Embedded device fault diagnosis prediction method and system

Abstract

The invention discloses an embedded equipment fault diagnosis prediction method and system, and belongs to the technical field of equipment fault management. Uniformly coding the embedded equipment to obtain a fault chain relation and a fault frequency; constructing a fault feature scatter point and chain reaction curve, and iteratively adjusting a checking period; predicting a fault cascading interval period based on the curve similarity; and cooperatively calibrating the period range, generating a list and indicating synchronous diagnosis. The system correspondingly comprises a fault data acquisition and coding module, a fault characteristic curve construction module, a fault interval period prediction module and a cooperative calibration and diagnosis module. The method breaks through the limitation of single equipment diagnosis, improves the fault prediction accuracy in combination with the assembly line cooperative relationship, improves the efficiency through synchronous diagnosis planning, is suitable for fault management of various assembly line embedded equipment, and effectively reduces the missed diagnosis rate and the diagnosis cost.

Description

Technical Field

[0001] This invention relates to the field of equipment fault management technology, specifically to an embedded equipment fault diagnosis and prediction method and system. Background Technology

[0002] In automated production lines, embedded devices are the core carriers for achieving precise control and data interaction, such as pick-and-place machines in SMT production lines and robotic arm controllers in automotive welding lines. These devices do not operate independently but are interconnected through industrial buses and Ethernet. A failure in any of these devices can trigger a chain reaction, causing subsequent equipment to stop or even the entire production line to shut down, resulting in significant production losses.

[0003] Existing embedded device fault diagnosis technologies often neglect the interrelationship between devices. Most solutions only perform fault detection on a single device (such as alarm based on sensor thresholds) and fail to identify the chain fault paths of devices in the pipeline. For example, a fault in the pick-and-place machine may cause the reflow oven to have no workpieces to be processed. However, if only the status of the reflow oven itself is monitored, it may be misjudged as "equipment idle" rather than "chain fault", leading to missed diagnosis.

[0004] Furthermore, in most existing technologies, the prediction cycle is fixed, and the fault verification cycle is mostly a fixed value preset by humans (such as inspection every 4 hours). This cannot match the actual fault propagation pattern. If the cycle is too short, it will increase the cost of redundant diagnosis. If the cycle is too long, it may miss the fault warning window. For example, if the chain propagation of a fault in a certain device only takes 2 hours, a fixed 4-hour inspection will cause the fault to be discovered only after it has spread.

[0005] Furthermore, each device executes the diagnostic process independently, and the lack of collaborative planning in diagnosis may lead to resource conflicts caused by "multiple devices needing diagnosis at the same time" or cascading failures caused by "disconnection in diagnosis time between devices". For example, if device A fails 1 hour after diagnosis, it triggers a failure in device B, but device B's diagnosis time is 2 hours later, and it cannot respond in time.

[0006] The aforementioned problems result in a high rate of missed fault diagnosis (typically 15%-20%) and low diagnostic efficiency in existing technologies, making it difficult to meet the fault management needs of continuous production lines. There is an urgent need for a technical solution that combines equipment collaboration, dynamic cycle prediction, and supports synchronous diagnosis. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for fault diagnosis and prediction of embedded devices to solve the problems mentioned in the background art.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] An embedded device fault diagnosis and prediction system includes: a fault data acquisition and encoding module, a fault characteristic curve construction module, a fault interval period prediction module, and a collaborative calibration and diagnosis module. These modules work together to diagnose and predict faults in embedded devices. Specifically, the fault data acquisition and encoding module acquires basic device identifiers and fault-related data; the fault characteristic curve construction module generates curves reflecting fault chain reactions; the fault interval period prediction module calculates the interval period of the device's fault chain reaction; and the collaborative calibration and diagnosis module determines the optimal inspection cycle for the device and plans synchronous diagnosis.

[0010] As a preferred embodiment of the present invention, the fault data acquisition and encoding module includes a unified equipment encoding unit, a cascading fault relationship set acquisition unit, and a fault count unit; the unified equipment encoding unit is used to uniquely identify and encode embedded devices in the production line chain; the cascading fault relationship set acquisition unit is used to extract the equipment cascading fault relationship set under the fault cascading reaction verification cycle based on the collaborative relationship of the devices in the production line; the fault count unit is used to count the number of faults of each embedded device in the cascading fault relationship set within the verification cycle.

[0011] As a preferred embodiment of the present invention, the fault feature curve construction module includes a fault feature scatter point construction unit, a chain reaction curve generation unit, and a periodic iteration adjustment unit. The fault feature scatter point construction unit is used to construct a two-dimensional fault feature coordinate scatter point with equipment identification and fault count as dimensions based on the equipment chain fault relationship set and the number of faults. The chain reaction curve generation unit is used to smoothly connect the two-dimensional fault feature coordinate scatter points in a logical order to generate a fault chain reaction curve function for the embedded device. The periodic iteration adjustment unit is used to gradually increase the scale value of the fault chain reaction verification period to trigger the chain reaction curve generation unit to generate fault chain reaction curve functions under different period scales.

[0012] As a preferred embodiment of the present invention, the fault interval period prediction module includes a fault condition similarity evaluation unit, an optimal period selection unit, and an interval period calculation unit; the fault condition similarity evaluation unit is used to compare the consistency of the fault chain reaction curves under different period scales, evaluate and output the similarity of the fault chain reaction conditions between period scales; the optimal period selection unit is used to select the verification period scale corresponding to the maximum value from all similarity results when the fault chain reaction verification period scale reaches a set upper limit; the interval period calculation unit is used to calculate the difference between the optimal verification period scale and the initial verification period scale as the fault chain reaction interval period of the embedded device.

[0013] As a preferred embodiment of the present invention, the collaborative calibration and diagnosis module includes a periodic collaborative calibration unit, a verification cycle sorting unit, and a synchronous diagnosis indication unit. The periodic collaborative calibration unit is used to perform collaborative calibration on the predicted interval cycle of a single device by combining the maximum and minimum predicted interval cycles of all embedded devices in the pipeline, and to determine the optimal fault cascading reaction verification cycle range for each device. The verification cycle sorting unit is used to arrange the optimal verification cycle ranges of each device in descending order of numerical value, generate a verification cycle list, and feed it back to the operator's port. The synchronous diagnosis indication unit is used to detect the overlapping area of ​​the optimal verification cycle ranges of different devices. If there is an overlap, a synchronous diagnosis instruction is generated to instruct the corresponding device to perform fault diagnosis within the overlapping area.

[0014] An embedded device fault diagnosis and prediction method, the method includes the following steps:

[0015] Step S1: Perform unified coding on embedded devices, and based on the collaborative relationship of devices in the production line, obtain the device chain failure relationship and the number of failures of each device under the fault chain reaction verification cycle;

[0016] Step S2: Based on the obtained equipment chain failure relationship and failure number, construct coordinate scatter points related to failure characteristics and generate a chain failure reaction curve. At the same time, iteratively adjust the chain failure reaction verification cycle to generate chain failure reaction curves under different chain failure reaction verification cycles.

[0017] Step S3: Based on the fault chain reaction curves under different fault chain reaction verification cycles, evaluate the similarity of fault chain reaction status between each fault chain reaction verification cycle, select the fault chain reaction verification cycle corresponding to the maximum similarity, and predict the fault chain reaction interval cycle of the embedded device.

[0018] Step S4: Based on the predicted fault cascading reaction interval, and combined with the synergy of the production line, the optimal fault cascading reaction check cycle range for each embedded device is determined, a check cycle list is generated, and synchronous fault diagnosis is performed on the embedded devices. If there is an overlap in the optimal fault cascading reaction check cycle range, synchronous diagnosis is performed within the overlapping range.

[0019] As a preferred embodiment of the present invention, the specific implementation process of step S1 includes:

[0020] Unified coding is performed on embedded devices. Based on the collaborative relationships of embedded devices in a pipelined production chain, a set of device cascading failure relationships under the fault cascading reaction verification cycle is obtained, denoted as... In the formula, This indicates the embedded device under the fault cascading response verification cycle scale k. The set of all embedded devices that cause a chain reaction of failure when a fault occurs, where i, x, and I all represent the coded serial numbers of the embedded devices, and i ≠ x and i ≠ I;

[0021] Under the fault cascading reaction verification cycle scale k, obtain the set of equipment cascading fault relationships. The number of failures for each embedded device is recorded to form a set of failure counts, denoted as . ,in, Indicates embedded device The number of failures;

[0022] It should be noted that by using unified coding to achieve unique identification of equipment, and combining the collaborative relationship of the production line to extract the correspondence between "fault-interlocking equipment" and the number of faults, the problem of "isolated data of single equipment" is solved.

[0023] As a preferred embodiment of the present invention, the specific implementation process of step S2 includes:

[0024] Under the fault cascading reaction verification cycle scale k, based on the equipment cascading fault relationship set and failure count set Construct a two-dimensional fault feature coordinate scatter point, denoted as . The fault characteristic coordinates are then smoothly connected sequentially to obtain the embedded device. The fault chain reaction curve function is denoted as ;

[0025] Iterative adjustment of the fault chain reaction verification cycle scale value ,and This yields the results on the fault cascading reaction verification cycle scale. Embedded devices Fault chain reaction curve function ;

[0026] It should be noted that by converting "equipment-number of failures" into two-dimensional scatter points and generating curves, and by iteratively adjusting the verification cycle, multiple sets of curves are generated. The changes in the shape of the curves reflect the stability of the failure chain under different cycles. For example, a high degree of curve overlap indicates that the cycle has little impact on the failure pattern, providing a visual basis for cycle optimization.

[0027] As a preferred embodiment of the present invention, the specific implementation process of step S3 includes:

[0028] Based on fault chain reaction curve function and fault chain reaction curve function Predicting embedded devices The interval between failure chain reactions:

[0029] Assessing the cascading effects of failures and checking the cycle scale k and Similarity of fault chain reaction conditions between ;

[0030] Change the scale value of the fault chain reaction verification cycle The size of the scale value of the inspection cycle will generate new chain reactions of failures. ,and The fault chain reaction curve function is obtained. And continue to assess the similarity of the failure chain reaction conditions. When the cycle scale value for checking fault chain reactions reaches its upper limit, the similarity scores of all assessed fault chain reaction conditions are retrieved to form a similarity set, denoted as . Where R represents the number of iterations when the fault chain reaction verification cycle scale value reaches the upper limit;

[0031] Select the fault chain reaction verification cycle scale value when the similarity of the fault chain reaction conditions is the highest. To obtain the predicted embedded device Fault chain reaction interval period ;

[0032] It should be noted that by evaluating the similarity of different period curves, the period that best reflects the stable law of fault chain reaction (with the highest similarity) is selected. The difference between the "optimal period and the initial period" is used as the interval period to avoid the blindness of fixed periods and improve the accuracy of prediction.

[0033] As a preferred embodiment of the present invention, the specific implementation process of step S4 includes:

[0034] Based on the predicted fault cascading reaction interval, the predicted embedded devices in the assembly line production chain are... The fault cascading reaction interval is periodically calibrated collaboratively to obtain the embedded device. Optimal fault chain reaction verification cycle range In the formula, This represents the maximum predicted fault cascading response interval for each embedded device. This represents the minimum predicted fault cascading reaction interval for each embedded device.

[0035] The optimal fault cascading response check cycle ranges for each embedded device are arranged in descending order, generating a check cycle list and feeding it back to the staff interface to instruct for synchronous fault diagnosis of the embedded devices. If the embedded device... Optimal fault chain reaction verification cycle range With embedded devices Optimal fault chain reaction verification cycle range If overlap exists, then within the optimal fault cascading response check period for the overlapping device, the embedded device should be checked. With embedded devices Perform synchronous fault diagnosis;

[0036] It should be noted that, based on the overall collaborative calibration of the production line, the optimal inspection cycle range of each device is calibrated. By sorting and judging the range, the timing of simultaneous diagnosis of multiple devices is planned, which solves the problems of "diagnosis resource conflict" and "chain failure omission", forming a closed loop from prediction to diagnosis.

[0037] Compared with existing technologies, the beneficial effects achieved by this invention are: unified encoding of embedded devices to obtain fault cascading relationships and fault counts; construction of fault feature scatter plots and chain reaction curves, iterative adjustment of the verification cycle; prediction of fault cascading interval cycles based on curve similarity; collaborative calibration of cycle ranges, generation of lists, and indication of synchronous diagnosis. The system includes modules for fault data acquisition and encoding, fault feature curve construction, fault interval cycle prediction, and collaborative calibration and diagnosis. This invention overcomes the limitations of single-device diagnosis, improves fault prediction accuracy by combining pipeline collaborative relationships, and enhances efficiency through synchronous diagnostic planning. It is applicable to fault management of various pipeline embedded devices, effectively reducing the missed diagnosis rate and diagnostic costs. Attached Figure Description

[0038] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0039] Figure 1 This is a schematic diagram illustrating the steps of an embedded device fault diagnosis and prediction method according to the present invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] In this first embodiment, an embedded device fault diagnosis and prediction system is provided. The system includes: a fault data acquisition and encoding module, a fault characteristic curve construction module, a fault interval period prediction module, and a collaborative calibration and diagnosis module. The modules work together to realize the diagnosis and prediction of embedded device faults.

[0042] The fault data acquisition and coding module is used to obtain basic equipment identification and fault-related data;

[0043] The fault data acquisition and coding module includes a unified equipment coding unit, a cascading fault relationship set acquisition unit, and a fault count unit.

[0044] The unified coding unit for equipment is used to uniquely identify embedded devices in the production line chain.

[0045] The chain failure relationship set acquisition unit is used to extract the equipment chain failure relationship set under the failure chain reaction verification cycle based on the cooperative relationship of equipment in the production line.

[0046] The fault count unit is used to count the number of faults of each embedded device in the cascading fault relationship set within the inspection period.

[0047] The fault characteristic curve construction module is used to generate curves that reflect the chain reaction of faults;

[0048] The fault characteristic curve construction module includes a fault characteristic scatter plot construction unit, a chain reaction curve generation unit, and a periodic iteration adjustment unit.

[0049] The fault feature scatter point construction unit is used to construct two-dimensional fault feature coordinate scatter points with equipment identification and fault count as dimensions, based on the set of equipment cascading fault relationships and the number of faults.

[0050] The chain reaction curve generation unit is used to smoothly connect the scattered points of two-dimensional fault feature coordinates in a logical order to generate the fault chain reaction curve function of the embedded device.

[0051] The periodic iterative adjustment unit is used to gradually increase the scale value of the fault chain reaction verification period, triggering the chain reaction curve generation unit to generate fault chain reaction curve functions under different period scales.

[0052] The fault interval prediction module is used to calculate the interval period of equipment fault chain reaction;

[0053] The fault interval cycle prediction module includes a fault condition similarity evaluation unit, an optimal cycle screening unit, and an interval cycle calculation unit.

[0054] The fault condition similarity assessment unit is used to compare the consistency of fault chain reaction curves under different periodic scales, and to evaluate and output the similarity of fault chain reaction conditions between periodic scales.

[0055] The optimal cycle screening unit is used to select the verification cycle scale corresponding to the maximum value from all similarity results when the fault chain reaction verification cycle scale reaches the set upper limit.

[0056] The interval cycle calculation unit is used to calculate the difference between the optimal verification cycle scale and the initial verification cycle scale, which serves as the fault cascading response interval cycle for the embedded device.

[0057] The collaborative calibration and diagnostics module is used to determine the optimal equipment verification cycle and plan synchronous diagnostics;

[0058] The collaborative calibration and diagnosis module includes a periodic collaborative calibration unit, a verification period sorting unit, and a synchronous diagnosis indication unit.

[0059] The periodic co-calibration unit is used to combine the maximum and minimum predicted interval periods of all embedded devices in the pipeline to perform co-calibration of the predicted interval period of a single device, and determine the optimal fault chain reaction verification period range for each device.

[0060] The verification cycle sorting unit is used to sort the optimal verification cycle range of each device in descending order of numerical value, generate a verification cycle list and feed it back to the staff operation port;

[0061] The synchronous diagnostic instruction unit is used to detect the overlapping area of ​​the optimal inspection cycle range of different devices. If there is an overlap, a synchronous diagnostic instruction is generated to instruct the corresponding device to perform fault diagnosis within the overlapping range.

[0062] Please see Figure 1 In this second embodiment, an embedded device fault diagnosis and prediction method is provided to be applicable to the first embodiment. In this embodiment, an automotive parts assembly line is used as the application scenario. The assembly line includes 6 core embedded devices: a material conveying robot (device 1), a bolt tightening machine (device 2), a vision inspection module (device 3), a glue applicator (device 4), a press fitting machine (device 5), and a finished product sorting machine (device 6). The devices coordinate through the Profinet bus. On average, a single device failure will cause 2-3 subsequent devices to stop in a chain reaction. The daily production is 1200 parts.

[0063] The method includes the following steps:

[0064] Step S1: Perform unified coding on embedded devices, and based on the collaborative relationship of devices in the production line, obtain the device chain failure relationship and the number of failures of each device under the fault chain reaction verification cycle;

[0065] For example, embedded devices are uniformly coded, and based on the collaborative relationships of embedded devices in a production line, a set of device cascading failure relationships under the fault cascading reaction verification cycle is obtained, denoted as... In the formula, This indicates the embedded device under the fault cascading response verification cycle scale k. The set of all embedded devices that cause a chain reaction of failure when a fault occurs, where i, x, and I all represent the coded serial numbers of the embedded devices, and i ≠ x and i ≠ I;

[0066] Under the fault cascading reaction verification cycle scale k, obtain the set of equipment cascading fault relationships. The number of failures for each embedded device is recorded to form a set of failure counts, denoted as . ,in, Indicates embedded device The number of failures;

[0067] For example, using the rule of "production line number - function type - equipment serial number", the codes are AQ-TR-01 (equipment 1), AQ-TG-02 (equipment 2), AQ-JC-03 (equipment 3), AQ-TJ-04 (equipment 4), AQ-YZ-05 (equipment 5), and AQ-FJ-06 (equipment 6).

[0068] Based on the assembly cycle time (3 minutes per piece), the initial cycle k = 1.5 hours (checked once every 1.5 hours);

[0069] By retrieving 1.5 hours of fault records from the MES system of the production line, when equipment 2 (tightening machine) malfunctions, equipment 3 (visual inspection) stops because there is no workpiece to be inspected, and equipment 4 (glue applicator) has no workpiece to be glued. Therefore, the chain relationship set of equipment 2 is {equipment 3, equipment 4}.

[0070] Equipment 2 malfunctioned once within 1.5 hours, and equipment 3 and 4 each stopped once due to interlocking. Record the number of malfunctions.

[0071] Step S2: Based on the obtained equipment chain failure relationship and failure number, construct coordinate scatter points related to failure characteristics and generate a chain failure reaction curve. At the same time, iteratively adjust the chain failure reaction verification cycle to generate chain failure reaction curves under different chain failure reaction verification cycles.

[0072] For example, under the fault cascading failure verification cycle scale k, based on the equipment cascading failure relationship set and failure count set Construct a two-dimensional fault feature coordinate scatter point, denoted as . The fault characteristic coordinates are then smoothly connected sequentially to obtain the embedded device. The fault chain reaction curve function is denoted as ;

[0073] Iterative adjustment of the fault chain reaction verification cycle scale value ,and This yields the results on the fault cascading reaction verification cycle scale. Embedded devices Fault chain reaction curve function ;

[0074] For example, the horizontal axis represents the equipment code (AQ-TG-02, AQ-JC-03, AQ-TJ-04), and the vertical axis represents the number of failures (1, 1, 1). Connecting the scatter points generates the failure cascading curve for equipment 2 at k=1.5 hours.

[0075] Adjust the period k' = 3 hours (> 1.5 hours) to generate a curve (equipment 2 fails once, equipment 3 and 4 each fail once in a chain, the curve shape is consistent with k = 1.5 hours); continue to adjust k'' = 4.5 hours, due to the excessive period, equipment 3 experiences one independent failure (lens contamination), and the curve shape changes.

[0076] Step S3: Based on the fault chain reaction curves under different fault chain reaction verification cycles, evaluate the similarity of fault chain reaction status between each fault chain reaction verification cycle, select the fault chain reaction verification cycle corresponding to the maximum similarity, and predict the fault chain reaction interval cycle of the embedded device.

[0077] For example, based on the fault cascading reaction curve function and fault chain reaction curve function Predicting embedded devices The interval between failure chain reactions:

[0078] Assessing the cascading effects of failures and checking the cycle scale k and Similarity of fault chain reaction conditions between ;

[0079] Change the scale value of the fault chain reaction verification cycle The size of the scale value of the inspection cycle will generate new chain reactions of failures. ,and The fault chain reaction curve function is obtained. And continue to assess the similarity of the failure chain reaction conditions. When the cycle scale value for checking fault chain reactions reaches its upper limit, the similarity scores of all assessed fault chain reaction conditions are retrieved to form a similarity set, denoted as . Where R represents the number of iterations when the fault chain reaction verification cycle scale value reaches the upper limit;

[0080] Select the fault chain reaction verification cycle scale value when the similarity of the fault chain reaction conditions is the highest. To obtain the predicted embedded device Fault chain reaction interval period ;

[0081] For example, the curves for k=1.5 and k'=3 hours show the same pattern (no change in interlocking equipment or number of cycles), with a similarity of 95%; the curves for k=1.5 and k''=4.5 hours show a similarity of 72% due to the independent failure of equipment 3.

[0082] The maximum period is set to 6 hours (k'''=6 hours, similarity 68%), and the period corresponding to the maximum similarity is k'=3 hours.

[0083] t = 3 - 1.5 = 1.5 hours, meaning the fault interlocking interval of device 2 is 1.5 hours.

[0084] Step S4: Based on the predicted fault cascading reaction interval, and combined with the synergy of the production line, the optimal fault cascading reaction check cycle range for each embedded device is determined, a check cycle list is generated, and synchronous fault diagnosis is performed on the embedded devices. If there is an overlap in the optimal fault cascading reaction check cycle range, synchronous diagnosis is performed within the overlapping range.

[0085] For example, based on the predicted fault cascading reaction interval, the predicted embedded devices in the assembly line production chain... The fault cascading reaction interval is periodically calibrated collaboratively to obtain the embedded device. Optimal fault chain reaction verification cycle range In the formula, This represents the maximum predicted fault cascading response interval for each embedded device. This represents the minimum predicted fault cascading reaction interval for each embedded device.

[0086] The optimal fault cascading response check cycle ranges for each embedded device are arranged in descending order, generating a check cycle list and feeding it back to the staff interface to instruct for synchronous fault diagnosis of the embedded devices. If the embedded device... Optimal fault chain reaction verification cycle range With embedded devices Optimal fault chain reaction verification cycle range If overlap exists, then within the optimal fault cascading response check period for the overlapping device, the embedded device should be checked. With embedded devices Perform synchronous fault diagnosis;

[0087] For example, the predicted intervals for 6 devices are: Device 1 (1 hour), Device 2 (1.5 hours), Device 3 (2 hours), Device 4 (1.2 hours), Device 5 (1.8 hours), and Device 6 (2.2 hours); the maximum value is 2.2 hours, and the minimum value is 1 hour; the optimal range after calibration for Device 2 is: lower limit = 1.5 - (1.5 - 1) / (2.2 - 1) = 1.1 hours, upper limit = 1.5 + (2.2 - 1.5) / (2.2 - 1) = 2.0 hours, i.e., [1.1, 2.0] hours;

[0088] Optimal sorting range: Device 6 (1.8-2.2), Device 3 (1.6-2.4), Device 5 (1.4-2.2), Device 2 (1.1-2.0), Device 4 (1.0-1.4), Device 1 (0.8-1.2); Overlapping range [1.8, 2.0] hours (Device 2, 3, 5, 6) indicates synchronous diagnosis during this period; [1.4, 1.8] hours (Device 2, 3, 5) synchronous diagnosis improves efficiency.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0090] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for fault diagnosis and prediction in embedded devices, characterized in that, The method includes the following steps: Step S1: Perform unified coding on embedded devices, and based on the collaborative relationship of devices in the production line, obtain the device chain failure relationship and the number of failures of each device under the fault chain reaction verification cycle; Step S2: Based on the obtained equipment chain failure relationship and failure number, construct coordinate scatter points related to failure characteristics and generate a chain failure reaction curve. At the same time, iteratively adjust the chain failure reaction verification cycle to generate chain failure reaction curves under different chain failure reaction verification cycles. Step S3: Based on the fault chain reaction curves under different fault chain reaction verification cycles, evaluate the similarity of fault chain reaction status between each fault chain reaction verification cycle, select the fault chain reaction verification cycle corresponding to the maximum similarity, and predict the fault chain reaction interval cycle of the embedded device. Step S4: Based on the predicted fault cascading reaction interval, and combined with the synergy of the production line, the optimal fault cascading reaction check cycle range for each embedded device is determined, a check cycle list is generated, and synchronous fault diagnosis is performed on the embedded devices. If there is an overlap in the optimal fault cascading reaction check cycle range, synchronous diagnosis is performed within the overlapping range.

2. The embedded device fault diagnosis and prediction method according to claim 1, characterized in that, The specific implementation process of step S1 includes: Unified coding is performed on embedded devices. Based on the collaborative relationships of embedded devices in a pipelined production chain, a set of device cascading failure relationships under the fault cascading reaction verification cycle is obtained, denoted as... In the formula, This indicates the embedded device under the fault cascading response verification cycle scale k. The set of all embedded devices that cause a chain reaction of failure when a fault occurs, where i, x, and I all represent the coded serial numbers of the embedded devices, and i ≠ x and i ≠ I; Under the fault cascading reaction verification cycle scale k, obtain the set of equipment cascading fault relationships. The number of failures for each embedded device is recorded to form a set of failure counts, denoted as . ,in, Indicates embedded device The number of failures.

3. The embedded device fault diagnosis and prediction method according to claim 2, characterized in that, The specific implementation process of step S2 includes: Under the fault cascading reaction verification cycle scale k, based on the equipment cascading fault relationship set and failure count set Construct a two-dimensional fault feature coordinate scatter point, denoted as . The fault characteristic coordinates are then smoothly connected sequentially to obtain the embedded device. The fault chain reaction curve function is denoted as ; Iterative adjustment of the fault chain reaction verification cycle scale value ,and This yields the results on the fault cascading reaction verification cycle scale. Embedded devices Fault chain reaction curve function .

4. The embedded device fault diagnosis and prediction method according to claim 3, characterized in that, The specific implementation process of step S3 includes: Based on fault chain reaction curve function and fault chain reaction curve function Predicting embedded devices The interval between failure chain reactions: Assessing the cascading effects of failures and checking the cycle scale k and Similarity of fault chain reaction conditions between ; Change the criterion value of the fault chain reaction verification cycle The size of the scale value of the inspection cycle will generate new chain reactions of failures. ,and The fault chain reaction curve function is obtained. And continue to assess the similarity of the failure chain reaction conditions. When the cycle scale value for checking fault chain reactions reaches its upper limit, the similarity scores of all assessed fault chain reaction conditions are retrieved to form a similarity set, denoted as . Where R represents the number of iterations when the fault chain reaction verification cycle scale value reaches the upper limit; Select the fault chain reaction verification cycle scale value when the similarity of the fault chain reaction conditions is the highest. To obtain the predicted embedded device Fault chain reaction interval period .

5. The embedded device fault diagnosis and prediction method according to claim 4, characterized in that, The specific implementation process of step S4 includes: Based on the predicted fault cascading reaction interval, the predicted embedded devices in the assembly line production chain are... The fault cascading reaction interval is periodically calibrated collaboratively to obtain the embedded device. Optimal fault chain reaction verification cycle range In the formula, This represents the maximum predicted fault cascading response interval for each embedded device. This represents the minimum predicted fault cascading reaction interval for each embedded device. The optimal fault cascading response check cycle ranges for each embedded device are arranged in descending order, generating a check cycle list and feeding it back to the staff interface to instruct for synchronous fault diagnosis of the embedded devices. If the embedded device... Optimal fault chain reaction verification cycle range With embedded devices Optimal fault chain reaction verification cycle range If overlap exists, then within the optimal fault cascading response check period for the overlapping device, the embedded device should be checked. With embedded devices Perform synchronous fault diagnosis.

6. An embedded device fault diagnosis and prediction system, executing the embedded device fault diagnosis and prediction method as described in any one of claims 1-5, characterized in that, The system includes: a fault data acquisition and encoding module, a fault characteristic curve construction module, a fault interval cycle prediction module, and a collaborative calibration and diagnosis module. These modules work together to diagnose and predict faults in embedded devices. Specifically, the fault data acquisition and encoding module acquires basic device identifiers and fault-related data; the fault characteristic curve construction module generates curves reflecting fault chain reactions; the fault interval cycle prediction module calculates the interval cycle of device fault chain reactions; and the collaborative calibration and diagnosis module determines the optimal inspection cycle for the device and plans synchronous diagnosis.

7. The embedded device fault diagnosis and prediction system according to claim 6, characterized in that, The fault data acquisition and encoding module includes a unified equipment encoding unit, a chain fault relationship set acquisition unit, and a fault count unit; the unified equipment encoding unit is used to uniquely identify and encode embedded devices in the production line chain. The chain failure relationship set acquisition unit is used to extract the equipment chain failure relationship set under the failure chain reaction verification cycle based on the cooperative relationship of equipment in the production line. The fault count unit is used to count the number of faults of each embedded device in the cascading fault relationship set within the inspection period.

8. The embedded device fault diagnosis and prediction system according to claim 6, characterized in that, The fault feature curve construction module includes a fault feature scatter plot construction unit, a chain reaction curve generation unit, and a periodic iteration adjustment unit. The fault feature scatter point construction unit is used to construct two-dimensional fault feature coordinate scatter points with equipment identification and fault count as dimensions, based on the set of equipment cascading fault relationships and the number of faults. The chain reaction curve generation unit is used to smoothly connect the scattered points of two-dimensional fault feature coordinates in a logical order to generate the fault chain reaction curve function of the embedded device. The periodic iterative adjustment unit is used to gradually increase the scale value of the fault chain reaction verification period, triggering the chain reaction curve generation unit to generate fault chain reaction curve functions under different period scales.

9. The embedded device fault diagnosis and prediction system according to claim 6, characterized in that, The fault interval cycle prediction module includes a fault condition similarity evaluation unit, an optimal cycle screening unit, and an interval cycle calculation unit. The fault condition similarity evaluation unit is used to compare the consistency of the fault chain reaction curves under different cycle scales, evaluate and output the similarity of the fault chain reaction conditions between cycle scales. The optimal cycle screening unit is used to screen out the verification cycle scale corresponding to the maximum value from all similarity results when the fault chain reaction verification cycle scale reaches the set upper limit. The interval period calculation unit is used to calculate the difference between the optimal verification period scale and the initial verification period scale, which serves as the fault cascading response interval period for the embedded device.

10. The embedded device fault diagnosis and prediction system according to claim 6, characterized in that, The collaborative calibration and diagnosis module includes a periodic collaborative calibration unit, a verification period sorting unit, and a synchronous diagnosis indication unit. The periodic collaborative calibration unit is used to combine the maximum and minimum predicted interval periods of all embedded devices in the pipeline to perform collaborative calibration on the predicted interval period of a single device, and determine the optimal fault chain reaction verification period range for each device; the verification period sorting unit is used to sort the optimal verification period range of each device in descending order of numerical value, generate a verification period list and feed it back to the operator's operation port. The synchronous diagnostic instruction unit is used to detect the overlapping area of ​​the optimal inspection cycle range of different devices. If there is an overlap, a synchronous diagnostic instruction is generated to instruct the corresponding device to perform fault diagnosis within the overlapping range.

Images