle-mounted embedded devices.

[0048] This embodiment takes a convenient testing and analysis scheme for the electrical system of military armored vehicles as an example. By embedding a convenient data acquisition and storage chip on the vehicle-mounted terminal and reserving a data interface for external access, it facilitates data download and analysis. The hardware configuration uses a 16-bit microcontroller and Flash memory. To address the issue of long-term storage recording under limited storage capacity, the data processing method described in this embodiment is used in the Flash-embedded vehicle device to process transient electrical data from the vehicle. This not only significantly reduces the data storage space, meeting the requirements for long-term recording, but also facilitates the tracing and location of transient faults, abnormal operations, and abnormal system states.

[0049] Table 1 shows the standard CAN data frame format, where the data segment is 64 bits and can contain (0 to 8) bytes of data, output starting from the most significant bit.

[0050] Table 1. Schematic diagram of CAN standard data frame format

[0051]

[0052] To achieve efficient processing of vehicle transient data, only data segments are extracted from CAN standard data frames; the ID of each standard data frame no longer occupies storage space. To clearly identify the ID to which the stored data belongs and avoid confusion during subsequent parsing and analysis of the stored data frames, the recording order of transient data is pre-arranged according to the ID when arranging the data storage rules for each data frame; there is no priority difference between data frames with different IDs. Taking the electrical system data of an armored vehicle as an example, the transient data ID recording order is as follows: Figure 1 As shown.

[0053] The storage strategy uses a standard frame data segment of 64 bits (8 bytes) as a standard storage unit (this standard storage unit is the smallest storage unit). For example, Figure 2 The data frame shown only contains 7 bytes (0-6), which does not meet the minimum standard storage unit capacity of 8 bytes. Therefore, the data frame is still written as 8 bytes, with the 8 empty bits of the 7th byte padded with 0s. For example... Figure 3 The data frame shown only occupies 2 bytes of capacity. By merging 4 similar data frames, a standard 8-byte storage unit can be filled, greatly saving storage space.

[0054] In addition to offline analysis, the collected and stored data also requires real-time playback and retrieval. Therefore, the time point of data collection must be clearly marked. Before storing the data, the collection time must be stored, meaning the storage module consists of two main parts: a time field and a data field. The time field is stored in the order of year, month, day, hour, minute, and second. To facilitate analysis of the underlying data, the stored time uses decimal mode. The time field occupies one standard 8-byte storage unit, and the year occupies 2 bytes. For easy distinction, they are called the "lower byte" and the "higher byte," respectively. For example, when the year is 2023, the lower byte corresponds to 23, and the higher byte corresponds to 20. The time field, from low to high, represents: lower byte, higher byte, month, day, hour, minute, second, and blank space. Figure 4 As shown.

[0055] Based on the above ideas, a high-efficiency embedded vehicle data storage format was designed. The storage format of the data fields in the data storage module is as follows: Figure 5 As shown, in each standard storage unit, Data0 to Data7 are the actual data stored. The rest of the table contains comments explaining the data frame name to which each standard storage unit belongs. For example, the first row indicates the acquisition time (time field); the second row contains data from the power controller data frame; and the third row contains data from the power compartment fire suppression data frame. The order of the data frames is as follows: Figure 1 The prescribed recording order.

[0056] Taking the data storage format of a certain armored vehicle's electrical system as an example, the stored data at 16:55:32 on May 6, 2020 is as follows: Figure 6 As shown, each second of data occupies 64×8 bytes of capacity, the time field occupies 1×8 bytes of capacity, and the data field occupies 63×8 bytes of capacity. The data for 16:55:33 on May 6, 2020 occupies the next 64×8 bytes of capacity, and so on.

[0057] Based on data iteration and experience from engineering projects, a 1-second cycle was determined as the main program cycle for recording vehicle electrical system data. All data types are optimized based on this cycle, including: switch data, analog data, heartbeat data, operating time, and fault code information. The processing methods for these data types are described in detail below.

[0058] (1) Switch data

[0059] The baud rate for transmitting switch data is typically 100ms. A maximum of 10 switch data points can be received per second. Recording all switch data is meaningless and wastes significant storage space, but recording in a 1-second cycle results in the loss of transient data. Therefore, a data processing method targeting the changing trends of switch states was designed. This method records the change value in the next second when the switch state changes relative to the previous second, thus providing accurate feedback on the system state with minimal data overhead. Switch data typically occupies 2 bits, and multiple switch data points are stored in an 8-byte standard storage unit. Taking the switch data of a certain armored vehicle's electrical system as an example, its storage format is as follows: Figure 7 As shown in the diagram, each switch data point is assigned an ID location code based on its storage location, with the ID location representing each switch data point.

[0060] Methods for obtaining CAN information switch data status changes, such as Figure 8 As shown:

[0061] Step 1: The CAN data receiving and processing unit determines whether the current data is the first data at the specified ID position. If it is, it records the first switch status data at the specified ID position and then proceeds to Step 2; otherwise, it proceeds directly to Step 2.

[0062] Step 2: Determine if the switch data at the specified ID position has changed. If it has changed, proceed to Step 3; otherwise, proceed to Step 4.

[0063] Step 3: Set the state change flag to "1", and then invert the state value at the specified ID position;

[0064] Step 4: Determine if the 1-second countdown has ended. If not, return to Step 1; if it has, write a constant string for 1 second. (Note: Ten switch data points will be received within 1 second. Therefore, to fully control the state changes of the switch quantities within 1 second, a 1-second timer function is added to record state changes within 1 second. If less than 1 second has elapsed, repeat the above steps.)

[0065] (2) Analog data

[0066] To meet the need for long-term data trend monitoring, the analog data recording period is set to 1 second to avoid impacting backend data analysis. For analog data frames smaller than 4 bytes, a merging process is used, combining two data frames into a single 8-byte standard storage unit to reduce data storage space.

[0067] (3) Heart rate data

[0068] Heartbeat data is a signal indicating the validity of control box data. Its transmission period varies from 100ms to 1s, but is uniformly recorded as 1s. Heartbeat data is typically an independent data frame, occupying only 1 bit. To reduce storage capacity, eight heartbeat data frames from multiple components can be merged into one 8-byte data unit, thus occupying a standard 8-byte storage unit. (See reference...) Figure 9 Merging method.

[0069] (4) Working hours

[0070] The working time and version information are generally combined into a two-byte data frame, with a transmission period of 1 second, uniformly recorded as 1 second. Four working time data frames are merged into one 8-byte data frame, as shown in the example below. Figure 10 The arrangement shown is as follows.

[0071] (5) Fault code data

[0072] Taking a certain armored vehicle as an example, the entire vehicle consists of 20 nodes. Nodes in the network send data according to a defined protocol, and only send valid data when a fault occurs; otherwise, the data is meaningless. Based on this, the fault code data of the entire vehicle network does not participate in the above-set 1-second cycle recording, but adopts an independent storage format, and is only briefly recorded when a fault occurs.

[0073] Fault code data includes a time field, a fault code field, and a data field; its storage format is as follows: Figure 11 As shown in the figure. The storage order and rules for the time field are the same as those for the time field in the previous data storage.

[0074] Fault Code Field: Receives data reported by all components in SAE J1939 format. The fault code field occupies 6 bytes and includes indicator light, 0, lower 8 bits of SPN, higher 8 bits of SPN, highest 3 bits of SPN and FMI number, and the number of times the SPN conversion mode has occurred.

[0075] Data Fields: Based on the fault code indication, the data fields primarily provide essential parameter data records closely related to the fault code, used to assist in confirming the identification and assessment of the fault indicated by the fault code. To save storage space, when a fault code occurs, to avoid frequent recording of the same fault code and occupying storage space, a strategy is adopted to record only 5 times after each fault code occurs and then stop.

[0076] The above data processing method is designed for CAN network data, significantly reducing data storage space and possessing universal properties. Using this method, the stored data format is clear, and the storage capacity is reduced to 1 / 20th of the original bus data, greatly extending the storage time of bus monitoring data in microcontroller systems. This makes it highly suitable for applications in embedded diagnostic systems for small microcontroller systems.

[0077] In a certain application project, the 108 frames of CAN extended frames of the network are compressed into 49 frames of data using a specified format. Each 16MB capacity can store 8 hours of driving data, achieving 512MB / 16MB = 32 days of data recording. The fault code recording file uses 8MB of space and can record 524,288 faults, which can normally meet the fault recording requirements for the entire life cycle.

[0078] By embedding a convenient data acquisition and storage chip on the armored vehicle's onboard unit, a fast-access recording device for armored vehicle electrical system parameters was built with minimal hardware configuration. Based on the proposed high-efficiency embedded vehicle data storage format, the problem of long-term storage recording under limited storage capacity was cleverly solved. Combined with the subsequent development of universal data parsing files, this perfectly achieved continuous support for the storage, parsing, and problem tracing of data from multiple research and development projects. In the early stages of system development, the mid-stage of system monitoring, and the later stages of system maintenance for armored vehicle electrical systems, the accuracy and reliability of communication data between electrical components and the operational status of each component were determined by recording and analyzing CAN bus data. The device boasts high recording accuracy and efficiency, can be installed in a storage plug, and is quick to disassemble and use, facilitating its use by relevant maintenance personnel. It can provide objective data for armored vehicle electrical performance evaluation, maintenance, and other activities.

[0079] For example, during a routine retrieval of historical fault code data, an overvoltage alarm was detected in the system. The stored driving data was retrieved, and the voltage data at that time point was found to be 27.3V, which did not meet the system protection (battery voltage) after the overvoltage alarm. It was determined to be a false alarm. After modifying the judgment logic of the control box and re-verifying, the fault was eliminated after several iterations. The procedure for handling this issue has been promoted to other projects with false alarm phenomena.

[0080] The processing method for long-term recording of transient data in embedded devices has been applied to the design of multiple models. Based on this, a mature supporting tool has been developed, which can easily realize data download, parsing and report generation. It is the foundation for intelligent health management in the later stage and plays an important supporting role in the research, development and maintenance of vehicle electrical systems and electromechanical systems.

[0081] CAN bus is widely used in automotive electronics, with comprehensive specifications, standards, testing tools, and diagnostic tools. Military armored vehicle chassis propulsion systems and electrical systems commonly use CAN bus. Drawing on advanced experience from the automotive industry, this invention adopts the research approach of reserving an OBD-II data interface for vehicle diagnostics in every vehicle, thereby improving the testing and diagnostic capabilities of armored vehicles. During the research and design phase of armored vehicles, a convenient data acquisition and storage chip is embedded in the vehicle's onboard unit, providing a standard data download interface for easy data download and analysis.

[0082] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A transient data processing method for long-period data recording, characterized in that: A standard 8-byte frame data segment is used as the standard storage unit; When storing data, the storage module includes a time field and a data field; The time field occupies one standard storage unit and is used to record the data acquisition time; The data field occupies one standard storage unit and is used to record data segments extracted from CAN standard data frames; Only data segments are extracted and stored from the CAN standard data frame. The ID of the CAN standard data frame does not occupy storage space. The storage locations of each data frame are arranged according to the pre-set ID record order. In the data field, for data frames that are less than 4 bytes, a merging process is adopted to merge multiple data frames of the same type to fill an 8-byte standard storage unit; When the stored data is switch-type data: Step 1: The CAN data receiving and processing unit determines whether the current data is the first data at the specified ID position. If it is, it records the first switch status data at the specified ID position and then proceeds to Step 2; otherwise, it proceeds directly to Step 2. Step 2: Determine if the switch data at the specified ID position has changed. If it has changed, proceed to Step 3; otherwise, proceed to Step 4. Step 3: Set the state change flag to "1", and then invert the state value at the specified ID position; Step 4: Determine if 1 second has elapsed. If not, return to Step 1; if so, write a constant string for 1 second.

2. The transient data processing method for long-period data recording as described in claim 1, characterized in that: The eight bytes of the time field, from low to high, are: year low, year high, month, day, hour, minute, second, and empty space. The year low and year high bytes together form the year.

3. The transient data processing method for long-period data recording as described in claim 1, characterized in that: In the standard storage unit used to store the data field, Data0 to Data7 are the actual data stored, and the rest are comments to help explain the data frame name to which the data stored in the corresponding standard storage unit belongs.

4. The transient data processing method for long-period data recording as described in any one of claims 1-3, characterized in that: When the stored data is analog data: In the analog data, for data frames that are less than 4 bytes, a merging process is adopted to combine two data frames into a standard 8-byte storage unit.

5. The transient data processing method for long-period data recording as described in any one of claims 1-3, characterized in that: When the stored data is heartbeat data: The heartbeat data is an independent data frame, occupying only 1 byte; 8 heartbeat data frames are merged into 8-byte data, occupying a standard 8-byte storage unit.

6. The transient data processing method for long-period data recording as described in any one of claims 1-3, characterized in that: When the stored data is from working hours: Four working time data frames are combined into one 8-byte data, which occupies a standard 8-byte storage unit.

7. The transient data processing method for long-period data recording as described in any one of claims 1-3, characterized in that: When the stored data is fault code data: The fault code data uses an independent storage format and is only briefly recorded when a fault occurs. The fault code data includes a time field, a fault code field, and a data field.

Images