System maintenance strategy design method based on data flow fault tree

Abstract

The invention belongs to the technical field of maintenance, reliability, testability and maintainability of avionics, and discloses a system maintenance strategy design method based on a data stream fault tree. Comprising the following steps: step 1, designing a data stream according to a system; 2, identifying public resources and special resources of the system; 3, generating a fault tree based on the data flow, the public resources and the special resources; 4, generating a fault correlation matrix based on the fault tree; and 5, generating a fault diagnosis sequence table based on the fault correlation matrix.

Description

Technical Field

[0001] The invention belongs to the technical field of avionics, and in particular relates to a system maintenance strategy design method based on a data flow fault tree. Background Art

[0002] With the increasing integration of aviation products, the hardware and software components of avionics systems and their failure mechanisms are becoming more complex. Large-scale, multi-level, and strongly coupled fault information poses a huge challenge to improving maintenance efficiency and ensuring maintenance quality. The current fault detection data of airborne electronic products is complex, but the intrinsic correlation between data at different object levels and different faults is not explicit, and fault isolation and positioning are difficult. They mainly rely on the experience of maintenance personnel, and the fault troubleshooting process is not standardized. Fault tree analysis is one of the tools for reliability analysis. The generated fault tree clearly represents the relationship between fault events, fault causes, and cause combinations. Fault isolation based on the fault tree can optimize the maintenance process and ensure maintenance quality. The correlation of fault information can improve the rationality of the fault isolation and positioning troubleshooting process and improve maintenance efficiency. Summary of the invention

[0003] The purpose of the present invention is to optimize the difficulty of isolating and locating faults in avionics systems and the problem of irregular troubleshooting process, and to improve the efficiency of using fault record data in the process of fault analysis and isolation and locating. The present invention proposes a method for designing avionics product maintenance strategy based on a fault tree.

[0004] The technical solution of the present invention:

[0005] A system maintenance strategy design method based on data flow fault tree includes the following steps:

[0006] Step 1: Design data flow according to the system;

[0007] Step 2: Identify the system's public and private resources;

[0008] Step 3: Generate a fault tree based on data flow and public resources and dedicated resources;

[0009] Step 4: Generate a fault correlation matrix based on the fault tree;

[0010] Step 5: Generate a fault diagnosis sequence table based on the fault correlation matrix.

[0011] Furthermore, in step 1, each function of the system corresponds to a data stream;

[0012] Specifically including: data processing data flow, data storage data flow, video compression processing data flow, control sampling processing data flow;

[0013] Each data flow consists of: an input part, a processing part, and an output part.

[0014] The data flow drawn according to this method can lay the foundation for the subsequent standardized and quick drawing of the fault tree.

[0015] Furthermore, the data flow design process is as follows: find the input interface, processing node, and output interface corresponding to the function on the system hardware topology diagram;

[0016] Identify support nodes and their support relationships with input interfaces, processing nodes, and output interfaces. Support nodes include power supply nodes, clock nodes, storage nodes, state control nodes, and timing configuration nodes.

[0017] Different functions use different forms of lines to connect input interfaces, processing nodes, and output interfaces according to topological paths to form different data flows;

[0018] Identify common resources and supporting nodes at the intersection of different data flows;

[0019] Identify dedicated resources and supporting nodes of dedicated resources in different data flows;

[0020] Public resources and their corresponding supporting node external resources are dedicated resources.

[0021] Identifying common resources and their corresponding supporting nodes can lay the foundation for repeated application of the same content when drawing fault trees in the future.

[0022] Furthermore, each data flow corresponds to a functional fault tree, and the first-level nodes of each fault tree include: common resource failure modes and dedicated resource failure modes in the data flow;

[0023] The second level nodes of each fault tree include: supporting node failure modes of public resources and supporting node failure modes of dedicated resources, and resource failure modes with supporting relationships are connected with their supporting node failure modes according to causal relationships;

[0024] There is no upper limit on the number of levels of each fault tree. If the supporting node of the i-th level still has supporting nodes, the supporting node of the supporting node is regarded as the i+1th level, i≥2;

[0025] When a resource has multiple supporting nodes, the resource failure mode and the supporting node failure mode are connected through logic gates according to the causal relationship.

[0026] The fault tree drawn by this method is directly mapped to the data flow, and the drawing is simple and intuitive.

[0027] Furthermore, the top event of each fault tree is the failure mode of the function;

[0028] The top event is connected to all first-level events through logic gates according to causal relationships.

[0029] Furthermore, causation is the use of logic gate symbols to describe the cause-effect relationship between various events in the system.

[0030] Furthermore, each row of the fault correlation matrix corresponds to a top event of the fault tree;

[0031] Each column of the fault correlation matrix corresponds to a node that has no connection relationship with the next level. Duplicate nodes need to be merged and cannot appear repeatedly.

[0032] Furthermore, the data in the i-th row and j-th column of the fault correlation matrix is ​​0 or an integer not less than 1, representing: whether there is a causal relationship between the failure mode of the i-th function and the failure mode of the j-th node;

[0033] 0 means irrelevant, other values ​​mean relevant, and the larger the value, the greater the correlation;

[0034] i takes values ​​from 1 to m, where m is the number of failure modes of the system function;

[0035] j ranges from 1 to n, where n is the number of non-repeated nodes in the fault tree that have no connection to the next level;

[0036] Further, the process of generating the fault diagnosis sequence table is as follows:

[0037] Match the top event of each function with the fault phenomenon that occurred;

[0038] Divide the matched top event corresponding rows and the unmatched top event corresponding rows into a matched matrix and an unmatched matrix;

[0039] The matching matrix is ​​summed column by column;

[0040] The unmatched matrix is ​​summed column by column;

[0041] Multiply the sum of the j-th column of the matching matrix by the sum of the j-th column of the unmatched matrix;

[0042] Eliminate the data that is non-zero after multiplication in the matching matrix summation result and the data that is zero in the matching matrix summation result, and sort the remaining data from large to small;

[0043] And perform fault detection on the corresponding nodes according to the sorting results.

[0044] Beneficial effects of the present invention:

[0045] The present invention provides a system maintenance strategy design method based on a data flow fault tree for avionics system maintenance. By drawing functional data flows, identifying corresponding contracted physical nodes, identifying public resources, dedicated resources and supporting nodes based on data flow paths, and compiling fault trees, it is a standardized engineering method that is easy to implement, and the relationship between the upper and lower levels of the fault tree is intuitive and clear. Public resources and their supporting nodes will be used by multiple fault trees, and can be reused after being drawn once, which shortens the time for drawing fault trees. A fault correlation matrix is ​​generated based on the fault tree. When a fault occurs, the fault correlation matrix is ​​divided into a matching matrix and an unmatched matrix by matching the fault phenomenon. After the fault source is reversely eliminated by the unmatched fault, the scope of fault source investigation is further narrowed, and then the fault source is investigated in order from high to low according to the correlation value. The method can effectively improve the maintenance quality and efficiency in the current highly integrated avionics system, solve the problems of the standardization of the fault isolation method and the insufficient correlation between the fault information and the fault source, and can be widely used in the field of avionics, with strong practical and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Hardware topology diagram of the video data processing module.

[0047] Figure 2 Data flow diagram of the video data processing module.

[0048] Figure 3 Fault tree for loss of FC bus data processing function.

[0049] Figure 4 Fault tree for loss of Ethernet data storage capability.

[0050] Figure 5 Fault tree for loss of compression in video data processing.

[0051] Figure 6 Schematic diagram of the fault correlation matrix of the video data processing module.

[0052] Figure 7 Schematic diagram of the fault correlation matched matrix and the unmatched matrix.

[0053] Figure 8 Schematic diagram of fault correlation matching matrix summation.

[0054] Fig. 9 Schematic diagram of fault correlation matching matrix and mismatching matrix operations.

[0055] Fig.10 Diagram of fault repair diagnosis sequence. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Taking a video data processing module as an example, its functions are as follows:

[0058] FC bus data processing function;

[0059] Ethernet data storage function;

[0060] Video data processing and compression function;

[0061] The hardware topology of this module is as follows Figure 1 As shown, the corresponding nodes include:

[0062]

[0063] The data flow diagram formed according to the function is as follows Figure 2 shown.

[0064] FC bus data processing data flow:

[0065] FC interface input interface → CPU processing node → Ethernet interface output interface. Ethernet data storage data flow:

[0066] Ethernet interface input interface → CPU processing node → SSD disk output interface.

[0067] Video data processing compressed data stream:

[0068] ARINC818 interface - input interface → FPGA processing node → CPU processing node → FPGA processing node → PCIE interface output interface.

[0069] Public resources: CPU, Ethernet interface

[0070] Dedicated resources: FC interface, ARINC818 interface, SSD disk, FPGA, PCIE interface

[0071] Support nodes: DC / DC1, DC / DC2, clock 1, clock 2, CPLD, FLASH storage node, DDR3, storage node 1, DDR3 storage node 2, SSD disk storage node.

[0072] According to the data flow, draw the FC bus data processing function loss fault tree, Ethernet data storage function loss fault tree, video data processing compression function loss fault tree, such as Figure 3-Figure 5 shown.

[0073] According to the three fault trees, the fault correlation matrix is ​​as follows: Figure 6 shown.

[0074] If the FC bus data processing function loss and Ethernet data storage function loss fault occur, but the video data processing compression function loss fault does not occur, the fault correlation matrix is ​​divided into a matching matrix and an unmatched matrix. The matching matrix consists of the FC bus data processing function loss and Ethernet data storage function loss data rows, and the matching matrix consists of the video data processing compression function loss data rows. Figure 7 shown.

[0075] Sum the matching matrix column by column and the unmatched matrix column by column. The result is as follows Figure 8 shown.

[0076] Multiply the sum of the matched matrix and the sum of the unmatched matrix;

[0077] Eliminate the non-zero data in the multiplication result and the zero data in the matching matrix summation result. Fig. 9 shown.

[0078] The remaining nodes are sorted according to the numerical order of the matching matrix summation results for fault detection, such as Fig.10 shown.

[0079] The order is:

[0080] The Ethernet interface loses functionality;

[0081] The FC interface loses its function;

[0082] The SSD storage node function is lost.

[0083] The above is only a specific embodiment of the present invention, and the present invention is described in detail. The unexplained part is a conventional technology. However, the protection scope of the present invention is not limited to this. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A system maintenance strategy design method based on data flow fault tree, characterized by: The following steps are involved: Step 1: Design data flow according to the system; Step 2: Identify the system's public and private resources; Step 3: Generate a fault tree based on data flow and public resources and dedicated resources; Step 4: Generate a fault correlation matrix based on the fault tree; Step 5: Generate a fault diagnosis sequence table based on the fault correlation matrix.

2. The method according to claim 1, characterized in that: In step 1, each function of the system corresponds to a data flow; Specifically including: data processing data flow, data storage data flow, video compression processing data flow, control sampling processing data flow; Each data flow consists of: an input part, a processing part, and an output part.

3. The method according to claim 2, characterized in that: The process of designing data flow is as follows: find the input interface, processing node, and output interface corresponding to the function on the system hardware topology diagram; Identify support nodes and their support relationships with input interfaces, processing nodes, and output interfaces. Support nodes include power supply nodes, clock nodes, storage nodes, state control nodes, and timing configuration nodes. Different functions use different forms of lines to connect input interfaces, processing nodes, and output interfaces according to topological paths to form different data flows; Identify common resources and supporting nodes at the intersection of different data flows; Identify dedicated resources and supporting nodes of dedicated resources in different data flows; Public resources and their corresponding supporting node external resources are dedicated resources.

4. The method according to claim 3, characterized in that: Each data flow corresponds to a functional fault tree, and the first-level nodes of each fault tree include: common resource failure modes and dedicated resource failure modes in the data flow; The second level nodes of each fault tree include: supporting node failure modes of public resources and supporting node failure modes of dedicated resources, and resource failure modes with supporting relationships are connected with their supporting node failure modes according to causal relationships; There is no upper limit on the number of layers of each fault tree. If the supporting node of the i-th level still has supporting nodes, the supporting node of the supporting node is regarded as the i+1th level, i≥2; When a resource has multiple supporting nodes, the resource failure mode and the supporting node failure mode are connected through logic gates according to the causal relationship.

5. The method according to claim 4, characterized in that: The top event of each fault tree is the failure mode of the function; The top event is connected to all first-level events through logic gates according to causal relationships.

6. The method according to claim 5, characterized in that: Causation is the use of logic gate symbols to describe the cause-effect relationship between various events in the system.

7. The method according to claim 6, characterized in that: Each row of the fault correlation matrix corresponds to a top event of the fault tree; Each column of the fault correlation matrix corresponds to a node that has no connection relationship with the next level. Duplicate nodes need to be merged and cannot appear repeatedly.

8. The method according to claim 7, characterized in that: The data in the i-th row and j-th column of the fault correlation matrix is ​​0 or an integer not less than 1, representing whether there is a causal relationship between the failure mode of the i-th function and the failure mode of the j-th node; 0 means irrelevant, other values ​​mean relevant, and the larger the value, the greater the correlation; i takes values ​​from 1 to m, where m is the number of failure modes of the system function; j takes a value from 1 to n, where n is the number of non-repeated nodes in the fault tree that have no connection with the next level.

9. The method according to claim 8, characterized in that: The process of generating a fault diagnosis sequence table is as follows: Match the top event of each function with the fault phenomenon that occurred; Divide the matched top event corresponding rows and the unmatched top event corresponding rows into a matched matrix and an unmatched matrix; The matching matrix is ​​summed column by column; The unmatched matrix is ​​summed column by column; Multiply the sum of the j-th column of the matching matrix by the sum of the j-th column of the unmatched matrix; Remove the non-zero data after multiplication in the sum of the matching matrix, and sort the remaining data from large to small; And perform fault detection on the corresponding nodes according to the sorting results.

Images