Ship electronic information system fault propagation analysis method based on public computing environment

Abstract

The invention belongs to the technical field of ship electronic information system fault propagation analysis, and provides a ship electronic information system fault propagation analysis method based on a public computing environment. Comprising the steps of hardware equipment and software resource identification, function classification and attribute evaluation, hardware resource connection diagram construction, dependency extraction, physical network topology, function logic network construction, association mapping and integration, dependent network construction, node normal threshold setting, failure node determination and load transfer formula setting. And determining a fault propagation mechanism and carrying out k-step iterative propagation. According to the method, the relationship and dependence between each functional module and hardware resources are more intuitive by constructing the system topology model, and an analysis basis is provided for subsequent fault propagation analysis; fault propagation analysis is performed on a physical domain and an information domain, and interaction between the physical domain and the information domain is considered, so that analysis of a fault propagation path is more comprehensive; and the fault propagation calculation process is simplified through a k-step method.

Description

Technical Field

[0001] The present invention relates to the technical field of fault propagation analysis of ship electronic information systems, and in particular to a method for fault propagation analysis of ship electronic information systems based on a public computing environment. Background Art

[0002] With the continuous improvement of the informationization of ships, the integration depth and breadth of ship electronic information systems have been significantly enhanced. Horizontally, the system covers all major functional modules of the ship; vertically, it covers the entire information chain such as data acquisition control, network communication transmission, computing and storage. The system architecture is becoming more simplified and more integrated. However, although this new flat and resource-sharing architecture has performed well in reducing system load and improving information processing efficiency, it also increases the difficulty of fault propagation analysis. The public computing environment of the ship's electronic information system is a highly integrated resource hardware and functional software system designed to process, store, analyze and transmit various electronic information generated during the operation of the ship. The core of the public computing environment lies in resource sharing. It provides basic service support for the normal operation of various functional systems of the ship and is closely integrated with each system. When a node fails, the fault may spread rapidly, affecting other functional systems, thereby reducing or interrupting their functions. The task of fault propagation analysis is to evaluate the possible fault propagation path and its impact range by conducting a detailed analysis of the fault information. Researchers have accumulated a lot of experience in the field of fault diagnosis, and in the early days they mainly relied on experience judgment and manual analysis. With the continuous advancement of technology, more and more advanced scientific methods are being introduced to improve the accuracy and efficiency of fault analysis. At present, model-driven fault propagation analysis methods have become the mainstream. This type of method first builds a topological model of the system through an in-depth analysis of the system structure and fault information, and then conducts fault propagation analysis on this basis to identify possible fault propagation paths. The application of heuristic algorithms further improves the efficiency of fault propagation analysis in ship electronic information systems. By analyzing the task links in a public computing environment, system administrators can identify potential fault propagation paths in a timely manner, thereby effectively preventing and handling them, and reasonably arranging fault detection points, ultimately improving the overall reliability of the system.

[0003] However, the relationship between the modules of the existing system is complex, which makes it difficult to identify the fault propagation path; the existing analysis method does not take into account key factors such as hardware resources, network bandwidth and computing power, making the fault propagation assessment rather one-sided; in addition, the analysis efficiency and response speed of the existing analysis algorithm are poor, resulting in insufficient system reliability and fault resistance. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method for fault propagation analysis of a ship electronic information system based on a public computing environment. By constructing a system topology model, the relationship and dependency between various functional modules and hardware resources are made more intuitive, and an analysis basis is provided for subsequent fault propagation analysis. By performing fault propagation analysis on the physical domain and the information domain respectively and considering the interaction between the two, the analysis of the fault propagation path is made more comprehensive. The fault propagation calculation process is simplified by the k-step method.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for fault propagation analysis of a ship electronic information system based on a public computing environment, comprising:

[0007] Identify the hardware equipment and software resources in the target ship and obtain hardware information and software information;

[0008] Performing functional classification and attribute evaluation on the hardware information to obtain evaluation data;

[0009] Constructing a hardware resource connection diagram of the target ship according to the hardware information;

[0010] Extracting the dependency relationship between the software in the target ship according to the software information to obtain software dependency data;

[0011] Determine the communication and data transmission channels between hardware resources according to the hardware resource connection diagram, and construct a physical network topology to obtain a resource hardware subnet;

[0012] Construct a functional logic network according to the software dependency data to obtain a functional software subnet;

[0013] Associatively mapping and integrating the resource hardware subnet and the functional software subnet to obtain a system topology model;

[0014] A dependent network is constructed according to the system topology model; the calculation formula of the dependent network is: G AB = {G A ,G B ,E AB}; Among them, G AB is the dependent network; G A is the physical domain; G B is the information domain; E AB is an edge connecting the physical domain and the information domain;

[0015] Setting a node normal threshold; the node normal threshold is: computing load, storage load and bandwidth load are all within the range of 0 to 1;

[0016] Applying an initial load to all nodes in the dependent network to determine failed nodes that do not meet the normal threshold of the nodes;

[0017] Set a load transfer formula; the load transfer formula is: Wherein, ΔL is the load amount distributed by the failed node to the normal nodes in the dependent network; P is the load amount received by the normal node from the failed node; Len is the shortest path length between the failed node and the normal node; ω is an adjustment parameter;

[0018] According to the failed node, the node status of each node in the dependent network is determined by using the load transfer formula, and the internal fault propagation mechanism of the first subnet of the physical domain is determined according to the node status;

[0019] Determine the internal fault propagation mechanism of the second subnet of the information domain and the system fault propagation mechanism of the dependent network according to the internal fault propagation mechanism of the first subnet;

[0020] According to the internal fault propagation mechanism of the first subnet, the internal fault propagation mechanism of the second subnet and the system fault propagation mechanism, the dependent network is iterated and propagated for k steps using the k-step method to obtain a set of iteratively completed fault propagation paths; each step of the iterative propagation process includes:

[0021] Performing fault propagation on nodes in the physical domain and the information domain respectively;

[0022] updating a propagation node set according to the first subnet internal fault propagation mechanism and the second subnet internal fault propagation mechanism respectively;

[0023] Fault path extraction is performed according to the system fault propagation mechanism of the dependent network and the newly completed propagation node set to obtain the fault propagation path set in iteration.

[0024] Preferably, the communication and data transmission channels between hardware resources are determined according to the hardware resource connection diagram, and a physical network topology is constructed to obtain a resource hardware subnet, including:

[0025] The hardware device topology in the hardware resource connection diagram is converted into hardware nodes;

[0026] The physical links between the hardware nodes are determined and the link bandwidth, link communication protocol and link delay of the physical links are recorded to obtain the resource hardware subnet; the physical links include: cable connection, optical fiber link and wireless connection.

[0027] Preferably, a functional logic network is constructed according to the software dependency data to obtain a functional software subnet, including:

[0028] Determining a logical node according to the software dependency data;

[0029] Adding attribute parameters to the logical node; the attribute parameters include: average execution time, computing resources, software storage resources and bandwidth resources;

[0030] The data flow diagram is used to confirm the logical interaction and data flow between the logical nodes, and the direction of the data flow is abstracted as the edge of the logical node to obtain the functional software subnet.

[0031] Preferably, the node status includes: normal status, abnormal failure status, overload status, isolated status and dependent failure status.

[0032] Preferably, the hardware information includes: network equipment, communication equipment, computing nodes and storage equipment; the software information includes: operating system, database, management software, control system and application program.

[0033] Preferably, the attribute evaluation includes: computing capability evaluation, storage capability evaluation and communication capability evaluation; the evaluation data includes: operation speed, hardware storage resources and network bandwidth.

[0034] Preferably, the hardware resource connection diagram includes: port information, communication protocol, bandwidth attributes and delay data.

[0035] Preferably, the second subnet internal fault propagation mechanism includes:

[0036] When a first node to be analyzed in the physical domain is attacked, determining that the node state of the node to be analyzed is the abnormal failure state;

[0037] Remove the first analysis node and the connected edges and dependent edges where the analysis node is located;

[0038] When the second node to be analyzed in the information domain loses the connected edge, removing the second node to be analyzed and the connected edge where the second node to be analyzed is located;

[0039] When the connected edge does not exist between the third node to be analyzed and the fourth node to be analyzed in the physical domain, remove the third node to be analyzed, the fourth node to be analyzed, and the nodes in the information domain corresponding to the third node to be analyzed and the fourth node to be analyzed respectively;

[0040] When the fifth node to be analyzed in the information domain does not have the connected edge, the fifth node to be analyzed and the node corresponding to the fifth node to be analyzed in the physical domain are removed.

[0041] The present invention discloses the following technical effects:

[0042] The present invention provides a method for fault propagation analysis of a ship electronic information system based on a public computing environment. By performing fault propagation analysis on the physical domain and the information domain respectively and considering the interaction between the two, the defect of a single analysis line in conventional technology is solved, and a comprehensive fault propagation path analysis is achieved. Through the k-step method, the problem of complex calculation process of conventional technology is solved, and the fault propagation calculation process is simplified. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 A schematic diagram of a fault propagation analysis process of a ship electronic information system based on a public computing environment provided by an embodiment of the present invention;

[0045] Figure 2 A flowchart of building a topology model provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of load fluctuation when a load theoretical node fails provided by an embodiment of the present invention;

[0047] Figure 4 A state transition diagram of a node under cascading failure provided by an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of the inter-subnet cascading failure process provided by an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of a system fault propagation mechanism provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] 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.

[0051] The purpose of the present invention is to provide a method for fault propagation analysis of a ship electronic information system based on a public computing environment. By constructing a system topology model, the relationship and dependency between various functional modules and hardware resources are made more intuitive, and an analysis basis is provided for subsequent fault propagation analysis. By performing fault propagation analysis on the physical domain and the information domain respectively and considering the interaction between the two, the analysis of the fault propagation path is made more comprehensive. The fault propagation calculation process is simplified by the k-step method.

[0052] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] Figure 1 A schematic diagram of a fault propagation analysis process of a ship electronic information system based on a public computing environment provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the present invention provides a method for analyzing fault propagation of a ship electronic information system based on a public computing environment, comprising:

[0054] Identify the hardware equipment and software resources in the target ship and obtain hardware information and software information;

[0055] Performing functional classification and attribute evaluation on the hardware information to obtain evaluation data;

[0056] Constructing a hardware resource connection diagram of the target ship according to the hardware information;

[0057] Extracting the dependency relationship between the software in the target ship according to the software information to obtain software dependency data;

[0058] Determine the communication and data transmission channels between hardware resources according to the hardware resource connection diagram, and construct a physical network topology to obtain a resource hardware subnet;

[0059] Construct a functional logic network according to the software dependency data to obtain a functional software subnet;

[0060] Associatively mapping and integrating the resource hardware subnet and the functional software subnet to obtain a system topology model;

[0061] A dependent network is constructed according to the system topology model; the calculation formula of the dependent network is: in, is the dependent network; G A is the physical domain; G B is the information domain; is an edge connecting the physical domain and the information domain;

[0062] Setting a node normal threshold; the node normal threshold is: computing load, storage load and bandwidth load are all within the range of 0 to 1;

[0063] Applying an initial load to all nodes in the dependent network to determine failed nodes that do not meet the normal threshold of the nodes;

[0064] Set a load transfer formula; the load transfer formula is: Wherein, ΔL is the load amount distributed by the failed node to the normal nodes in the dependent network; P is the load amount received by the normal node from the failed node; Len is the shortest path length between the failed node and the normal node; ω is an adjustment parameter;

[0065] According to the failed node, the node status of each node in the dependent network is determined by using the load transfer formula, and the internal fault propagation mechanism of the first subnet of the physical domain is determined according to the node status;

[0066] Determine the internal fault propagation mechanism of the second subnet of the information domain and the system fault propagation mechanism of the dependent network according to the internal fault propagation mechanism of the first subnet;

[0067] According to the internal fault propagation mechanism of the first subnet, the internal fault propagation mechanism of the second subnet and the system fault propagation mechanism, the dependent network is iterated and propagated for k steps using the k-step method to obtain a set of iteratively completed fault propagation paths; each step of the iterative propagation process includes:

[0068] Performing fault propagation on nodes in the physical domain and the information domain respectively;

[0069] updating a propagation node set according to the first subnet internal fault propagation mechanism and the second subnet internal fault propagation mechanism respectively;

[0070] Fault path extraction is performed according to the system fault propagation mechanism of the dependent network and the newly completed propagation node set to obtain the fault propagation path set in iteration.

[0071] Specifically, the communication and data transmission channels between hardware resources are determined according to the hardware resource connection diagram, and a physical network topology is constructed to obtain a resource hardware subnet, including:

[0072] The hardware device topology in the hardware resource connection diagram is converted into hardware nodes;

[0073] The physical links between the hardware nodes are determined and the link bandwidth, link communication protocol and link delay of the physical links are recorded to obtain the resource hardware subnet; the physical links include: cable connection, optical fiber link and wireless connection.

[0074] Furthermore, a functional logic network is constructed according to the software dependency data to obtain a functional software subnet, including:

[0075] Determining a logical node according to the software dependency data;

[0076] Adding attribute parameters to the logical node; the attribute parameters include: average execution time, computing resources, software storage resources and bandwidth resources;

[0077] The data flow diagram is used to confirm the logical interaction and data flow between the logical nodes, and the direction of the data flow is abstracted as the edge of the logical node to obtain the functional software subnet.

[0078] Preferably, the node status includes: normal status, abnormal failure status, overload status, isolated status and dependent failure status.

[0079] Specifically, the hardware information includes: network equipment, communication equipment, computing nodes and storage equipment; the software information includes: operating system, database, management software, control system and application program.

[0080] Preferably, the attribute evaluation includes: computing capability evaluation, storage capability evaluation and communication capability evaluation; the evaluation data includes: operation speed, hardware storage resources and network bandwidth.

[0081] Specifically, the hardware resource connection diagram includes: port information, communication protocol, bandwidth attributes and delay data.

[0082] Furthermore, the second subnet internal fault propagation mechanism includes:

[0083] When a first node to be analyzed in the physical domain is attacked, determining that the node state of the node to be analyzed is the abnormal failure state;

[0084] Remove the first analysis node and the connected edges and dependent edges where the analysis node is located;

[0085] When the second node to be analyzed in the information domain loses the connected edge, removing the second node to be analyzed and the connected edge where the second node to be analyzed is located;

[0086] When the connected edge does not exist between the third node to be analyzed and the fourth node to be analyzed in the physical domain, remove the third node to be analyzed, the fourth node to be analyzed, and the nodes in the information domain corresponding to the third node to be analyzed and the fourth node to be analyzed respectively;

[0087] When the fifth node to be analyzed in the information domain does not have the connected edge, the fifth node to be analyzed and the node corresponding to the fifth node to be analyzed in the physical domain are removed.

[0088] refer to Figure 2 , the ship's electronic information system can be divided into four components in its composition structure: perception equipment, command and control software, basic support equipment and weapon equipment. According to their different focuses, they are divided into physical domain and information domain. The physical domain focuses more on the physical connection and transmission capabilities of the network, while the information domain focuses more on the logical flow and processing of data. The physical domain provides the infrastructure required for the operation of the information domain, and the information domain implements higher-level functions and services based on the physical domain. In this way, the basic support equipment is constructed into the physical domain, and other perception equipment, command and control software and weapon equipment are jointly constructed into the information domain model. The steps for building the topology model are as follows:

[0089] 1) Public computing environment resource hardware extraction

[0090] Extract and organize all hardware-related resource information in the ship's electronic information system, and clarify the composition and connection relationship of each hardware node in the system. It is mainly divided into the following steps:

[0091] S1: Hardware resource identification: Identify the hardware related to the electronic information system from various functional systems of the ship, including: network equipment, communication equipment, computing nodes, storage devices, etc.

[0092] S2: Hardware resource classification: The identified hardware resources are classified according to their functions, such as network transmission equipment, computing equipment and storage equipment.

[0093] S3: Evaluation and analysis: In the resource hardware extraction phase, the hardware's computing power, storage capacity, and communication capacity need to be evaluated in detail to ensure a comprehensive understanding of the performance of each node. These evaluations are the core of building a topology model. Attributes are added to nodes based on the computing speed, storage resources, network bandwidth, and other resources of the resource hardware, such as C1 = <54TFLOPS, 42GB, 36Gbps>, where C1 is a physical domain hardware resource node.

[0094] S4: Physical connection relationship: Describe the physical connection between hardware devices, record port information, communication protocol, bandwidth, delay and other parameters, and form a preliminary hardware resource connection diagram.

[0095] 2) Public computing environment resource software extraction

[0096] Clarify the software resources running in the ship's electronic information system, especially the software structure and dependencies of each functional module. It is mainly divided into the following steps:

[0097] S1: Software resource identification: Extract software resources in electronic information systems, including operating systems, databases, management software, control systems, applications, etc.

[0098] S2: Software dependency analysis: Analyze the interdependencies between software and make it clear whether software modules communicate with each other or rely on specific hardware resources when the system is running.

[0099] 3) Resource hardware subnet construction

[0100] Build a physical network topology based on hardware devices and define the communication and data transmission channels between hardware resources. The specific steps are as follows:

[0101] S1: Hardware node definition: Based on the results of the hardware extraction phase, various hardware devices are used as nodes in the subnet.

[0102] S2: Physical link construction: Determine the physical link between hardware nodes, such as cable connection, fiber link, wireless connection, etc., and record the link's bandwidth, communication protocol, delay and other parameters.

[0103] 4) Functional software subnet construction

[0104] Based on the dependencies of software operation, build the functional logic network of the system and analyze the data flow and interaction relationship between software modules.

[0105] S1: Functional node definition: Based on the functional modules, define the logical nodes in each subnet, such as control module, data analysis module, decision support module, etc. And add attributes such as average execution time, required computing resources, storage resources, bandwidth resources, etc. to the functional software nodes.

[0106] P1=<100ms,10TFLOPS,5GB,36Gbps>

[0107] Among them, P1 is the functional software subnet software node;

[0108] The analysis system assists in the data flow diagram (DFD) of the functional software when executing specific tasks, abstracts the data flow direction between the functional software into the edges of the network nodes, and then completes the construction of the functional software subnet.

[0109] S2: Software logical connection: Analyze the auxiliary data flow diagram (DFD) of the functional software when the system performs specific tasks to determine the logical interaction and data flow between software modules, and record its dependent links in task execution. The data flow direction between functional software is abstracted as the connection edge of the network node, and then the functional software subnet is constructed.

[0110] 5) Topology model construction

[0111] The hardware subnet and software subnet are combined to form a complete system topology model to describe the structure of the entire ship's electronic information system and its internal communication, data transmission and resource dependencies. The specific steps are as follows:

[0112] S1: Hardware and software mapping: Associate and map the hardware subnet with the software subnet to clarify the hardware resources and network connections that each software module depends on.

[0113] S2: Global topology formation: Integrate the topology diagrams of hardware and software to form the global topology of the system, and clarify all nodes, links and their transmission paths.

[0114] Through the above five main links, a complete topological model of the ship electronic information system is constructed. In the ship electronic information system, the two interdependent networks have different behaviors. AB = {G A ,G B ,E AB} Where G A is the physical domain and G B It is the information domain. The model not only reflects the hardware and software resources of the system, but also includes its communication paths, resource dependencies, and fault propagation possibilities. This topology model provides a solid foundation for subsequent fault propagation analysis and path identification algorithms, ensuring that the system can maintain efficient and reliable operation in actual operations.

[0115] Specifically, after constructing the topological model of the ship's electronic information system, the next step is to analyze the fault propagation theory of the functional software nodes and resource hardware nodes in the topological model to provide a theoretical basis for applying the fault propagation path identification solution on the topological model. Specifically, it includes three parts: analysis of the fault propagation mechanism within the subnet, analysis of the fault propagation mechanism between subnets, and analysis of the system fault propagation mechanism. The fault propagation analysis specifically includes the following implementation steps:

[0116] 1) Analysis of the fault propagation mechanism within the subnet

[0117] S1: Fault propagation mechanism within the physical domain subnet

[0118] In the topology model building step, a dependency network is established Among them, G A =(V A ,E A ), V A The nodes define the load and capacity of the resource hardware V A =<Load,Capacity>. Among them, Load=<L Computing ,L storage ,Lbandwidth >, Capacity=<C Computing ,C storage ,C bandwidth >;V A is the physical domain node, Load is the current load of the physical domain node, where L Computing , L storage , L bandwidth They represent computing, storage, and bandwidth loads respectively. Capacity is the load capacity of the physical domain node. Computing , C storage , C bandwidth They represent computing, storage, and bandwidth capacity respectively. When a node in a subnet fails, the load in the subnet will flow. Some nodes will be overloaded due to the influence of the failed node's load, which will cause new node failures. Initially, all nodes have an initial load (each node has a different initial load), and the constant 1 refers to the load capacity.

[0119]

[0120] Among them, L Computing (t), L storage (t), L bandwidth (t) are the computing load, storage load and bandwidth load on the node at time t, and under normal circumstances, 0 <= L Computing (t)<=1,0<=L storage (t)<=1,0<=L bandwidth (t)<=1.

[0121] At the initial moment, an initial load D is applied to all nodes to determine whether there are nodes with a computing load, storage load, or bandwidth load exceeding 1. If so, these nodes fail. After the node fails, this part of the load will be transferred through the edges of the node. The load transfer of nodes in the network satisfies the formula:

[0122]

[0123] Among them, ΔL is the load allocated to the remaining nodes after the node fails, the load obtained by the nearest neighbor from the failed node is represented by P, the shortest path length between other nodes and the failed node is represented by Len, and ω is an adjustable parameter. Therefore, there can be multiple allocation rules.

[0124] refer to Figure 3When a node fails due to overload or attack, the load of the failed node will flow to the nodes that are still working normally in the network in a certain proportion, causing load fluctuations in the network. A local network load fluctuation is shown in the figure below. The nodes pointed by the red arrows in the figure are nodes that are attacked or overloaded, the red nodes are failed nodes, and the blue nodes are nodes that are working normally. The orange line in the figure shows the flow of load after the node fails.

[0125] refer to Figure 4 , five possible states of nodes in the physical domain network, namely node states. hi = 0 means that node i is in a normal and non-overloaded state and the node is in the largest connected subgraph GCC of the network; s hi =1 indicates a critical state, the node is still working normally but showing an abnormal overload state. These nodes will either fail due to inability to alleviate the overload or the load will be reduced to normal due to topology changes or node recovery; hi =2 means that the node maintains an overload state, causing the overload pressure L(t) to exceed the overload pressure threshold O T , and eventually failed due to overload; hi =3 means that the node fails because it cannot connect to GCC. This is because the failure of some nodes causes the change of network topology and makes the node unable to connect to GCC. hi =4 indicates that the node fails due to the loss of dependency on nodes in another layer of the network.

[0126] S2: Internal fault propagation mechanism of information domain subnet

[0127] There is no load flow in the information domain subnet. Compared with the load, more attention is paid to the connectivity of the network. As long as all the predecessor nodes of the node can work normally and the node can work normally, the node can work normally. si ∈{0,3,4} where s si =0 means that node i is in normal state and all the predecessor nodes of the node in the network are in normal state; s si =3 means that node i fails due to the failure of the predecessor node; s si =4 indicates that the node fails due to the failure of the corresponding dependent node.

[0128] 2) Analysis of fault propagation mechanism between subnets

[0129] The nodes in the physical domain subnet and the information domain subnet correspond one to one. In the cascade propagation process, all nodes must meet the following requirements: In a pair of coupled nodes connected by dependent edges, the failure of any node will inevitably lead to the failure of the corresponding coupled node; only nodes existing in the maximal connected graph can maintain their original functions, and nodes that are out of the maximal connected graph are considered to be failed.

[0130] refer to Figure 5 , this cascade process is abstracted into a simplified model. The connected edges are represented by black solid lines, and the dependent edges are represented by black dotted lines with bidirectional arrows. The physical domain subnet and the information domain subnet each have six nodes and correspond one to one. The yellow nodes are nodes in the physical domain subnet, and the green nodes are nodes in the information domain subnet. Four stages are used to represent the four steps in the cascade process. The cascade propagation process is as follows:

[0131] Stage 0: Node a5 in the physical domain subnet is attacked (Attack), the node fails and is removed;

[0132] Stage 1: In the physical domain subnet, since node a5 is removed, all connected edges and dependent edges of a5 are removed at the same time. In the information domain subnet, node b5 loses its coupling edge and fails, so it is removed. The connected edges of b5 are also removed.

[0133] Stage 2: Nodes a4 and a6 in the physical domain subnet fail due to the loss of connected edges, which leads to their separation from the maximum connectivity graph. Their corresponding coupling nodes b4 and b6 also fail one after another and their connected edges are removed.

[0134] Stage 3: Node b3 in the information domain subnet loses its connected edge, causing it to be out of the maximal connected graph and fail. Its coupling node a2 also fails at the same time. At this time, there is no new failed node in the entire network system, reaching a stable state, and the cascade ends.

[0135] 3) Analysis of system fault propagation mechanism

[0136] refer to Figure 6 , the cascading failure mechanism of the dependent network is as follows:

[0137] S1: Initial attack: Randomly or intentionally remove a node from the physical domain subnet to trigger a cascading failure of dependent networks;

[0138] S2: Internal failure of the physical domain subnet: The physical domain subnet is a load network. The failure of a node will cause the node load to be redistributed within a certain range of the failed node, causing more nodes to fail due to overload. At the same time, the topology of the network has also changed, which may cause the entire network to be broken, and some nodes cannot be connected to the GCC (maximum connected subgraph), causing these nodes to fail.

[0139] S3: External failure of the information domain subnet: Some nodes in the physical domain subnet fail, causing the nodes that are interdependent with these nodes in the information domain subnet to fail due to loss of dependence. This can be seen as an "attack" on the information domain subnet.

[0140] S4: Internal failure of the information domain subnet: The information domain subnet is a percolation-based network, where nodes will only fail due to the failure of their predecessor nodes and loss of dependency. Similar to the above, the failure of some nodes causes the network topology to change, causing some nodes to fail due to the failure of their predecessor nodes.

[0141] S5: External failure of the physical domain subnet: The failure of a node in the network causes the nodes in the network to fail due to loss of dependence, which can be regarded as an "attack" by the network on the network. However, this is not a real attack, and such nodes cannot be actively restored. This process will continue to cycle until the network reaches a stable state.

[0142] Specifically, the fault propagation path of the dual-domain subnet is analyzed based on the k-step method. The hierarchical propagation of the ship electronic information system fault is carried out step by step on the basis of the system stratification. The fault originates from the bottom node and propagates to the nodes of the same layer associated with it. After k steps of propagation within the layer, it propagates from a certain node to the upper layer and completes one step of inter-layer propagation.

[0143] Furthermore, the pseudo code and description of the dual-domain subnet fault propagation path analysis based on the k-step method are as follows:

[0144] 1) Input: G P (V P ,E P ): physical domain subnet, V P is a node set, E P is the edge set;

[0145] G I (V I ,E I ): Information domain subnet, V I is a node set, E I is the edge set;

[0146] S P : Physical domain initial fault set;

[0147] S I : Information domain initial fault set;

[0148] k: step limit;

[0149] T P : Load threshold of physical domain nodes;

[0150] F: Fault propagation rules between the physical domain and the information domain (based on percolation theory).

[0151] Output: Fault propagation path set P.

[0152] 2) Initialization:

[0153] Set an empty collection Used to store fault propagation paths;

[0154] Set the current propagation node set for the physical domain The current set of propagation nodes in the information domain

[0155] Set the step counter step=0.

[0156] 3) Iterative propagation:

[0157] While step<k and ( or )do

[0158] 4) Physical domain propagation:

[0159]

[0160] Detection node v P Whether the load exceeds the threshold T P If it exceeds, the node fails and the path is recorded. P =v P →fail, and join the set P;

[0161] Find the v P The set of adjacent non-faulty nodes N P (v P ), if N P (v P ) has a node load exceeding the threshold T P , then these nodes fail and update the physical domain propagation node set

[0162] 5) Information domain dissemination:

[0163]

[0164] Detection node v I Whether the failure is caused by loss of physical domain support or isolation. If isolated or lost support, the path path is recorded I =v I →fail, and join the set P;

[0165] Find the v P The set of adjacent non-faulty nodes N I (v I ), if N I (v I ) If a node fails due to loss of physical domain support or isolation, the physical domain propagation node set is updated

[0166] 6) Seepage fault propagation:

[0167] According to the coupling rule F between the physical domain and the information domain (based on the percolation theory), the fault propagates between the faulty nodes in the physical domain and the faulty nodes in the information domain. P Leading to the information domain corresponding to node v I Failed, record path path = v P →v I →fail, and add the path to the set P.

[0168] The number of steps increases by step=step+1;

[0169] Termination condition: The algorithm terminates when the number of steps reaches k or the current propagation node sets of both domains are empty.

[0170] Return,the set P of fault propagation paths.

[0171] The algorithm analyzes the fault propagation path of the dual-domain subnet based on the k-step method, handles the fault propagation of the physical domain subnet and the information domain subnet respectively, and considers the mutual influence between the two. The specific steps are as follows:

[0172] 1) Initialization: Select the initial faulty node from the physical domain and information domain subnets, and set the step limit k. The fault propagation in the physical domain is mainly based on the load theory, while the fault propagation in the information domain considers the failure of nodes due to isolation or loss of physical domain support.

[0173] 2) Iterative propagation: In each step, fault propagation is performed on nodes in the physical domain and information domain respectively.

[0174] S1: Physical domain propagation: Check whether the load of each physical domain node exceeds the threshold. If overloaded, the node fails and propagates to adjacent nodes, updating the propagation node set.

[0175] S2: Information domain propagation: Check whether the information domain node fails due to isolation or lack of physical domain support, and propagate it to adjacent nodes to update the propagation node set.

[0176] S3: Percolation fault propagation: Through the coupling relationship between the physical domain and the information domain (based on percolation theory), if a faulty node in the physical domain affects the corresponding node in the information domain, the information domain node will also fail and the fault path will be recorded.

[0177] 3) Termination condition: When the number of steps reaches k or there is no new fault node propagation, the algorithm terminates and outputs a complete set of fault propagation paths.

[0178] The beneficial effects of the present invention are as follows:

[0179] The present invention makes the relationship and dependency between various functional modules and hardware resources more intuitive by constructing a system topology model, and provides an analysis basis for subsequent fault propagation analysis; by performing fault propagation analysis on the physical domain and information domain respectively and considering the interaction between the two, the analysis of the fault propagation path is more comprehensive; and the fault propagation calculation process is simplified by the k-step method.

[0180] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0181] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for analyzing fault propagation of a ship electronic information system based on a public computing environment, characterized in that: include: Identify the hardware equipment and software resources in the target ship and obtain hardware information and software information; Performing functional classification and attribute evaluation on the hardware information to obtain evaluation data; Constructing a hardware resource connection diagram of the target ship according to the hardware information; Extracting the dependency relationship between the software in the target ship according to the software information to obtain software dependency data; Determine the communication and data transmission channels between hardware resources according to the hardware resource connection diagram, and construct a physical network topology to obtain a resource hardware subnet; Construct a functional logic network according to the software dependency data to obtain a functional software subnet; Associatively mapping and integrating the resource hardware subnet and the functional software subnet to obtain a system topology model; A dependent network is constructed according to the system topology model; the calculation formula of the dependent network is: in, is the dependent network; G A is the physical domain; G B is the information domain; is an edge connecting the physical domain and the information domain; Setting a node normal threshold; the node normal threshold is: computing load, storage load and bandwidth load are all within the range of 0 to 1; Applying an initial load to all nodes in the dependent network to determine failed nodes that do not meet the normal threshold of the nodes; Set a load transfer formula; the load transfer formula is: Wherein, ΔL is the load amount distributed by the failed node to the normal nodes in the dependent network; P is the load amount received by the normal node from the failed node; Len is the shortest path length between the failed node and the normal node; ω is an adjustment parameter; According to the failed node, the node status of each node in the dependent network is determined by using the load transfer formula, and the internal fault propagation mechanism of the first subnet of the physical domain is determined according to the node status; Determine the internal fault propagation mechanism of the second subnet of the information domain and the system fault propagation mechanism of the dependent network according to the internal fault propagation mechanism of the first subnet; According to the fault propagation mechanism within the first subnet, the fault propagation mechanism within the second subnet, and the system fault propagation mechanism, the dependent network is iterated and propagated for k steps using the k-step method to obtain a set of iteratively completed fault propagation paths; each step of the iterative propagation process includes: Performing fault propagation on nodes in the physical domain and the information domain respectively; updating a propagation node set according to the first subnet internal fault propagation mechanism and the second subnet internal fault propagation mechanism respectively; Fault path extraction is performed according to the system fault propagation mechanism of the dependent network and the newly completed propagation node set to obtain the fault propagation path set in iteration.

2. According to the method for analyzing the fault propagation of a ship electronic information system based on a public computing environment in claim 1, it is characterized in that: According to the hardware resource connection diagram, the communication and data transmission channels between the hardware resources are determined, and the physical network topology is constructed to obtain the resource hardware subnet, including: The hardware device topology in the hardware resource connection diagram is converted into hardware nodes; The physical links between the hardware nodes are determined and the link bandwidth, link communication protocol and link delay of the physical links are recorded to obtain the resource hardware subnet; the physical links include: cable connection, optical fiber link and wireless connection.

3. According to the method for fault propagation analysis of a ship electronic information system based on a public computing environment in claim 1, it is characterized in that: A functional logic network is constructed according to the software dependency data to obtain a functional software subnet, including: Determining a logical node according to the software dependency data; Adding attribute parameters to the logical node; the attribute parameters include: average execution time, computing resources, software storage resources and bandwidth resources; The data flow diagram is used to confirm the logical interaction and data flow between the logical nodes, and the direction of the data flow is abstracted as the edge of the logical node to obtain the functional software subnet.

4. According to the method for analyzing fault propagation of a ship electronic information system based on a public computing environment in claim 1, it is characterized in that: The node status includes: normal status, abnormal failure status, overload status, isolated status and dependent failure status.

5. According to the method for fault propagation analysis of ship electronic information system based on public computing environment in claim 1, it is characterized in that: The hardware information includes: network equipment, communication equipment, computing nodes and storage equipment; the software information includes: operating system, database, management software, control system and application program.

6. According to the method for fault propagation analysis of a ship electronic information system based on a public computing environment in claim 1, it is characterized in that: The attribute evaluation includes: computing capability evaluation, storage capability evaluation and communication capability evaluation; the evaluation data includes: operation speed, hardware storage resources and network bandwidth.

7. According to the method for analyzing fault propagation of a ship electronic information system based on a public computing environment in claim 1, it is characterized in that: The hardware resource connection diagram includes: port information, communication protocol, bandwidth attributes and delay data.

8. A method for analyzing fault propagation of a ship electronic information system based on a public computing environment according to claim 4, characterized in that: The second subnet internal fault propagation mechanism includes: When a first node to be analyzed in the physical domain is attacked, determining that the node state of the node to be analyzed is the abnormal failure state; Remove the first analysis node and the connected edges and dependent edges where the analysis node is located; When the second node to be analyzed in the information domain loses the connected edge, removing the second node to be analyzed and the connected edge where the second node to be analyzed is located; When the connected edge does not exist between the third node to be analyzed and the fourth node to be analyzed in the physical domain, remove the third node to be analyzed, the fourth node to be analyzed, and the nodes in the information domain corresponding to the third node to be analyzed and the fourth node to be analyzed respectively; When the fifth node to be analyzed in the information domain does not have the connected edge, the fifth node to be analyzed and the node corresponding to the fifth node to be analyzed in the physical domain are removed.

Images