System failure minimum cut set generation method based on GO-FLOW
By using graph theory traversal and binary tree reverse traversal methods based on GO-FLOW, the minimum cut set of system failure is generated, which solves the problem of low generation efficiency in the existing technology and realizes efficient and accurate system failure analysis.
Patent Information
- Application Number
- CN202410908138.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-08
AI Technical Summary
Existing GO-FLOW methods cannot efficiently generate the minimum cut set of system failures, and existing methods require secondary refinement, which is cumbersome and error-prone.
By using a graph theory traversal search based on GO-FLOW, common signals are identified and renamed to generate the minimum path set. Through repeated signal decomposition and Boolean logic operations, combined with reverse traversal of a binary tree, the minimum cut set of system failure is generated.
It achieves efficient generation of the minimum cut set of system failure, improves the accuracy and compatibility of analysis results, simplifies the calculation process, and improves analysis efficiency.
Smart Images

Figure CN119005772A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of reliability engineering and system safety analysis, and in particular relates to a method for generating a system failure minimum cut set based on GO-FLOW. Background Art
[0002] The GO-FLOW method is a success-oriented system reliability and safety analysis method that can describe the complex dynamic sequential operation characteristics of the system under multi-stage mission profiles and simulate the life aging effects and maintenance strategies of equipment. GO-FLOW describes the potential or actual existence of actual physical quantities or information with sequential signal flows (such as current, water flow, control signal flow, etc.), and constructs a system dynamic reliability and risk model through the concrete expression of operators and signal lines. The generated GO-FLOW model is similar to the system flow chart, which is convenient for model verification and confirmation. At the same time, the GO-FLOW model supports multi-target signal output and can be integrated into the online monitoring of system process parameters. It has a strong model expandability and generalization capability and is currently widely used in nuclear energy, aerospace, transportation, petrochemicals, logistics and other fields.
[0003] However, the prior art found that the existing GO-FLOW method program lacks a qualitative analysis expression of the minimum cut set of system failure (ZYHe, J.Yang, YMHong. A flow-directed minimal path sets method for success path planning and performance analysis. Nuclear Engineering and Technology, 2024, 56(5): 1603-1618.), and instead the analysis results must be refined twice, which is a cumbersome and error-prone process. Summary of the invention
[0004] Aiming at the problems that the existing GO-FLOW method program cannot obtain the minimum cut set of system failure and the generation efficiency of the whole set of system minimum cut sets is low due to the state combination method, the present invention proposes a method for generating the minimum cut set of system failure based on GO-FLOW. The method realizes the common signal recognition and renaming processing through the GO-FLOW graph traversal search, and obtains the whole set of minimum path sets for the success of system tasks; obtains the whole set of non-intersecting minimum path sets after decomposition of repeated signals in the whole set of minimum path sets; obtains the combination of system failure cut sets according to the inversion of the whole set of minimum path sets signals and Boolean logic operations; according to the combination of system failure cut sets, realizes the reconstruction analysis of repeated signals through the reverse traversal of binary trees from bottom to top, and realizes the generation of the whole set of system modular minimum cut sets in combination with the inductive merging calculation of each sub-node; for the modular minimum cut sets, derives the system minimum cut sets and their occurrence probability through signal renaming back restoration and signal strength calculation, and then realizes the comprehensive analysis and expression of the system success minimum path sets and the system failure minimum cut sets.
[0005] The first object of the present invention is to provide a method for obtaining the complete set of system minimum path sets based on graph theory traversal search and ideal signal renaming, which can realize system reliability analysis from the perspective of system mission success.
[0006] The second object of the present invention is to provide a method for efficiently generating system failure minimum cut sets based on direct conversion of the full set of minimum path sets, so as to ensure that the GO-FLOW analysis results are well compatible with the existing failure tree analysis method and probabilistic safety evaluation system.
[0007] The present invention is achieved by at least one of the following technical solutions.
[0008] A method for generating a system failure minimum cut set based on GO-FLOW, comprising the following steps:
[0009] S1. Based on the system GO-FLOW model graph, through graph theory traversal search, common signal recognition and renaming processing are identified to obtain the complete set of minimum paths for system task success;
[0010] S2, according to the order of repeated signals in the minimum path set, perform binary tree decomposition on the repeated signals in the minimum path set from top to bottom to obtain the non-overlapping minimum path set after deduplication;
[0011] S3, based on the complete set of disjoint minimum path sets, sequentially negate the logic operations of the signals in the complete set of minimum path sets, and obtain the combination of system failure cut sets through Boolean logic operations;
[0012] S4. Based on the combination of system failure cut sets, the reconstructed analysis of repeated signals is realized through reverse traversal of the binary tree, and the induction and merging of each child node in the binary tree are combined to simplify the calculation and generate the complete set of system modular minimum cut sets;
[0013] S5. For modular minimum cut sets, the final set of system failure minimum cut sets and their occurrence probability are derived through signal renaming, backtracking and signal strength calculation.
[0014] Furthermore, in step S1, the GO-FLOW model diagram is a literal translation of the reliability logic of the system flow structure diagram P&ID diagram. The GO-FLOW model diagram consists of a series of GO-FLOW operators and signal lines. The GO-FLOW operator is used to simulate the failure of equipment function requirements or operation time-related failures. The GO-FLOW signal line simulates the real working fluid flow of the system through the signal flow.
[0015] Furthermore, the GO-FLOW model graph traversal search includes the following steps:
[0016] S1.1. According to the final signal identification in the GO-FLOW model file, extract the final output signal of the system for final signal marking;
[0017] The GO-FLOW model file includes the data structure of the GO-FLOW model diagram, and supports the analysis of various data information in the GO-FLOW model diagram; the final output signal of the system is an associated embodiment of the system function or state target, and multiple final signals indicate that there are multiple target outputs in the system;
[0018] S1.2, taking the final output signal of the system as the starting node, reversely follow the GO-FLOW signal flow to realize the final signal marking of all signals, and the signal flow is traced back to the source signal;
[0019] The nodes refer to network nodes represented by various GO-FLOW operators, and the network nodes are connected to each other through signal lines, i.e., edges between nodes, to form a GO-FLOW model graph;
[0020] S1.3, according to the final signal marking of the GO-FLOW signal flow, combined with the signal flow branching and merging criteria, the system common signal identification is realized;
[0021] The common signal refers to the output signal of an operator that is simultaneously used as the input signal of multiple operators and reunited at a downstream node; the common signal is used to mark the signal that appears multiple times in the whole set of system minimum path sets. Affected by the common signal, there is a correlation between the system minimum path sets, and renaming and full probability decomposition processing are required to obtain the correct quantitative analysis results;
[0022] S1.4, according to the signal flow backtracking analysis, identify the GO-FLOW signal composition in each branch flow, and realize the generation of the minimum path set through signal renaming processing;
[0023] The minimum path set refers to the set of all minimum path sets that ensure the success of the system. The minimum path set in GO-FLOW refers to all minimum path set combinations starting from all source signal generators, i.e., GO-FLOW operator No. 25, leading to the final node. During the analysis and calculation process of GO-FLOW, both the path set and the path set global set need to be simplified through Boolean logic to obtain the simplest structural expression.
[0024] Furthermore, the signal renaming process refers to the merging process of one or more ideal signals. Signal renaming is used to simplify the GO-FLOW calculation process and distinguish signals of different states. The renaming rules of the GO-FLOW ideal signal include:
[0025] i) Post-rename: common output signal processing for common functional operators; the common functional operators are GO-FLOW operators No. 21, 26, 27, 39, 35, 37 and 38, which are used to describe functional requirement failures, runtime-related failures and maintenance processes of physical components;
[0026] ii) Pre-rename: rename all input signals of logic operators and special function operators, and the input signals of operator No. 35 where the main input signal has multiple path sets; the logic operators refer to logic gates including AND, OR, and NOT, and the special function operators specifically refer to GO-FLOW operators No. 24, No. 28, and No. 40, which respectively implement differential calculation of signal values, signal delay simulation, and signal multi-stage conversion;
[0027] iii) Process renaming: signal process processing for special function operators and logic NOT gates; the process renaming needs to assign a new number to the signal again on the basis of the pre / post renaming, indicating the signal's differential, delay, inversion or stage interception process; if the second renamed signal and the first renamed signal appear simultaneously in the full set of the system's minimum path set, they need to be marked and distinguished by the second renaming.
[0028] Furthermore, in step S2, the repeated signal refers to the same signal appearing multiple times in multiple different system minimum path sets, and the repeated signal is removed by full probability decomposition to obtain the complete set of non-intersecting system minimum cut sets and the exact solution of system reliability;
[0029] The process of removing repeated signals based on full probability decomposition includes the following steps:
[0030] 2.1. Sort the repeated signals by their frequency in the minimum path set of the system, and extract the signals with the most repetitions from the minimum path set in turn for priority processing;
[0031] 2.2. Treat the repeated signal as a deterministic event Y, and consider two mutually exclusive states: the repeated signal exists (i.e. the event occurs) and the repeated signal does not exist (i.e. the event does not occur);
[0032] When a repeated signal exists, it means that the device function mode or basic event represented by the signal is successful. At this time, P(Y)=1 exists. The repeated signal in the system minimum path set is deleted to obtain the reduced minimum path set. The reduced path set after deleting the repeated signal is not necessarily the minimum path set. It needs to be further simplified by Boolean logic to obtain the minimum path set.
[0033] When the duplicate signal does not exist, it indicates that the device function mode or basic event represented by the signal fails. At this time, P(Y)=0, and all minimum path sets containing the duplicate signal in the system minimum path set set are removed. The minimum path set removed by negating the duplicate signal can no longer lead to the success of the system task. If the duplicate signal exists in all the system minimum path sets, the system minimum path set set after the duplicate signal is deleted will become an empty set, indicating that the system fails when the duplicate signal does not exist.
[0034] 2.3. Perform similar binary tree recursive decomposition on other repeated signals in turn until all repeated signals have been eliminated, and finally obtain the complete set of non-overlapping minimum path sets after deduplication.
[0035] Further, step S3 includes the following steps:
[0036] 3.1. Based on the complete set of the deduplicated minimal path sets obtained in step S2, extract signals from each of the minimal path sets in turn, and construct the inverse event of the basic event by logically negating the signals;
[0037] 3.2. By logically negating the sum of each minimum path set in the complete set of non-intersecting minimum path sets, it is converted into a sum-intersection Boolean logic operation between each negated minimum path set, and then the system failure cut set combination is obtained.
[0038] Further, in step S4, the system failure cut set refers to a disjoint minimum cut set without repeated signal expressions obtained by logically inverting the signals in each component minimum path set in the full list of system minimum path sets and performing random combinations; the actual full list of system failure minimum cut sets should be a combination of system cut sets and repeated signal state decomposition, obtained by reverse traversal of the binary tree.
[0039] Furthermore, the binary tree reverse traversal process includes the following steps:
[0040] 4.1. Based on the combination of system failure cut sets obtained in step S3, the consideration of the success and failure of repeated signals is included again in the reverse order of repeated signal decomposition;
[0041] Furthermore, the left branch represents the failure of the repeated signal, and the minimum modular cut set of the current node is directly obtained by the intersection operation ∩ of the minimum modular cut set of the child node and the repeated signal failure event; on the contrary, the right branch represents the success of the repeated signal, and the minimum modular cut set generated by the child node is directly obtained;
[0042] 4.2. Combine the complete set of minimum cut sets in the left branch nodes with the complete set of minimum cut sets in the right branch nodes, and simplify them through Boolean logic to obtain the complete set of minimum modular cut sets of the intermediate nodes in the previous layer;
[0043] 4.3. Through layer-by-layer iterative calculation and merging simplification, the complete set of minimum modular cut sets for top node system failure is finally obtained.
[0044] Further, in step S5, the modular minimum cut set is a combination of failure modes of the combined signal module;
[0045] The combined signal module refers to a signal that renames multiple unnamed signals together during the signal renaming process; the signals in the combined signal module are traced back and restored to identify the specific failure mode of the device, and the minimum cut set is generated through their combination to achieve the final acquisition of the complete set of system failure minimum cut sets.
[0046] Furthermore, the modular minimum cut set decomposition process includes the following steps:
[0047] S5.1. According to the signal renaming mapping list, trace back and obtain the initial original signal composition of the combined signal module before renaming, and extract the device function and failure mode corresponding to the original signal;
[0048] S5.2. Based on the original signal composition in the modular minimum cut set, a system failure minimum cut set is derived through Boolean logic operation; the minimum cut set is a minimum device failure mode combination for system failure, and no further simplification is required;
[0049] S5.3. By calculating and analyzing the signal strength, the occurrence probability of each minimum cut set is obtained, and quantitative analysis of the minimum cut set is achieved.
[0050] The present invention has the following beneficial effects compared with the prior art:
[0051] 1. The GO-FLOW minimum path set acquisition method based on graph theory traversal search provided by the present invention can not only provide new insights and insights from the perspective of system mission success in system reliability and safety analysis, but also has unique advantages such as small number of path sets, low computational complexity, and concrete expression of signal flow (working fluid flow). The analysis efficiency of the algorithm engine can be greatly improved.
[0052] 2. The GO-FLOW minimum cut set generation method based on repeated signal decomposition and reconstruction provided by the present invention realizes the rapid transformation from the complete set of system success minimum path sets to the complete set of system failure minimum cut sets through the signal inversion logic operation, cutting combination and inductive merging calculation in the complete set of minimum path sets, which breaks through the qualitative analysis expression and quantitative calculation of the system failure minimum cut sets by the GO-FLOW method, and can promote the compatibility of the GO-FLOW method with the existing failure tree analysis method and the probabilistic safety evaluation system. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 is a system schematic diagram of an embodiment;
[0054] Figure 2 GO-FLOW model diagram of the embodiment system;
[0055] Figure 3 It is the decomposition diagram of the repeated signal of the whole set of the minimum path set of the system;
[0056] Figure 4 This is a schematic diagram of the minimum cut set transformation principle of the embodiment;
[0057] Figure 5 It is a reverse traversal graph of a binary tree of an embodiment;
[0058] Figure 6 The present invention is a flowchart of a method for generating a minimum cut set of system failure based on GO-FLOW. DETAILED DESCRIPTION
[0059] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in 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. It should be understood that the specific embodiments described are only used to explain the present application and are not used to limit the present application.
[0060] like Figure 6The method for generating the minimum cut set of system failure based on GO-FLOW is shown, which finally obtains the minimum cut set of system failure through repeated signal decomposition in the minimum path set, negation logic operation of the non-intersecting minimum path set signal, minimum cut set cutting combination, binary tree reverse traversal merging calculation, and original signal backtracking decomposition, and then establishes a comprehensive analysis expression of the system success minimum path set and the system failure minimum cut set, which specifically includes the following steps:
[0061] S1. Based on the system GO-FLOW model graph, through graph theory traversal search, common signal recognition and renaming processing are realized to obtain the complete set of minimum paths for system task success;
[0062] The GO-FLOW model diagram is a direct translation of the reliability logic of the system flow structure diagram (PID diagram). The GO-FLOW model diagram consists of a series of GO-FLOW operators and signal lines. The GO-FLOW operator is used to simulate the failure of equipment function requirements or operation time-related failures. The GO-FLOW signal line simulates the real working fluid flow of the system through the signal flow.
[0063] Furthermore, the GO-FLOW graph traversal search is implemented by the following steps:
[0064] S2.1. According to the final signal identification in the GO-FLOW model file, extract the final output signal of the system for final signal labeling;
[0065] The GO-FLOW model file is a data structure embodiment of the GO-FLOW model diagram. It is a structured data file that supports the analysis of various types of data information in the GO-FLOW model diagram. The system final output signal is an associated embodiment of the system function or state target. Multiple final signals indicate that there are multiple target outputs in the system.
[0066] S2.2, taking the final output signal of the system as the starting node, reversely follow the GO-FLOW signal flow to realize the final signal marking of all signals, and the signal flow is traced back to the source signal (GO-FLOW operator No. 25);
[0067] The node refers to a network node abstracted from each GO-FLOW operator, which is connected to each other through signal lines (edges between nodes) to form a GO-FLOW model graph; in order to facilitate the effective identification of the GO-FLOW operator, the GO-FLOW operator encoding is consistent with the GO-FLOW signal flow encoding.
[0068] S2.3, according to the final signal marking of the GO-FLOW signal flow, combined with the signal flow branching and merging criteria, the system common signal identification is realized;
[0069] The common signal refers to the output signal of an operator that is simultaneously used as the input signal of multiple operators and reunited at a downstream node. The common signal is used to mark signals that appear multiple times in the entire set of system minimum paths. Affected by the common signal, there is a correlation between the minimum paths of the system, and renaming and full probability decomposition are required to obtain the correct quantitative analysis results.
[0070] S2.4, according to the signal flow backtracking analysis, identify the GO-FLOW signal composition in each branch flow, and realize the generation of the minimum path set through signal renaming processing;
[0071] The signal renaming process refers to the merging process of one or more ideal signals. Signal renaming can be used to simplify the GO-FLOW calculation process and distinguish signals in different states. The renaming rules of GO-FLOW ideal signals include:
[0072] i) Post-rename: Common output signal processing for general function operators. The general function operators are GO-FLOW operators No. 21, 26, 27, 39, 35, 37 and 38, which are usually used to describe functional requirement failures, runtime-related failures and maintenance processes of physical components.
[0073] ii) Pre-rename: Rename all input signals of logic operators and special function operators, as well as the input signals of operator No. 35 where the main input signal has multiple path sets. The logic operators refer to logic gates including "AND", "OR" and "NOT", while the special function operators specifically refer to GO-FLOW operators No. 24, No. 28 and No. 40, which respectively implement differential calculation of signal values, signal delay simulation and signal multi-stage conversion.
[0074] iii) Process renaming: signal process processing for special function operators and logic "not" gates. The process renaming needs to assign a new number to the signal again on the basis of the pre- / post-renaming, indicating the signal's differential, delay, inversion or stage interception process. Since there is a strong correlation between the second renamed signal and the first renamed signal, and they may appear at the same time in the system's minimum path set, they need to be marked and distinguished through the second renaming.
[0075] The minimum path set is the set of all minimum path sets that ensure the success of the system. The minimum path set in GO-FLOW refers to all minimum path set combinations that start from all source signal generators (GO-FLOW operator No. 25) and lead to the final node. During the analysis and calculation process of GO-FLOW, both the path set and the path set need to be simplified through Boolean logic to obtain the simplest structural expression.
[0076] S2, according to the order of repeated signals in the minimum path set, perform binary tree decomposition on the repeated signals in the minimum path set from top to bottom to obtain the non-overlapping minimum path set after deduplication;
[0077] The repeated signal refers to the same signal that appears multiple times in multiple different system minimum path sets. Due to the correlation of repeated signals, the complete set of system minimum path sets cannot be directly calculated. It is necessary to first remove the repeated signals through full probability decomposition to obtain the complete set of non-intersecting system minimum cut sets and the exact solution of system reliability.
[0078] Furthermore, the repeated signal removal process based on full probability decomposition is specifically implemented by the following steps:
[0079] 3.1. Sort the repeated signals by their frequency in the minimum path set of the system, and extract the signals with the most repetitions from the minimum path set in turn for priority processing;
[0080] 3.2. Treat the repeated signal as a deterministic event (event Y), considering two mutually exclusive states: the repeated signal exists (event occurs) and does not exist (event does not occur);
[0081] according to Figure 6 As shown in step 1, MPS represents the complete set of minimum paths in the system. Sn represents the minimum path set contained in the signal flow Sn, S 1 =0 and S 1 =1 respectively indicates the failure (non-existence) and success (existence) branches of the duplicate signal (#1) to be decomposed. S1 In, MPS S1 =MPS S2 ∪(S 1 ∩MPS S2 ) represents the parent node MPS S1 Decomposition into MPS via binary tree S2 and MPS S5 , because the left leaf node indicates that the repeated signal fails, and the right leaf node indicates that the repeated signal succeeds, so S 1 =1 to be connected with the right leaf node MPS S5 Similarly, the subsequent leaf nodes are decomposed into the minimum path set MPS of the irredecomposable system through binary tree decomposition and Boolean logic simplification. S3 、MPS S4 、MPS S6 、MPS S7 .
[0082] When a repeated signal exists, it means that the device function mode or basic event represented by the signal is successful. At this time, P(Y)=1 exists, and the repeated signal in the system minimum path set can be deleted to obtain the reduced minimum path set. It is worth noting that the reduced (deleted repeated signal) path set is not necessarily the minimum path set, and further Boolean logic simplification is required to obtain the minimum path set.
[0083] When the duplicate signal does not exist, it means that the device function mode or basic event represented by the signal fails. At this time, P(Y) = 0, and all the minimum path sets containing the duplicate signal in the system minimum path set are removed. The removed minimum path sets (after the duplicate signal is negated) can no longer lead to the success of the system task. If the duplicate signal exists in all the system minimum path sets, the system minimum path set after the duplicate signal is deleted will become an empty set, indicating that the system fails when the duplicate signal does not exist.
[0084] 3.3. Perform similar binary tree recursive decomposition on other repeated signals in turn until all repeated signals have been eliminated, and finally obtain the complete set of non-overlapping minimum path sets after deduplication.
[0085] S3, based on the complete set of disjoint minimum path sets, sequentially negate the logic operations of the signals in the complete set of minimum path sets, and obtain the system failure cut set combination through Boolean logic operations; the minimum cut set conversion process is implemented by the following steps:
[0086] 4.1. Based on the complete set of the deduplicated minimal path sets obtained in step S2, extract signals from each of the minimal path sets in turn, and construct the inverse event of the basic event by logically negating the signals;
[0087] 4.2. By logically negating the sum of each minimum path set in the complete set of non-intersecting minimum path sets, it is converted into a "sum-to-intersection" Boolean logic operation between each negated minimum path set, and then the system failure cut set combination is obtained;
[0088] S4. According to the combination of system failure cut sets, the reconstruction analysis of repeated signals is realized through the reverse traversal of the binary tree. The induction and merging of each child node in the binary tree are combined to simplify the calculation and generate the complete set of system modular minimum cut sets.
[0089] In step S4, the system failure cut set refers to a disjoint minimum cut set without repeated signal expression obtained by logically inverting the signals in each component minimum path set in the system minimum path set list and performing random combinations. The actual system failure minimum cut set should be a combination of the system cut set and the repeated signal state decomposition, which can be obtained by reverse traversing the binary tree.
[0090] according to Figure 6 As shown in step 4 of MCSs represents the complete set of minimal cut sets of the system, S 2 =0 and S 2 =1 respectively indicates the failure (non-existence) and success (existence) branches of the repeated signal to be decomposed (#2). System minimum cut set MCS s3、 MCS s4、 MCS s6、 MCS s7 The system minimum path set MPS s3、 MPS s4、 MPS s6、 MPS s7 Obtained by logical inversion. MCS S2 =(MCS S3 ∩S 2 )∪MCS S4 It represents the minimum cut set MCS of the left leaf node of the system failure S3 After taking the intersection with the repeated signal #2, it is successfully combined with the system's right leaf node minimum cut set MCS S4 The system cut set obtained after taking the union may not be the minimum cut set. It is necessary to use Boolean logic simplification to obtain the system minimum cut set MCS S2 Similarly, the complete set of minimum cut sets of the system MCS can be obtained by reverse traversal of the binary tree and Boolean logic simplification. S .
[0091] Furthermore, the binary tree reverse traversal process is implemented by the following steps:
[0092] 5.1. Based on the complete set of system cuts obtained in step S3, the consideration of the success and failure of the repeated signal is included again in the reverse order of the repeated signal decomposition;
[0093] Furthermore, the left branch represents the failure of the repeated signal, and the minimum modular cut set of the current node can be directly obtained by the intersection operation (∩) of the minimum modular cut set of the child node and the repeated signal failure event; on the contrary, the right branch represents the success of the repeated signal, and the minimum modular cut set generated by the child node is directly obtained;
[0094] 5.2. Combine the complete set of minimum cut sets in the left branch nodes with the complete set of minimum cut sets in the right branch nodes (∪), and obtain the complete set of minimum modular cut sets of the intermediate nodes in the previous layer through Boolean logic simplification;
[0095] 5.3. Through layer-by-layer iterative calculation and merging simplification, the complete set of minimum modular cut sets for top node system failure is finally obtained.
[0096] S5. For modular minimum cut sets, derive the final set of system failure minimum cut sets and their occurrence probability through signal renaming, backtracking and signal strength calculation. The modular minimum cut set is a combination of failure modes of combined signal modules;
[0097] Furthermore, the combined signal module refers to a signal that renames multiple unnamed signals (such as multiple GO-FLOW operator output signals in a series structure) together in order to simplify calculations during the signal renaming process. In order to achieve the final acquisition of the complete set of system failure minimum cut sets, it is necessary to further trace back and restore the signals in the combined signal module, identify the specific failure mode of the given equipment, and generate the minimum cut set through their combination.
[0098] Furthermore, the modular minimum cut set decomposition process is implemented by the following steps:
[0099] S6.1. According to the signal renaming mapping list, trace back and obtain the initial original signal composition of the combined signal module before renaming, and extract the device function and failure mode corresponding to the original signal;
[0100] S6.2. According to the original signal composition in the modular minimum cut set, the system failure minimum cut set is derived through Boolean logic operation; the minimum cut set is the minimum device failure mode combination for system failure, and no further simplification processing is required;
[0101] S6.3. By calculating and analyzing the signal strength, the occurrence probability of each minimum cut set is obtained, and quantitative analysis of the minimum cut set is achieved.
[0102] The system minimum cut set consists of the minimum combination of equipment failure modes that cause system failure. By reading the characteristic parameters of the GO-FLOW operator, the reliability parameters of the system equipment can be obtained, including the success probability and the failure probability. The occurrence probability of each minimum cut set is equal to the cumulative multiplication of the failure probabilities of the equipment included in the cut set.
[0103] As a specific embodiment, this embodiment takes a hypothetical water supply system as an example to illustrate the relevant method proposed by the present invention. Figure 1 As shown in the figure, the system of this embodiment is composed of a water supply storage tank (CCST), three water supply pumps (MDP-a, MDP-b, TDP), four electric valves (MOV-1 to MOV-4) and several check valves (CV-0 to CV-4). The water supply system supplies water to users through four pipelines. When the system is operating normally, it should ensure that at least one pipeline is successfully put into operation.
[0104] Step 1: Obtain the complete set of system minimum paths
[0105] Refer to the system flow chart and establish the system GO-FLOW model as follows Figure 2 As shown. The GO-FLOW model uses operator No. 25 to represent the intact state of the water source, storage tank, check valve (CV-0~CV-4) and other equipment in the storage tank, which is represented by operator No. 21. Assuming that the water supply system is not put into operation under normal circumstances and is only put into use in an emergency, the water supply system can be regarded as a standby state, and the water supply pump and electric valve can be simulated in a normally closed state through operator No. 26. Considering the relatively low reliability of the water pump and its failure over time, it is simulated in the model by operator No. 35. The final signal of the system is 20. Through graph theory traversal search, the complete set of the minimum path set of the system is obtained, see Table 1. The signal renaming and restoration traceability records in the process of obtaining the complete set of the minimum path set of the system are shown in Table 2. The success rate and failure rate of each equipment component are shown in Table 6.
[0106] Table 1 The complete set of system minimum paths
[0107]
[0108] Table 2 Signal renaming and restoration traceback records
[0109] Original signal Rename signal 1,2,3 24 5,8 25 11 26 4,7,10 27 6,9,13 28 12 29 19 30 18 31 17 32 16 33
[0110] Step 2: Minimum path set full set repeated signal decomposition
[0111] According to the analysis results in Table 1, the system's final output signal #20 contains the following 8 minimum path sets: {[24,28,30], [24,25,29,30], [24,28,31], [24,25,29,31], [24,25,26,32], [24,27,32], [24,25,26,33], [24,27,33]}.
[0112] The system minimum path set full set repeated signal decomposition process refers to Figure 3 and Figure 6. According to the above minimum path set (AR1), #24 is the most frequent repeated signal (repeated 8 times). According to the full probability decomposition theory, considering the existence and non-existence of signal #24, the minimum path set is decomposed into the left node AR3 and the right node AR2. The right branch node indicates that the repeated signal #24 exists (that is, the event represented by it successfully occurs). In this case, the repeated signal #24 can be deleted from all minimum path sets. Therefore, the minimum path set of the right branch node AR2 becomes {[28,30], [25,29,30], [28,31], [25,29,31], [25,26,32], [27,32], [25,26,33], [27,33]}. The left branch node indicates that the repeated signal #24 does not exist (that is, the event represented by it fails to occur), so that all minimum path sets containing the repeated signal #24 will be eliminated, so the left branch node AR3 becomes an empty set.
[0113] Furthermore, by identifying repeated signals in the minimum path set in AR2, it is found that signal #25 has the most repetitions (repeated 4 times). Similarly, according to the presence or absence of signal #25, AR2 is decomposed into two branches, AR4 and AR5, and the minimum path set of the right node AR4 is obtained as follows: {[28,30], [29,30], [28,31], [29,31], [26,32], [27,32], [26,33], [27,33]}, and the minimum path set of the left node AR5 is obtained as follows: {[28,30], [28,31], [27,32], [27,33]}. Traverse downward in sequence. If a path set is included in another path set in the path set of a certain node, it is necessary to first use Boolean logic simplification to absorb the included path set, so as to obtain the minimum path set of the node. The complete set of minimum path sets of sub-nodes at each level obtained in the repeated signal decomposition process is shown in Table 3.
[0114] Table 3 The complete set of minimum path sets of sub-nodes at all levels obtained during repeated signal decomposition
[0115]
[0116]
[0117] Step 3: Minimum cut set conversion
[0118] After decomposing all duplicate signals in the minimum path set of the system, all decomposed branch nodes will no longer contain duplicate signals. At this time, a device failure (logical negation) can be selected from the minimum path set in each leaf node after decomposition. These device failure combinations will form a cut set. For the principle of combined cutting, see Figure 4 . Figure 4In the above, MPS#1 and MPS#2 represent the first and second minimum paths in the system minimum path set, respectively. If the system minimum path set contains N minimum paths, then MPS#N represents the Nth minimum path in the system minimum path set. Similarly, MCS#1 represents the first minimum cut set in the system minimum cut set. If the system minimum cut set contains M minimum equipment failure mode combinations that cause system failure, then MCS#M represents the Mth minimum cut set in the system minimum cut set.
[0119] For example, the node AR26 whose original minimum path set is {
[30] ,
[31] ,
[32] ,
[33] }, obtains the cut set after negation as follows: The original minimum path set is an empty set The node AR19 is negated and the cut set obtained is the whole set U. By traversing all possible failure combinations, the whole set of the minimum cut sets of the reduced system failure after decomposition is obtained. If the whole set of the minimum path set of the decomposed system is an empty set, it means that there is no successful path set in the system after the repeated signal is decomposed, which leads to system failure, and the minimum cut set of the system is also an empty set.
[0120] Step 4: Obtain the complete set of system minimum modular cut sets
[0121] The conversion in step 3 above obtains the complete set of simplified minimum cut sets without repeated signal expressions. The complete set of actual system failure minimum cut sets is a combination of repeated signals and the complete set of simplified minimum cut sets after deduplication. It is necessary to perform a reverse traversal of the binary tree (bottom-up inductive analysis) based on the complete set of simplified minimum cut sets after deduplication and combined with the decomposition of repeated signal states. For the reverse traversal process of the binary tree, see Figure 5 During the reverse traversal of the binary tree, the minimum cut set subset recorded in the left node is intersected with the repeated signal failure, and then merged and simplified with the minimum cut set subset in the right node (coupled with the repeated signal failure), iterated layer by layer, and finally the minimum cut sets of all leaf nodes are gathered to the top node, and the complete minimum cut set of the system can be obtained.
[0122] As a specific embodiment, the reverse traversal of the binary tree is performed starting from leaf nodes AR22 and AR23. The cut set of leaf node AR22 is The cut set of leaf node AR23 is Failure with repeated signal After taking the intersection Then Added to node AR21, the cut set of node AR21 is Similarly, disable the cut set and repeated signal of AR21 After taking the intersection Cut set with AR20 Add to node AR14, and obtain the cut set of AR14: At this point, the cut set of AR14 is not the minimum cut set. It is necessary to simplify the minimum modular cut set of AR14 through Boolean logic: In this way, we iterate upward layer by layer, and finally gather the minimum cut sets of all leaf nodes to the top node AR1, and then we can get the minimum modular cut set of the system as The complete set of minimum modular cut sets for each child node during the reverse traversal of the binary tree is shown in Table 4.
[0123] Table 4 Minimum modular cut sets for each node in binary tree traversal
[0124]
[0125]
[0126] Step 5: Generate the complete set of system failure minimum cut sets
[0127] According to the signal composition in the system minimum cut set and the signal renaming restoration traceability record chain list (see Table 2), the minimum equipment failure mode combination in the minimum cut set is obtained by analysis, see Table 5. The system equipment success probability and failure probability parameters used for the minimum cut set occurrence probability calculation are shown in Table 6.
[0128] Table 5. The complete set of minimum cut sets for system failure
[0129]
[0130]
[0131]
[0132] Table 6 System equipment success probability and failure probability
[0133] Signal number Device Name Success rate Failure rate 2 CCST 0.999998 0.000002 3 CV-0 0.999996 0.000004 10 CV-1 0.999996 0.000004 11 CV-2 0.999996 0.000004 12 CV-3 0.999996 0.000004 13 CV-4 0.999996 0.000004 4 MDP-a 0.99835 0.00165 6 MDP-b 0.99835 0.00165 5 TDP 0.9685 0.0315 16 MOV-1 0.9957 0.0043 17 MOV-2 0.9957 0.0043 18 MOV-3 0.9957 0.0043 19 MOV-4 0.9957 0.0043
[0134] The minimum cut sets of the system equipment reliability parameters are finally obtained, which are 19, namely: {[CCST failure], [CV-0 failure], [MDP-b failure, MDP-a failure, TDP failure], [MDP-b failure, CV-1 failure, TDP failure], [CV-4 failure, MDP-a failure, TDP failure], [CV-4 failure, CV-1 failure, TDP failure], [MOV-4 failure, MOV-3 failure, MDP-a failure, TDP failure], [MOV-4 failure, MOV-3 failure, CV-1 failure, TDP failure], [MOV-2 failure, MOV-1 failure, MDP-b failure, TDP failure], [MOV-2 failure, MOV-1 failure, CV-4 failure, TDP failure], [CV-3 failure, MDP-4 failure, TDP failure], [MDP-2 failure, MDP-1 failure, TDP failure], [MDP-2 failure, MDP-1 failure, TDP failure], [MDP-2 failure, CV-1 failure, TDP failure], [MDP-3 failure, CV-4 failure, TDP failure], [MDP-4 failure, MDP-3 failure, TDP failure], [MDP- -b failure, MOV-1 failure, MOV-2 failure], [CV-3 failure, CV-4 failure, MOV-1 failure, MOV-2 failure], [MOV-4 failure, MOV-3 failure, MOV-1 failure, MOV-2 failure], [CV-2 failure, MDP-a failure, CV-3 failure, MDP-b failure], [CV-2 failure, MDP-a failure, CV-3 failure, CV-4 failure], [CV-2 failure, CV-1 failure, MDP-b failure, CV-3 failure], [CV-1 failure, CV-2 failure, CV-3 failure, CV-4 failure], [MOV-4 failure, MOV-3 failure, CV-2 failure, MDP-a failure], [MOV-4 failure, MOV-3 failure, CV-2 failure, CV-1 failure]}.
[0135] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well.
Claims
1. A method for generating a minimum cut set of system failure based on GO-FLOW, characterized in that: The following steps are involved: S1. Based on the system GO-FLOW model graph, through graph theory traversal search, common signal recognition and renaming processing are identified to obtain the complete set of minimum paths for system task success; S2, according to the order of repeated signals in the minimum path set, perform binary tree decomposition on the repeated signals in the minimum path set from top to bottom to obtain the non-overlapping minimum path set after deduplication; S3, based on the complete set of disjoint minimum path sets, sequentially negate the logic operations of the signals in the complete set of minimum path sets, and obtain the combination of system failure cut sets through Boolean logic operations; S4. Based on the combination of system failure cut sets, the reconstructed analysis of repeated signals is realized through reverse traversal of the binary tree, and the induction and merging of each child node in the binary tree are combined to simplify the calculation and generate the complete set of system modular minimum cut sets; S5. For modular minimum cut sets, the final set of system failure minimum cut sets and their occurrence probability are derived through signal renaming, backtracking and signal strength calculation.
2. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 1, characterized in that: In step S1, the GO-FLOW model diagram is a direct translation of the reliability logic of the system flow structure diagram P&ID diagram. The GO-FLOW model diagram consists of a series of GO-FLOW operators and signal lines. The GO-FLOW operator is used to simulate the failure of equipment function requirements or operation time related failures. The GO-FLOW signal line simulates the real working fluid flow of the system through the signal flow.
3. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 1, characterized in that: The GO-FLOW model graph traversal search includes the following steps: S1.
1. According to the final signal identification in the GO-FLOW model file, extract the final output signal of the system for final signal marking; The GO-FLOW model file includes the data structure of the GO-FLOW model diagram, and supports the analysis of various data information in the GO-FLOW model diagram; the final output signal of the system is an associated embodiment of the system function or state target, and multiple final signals indicate that there are multiple target outputs in the system; S1.2, taking the final output signal of the system as the starting node, reversely follow the GO-FLOW signal flow to realize the final signal marking of all signals, and the signal flow is traced back to the source signal; The nodes refer to network nodes represented by various GO-FLOW operators, and the network nodes are connected to each other through signal lines, i.e., edges between nodes, to form a GO-FLOW model graph; S1.3, according to the final signal marking of the GO-FLOW signal flow, combined with the signal flow branching and merging criteria, the system common signal identification is realized; The common signal refers to the output signal of an operator that is simultaneously used as the input signal of multiple operators and reunited at a downstream node; the common signal is used to mark the signal that appears multiple times in the whole set of system minimum path sets. Affected by the common signal, there is a correlation between the system minimum path sets, and renaming and full probability decomposition processing are required to obtain the correct quantitative analysis results; S1.4, according to the signal flow backtracking analysis, identify the GO-FLOW signal composition in each branch flow, and realize the generation of the minimum path set through signal renaming processing; The minimum path set refers to the set of all minimum path sets that ensure the success of the system. The minimum path set in GO-FLOW refers to all minimum path set combinations starting from all source signal generators, i.e., GO-FLOW operator No. 25, leading to the final node. During the analysis and calculation process of GO-FLOW, both the path set and the path set global set need to be simplified through Boolean logic to obtain the simplest structural expression.
4. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 3, characterized in that: The signal renaming process refers to the merging process of one or more ideal signals. Signal renaming is used to simplify the GO-FLOW calculation process and distinguish signals of different states. The renaming rules of GO-FLOW ideal signals include: i) Post-rename: common output signal processing for common functional operators; the common functional operators are GO-FLOW operators No. 21, 26, 27, 39, 35, 37 and 38, which are used to describe functional requirement failures, runtime-related failures and maintenance processes of physical components; ii) Pre-rename: rename all input signals of logic operators and special function operators, and the input signals of operator No. 35 where the main input signal has multiple path sets; the logic operators refer to logic gates including AND, OR, and NOT, and the special function operators specifically refer to GO-FLOW operators No. 24, No. 28, and No. 40, which respectively implement differential calculation of signal values, signal delay simulation, and signal multi-stage conversion; iii) Process renaming: signal process processing for special function operators and logic NOT gates; the process renaming needs to assign a new number to the signal again on the basis of the pre / post renaming, indicating the signal's differential, delay, inversion or stage interception process; if the second renamed signal and the first renamed signal appear simultaneously in the full set of the system's minimum path set, they need to be marked and distinguished by the second renaming.
5. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 1, characterized in that: In step S2, the repeated signal refers to the same signal appearing multiple times in multiple different system minimum path sets. The repeated signal is removed by full probability decomposition to obtain the complete set of non-intersecting system minimum cut sets and the exact solution of system reliability; The process of removing repeated signals based on full probability decomposition includes the following steps: 2.
1. Sort the repeated signals by their frequency in the minimum path set of the system, and extract the signals with the most repetitions from the minimum path set in turn for priority processing; 2.
2. Treat the repeated signal as a deterministic event Y, and consider two mutually exclusive states: the repeated signal exists (i.e. the event occurs) and the repeated signal does not exist (i.e. the event does not occur); When a repeated signal exists, it means that the device function mode or basic event represented by the signal is successful. At this time, P(Y)=1 exists. The repeated signal in the system minimum path set is deleted to obtain the reduced minimum path set. The reduced path set after deleting the repeated signal is not necessarily the minimum path set. It needs to be further simplified by Boolean logic to obtain the minimum path set. When the duplicate signal does not exist, it indicates that the device function mode or basic event represented by the signal fails. At this time, P(Y)=0, and all minimum path sets containing the duplicate signal in the system minimum path set set are removed. The minimum path set removed by negating the duplicate signal can no longer lead to the success of the system task. If the duplicate signal exists in all the system minimum path sets, the system minimum path set set after the duplicate signal is deleted will become an empty set, indicating that the system fails when the duplicate signal does not exist. 2.
3. Perform similar binary tree recursive decomposition on other repeated signals in turn until all repeated signals have been eliminated, and finally obtain the complete set of non-overlapping minimum path sets after deduplication.
6. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 1, characterized in that: Step S3 includes the following steps: 3.
1. Based on the complete set of the deduplicated minimal path sets obtained in step S2, extract signals from each of the minimal path sets in turn, and construct the inverse event of the basic event by logically negating the signals; 3.
2. By logically negating the sum of each minimum path set in the complete set of non-intersecting minimum path sets, it is converted into a sum-intersection Boolean logic operation between each negated minimum path set, and then the system failure cut set combination is obtained.
7. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 1, characterized in that: In step S4, the system failure cut set refers to a disjoint minimum cut set without repeated signal expressions obtained by logically inverting the signals in each component minimum path set in the full list of system minimum path sets and performing random combinations; the actual full list of system failure minimum cut sets should be a combination of system cut sets and repeated signal state decomposition, obtained by reverse traversal of the binary tree.
8. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 1, characterized in that: The binary tree reverse traversal process, The following steps are involved: 4.
1. Based on the combination of system failure cut sets obtained in step S3, the consideration of the success and failure of repeated signals is included again in the reverse order of repeated signal decomposition; Furthermore, the left branch represents the failure of repeated signals, and the complete set of the minimum modular cut sets of the current node is obtained directly through the intersection operation ∩ of the complete set of the minimum modular cut sets of the child nodes and the repeated signal failure event; In contrast, the right branch represents the success of the repeated signal, and directly obtains the full set of minimum modular cut sets generated by the child nodes; 4.
2. Combine the complete set of minimum cut sets in the left branch nodes with the complete set of minimum cut sets in the right branch nodes, and simplify them through Boolean logic to obtain the complete set of minimum modular cut sets of the intermediate nodes in the previous layer; 4.
3. Through layer-by-layer iterative calculation and merging simplification, the complete set of minimum modular cut sets for top node system failure is finally obtained.
9. A method for generating a system failure minimum cut set based on GO-FLOW according to any one of claims 1 to 8, characterized in that: In step S5, the modular minimum cut set is a combination of failure modes of the combined signal module; The combined signal module refers to a signal that renames multiple unnamed signals together during the signal renaming process; the signals in the combined signal module are traced back and restored to identify the specific failure mode of the device, and the minimum cut set is generated through their combination to achieve the final acquisition of the complete set of system failure minimum cut sets.
10. The method for generating a system failure minimum cut set based on GO-FLOW according to claim 9, characterized in that: The modular minimum cut set deconstruction process includes the following steps: S5.
1. According to the signal renaming mapping list, trace back and obtain the initial original signal composition of the combined signal module before renaming, and extract the device function and failure mode corresponding to the original signal; S5.
2. Based on the original signal composition in the modular minimum cut set, a system failure minimum cut set is derived through Boolean logic operation; the minimum cut set is a minimum device failure mode combination for system failure, and no further simplification is required; S5.
3. By calculating and analyzing the signal strength, the occurrence probability of each minimum cut set is obtained, and quantitative analysis of the minimum cut set is achieved.
Citation Information
Patent Citations
Minimum cut set analysis method-based power system reliability analysis method for more-electric aircraft
CN106383960A
A path set combination method for multi-stage system reliability analysis based on cumulative damage model
CN109255172A
System success path planning method based on stream signal structured analysis
CN118036450A
Method and system for dynamic probabilistic risk assessment
US20110137703A1
