A fault tree diagnosis method of a ground measurement and control device

Abstract

The application discloses a fault tree diagnosis method of ground TT&C equipment, which comprises the following steps: S1, a fault tree matching step, wherein fault tree matching is completed according to a diagnosis object and a fault phenomenon; S2, a fault positioning step, wherein equipment fault positioning diagnosis is completed based on the matched fault tree, and the equipment fault positioning diagnosis process is completed based on event reasoning steps to judge fault tree node events, and a fault tree search branch is determined according to a judgment result to complete equipment fault positioning; S3, a fault synthesis step, wherein whether the fault positioning result obtained in S2 is a cut set of a matching node is judged, if yes, the matching node event occurs, then the fault positioning is consistent with the fault synthesis; if no, the matching node event does not occur, that is, the fault positioning is inconsistent with the fault synthesis; S4, an output fault diagnosis conclusion step, wherein a fault diagnosis conclusion is given based on a fault mode positioned in S2 and a fault synthesis result in S3. The problem of how to efficiently diagnose faults of ground TT&C equipment is solved.

Description

Technical Field

[0001] This invention belongs to the field of health management of aerospace ground telemetry and control equipment, and specifically relates to a fault tree diagnosis method for ground telemetry and control equipment. Background Technology

[0002] In recent years, with the explosive growth of spacecraft in orbit and planned launches, the demand for ground tracking, telemetry, and command (TT&C) has also increased rapidly. To meet this massive demand, ground-based TT&C equipment is constantly being expanded and constructed, posing significant challenges to its operational support. Therefore, establishing an automated and intelligent central-level Prognostic and Health Management (PHM) system has become urgent. Historically, the research and development and production of ground-based TT&C equipment have been conducted through research projects, resulting in customized equipment that lacks standardization, systematization, and large-scale production, hindering the development of health management research and applications in related fields.

[0003] Fault diagnosis is a crucial component of a PHM (Prognostics and Health Management) system. Its main purpose is to locate underlying fault modes, analyze fault mechanisms, and provide remedial recommendations, providing key inputs for functions such as false alarm elimination, health assessment, fault prediction, and maintenance support. Commonly used fault diagnosis methods include expert knowledge systems, machine learning, and deep learning, all of which have been successfully applied in PHM systems across various fields. However, the application of these methods in fault diagnosis of ground-based telemetry and control equipment has not been fully explored: on the one hand, the lack of data acquisition, transmission, storage, and processing platforms prevents the use of machine learning and deep learning methods to train and utilize fault diagnosis models; on the other hand, because equipment is customized on a project-by-project basis, there is a lack of conditions for accumulating expert knowledge from historical data.

[0004] Therefore, current research on fault diagnosis methods for ground-based telemetry and control equipment is insufficient, failing to adapt to the increasingly strained ground-based telemetry and control resources and unable to meet users' urgent PHM (Prognostics and Monitoring) needs. To address this, this invention proposes a rule-based knowledge-based fault diagnosis method for ground-based telemetry and control equipment. Summary of the Invention

[0005] The purpose of this invention is to overcome the problems of the prior art by disclosing a fault tree diagnosis method for ground measurement and control equipment, thereby solving the problem of how to efficiently diagnose faults in ground measurement and control equipment.

[0006] The objective of this invention is achieved through the following technical solution:

[0007] A fault tree diagnosis method for ground-based telemetry and control equipment, the fault tree diagnosis method comprising at least the following steps:

[0008] S1: Fault tree matching step, completes fault tree matching based on the diagnostic object and fault phenomenon;

[0009] S2: Fault location step, completes equipment fault location diagnosis based on the matched fault tree, and the equipment fault location diagnosis process completes fault tree node event judgment based on event reasoning steps, and determines the fault tree search branch based on the judgment result to complete equipment fault location;

[0010] S3: Fault synthesis step, determine whether the fault location result obtained in S2 is a cut set of the matching node. If yes, it supports the occurrence of the matching node event, and the fault location and fault synthesis are consistent; if no, it does not support the occurrence of the matching node event, that is, the fault location and fault synthesis are inconsistent.

[0011] S4: Output fault diagnosis conclusion step, based on the fault mode located in S2 and the fault synthesis result in S3, to give the fault diagnosis conclusion.

[0012] According to a preferred embodiment, step S1, the fault tree matching step, includes:

[0013] S101: Query the fault tree associated with the diagnostic object;

[0014] S102: Calculate the matching degree of each fault tree; calculate the matching degree between each fault tree and the fault phenomenon keywords in turn, and sort them in descending order of matching degree;

[0015] S103: Fault tree filtering. Fault trees with a matching degree of 0 are called unmatched fault trees and are excluded from the matched fault trees. The remaining ones are the matched fault trees.

[0016] According to a preferred embodiment, in step S102, the calculation of the matching degree between each fault tree and the fault phenomenon keyword includes: calculating the matching degree between each node of the fault tree and the keyword respectively, marking the node with the largest matching degree as the matching node, and taking the matching degree of the matching node as the fault tree matching degree; the matching degree value ranges from 0 to 1.

[0017] According to a preferred embodiment, the event reasoning step in step S2 includes fault determination being completed automatically or interactively.

[0018] According to a preferred embodiment, when performing event reasoning automatically:

[0019] First, obtain monitoring point data based on the input configuration;

[0020] Then, the rule logic is invoked to obtain the boolean output;

[0021] Finally, the output of the judgment rule determines whether a fault has occurred, where true indicates that a fault has occurred and false indicates that a fault has not occurred.

[0022] According to a preferred implementation, when performing event reasoning in an interactive manner:

[0023] First, collect data from monitoring points that can be acquired automatically;

[0024] Next, prompts will be pushed to the user regarding the monitoring points that require interactive data acquisition;

[0025] Then, users obtain monitoring point data by visual inspection or instruments, based on the prompts.

[0026] Next, the rule logic is invoked to obtain the boolean output;

[0027] Finally, the output of the judgment rule determines whether a fault has occurred, where true indicates that a fault has occurred and false indicates that a fault has not occurred.

[0028] According to a preferred embodiment, in step S2, during the process of locating equipment faults, the search for descendant nodes of the fault tree is based on a depth-first strategy. When a basic event or an event that has not occurred is encountered, backtracking is performed. When a matching node is found and all its child nodes have been traversed, the entire process ends.

[0029] According to a preferred embodiment, step S3, the fault synthesis step, includes:

[0030] S301: All fault tree nodes corresponding to all fault modes in the fault location results are marked and denoted as set A, representing the occurrence of the event they represent;

[0031] S302: Get all unlabeled parent nodes of nodes in A, denoted as set C;

[0032] S303: Take out node c from C in sequence, and determine whether the nodes in A support the event c being true.

[0033] The conditions for the condition to be met are: the logic gate of c is AND, and all child nodes of c belong to A; the logic gate of c is OR, and c has a child node that belongs to A; the logic gate of c is XOR, and c has one and only one child node that belongs to A; the logic gate of c is voting, and more than half of the nodes of c belong to A.

[0034] S304: Merge the nodes that are valid in C into A;

[0035] S305: Repeat steps S302 to S304 until A remains unchanged or C is empty, where the fault location conclusion supports the occurrence of the matching node event only if the matching node belongs to A.

[0036] According to a preferred embodiment, the failure mode includes a name, details, device node, mechanism analysis, impact domain, and remedial recommendations.

[0037] According to a preferred embodiment, the fault phenomenon includes two attributes: fault description and keywords.

[0038] The aforementioned main solution of the present invention and its various further alternative solutions can be freely combined to form multiple solutions, all of which are solutions that can be adopted and are claimed by the present invention. Those skilled in the art, after understanding the solution of the present invention, will realize that there are many combinations based on existing technology and common knowledge, all of which are technical solutions to be protected by the present invention, and will not be exhaustively listed here.

[0039] The beneficial effects of this invention are as follows: The fault tree diagnosis method of this invention solves the problem of how to efficiently diagnose faults in ground-based telemetry and control equipment, and supports needs such as fault location, auxiliary information provision, and interactive diagnosis. The fault tree structure can be dynamically configured online, the diagnostic knowledge update operation is simple, and the diagnostic method can be iteratively upgraded in actual use. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the topology of the fault tree of the present invention;

[0041] Figure 2 This is a schematic diagram of the fault tree event or node composition structure of the present invention.

[0042] Figure 3 This is a schematic diagram of the specific process of the fault tree matching step in the fault tree diagnosis method of the present invention;

[0043] Figure 4 This is a schematic diagram of the specific event reasoning steps in the fault tree diagnosis method of the present invention;

[0044] Figure 5 This is a schematic diagram of the specific process of fault location steps in the fault tree diagnosis method of the present invention;

[0045] Figure 6 This is a schematic diagram of the specific steps in the fault synthesis process of the fault tree diagnosis method of the present invention. Detailed Implementation

[0046] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0047] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0048] The fault tree topology of this invention is as follows: Figure 1 As shown, it consists of events and logic gates. The top event has no parent node, the basic events have no child nodes or logic gates, and the logic gates include four types: AND, OR, XOR, and voting.

[0049] like Figure 2 As shown, the fault tree event (i.e. node) of this invention consists of fault mode, judgment rule and node structure.

[0050] The failure mode consists of fields such as name, details, device node, mechanism analysis, impact domain, and remedial recommendations.

[0051] Among them, the fault mode name is a concise description of the fault content; the details describe the fault content in detail; the equipment node hard-associates the fault mode to the equipment part; the mechanism analysis explains the mechanism of the fault occurrence, providing assistance to non-professionals in understanding the fault; the impact domain explains the impact of the fault on equipment functions and task capabilities, providing a reference for users to handle the fault; the handling suggestions provide suggestions for subsequent fault handling, which need to be filled in based on expert advice and the fault handling manual.

[0052] The decision rules consist of fields such as automatic flags, rule logic, input configuration, feature extraction algorithm, time slice length, and prompt information. These are the parameters for event reasoning.

[0053] The system includes several components: Automatic flags indicate whether an event has occurred; if true, the event is automatically judged; otherwise, it is judged interactively. The rule logic consists of mixed logical operations from several monitoring points. Monitoring point data can be automatically collected by sensors or obtained manually through visual inspection or instruments. Logical operations support basic operations such as AND, OR, NOT, XOR, and voting, with results being Boolean values. The input configuration explains the monitoring points and their data acquisition methods in the rule logic, facilitating the organization of rule logic input. The feature extraction algorithm maps time-series data into Boolean values ​​for participation in logical operations. The time slice length is used to divide the time-series data into several segments for rule judgment. The prompt information stores prompts for interactive input acquisition, assisting in obtaining some or all rule inputs.

[0054] The node structure consists of a parent event, logic gates, and child events. There is at most one parent event, and only the top event has no parent event; the logic gates express the connection logic from the child event to the current event, supporting logic gates such as AND, OR, XOR, and voting; there are several child events, which express the conditional relationship of reasoning from the event to the current event, while the basic event has no child events.

[0055] This invention discloses a fault tree diagnosis process for ground-based measurement and control equipment, including fault tree matching, fault location, fault synthesis, and output of fault diagnosis conclusions.

[0056] First, a suitable fault tree is matched using a fault tree matching step. Then, based on the matched fault tree, a fault location step is used to diagnose the equipment fault. During this process, an event reasoning step is invoked to determine the events of the fault tree nodes, and the fault location process determines the fault tree search branch based on the results of the event reasoning step. Finally, a fault synthesis process is used to determine whether the fixed location results support the top event of the fault location process. After the diagnostic process is completed, a fault diagnosis conclusion is given by combining the fault mode located and the fault synthesis results.

[0057] The fault tree matching step of this invention achieves fault tree matching based on the diagnostic object and fault phenomenon. The diagnostic object is the ground-based measurement and control equipment system. Each fault tree needs to be associated with several diagnostic objects, and a diagnostic object will also be associated with several fault trees. The fault phenomenon includes two attributes: fault description and keywords. The fault description is a description of the observable information when the fault occurs; there can be multiple keywords, which provide fuzzy query conditions for matching fault trees.

[0058] Specifically, refer to Figure 3 As shown, the fault tree matching steps include:

[0059] Step 1: Query the fault tree associated with the diagnostic object. Retrieve the fault tree associated with the diagnostic object to use as a condition for subsequent filtering.

[0060] Step 2: Iterative Fault Tree. The matching degree between the fault phenomenon and each fault tree will be calculated sequentially.

[0061] Step 3: Loop through the fault tree nodes. Calculate the matching degree between the fault phenomenon and each node in turn, and use the maximum matching degree of a node as the matching degree of the fault tree. Then, select the node with the highest matching degree as the matching node.

[0062] Step 4: Calculate the matching degree of fault tree nodes. The steps to calculate the matching degree between the fault mode and the fault phenomenon keywords of the fault tree nodes are as follows:

[0063] (1) Let the total number of keywords be N, and initialize the number of keyword matches to m = 0;

[0064] (2) Keywords of cyclical fault phenomena;

[0065] (3) Set the length of the keyword string to l, and initialize it to the length of the keyword string L;

[0066] (4) Loop key l length string (the total number of substrings is L-l+1);

[0067] (5) Is the keyword substring a substring of the fault mode details? If so, let m = m + l / (L(L-l+1));

[0068] (6) Has the loop keyword substring ended? If yes, proceed to the next step; otherwise, jump to step (4).

[0069] (7) Is the number of matches m greater than 0? If yes, proceed to step (10); if no, proceed to the next step.

[0070] (8) Is l greater than 2? If yes, proceed to the next step; if no, skip to step (10);

[0071] (9) Let l = l-1, then jump to step (4);

[0072] (10) Return the matching degree m / N.

[0073] Step 5: Update the matching degree of the fault tree. If the matching degree of the current node is greater than the matching degree of the fault tree, update the matching degree of the fault tree to the matching degree of the current node, and update the matching node to the current node.

[0074] Step 6: Is the node loop over? If yes, proceed to the next step; otherwise, go back to step 3.

[0075] Step 7: Has the fault tree loop ended? If yes, proceed to the next step; otherwise, skip to step 2.

[0076] Step 8: Sort the fault tree. Sort the fault tree from highest to lowest according to the matching degree.

[0077] Step 9: Delete unmatched fault trees. Delete fault trees with a match rate of 0 from the results.

[0078] Step 10: Return the matching fault tree.

[0079] refer to Figure 4 As shown, the event reasoning steps of this invention are divided into two branches: automatic and interactive. The specific process is as follows:

[0080] Step 1: Receive an event reasoning request. The event reasoning request is issued by the fault tree reasoning process.

[0081] Step 2: Determine if the event will execute automatically. If yes, proceed to the next step; otherwise, skip to step 7.

[0082] Step 3: Automatically acquire time-series data from monitoring points. During automatic execution, inference data acquisition is performed automatically according to the configuration.

[0083] Step 4: Divide the time series data into data segments. Based on the time slice length, divide the time series data into disjoint data segments. Segments that are less than one segment at the end are considered as one data segment.

[0084] Step 5: Traverse the data fragments. The data fragments will be used sequentially to determine if an event has occurred.

[0085] Step 6: Extract Boolean features from the data. Based on the feature extraction algorithm configuration, extract the Boolean features from the data at each monitoring point sequentially, that is, map the data segments into Boolean vectors. Proceed to step 14.

[0086] Step 7: Display the interactive measurement point acquisition prompts to the user. In this case, some monitoring point data needs to be acquired interactively, and the configured prompts need to be displayed to the user to guide them in obtaining the measurement point status.

[0087] Step 8: Obtain Boolean features of monitoring points through interactive methods. Based on the prompts, the user obtains the Boolean features of the monitoring points through visual inspection, instruments, or other means.

[0088] Step 9: Obtain automatically acquired monitoring point data. In addition to interactively acquiring Boolean features of monitoring points, the judgment rules may also include automatically acquiring time-series data of monitoring points.

[0089] Step 10: Divide the automatically acquired time-series data from monitoring points into segments. Based on the time slice length, divide the time-series data into disjoint data segments. Segments that are less than one segment at the end are considered as one data segment.

[0090] Step 11: Traverse the data segments. The data segments will be used sequentially, along with interactive data from monitoring points, to determine if the event has occurred.

[0091] Step 12: Extract Boolean features from the data. Based on the feature extraction algorithm configuration, extract the Boolean features from each automatically acquired monitoring point in sequence, that is, map the data segments into Boolean vectors.

[0092] Step 13: Combine the Boolean features of the monitoring points obtained automatically and interactively. Combine the two into a single Boolean vector.

[0093] Step 14: Calculate the rule logic. Substitute the values ​​from the Boolean feature vector into the variables in the rule logic expression and perform logical operations.

[0094] Step 15: Is the logical operation result true? If yes, proceed to step 18; otherwise, proceed to the next step.

[0095] Step 16: Is the data fragment traversal complete? If yes, proceed to the next step; otherwise, depending on whether the rule logic is automatic or not, jump to step 5 or step 11 respectively.

[0096] Step 17: Assume the event has not occurred. Skip to step 19.

[0097] Step 18: Assert the event has occurred.

[0098] Step 19: End the process.

[0099] This invention implements a fault location process from matching nodes to underlying fault locations. The location process searches descendant nodes using a depth-first strategy. When a basic event or an event that has not yet occurred is encountered, backtracking is performed. The entire process ends when the matching node is reached and all its child nodes have been traversed. The conclusion of the fault location process is all the basic events found.

[0100] refer to Figure 5 As shown, the fault tree reasoning process of this invention starts from the matching node (default event occurrence) and searches for the descendant nodes of that node using a depth-first strategy. The specific process is as follows.

[0101] Step 1: Mark the matching node as the base node, mark the cursor on the matching node as well, and set the cursor node's event state to "occurring". Since only the descendant nodes of the starting node are searched, it is marked as the base node, serving as the endpoint for backtracking. The cursor node is used for traversal, and the node it marks is the current node. The matching node's event occurs by default, so the cursor node's event state is set to "occurring". During the reasoning process, it is assumed that the base node has no parent node.

[0102] Step 2: Retrieve unsearched child nodes of the cursor node? If they exist, proceed to the next step; otherwise, skip to step 8.

[0103] Step 3: Move the cursor node to its child nodes.

[0104] Step 4: Execute event reasoning for the current node. Invoke the event reasoning process of this invention to obtain the result of the judgment rule.

[0105] Step 5: Has the current node event occurred? If yes, proceed to the next step; if no, skip to step 8.

[0106] Step 6: Determine if the current node is a basic event. If yes, proceed to the next step; otherwise, go back to step 2.

[0107] Step 7: Save the fault modes of the basic events to the fault location results. In this invention, the fault modes in the basic events are the target of fault diagnosis, while the fault tree is mainly used to locate the fault modes in the basic events.

[0108] Step 8: Get the parent node of the current node. If the child node of the current node does not exist or the event has not occurred, backtrack to the parent node.

[0109] Step 9: Does the parent node exist? If yes, proceed to the next step; otherwise, skip to step 11.

[0110] Step 10: Move the cursor to the parent node and jump to step 2.

[0111] Step 11: Return the fault location results and end the process.

[0112] This invention enables fault synthesis to deduce faults from fault location conclusions to top-level matching nodes, based on the fault tree cut-set analysis method. The function of fault synthesis is to determine whether the fault location result is a cut set of the matching node: if so, it supports the occurrence of the matching node event, and the fault location and fault synthesis are consistent; if not, it does not support the occurrence of the matching node event, meaning the fault location and fault synthesis are inconsistent.

[0113] like Figure 6 As shown, the fault synthesis process of this invention is an analysis process that starts from basic events and gradually searches for matching nodes. The process is as follows.

[0114] Step 1: Initialize the event set. Assume that all basic events in the fault location results have occurred, and denote them as set A.

[0115] Step 2: Find the set to be confirmed. Obtain all parent nodes of nodes in node A that are not parent nodes of A, denoted as set C. The nodes in C will be the objects to be confirmed in the next step.

[0116] Step 3: Is C empty? If yes, skip to step 10; if no, proceed to the next step.

[0117] Step 4: Loop through set C. The purpose is to determine whether the event to be confirmed has occurred based on the set of events A.

[0118] Step 5: For c∈C, perform logic gate operations based on A. The condition for the operation result to be true is exactly one and only:

[0119] (1) The logic gate of c is AND, and all child nodes of c belong to A;

[0120] (2) The logic gate of c is OR, and c has a child node that belongs to A;

[0121] (3) The logic gate of c is XOR, and c has one and only one child node belonging to A;

[0122] (4) The logic gate of c is voting, and more than half of the nodes of c belong to A.

[0123] Step 6: Is the result of the logic gate execution true? If yes, proceed to the next step; otherwise, jump to step 8.

[0124] Step 7: Add event c to the event set A. When the logic gate executes and the result is true, it is considered that the event to be confirmed has occurred, so it is added to set A.

[0125] Step 8: Has the loop of the set of events to be confirmed ended? If yes, proceed to the next step; if no, skip to step 4.

[0126] Step 9: Has the event set A changed? If yes, go to step 2; otherwise, proceed to the next step.

[0127] Step 10: Does the matching node belong to the event set? If yes, proceed to the next step; if no, skip to step 12.

[0128] Step 11: Support the fault location conclusion, proceed to step 13.

[0129] Step 12: Fault location conclusion is not supported.

[0130] Step 13: Process ends.

[0131] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fault tree diagnosis method of a ground data control device, characterized by, The fault tree diagnosis method comprises at least the following steps: S1: a fault tree matching step, which matches a fault tree according to a diagnosis object and a fault phenomenon; S2: a fault positioning step, which completes device fault positioning diagnosis based on the matched fault tree, and completes fault tree node event judgment based on an event reasoning step, and determines a fault tree search branch according to a judgment result to complete device fault positioning; S3: a fault synthesis step, which judges whether the fault positioning result obtained in S2 is a cut set of a matching node, if yes, the matching node event occurs, and the fault positioning is consistent with the fault synthesis; if no, the matching node event does not occur, and the fault positioning is inconsistent with the fault synthesis; S4: an output fault diagnosis conclusion step, which gives a fault diagnosis conclusion based on the fault mode positioned in S2 and the fault synthesis result in S3; The fault tree matching step S1 comprises: S101: querying a fault tree associated with a diagnosis object; S102: calculating a matching degree of each fault tree; the matching degrees between each fault tree and a fault phenomenon keyword are calculated in sequence, and are sorted in descending order of the matching degrees; S103: fault tree screening, a fault tree with a matching degree of 0 is referred to as an unmatched fault tree, which is excluded from the matched fault trees, and the remaining fault trees are the matched fault trees; The fault synthesis step S3 comprises: S301: Mark all fault tree nodes corresponding to fault modes in the fault location result, denoted as set A , indicating the occurrence of the event represented by it; S302: Acquire A All unmarked parent nodes of the middle node, denoted as set C ; S303: Take out in sequence C Nodes in c ,judge A Does the node in support c The incident is established. The establishment conditions are: c the logic gate of the node is AND, and c all the child nodes of the node belong to A ; c the logic gate of the node is OR, and c there is a child node belonging to A ; c the logic gate of the node is XOR, and c only one child node belongs to A ; c the logic gate of the node is voting, and c more than half of the nodes belong to A ; S304: Incorporate the node established in S303 into the network C A ;​ S305: repeat steps S302 to S304 until A does not change or C is empty, wherein only when the matching node belongs to A the fault localization conclusion supports the occurrence of the matching node event.

2. The fault tree diagnosis method of a ground data control device according to Claim 1, wherein In S102, the calculation of the matching degrees between each fault tree and the fault phenomenon keyword comprises: the matching degrees between each node of the fault tree and the keyword are calculated respectively, the node with the largest matching degree is marked as a matching node, and the matching degree of the matching node is taken as the matching degree of the fault tree; the matching degree takes a value ranging from 0 to 1.

3. The fault tree diagnosis method of a ground data control device according to Claim 1, wherein The event reasoning step in S2 comprises completing fault judgment in an automatic manner or an interactive manner.

4. The fault tree diagnosis method of a ground data control device according to Claim 3, wherein When the event reasoning is performed in the automatic manner: first, monitoring point data is acquired according to input configuration; then, a Boolean output is acquired by calling a rule logic; finally, whether a fault occurs is judged according to the output of the judgment rule, wherein true represents that the fault occurs, and false represents that the fault does not occur.

5. The fault tree diagnosis method of a ground data control device according to Claim 3, wherein When the event reasoning is performed in the interactive manner: first, monitoring point data that can be automatically acquired is collected; then, prompt information of monitoring points that need to be interactively acquired is pushed to a user; then, the user acquires monitoring point data by visual inspection or instrument means according to the prompt information; then, a Boolean output is acquired by calling a rule logic; finally, whether a fault occurs is judged according to the output of the judgment rule, wherein true represents that the fault occurs, and false represents that the fault does not occur.

6. The fault tree diagnosis method of a ground data system according to claim 1, wherein, In the process of positioning the device fault in S2, the search of the child nodes and the grandchildren nodes of the fault tree is performed based on a depth-first strategy, when a basic event or an event non-occurrence event is encountered, backtracking is performed, when the matching node is reached and all the child nodes of the matching node have been traversed, the whole process is ended.

7. The fault tree diagnosis method of a ground data system according to claim 1, wherein, The fault mode comprises a name, details, a device node, mechanism analysis, an influence domain and disposal suggestions.

8. The fault tree diagnosis method of a ground data control device according to Claim 1, wherein The fault phenomenon comprises a fault description and a keyword.

Images