Fault analysis method based on common cause failure parameters

Abstract

The invention belongs to the field of system reliability design analysis, and particularly relates to a common cause failure parameter-based fault analysis method, which comprises the following steps: acquiring a fault tree model of an analysis object, and fusing the fault tree model with a common cause failure parameter model; constructing a fault tree model with a common cause failure parameter model; performing common cause failure analysis by using the common cause failure parameter model; expanding the constructed fault tree model with the common cause failure parameter model into an explicit expression common cause failure fault tree model by adopting a universal parameter model expansion method; solving the extended explicit expression common cause failure fault tree model to obtain a top event occurrence probability; according to the method, the implicit common cause failure parameter model is expanded into explicit expression, the automation of the reliability analysis process of the complex system considering the common cause failure influence is realized, the modeling difficulty is reduced, and the reliability analysis working efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of system reliability design and analysis, and particularly relates to a method for fault analysis based on common cause failure parameters. Background Art

[0002] During the operation of complex systems, the phenomenon of common cause failure is widespread and has a significant impact. Common cause failure refers to the situation where multiple units simultaneously lose their intended functions due to the same cause, which can significantly reduce the reliability of redundant systems below the design expectations. Taking the navigation system of a spacecraft as an example, if multiple sensors fail simultaneously due to a common electronic interference source, it may lead to a serious decline in the navigation accuracy of the spacecraft and even cause a flight accident; in the cooling system of a nuclear power plant, if multiple cooling pumps cannot operate properly due to the same pipeline rupture accident, it will pose a great threat to the safe operation of the nuclear power plant.

[0003] Currently, in the research on common cause failure, system failures are usually divided into two parts: system itself failures and common cause failures. Researchers use a combination of parametric models and reliability models to calculate the reliability or failure probability of the system. Common parametric models include the beta factor model, the alpha factor model, the multi-Greek letter model, etc. However, the existing research methods have obvious drawbacks. In the actual modeling process, the parametric model and the reliability model are independent of each other. Modelers need to manually use the parametric model to calculate the failure probability of the unit itself and the probability of the common cause failure part, and then manually add nodes representing common cause failure to the reliability model that originally does not consider common cause failure. This process is not only cumbersome, requires extremely high professional knowledge and skills of modelers, and is difficult for ordinary analysts to handle, but also seriously hinders the automation of the modeling and analysis process, resulting in low work efficiency, prone to human errors, and thus affecting the accuracy and timeliness of system reliability analysis. Summary of the Invention

[0004] To solve the problems of the existing technology, the present invention provides a method for fault analysis based on common cause failure parameters, including the following steps: S1: Obtain the fault tree model of the analysis object, and the fault tree model integrates the common cause failure parameter model; construct a fault tree model with a common cause failure parameter model; wherein, use the common cause failure parameter model for common cause failure analysis; S2: Expand the fault tree model with a common cause failure parameter model constructed in step S1 into an explicit expression of a common cause failure fault tree model by using a general parametric model expansion method; S3: Solve the expanded explicit expression of the common cause failure fault tree model to obtain the probability of the top event occurring.

[0005] Further, after constructing the fault tree model with a common cause failure parameter model in step S1, assign values, which specifically includes the following steps: S11: Assign the independent failure occurrence probability to the units not affected by the common cause unit; S12: Set the common cause failure parameter model and model parameters for the units affected by the common cause unit.

[0006] Furthermore, in step S2, a general parameter model extension method is adopted, which specifically includes the following steps: S21: Identify the sub-module fault tree where all common cause failure units are located, and save the sub-module fault tree as a template; S22: Create a new node with the same name as the top node of the sub-module fault tree and of the type "OR gate" to replace the top node of the atomic module fault tree; S23: Generate the fault tree model of the independent failure part based on the sub-module fault tree template, and set the occurrence probability of the bottom events in the template as the independent failure occurrence probability; S24: Generate an explicit expression of the common cause failure fault tree model.

[0007] Furthermore, in step S24, generating an explicit expression of the common cause failure fault tree model specifically includes: Assume that the number of redundant units in the common cause failure units is m, and the number of sub-fault tree models generated for the common cause failure part is , then the set of common cause failure units is , is the number of the common cause failure unit, k is the number of units that fail simultaneously; for each common cause unit in the set of common cause failure units, add one of the common cause units under the root node of the template, and mark the units whose numbers exist in the common cause unit numbers in the template fault tree as occurring, and simplify the fault tree model according to Boolean logic, where the occurrence probability of the units not in the set of common cause units is the independent failure occurrence probability; Take out all the common cause units in the set of common cause failure units in turn, and finally obtain the extended explicit expression of the common cause failure fault tree model.

[0008] Furthermore, in step S3, the method for solving the fault tree model specifically includes: the up-and-down method, the binary decision diagram method or the Bayesian network method.

[0009] Furthermore, the common cause failure parameter model includes the β-factor model, the multi-Greek letter model or the α-factor model.

[0010] Furthermore, in step S1, using the α-factor model for common cause failure analysis specifically includes: The α-factor model considers the case of any unit failure. For a system with m redundant units, m parameters are introduced, namely: , , …, , …, , so for a system with m redundant units, it is defined as as the probability proportion parameter, as shown in Equation (1); (1) In the formula, m is the number of redundant units in the common cause failure units, represents the number of times that k units fail among the observed common cause failure units; (2) represents the average order of common cause failure; Meanwhile, it is defined as : (3) In the formula, t is the time, Q t is the failure probability of the unit at time t, is the constant failure rate; Then, from Equations (1), (2) and (3), we can get: ; In the formula represents the probability that exactly k units have common cause failure. In the α factor model , , …, , …, and are all parameters, and the parameters , , …, , …, and can be determined according to the known failure data, so as to calculate .

[0011] Advantages of the present invention: The present invention adopts a general parameter model expansion method, realizes the automation of the modeling process through template replication and Boolean logic simplification, replaces the top node of the atomic module with an "OR gate" node, separates the independent failure and common cause failure paths, explicitly generates the fault trees of the independent failure and multi-order common cause failure sub-modules, expands the implicit common cause failure parameter model into an explicit expression, realizes the automation of the reliability analysis process of complex systems considering the influence of common cause failure, reduces the modeling difficulty, and improves the work efficiency of reliability analysis. The fault analysis method first constructs a fault tree model with a common cause failure parameter model and assigns values to the common cause failure parameter model and its bottom events; secondly, through a general parameter model expansion method, an expanded explicit expression common cause failure fault tree model is obtained. Finally, the top event occurrence probability is calculated based on the expanded fault tree model and the occurrence probabilities of its bottom events. Description of the drawings

[0012] Figure 1 This is the flowchart of the fault analysis method based on common cause failure parameters of the present invention; Figure 2 This is the fault tree model of the failure of the aircraft DC power supply system in the embodiment of the present invention; Figure 3 This is the explicit expression common cause failure fault tree model after the expansion of the common cause failure parameter model of the present invention; Figure 4 This is the fault tree model after node substitution in the process of expanding the fault tree model of the present invention; Figure 5 This is the fault tree model for generating the independent failure part in the process of expanding the fault tree model of the present invention; Figure 6 This is the fault tree model for generating the two - unit common cause failure part in the process of expanding the fault tree model in the embodiment of the present invention. Detailed implementation manners

[0013] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0014] Please refer to Figures 1-6 , the present invention provides a fault analysis method based on common cause failure parameters, including the following steps: S1: Obtain the fault tree model of the analysis object, and the fault tree model integrates the common cause failure parameter model; construct a fault tree model with a common cause failure parameter model; wherein, use the common cause failure parameter model for common cause failure analysis; S2: Expand the fault tree model with the common cause failure parameter model constructed in step S1 into an explicit expression common cause failure fault tree model by using a general parameter model expansion method; S3: Solve the expanded explicit expression common cause failure fault tree model to obtain the probability of the top event occurring.

[0015] Further, after constructing the fault tree model with the common cause failure parameter model in step S1, assign values, specifically including the following steps: S11: Assign the probability of independent failure occurrence to the units not affected by the common cause unit; It should be noted that in a complex system, the failure of some units is not affected by the common-cause failure of other units. For these units, according to their own physical characteristics, historical operation data, and relevant industry standards, the parameter values related to reliability are determined. For example, for some simple electronic components, the failure rate data provided in their product manuals can be referred to; for some devices that have undergone a large number of actual operation tests, parameters such as failure probability can be determined based on their historical failure statistics data. By accurately assigning values, reliable basic data can be provided for subsequent reliability analysis; S12: Set the common-cause failure parameter model and model parameters for the units affected by the common-cause units.

[0016] It should be noted that for those unit groups that may fail simultaneously due to the same reason, first, according to the characteristics of the system and the possible causes of common-cause failure, a suitable common-cause failure parameter model should be selected. If the failures of the units in the system are mainly affected by environmental factors and these factors show a certain proportional relationship, the factor model can be selected; if the failure mechanism of the system is relatively complex and involves the combined action of multiple factors, the multi-Greek-letter model can be considered. After the model is selected, through in-depth analysis of data from various aspects such as the system operation environment, manufacturing process, and maintenance records, combined with professional statistical methods and empirical formulas, the specific parameters required for the model are determined. The accurate setting of these parameters is the key to accurately simulating the common-cause failure situation; Furthermore, in step S2, a general parameter model expansion method is adopted, which specifically includes the following steps: S21: Identify the sub-module fault tree where all common-cause failure units are located and save the sub-module fault tree as a template; It should be noted that through a comprehensive and detailed analysis of the fault tree model constructed in step S1, using professional fault tree analysis software or algorithms, all sub-module fault trees containing common-cause failure units are accurately found. These sub-module fault trees reflect the local structure and logical relationships of the common-cause failure risks in the system. Saving these sub-module fault trees as templates for subsequent expansion operations can ensure the consistency of the original structure and logic during the expansion process and avoid errors or omissions; S22: Create a new node of the type "OR gate" with the same name as the top node of the sub-module fault tree to replace the top node of the atomic sub-module fault tree; It should be noted that, in order to clearly represent the relationship between common cause failure and independent failure, in the fault tree model, the top node of the sub-module fault tree is replaced by a newly created "OR gate" node. The logical characteristic of the "OR gate" determines that as long as one of its inputs occurs, the output unit will occur, which exactly conforms to the logical relationship between common cause failure and independent failure, that is, as long as any one of the common cause failure or independent failure occurs, the sub-module will fail. The newly created "OR gate" node becomes the core node for constructing the fault tree models of the independent failure part and the common cause failure part in the subsequent expansion operations; S23: Generate the fault tree model of the independent failure part based on the sub-module fault tree template, and set the occurrence probability of the bottom event in the template to the independent failure occurrence probability; It should be noted that based on the saved sub-module fault tree template, the fault tree model of the independent failure part is generated through a copy operation. In this newly generated model, the occurrence probability of the bottom event in the template is adjusted so that it only reflects the independent failure of the unit. This process requires removing the influence of the common cause failure factor on the bottom event probability and accurately setting it according to the unit independent failure occurrence probability data determined in the previous steps. In this way, the fault tree model of the independent failure part is constructed, clearly showing the failure logic and probability distribution of the sub-module without considering the common cause failure; S24: Generate an explicit expression of the common cause failure fault tree model.

[0017] It should be noted that for m redundant units, the common cause failure may involve the simultaneous failure of 2 to m units. Therefore, it is necessary to calculate all combinations where k≥2. Assuming the number of redundant units in the common cause failure unit is m, the number of sub-fault tree models of the common cause failure part is , common cause failure (CCF, Common Cause Failure), represents the combination number of selecting k from m units, then the set of common cause failure units is , Let \(i\) be the number of the common - cause failure unit, and \(k\) be the number of units that fail simultaneously. When generating the fault - tree model for the common - cause failure part, take the case of generating a fault - tree model with \(k = 2\) common - cause failure units as an example. Take a common - cause unit with \(k = 2\) from the set of common - cause failure units and add it under the root node of the template. The purpose of this operation is to simulate the impact of the common - cause failure unit on the fault tree. At the same time, mark all the units in the template fault tree whose numbers exist in the number of this common - cause unit as occurring. This is based on the definition of common - cause failure, that is, these units fail simultaneously due to a common cause. Then, simplify the fault - tree model according to Boolean logic. Boolean - logic simplification is an operation to remove redundant logic and duplicate units. By applying the basic rules of Boolean algebra, such as the absorption law and the distributive law, optimize the fault - tree model to make its structure clearer for subsequent analysis and calculation. During the simplification process, note that the occurrence probability of the units not in the set of common - cause units remains the independent - failure occurrence probability. Repeat the above operations, successively take out all the common - cause units in the set of common - cause failure units, and generate the corresponding fault - tree models in the same way. By traversing all the common - cause units, finally obtain the complete fault - tree model for the common - cause failure part, which is combined with the previously constructed fault - tree model for the independent - failure part to form an extended fault - tree model, which can comprehensively and accurately reflect the impact of common - cause failure on the system.

[0018] Furthermore, in step S24, generating an explicit - expression common - cause - failure fault - tree model specifically includes: For each common - cause unit in the set of common - cause failure units, add one of the common - cause units under the root node of the template, and mark all the units in the template fault tree whose numbers exist in the number of the common - cause unit as occurring, and simplify the fault - tree model according to Boolean logic, where the occurrence probability of the units not in the set of common - cause units is the independent - failure occurrence probability; Successively take out all the common - cause units in the set of common - cause failure units, and finally obtain the extended explicit - expression common - cause - failure fault - tree model.

[0019] Furthermore, in step S3, the method for solving the fault - tree model specifically includes: the up - down method, the binary decision diagram method, or the Bayesian network method.

[0020] Furthermore, the common - cause failure parameter model includes the beta - factor model, the multi - Greek - letter model, or the alpha - factor model; Furthermore, using the alpha - factor model for common - cause failure analysis specifically includes: The alpha - factor model considers the case of any unit failure. For a system with \(m\) redundant units, \(m\) parameters are introduced, namely: , ,…, ,…, , so for a system with m redundant units, define as the probability proportionality parameter, as shown in Equation (1); (1) where m is the number of redundant units among the common-cause failure units, k is the number of units failing simultaneously, represents the number of times that k units fail among the observed common-cause failure units; (2) represents the average order of common-cause failure; At the same time, define : (3) where t is time, Q t is the failure probability of the unit at time t, is the constant failure rate; Then from Equations (1), (2) and (3), we can get: ; where represents the probability that exactly k units have common-cause failure. In the α-factor model , , …, , …, and are all parameters, and the parameters , , …, , …, and can be determined according to the known failure data, so as to obtain .

[0021] It should be noted that this application is based on the common-cause failure parameter fault analysis method, which is applicable to the high safety and redundancy design of the aircraft power supply system. By means of automated explicit modeling and multi-order common-cause failure analysis, it solves the problem of reliability quantification of complex systems and is applicable to scenarios with extremely high requirements for computational efficiency and accuracy. The present invention adopts a general parameter model extension method to automate the modeling process through template replication and Boolean logic simplification. It replaces the top node of the atomic module with an "OR gate" node, separates the independent failure and common-cause failure paths, explicitly generates the fault trees of the independent failure and multi-order common-cause failure sub-modules, extends the implicit common-cause failure parameter model to an explicit expression, and realizes the automation of the reliability analysis process of complex systems considering the impact of common-cause failures, reduces the modeling difficulty, and improves the efficiency of reliability analysis work. The fault analysis method first constructs a fault tree model with a common-cause failure parameter model and assigns values to the common-cause failure parameter model and its bottom events; secondly, through a general parameter model extension method, an extended explicit expression common-cause failure fault tree model is obtained; finally, the top event occurrence probability is calculated based on the extended fault tree model and the occurrence probabilities of its bottom events.

[0022] Example 1: Taking the aircraft DC power supply system as an example, the aircraft DC power supply system is a key part of the aircraft's power supply, and its reliability is directly related to the flight safety of the aircraft. This system consists of a main DC power supply system and a standby DC power supply system; The structure of the aircraft DC power supply system includes a main power supply and a standby power supply. The main power supply includes driving generators B1, B2, and auxiliary generator B3, and three step-down rectifiers B4, B5, B6; the standby power supply includes a main battery B7 and an auxiliary battery B8; among them, the main power supply is an independent unit, and the step-down rectifiers B4, B5, B6 are common-cause units; during the normal flight of the aircraft, the three parallel step-down rectifiers are responsible for converting the alternating current generated by the generators into direct current required by DC equipment, providing stable power support for numerous electronic devices and systems on the aircraft. The standby DC power supply system consists of the aircraft's main battery, auxiliary battery, and their control components. When the main DC power supply system fails, the standby DC power supply system can be quickly started to directly provide the corresponding direct current for DC equipment, ensuring the continuous operation of the aircraft's key systems; According to the functional principle of the aircraft DC power supply system and the failure modes of each component unit of the system, the basic fault model of the aircraft DC power supply system is analyzed, and further considering the common-cause failure factor, the fault tree model of the aircraft DC power supply system failure is constructed as Figure 2 shown, and the failure rates of each unit are shown in Table 1 below.

[0023] Table 1

[0024] In actual operation, due to the influence of manufacturing processes, maintenance, etc., the three variable voltage rectifiers B4, B5, and B6 may be subject to the same external factors (such as excessive ambient temperature, electromagnetic interference, etc.) or internal factors (such as design defects, component aging, etc.), resulting in simultaneous failures. This situation belongs to typical common cause failure. In order to accurately analyze the impact of this common cause failure on the reliability of the aircraft DC power supply system, the α-factor model is used to model this common cause failure problem. The α-factor model describes the probability distribution of common cause failure by introducing a series of parameters. Through statistical analysis of historical operation data, it is observed that B4, B5, and B6 have 201 single-unit failures, 39 two-unit failures, and 10 three-unit failures respectively. Calculated according to formula (1), the parameter values of the parameter model are shown in Table 2.

[0025] Table 2

[0026] According to the calculation formula of the α-factor model, the probability of common cause failure of k variable voltage rectifiers can be calculated; The α-factor model takes into account the situation of any unit failure. For a system with m redundant units, m parameters are introduced, namely: , , …, , …, Therefore, for a system with m redundant units, define as the probability ratio parameter, as shown in formula (1); (1) In the formula, m is the number of redundant units in the common cause failure units, k is the number of units failing simultaneously, represents the number of times that k units fail among the observed common cause failure units; (2) represents the average order of common cause failure; At the same time, define : (3) In the formula, t is time, Q t is the failure probability of the unit at time t, is the constant failure rate; Then from formulas (1), (2), and (3), we can get: ; In the formula represents the probability that exactly k units have common cause failure. In the α-factor model , , …, , …, and They are all parameters, and the parameters can be determined according to the known failure data , ,…, ,…, and , so as to obtain .

[0027] For the common cause failure analysis of the aircraft DC power supply system, first, construct a fault tree model with a common cause failure parameter model. The fault tree model is shown in Figure [Figure number not provided]. The unit shown is the DC power supply failure T, which includes the main power supply failure G1 and the standby power supply failure G2. Among them, G1 is composed of the generator system failure G3 and the transformer-rectifier system failure G4; second, assign values to the common cause units, and enable the α-factor model for the common cause units B4, B5, and B6 in the system; Figure 2 As shown, the unit is the DC power supply failure T, including the main power supply failure G1 and the standby power supply failure G2. Among them, G1 is composed of the generator system failure G3 and the transformer-rectifier system failure G4; secondly, assign values to the common cause units, and enable the α-factor model for the common cause units B4, B5, and B6 in the system; Finally, expand it into an explicit expression of the common cause failure fault tree, identify the fault tree (B4, B5, B6) of the transformer-rectifier system failure G4 sub-module as the common factor module, save it as a template, replace the original top node with the "OR gate" node G4, and separate the independent failure (G4_1) and common cause failure (G4_2 - G4_5) paths (as shown in Figure [Figure number not provided]), to achieve explicit modeling; Figure 3 As shown, to achieve explicit modeling; For independent failure, generate G4_1, and set the probabilities of the bottom events B4, B5, and B6 as the independent failure probabilities; For the common cause of two units, generate G4_2 (common cause of B4 + B5), G4_3 (common cause of B4 + B6), G4_4 (common cause of B5 + B6), mark the corresponding units as occurring, and retain the common cause unit node (such as CCF45) after simplification; For the common cause of three units, generate G4_5 (common cause of B4 + B5 + B6), mark all units as occurring, and simplify it to the CCF456 node; Adopt the binary decision diagram method to calculate the top event probability of the expanded explicit expression of the common cause failure fault tree; Among them, based on the above fault tree model, using the expansion method in step S2 of the present invention for operation, the fault tree model of the explicit common cause failure impact after expanding the implicit common cause failure parameter model as shown in Figure [Figure number not provided] can be obtained; Figure 3 As shown, the fault tree model of the explicit common cause failure impact after expanding the implicit common cause failure parameter model; First, through a professional fault tree analysis (Isograph FTA) tool, identify the sub-module fault tree G4 containing the bottom events B4, B5, and B6. Save the sub-module fault tree G4 completely as a template for subsequent expansion operations. Then, create a fault tree node with the same name as the top node of the sub-module fault tree G4 in the original fault tree model and set it as an "OR gate". After this operation, the model state is as shown in Figure [Figure number not provided] Figure 4As shown, the new "OR gate" node becomes the core hub for subsequent expansion operations, connecting the fault tree models of the independent failure part and the common cause failure part; Next, generate the sub-fault tree model of independent failure. Directly copy the previously saved template fault tree and name the newly generated model G4_1. In G4_1, set the occurrence probabilities of all the units belonging to the common cause failure units below it (i.e., B4, B5, B6) to the occurrence probabilities of independent failure. These probability values are accurately set according to the previously calculated independent failure occurrence probability data. After the setting is completed, add this independent failure sub-fault tree model under the newly created "OR gate" node, and the obtained model structure is as Figure 5 shown. Through this step, the fault tree model of the independent failure part is successfully constructed, clearly showing the failure logic and probability distribution of this sub-module without considering common cause failure; After that, generate the sub-module fault tree with 2 common cause units. Take the selection of B4 and B5 as common cause failure units as an example for detailed description. First, copy the template sub-module fault tree to get the sub-fault tree G4_2. Then add the unit CCF45 representing the common cause failure of B4 and B5 under the top node of G4_2, and set both B4 and B5 to the occurrence state. At this time, simplify this fault tree model according to Boolean logic. During the Boolean logic simplification process, follow the basic rules of Boolean algebra to remove redundant logical relationships and duplicate units. After simplification, the obtained sub-fault tree G4_2 is as Figure 6 shown. Follow the same steps to operate on the common cause unit combinations such as {B4,B6}, {B5,B6}, {B4,B5,B6} respectively, and obtain the sub-fault trees G4_3, G4_4 and G4_5 in turn. Through these operations, the fault tree model of the common cause failure part is comprehensively constructed, covering the impact of different common cause unit combinations on the system; Finally, use the solution method of the fault tree to solve the expanded fault tree model. In this example, methods such as the top-down and bottom-up method, the binary decision diagram method or the Bayesian network method can be selected for solution. Take the binary decision diagram method as an example. Convert the expanded fault tree model into a binary decision diagram structure, and through the traversal and calculation of the graph, obtain the occurrence probability of the top event of the fault tree. This probability value reflects the likelihood of the failure of the aircraft DC power supply system considering common cause failure. By comparing with the reliability index required by the design, analysts can evaluate the system.

[0028] It should be noted that the method of the present invention realizes the automated construction and solution process of the common cause failure parameter model and the fault tree model. From the setting of unit parameters, model expansion to the final result calculation, a large amount of manual intervention is not required. By writing an automated program or using professional software tools, the entire reliability analysis process can be completed quickly and accurately, greatly improving the efficiency of the reliability analysis of complex systems and reducing the time consumption caused by manual operations.

[0029] In traditional methods, the complex manual calculation and model integration processes have extremely high professional requirements for modelers. However, the automated method of the present invention simplifies the operation process. By adopting a general parameter model expansion method, the modeling process is automated through template replication and Boolean logic simplification. The top node of the atomic module is replaced by an "OR gate" node, the independent failure and common cause failure paths are separated, and the fault trees of the independent failure and multi-order common cause failure sub-modules are explicitly generated, expanding the implicit common cause failure parameter model into an explicit expression, realizing the automation of the reliability analysis process of complex systems considering the influence of common cause failure, reducing the modeling difficulty, and improving the efficiency of the reliability analysis work. The fault analysis method first constructs a fault tree model with a common cause failure parameter model and assigns values to the common cause failure parameter model and its bottom events; secondly, through a general parameter model expansion method, an expanded explicit expression of the common cause failure fault tree model is obtained; finally, the probability of the top event is calculated based on the expanded fault tree model and the occurrence probability of its bottom events.

[0030] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for fault analysis based on common cause failure parameters, characterized in that Including the following steps: S1: Obtain the fault tree model of the analysis object, and the fault tree model integrates a common cause failure parameter model; Construct a fault tree model with a common cause failure parameter model; wherein, perform common cause failure analysis using the common cause failure parameter model; S2: Expand the fault tree model with a common cause failure parameter model constructed in step S1 into an explicit common cause failure fault tree model by using a general parameter model expansion method; S3: Solve the expanded explicit common cause failure fault tree model to obtain the top event occurrence probability.

2. The method for fault analysis based on common cause failure parameters according to claim 1, wherein After constructing the fault tree model with a common cause failure parameter model in step S1, assign values, which specifically includes the following steps: S11: Assign the independent failure occurrence probability to the units not affected by the common cause unit; S12: Set the common cause failure parameter model and model parameters for the units affected by the common cause unit.

3. The fault analysis method based on common cause failure parameters according to claim 2, characterized in that In step S2, the general parameter model expansion method is adopted, which specifically includes the following steps: S21: Identify the sub-module fault trees where all common cause failure units are located, and save the sub-module fault trees as templates; S22: Create a node with the same name as the top node of the sub-module fault tree and of the type "OR gate" to replace the top node of the atomic module fault tree; S23: Generate a fault tree model for the independent failure part based on the sub-module fault tree template, and set the bottom event occurrence probability in the template as the independent failure occurrence probability; S24: Generate an explicit common cause failure fault tree model.

4. The method for fault analysis based on common cause failure parameters according to claim 3, wherein In step S24, generating an explicit common cause failure fault tree model specifically includes: Suppose the number of redundant units in the common cause failure unit is m, and the number of sub-fault tree models generating the common cause failure part is , then the set of common cause failure units is , is the number of the common cause failure unit, and k is the number of units that fail simultaneously; for each common cause unit in the set of common cause failure units, add one of the common cause units under the root node of the template, and mark all the units with numbers existing in the common cause unit numbers in the template fault tree as occurring, and simplify the fault tree model according to the Boolean logic, where the occurrence probability of the units not in the common cause unit set is the independent failure occurrence probability; Successively take out all common cause units in the common cause failure unit set, and finally obtain the expanded explicit common cause failure fault tree model.

5. The fault analysis method based on common cause failure parameters according to claim 1, wherein In step S3, the method for solving the fault tree model specifically includes: the up-and-down method, the binary decision diagram method or the Bayesian network method.

6. The method for fault analysis based on common cause failure parameters according to claim 1, wherein The common cause failure parameter model includes a β-factor model, a multi-Greek letter model or an α-factor model.

7. The method for fault analysis based on common cause failure parameters according to claim 6, wherein In step S1, performing common cause failure analysis using the α-factor model specifically includes: The α-factor model takes into account the situation of any unit failure, and introduces m parameters for a system with m redundant units, namely: , ,…, ,…, Therefore, for a system with m redundant units, define as the probability proportionality parameter, as shown in Equation (1); （1） Where m is the number of redundant units in the common-cause failure unit, represents the number of times that k units fail among the observed common-cause failure units; （2） represents the average order of common cause failure; Define simultaneously : （3） where t is the time, Q t is the failure probability of the unit at time t, is the constant failure rate; Then, from equations (1), (2) and (3), it can be obtained that: ； where represents the probability that exactly k units have common-cause failures. In the α-factor model , ,… ,… and are all parameters, and the parameters , ,… ,… and can be determined according to the known failure data, so as to obtain .

Images