A method for identifying risk factors of complex systems based on PRA framework

Abstract

This invention relates to a method for identifying risk factors in complex systems based on the Probabilistic Risk Assessment (PRA) framework, belonging to the field of risk factor assessment for complex systems. The types of risk factor identification for the system include internal risk factor identification, external risk factor identification, and human-related risk factor identification. The Probabilistic Risk Assessment (PRA) method includes event tree analysis, fault tree analysis, and expert subjective judgment. When identifying internal risk factors, the fault tree analysis method of PRA is used; when identifying external risk factors, the event tree analysis method of PRA is used; and when identifying human-related risk factors, the expert subjective judgment method is used. By integrating the three types of risk factors in the system, the system's risk factors are determined. This invention ensures the comprehensiveness of risk factor identification for complex systems, enabling a comprehensive analysis of internal, external, and human-related risk factors within complex systems.

Description

Technical Field

[0001] This invention belongs to the field of risk factor assessment for complex systems, and relates to a method for identifying risk factors in complex systems based on the PRA framework. Background Technology

[0002] Probabilistic risk assessment (PAR) frameworks, as a primary means of assessing lifecycle-related risks of complex systems, mainly employ event trees, fault trees, and various qualitative and quantitative information (such as experimental data, field data, simulation data, expert judgment, etc.) to construct system risk event chains and model system risks. For risk factor identification in complex systems, the numerous influencing factors contributing to system risk encompass both internal and external risks throughout the system's lifecycle, as well as risks involving human factors. Current risk analysis methods cannot achieve a comprehensive analysis of the risk factors in complex systems. Summary of the Invention

[0003] The technical problem solved by this invention is to overcome the shortcomings of the prior art and propose a method for identifying risk factors in complex systems based on the PRA framework. This method ensures the comprehensiveness of risk factor identification in complex systems and can comprehensively analyze the internal and external risk factors of complex systems as well as human-related risk factors.

[0004] The solution of the present invention is:

[0005] A method for identifying risk factors in complex systems based on the PRA framework, comprising:

[0006] The system's risk factor identification types include internal risk factor identification, external risk factor identification, and human risk factor identification; the probabilistic risk assessment and analysis method, namely PRA, includes event tree analysis, fault tree analysis, and expert subjective judgment; when identifying internal risk factors, the fault tree analysis method of PRA is used; when identifying external risk factors, the event tree analysis method of PRA is used; when identifying human risk factors, the expert subjective judgment method is used; by combining the three types of risk factors of the system, the system's risk factors are determined.

[0007] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the probabilistic risk assessment and analysis method (PRA) comprehensively utilizes event trees, fault trees, and expert subjective judgment to construct a risk event chain model, thereby summarizing and analyzing various qualitative and quantitative information of the project.

[0008] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the qualitative and quantitative information includes experimental data, field data, simulation data, and expert judgment.

[0009] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, when using PRA's fault tree analysis method to identify internal risk factors, the bottom events in the fault tree diagram represent the system's internal risk factors. When using PRA's event tree analysis method to identify external risk factors, the external risk events causing system risks are analyzed, with atomic events representing external system risks. When identifying human risk factors, human performance is unpredictable, and the system phenomena involved are relatively ambiguous. Data related to such risks can only be obtained through expert subjective judgment. By employing expert judgment methods, the human factor risks of the system are analyzed by integrating the opinions of multiple experts.

[0010] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the specific method for identifying internal risk factors using the fault tree analysis method of PRA is as follows:

[0011] First, the complex system is decomposed into multiple simple subsystems. Then, fault tree analysis is performed on each subsystem to analyze the failure modes of the system. Fault tree diagrams for each subsystem are listed. In the fault tree diagram, the bottom event is the risk factor of the system, the middle event is the risk event of the system, and the top event is the risk consequence of the system. The bottom event is used as the internal risk factor of the system.

[0012] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the specific method for identifying external risk factors using the PRA event tree analysis approach is as follows:

[0013] The event that causes system risk due to external factors is defined as the initial event; the related events are identified as the link events, that is, a series of other causal events that may cause accident consequences after the initial event; until the last link event, the last link time is the atomic event, which is the external risk factor of the system.

[0014] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the specific method for identifying human risk factors using expert judgment is as follows:

[0015] To reduce the subjectivity of expert estimates and the differences in judgment among different experts, the Analytic Hierarchy Process (AHP) is used to classify expert capabilities into four levels: personal knowledge, information sources, impartiality, and personal experience. Different weights are assigned to the four levels according to different systems. The capabilities of experts at the four levels are evaluated separately, and the results are combined to assign the optimal expert weight allocation. The results of multiple experts in identifying human risk factors and their corresponding weights are combined to obtain the human risk factors and their weight ratios.

[0016] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, when performing fault tree analysis on the decomposed subsystems:

[0017] Draw a fault tree diagram for the subsystem; for the bottom events in the subsystem fault tree diagram, define them as risk factors of the system; intermediate events are identified as risk events of the system; the top event is the risk consequence of the system; for a complex system composed of multiple subsystems, the top event of the subsystem is the risk event of the overall system, the top event of the system is the risk consequence of the system, and the sum of the risk factors of the subsystems is the internal risk factor of all complex systems.

[0018] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the external risk factors of the system are determined by identifying external risk events caused by external factors as the initial events of system risk events; for multiple initial events, the related loop events are identified to complete the system risk event chain; and the atomic events at the end of the final system risk event tree are taken as the external risk factors affecting system risk.

[0019] In the aforementioned method for identifying risk factors in complex systems based on the PRA framework, the results of different experts are combined with weights to determine the human risk factors evaluated by the experts and their weight ratios. Risk items that meet the weight ratio requirements are selected as human risk factors of the system.

[0020] The advantages of this invention compared to the prior art are:

[0021] (1) The method for identifying risk factors in complex systems based on the PRA framework proposed in this invention is applicable to complex systems involving the entire life cycle, such as satellites and spacecraft, and provides a more comprehensive method for identifying risk factors in complex systems.

[0022] (2) The present invention divides the risk factor identification types of the system into internal risk factor identification, external risk factor identification and human risk factor identification. After identification in three categories, the overall identification is finally carried out, which can meet the requirements of the comprehensiveness of the system risk factor identification.

[0023] (3) In this invention, the three analysis methods of Probabilistic Risk Assessment (PRA) – event tree analysis, fault tree analysis, and expert subjective judgment – ​​are used for internal risk factor identification, external risk factor identification, and human risk factor identification, respectively, thereby achieving efficient and accurate analysis of risk factors in complex systems. Attached Figure Description

[0024] Figure 1 This is a framework diagram of the method for identifying risk factors in complex systems based on the PRA framework of this invention;

[0025] Figure 2 This is a flowchart of the risk factor identification method based on the PRA framework of the present invention.

[0026] Figure 3 This is a fault tree diagram of a subsystem according to an embodiment of the present invention;

[0027] Figure 4 This is a risk event tree diagram of a subsystem according to an embodiment of the present invention;

[0028] Figure 5 This is a diagram of the hierarchical analysis model estimated by experts in an embodiment of the present invention;

[0029] Figure 6 This is a matrix diagram of expert comprehensive judgment in an embodiment of the present invention. Detailed Implementation

[0030] The present invention will be further described below with reference to the embodiments.

[0031] This method is primarily based on the PRA framework, employing FTA, ETA, and expert judgment methods to analyze internal, external, and human-caused risk factors in complex systems. The PRA-based approach to identifying risk factors in complex systems provides a more comprehensive analysis of the risk factors contributing to these risks, resulting in a more complete risk factor identification process.

[0032] A method for identifying risk factors in complex systems based on the PRA framework, specifically including the following steps:

[0033] The system's risk factor identification types include internal risk factor identification, external risk factor identification, and human risk factor identification; the probabilistic risk assessment and analysis method, or PRA, includes event tree analysis, fault tree analysis, and expert subjective judgment; when identifying internal risk factors, the PRA fault tree analysis method is used; when identifying external risk factors, the PRA event tree analysis method is used; and when identifying human risk factors, the expert subjective judgment method is used; by integrating the three types of risk factors in the system, the system's risk factors are determined, and the overall framework is as follows: Figure 1 As shown.

[0034] The Probabilistic Risk Assessment (PRA) method comprehensively utilizes event trees, fault trees, and expert subjective judgment to construct a risk event chain model, enabling the aggregation and analysis of various qualitative and quantitative information related to the project. The qualitative and quantitative information includes experimental data, field data, simulation data, and expert judgment.

[0035] Specifically, when using the PRA fault tree analysis method to identify internal risk factors, the bottom events in the fault tree diagram represent the internal risk factors of the system. When using the PRA event tree analysis method to identify external risk factors, the external risk events that cause system risks are analyzed, with atomic events representing external risks. When identifying human risk factors, human performance is unpredictable, and the system phenomena involved are relatively ambiguous. Data related to such risks can only be obtained through expert subjective judgment. The human factor risk of the system is analyzed by combining the opinions of multiple experts.

[0036] The specific method for identifying internal risk factors using the fault tree analysis approach of PRA is as follows:

[0037] First, the complex system is decomposed into multiple simple subsystems. Then, fault tree analysis is performed on each subsystem to analyze the failure modes of the system. Fault tree diagrams for each subsystem are listed. In the fault tree diagram, the bottom event is the risk factor of the system, the middle event is the risk event of the system, and the top event is the risk consequence of the system. The bottom event is used as the internal risk factor of the system.

[0038] The specific method for identifying external risk factors using the event tree analysis approach of PRA is as follows:

[0039] The event that causes system risk due to external factors is defined as the initial event; the related events are identified as the link events, that is, a series of other causal events that may cause accident consequences after the initial event; until the last link event, the last link time is the atomic event, which is the external risk factor of the system.

[0040] The specific method for identifying human risk factors using expert judgment is as follows:

[0041] To reduce the subjectivity of expert estimates and the differences in judgment among different experts, the Analytic Hierarchy Process (AHP) is used to classify expert capabilities into four levels: personal knowledge, information sources, impartiality, and personal experience. Different weights are assigned to the four levels according to different systems. The capabilities of experts at the four levels are evaluated separately, and the results are combined to assign the optimal expert weight allocation. The results of multiple experts in identifying human risk factors and their corresponding weights are combined to obtain the human risk factors and their weight ratios.

[0042] When performing fault tree analysis on the split subsystems:

[0043] Draw a fault tree diagram for the subsystem; for the bottom events in the subsystem fault tree diagram, define them as risk factors of the system; intermediate events are identified as risk events of the system; the top event is the risk consequence of the system; for a complex system composed of multiple subsystems, the top event of the subsystem is the risk event of the overall system, the top event of the system is the risk consequence of the system, and the sum of the risk factors of the subsystems is the internal risk factor of all complex systems.

[0044] The system's external risk factors are identified, and the external risk events caused by these external factors are determined as the initial events of the system's risk events. For multiple initial events, the related cyclic events are identified to complete the system risk event chain. The atomic events at the end of the final system risk event tree are taken as the external risk factors affecting the system's risk.

[0045] For the results of different experts, the weights are combined to determine the human risk factors assessed by the experts and their weight ratios. Risk items that meet the weight ratio requirements are selected as the human risk factors of the system.

[0046] Methods for identifying risk factors in complex systems based on the PRA framework, such as Figure 2 As shown, the detailed steps include the following:

[0047] S1. Based on the system's functions, the system structure is broken down into multiple related but independent subsystems.

[0048] S2. To address internal risk factors in the system, a fault tree approach is used for fault mode analysis. First, the system structure needs to be broken down into multiple interrelated subsystems based on system functions.

[0049] S3. For each subsystem, perform fault tree analysis and draw a fault tree diagram for the subsystem. The bottom event in the subsystem fault tree diagram is identified as the risk factor of the system, the intermediate events are identified as the risk events of the system, and the top event is the risk consequence of the system. For a complex system composed of multiple subsystems, the top event of the subsystem is the risk event of the overall system, the top event of the system is the risk consequence of the system, and the sum of the risk factors of the subsystems is the internal risk factor of the entire complex system.

[0050] S4. Regarding external risk factors of the system, first identify the external risk events caused by external factors as the initial events of system risk events.

[0051] S5. For multiple initial events, identify the related cyclic events to complete the system risk event chain. The atomic event at the end of the final system risk event tree is taken as the external risk factor affecting the system risk.

[0052] S6. To address human risk factors in the system, an expert judgment approach is adopted, employing multiple experts to comprehensively assess these factors. Furthermore, the Analytic Hierarchy Process (AHP) is used to assign different weights to the results of different experts, taking into account their varying abilities.

[0053] S7. For the results of different experts, take into account the weights, determine the human risk factors of the expert evaluation and their weight ratios, and select the risk items that meet the weight ratio requirements as the human risk factors of the system.

[0054] S8. Three types of risk factors for integrated systems, and identify risk factors for complex systems.

[0055] Example

[0056] Take the lower limb exoskeleton system as an example.

[0057] S1. Due to its complex system structure and numerous failure modes, it is divided into four subsystems based on its functions: power supply system, communication system, control system, and execution structure.

[0058] S2. Taking the power system of the lower limb exoskeleton system as an example, identify the fault modes of the power system and draw a fault tree diagram of the lower limb exoskeleton power system as follows: Figure 3 As shown.

[0059] S3. Based on step 2, the internal risk factors of the lower limb exoskeleton power system are: power switch damage risk, signal line open circuit risk, display screen failure risk, main control component failure risk, node module component failure risk, cable open circuit risk, power failure risk, cane switch damage risk, and electronic device failure risk.

[0060] S4. Using external risk events that cause malfunctions in the power system of the lower limb exoskeleton as initial events, construct an external risk event tree for the power system, such as... Figure 4 As shown.

[0061] S5. Based on the three atomic events a1, a2, and a3 in the external risk event tree of the lower limb exoskeleton power system obtained in step 4, the atomic events are used as external risk factors of the system, and the external risk factors of the lower limb exoskeleton power system are obtained as charging voltage, air humidity, and external load.

[0062] S6. Four experts were selected, and their abilities were stratified based on their differences in expertise. The resulting hierarchical expert estimation model is as follows: Figure 5 As shown.

[0063] S7. Regarding the experts' personal knowledge, information sources, impartiality, and personal experience, it was determined that in the evaluation process, the experts' personal knowledge was of paramount importance, information sources were of moderate importance, impartiality of opinions was of moderate importance, and personal experience and personal knowledge were of equal importance. Weights were assigned to each of the four indicators, and a comprehensive evaluation of the four experts' abilities in different aspects was conducted to obtain the weighting of the experts in this evaluation, as follows: Figure 6 As shown.

[0064] S8. Combining the results of the four experts and their corresponding weights, the weights for the same judgments are added together. Using a weight of 0.3 to determine the human risk factors that meet the conditions, the human risk factors for the lower limb exoskeleton power system are: improper use by the operator, prolonged battery charging without disconnection, inadequate circuit inspection, unreasonable switch design, and unreasonable main control module design.

[0065] S9. Based on the above analysis, the risk factor identification table for lower limb exoskeleton is shown in Table 1.

[0066] Table 1

[0067]

[0068] This invention discloses a method for identifying risk factors in complex systems based on the PRA framework. This method primarily aims to identify risk factors in complex systems, including internal factors, external factors, and human factors. It mainly employs the FTA and ET methods within the PRA framework, which involve risk factor analysis, as well as an expert judgment method combined with the AHP method, to analyze the internal, external, and human factors contributing to the occurrence of risks in complex systems. Specifically, the FTA method is used to perform fault tree analysis on the complex system. The bottom event in the fault tree diagram is identified as the system's internal risk factor, intermediate events as risk events, and the top event as the system's risk consequence, thus analyzing the system's internal risk. The ETA method is used to analyze external risk events leading to system risk. By decomposing the event tree, atomic events leading to system risk are identified, and these atomic events are considered external risk factors. For human risk factors in the system, an expert judgment method is used to assess human-related faults. Since the reliability of expert estimates depends on factors such as the expert's personal knowledge, experience in the field, and available information, an expert judgment method combined with AHP is used to comprehensively evaluate the expert's best estimate and assign appropriate weights based on the best estimate. This invention ensures the comprehensiveness of risk factor identification for complex systems, enabling a complete analysis of internal and external risk factors related to complex systems and human factors.

[0069] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A method for identifying risk factors in complex systems based on the PRA framework, characterized in that: include: The types of risk factor identification in the system include internal risk factor identification, external risk factor identification, and human risk factor identification; Probabilistic risk assessment and analysis (PRA) methods include event tree analysis, fault tree analysis, and expert subjective judgment. When identifying internal risk factors, fault tree analysis is used; when identifying external risk factors, event tree analysis is used; and when identifying human-caused risk factors, expert subjective judgment is used. By integrating these three types of risk factors, the system's overall risk factors are determined.

2. The method for identifying risk factors in complex systems based on the PRA framework according to claim 1, characterized in that: The Probabilistic Risk Assessment (PRA) method comprehensively utilizes event trees, fault trees, and expert subjective judgment to construct a risk event chain model, thereby summarizing and analyzing various qualitative and quantitative information of the project.

3. The method for identifying risk factors in complex systems based on the PRA framework according to claim 2, characterized in that: The qualitative and quantitative information includes experimental data, field data, simulation data, and expert judgment.

4. The method for identifying risk factors in complex systems based on the PRA framework according to claim 1, characterized in that: When using the fault tree analysis method of PRA to identify internal risk factors, the bottom events in the fault tree diagram are the internal risk factors of the system. When using the PRA event tree analysis method to identify external risk factors, the external risk events that cause system risks are analyzed, with atomic events as the external risks of the system. When identifying human risk factors, human performance is unpredictable, and the system phenomena involved are relatively ambiguous. Data related to this type of risk can only be obtained through expert subjective judgment. The human factor risk of the system is analyzed by combining the opinions of multiple experts.

5. The method for identifying risk factors in complex systems based on the PRA framework according to claim 4, characterized in that: The specific method for identifying internal risk factors using the fault tree analysis approach of PRA is as follows: First, the complex system is decomposed into multiple simple subsystems. Then, fault tree analysis is performed on each subsystem to analyze the failure modes of the system. Fault tree diagrams for each subsystem are then drawn up. In the fault tree diagrams, the bottom events are the risk factors of the system, the middle events are the risk events of the system, and the top event is the risk consequence of the system. Bottom events are used as internal risk factors for the system.

6. The method for identifying risk factors in complex systems based on the PRA framework according to claim 4, characterized in that: The specific method for identifying external risk factors using the event tree analysis approach of PRA is as follows: Define the event that causes system risk due to external factors as the initial event; identify the related events that are linked to the initial event, i.e., a series of other causal events that may lead to accident consequences after the initial event; Until the last event in the chain, the final event is an atomic event, which is also an external risk factor of the system.

7. The method for identifying risk factors in complex systems based on the PRA framework according to claim 4, characterized in that: The specific method for identifying human risk factors using expert judgment is as follows: To reduce the subjectivity of expert estimates and the differences in judgment among different experts, the Analytic Hierarchy Process (AHP) is used to classify expert capabilities into four levels: personal knowledge, information sources, impartiality, and personal experience. Different weights are assigned to the four levels according to different systems. The capabilities of experts at the four levels are evaluated separately, and the results are combined to assign the optimal expert weight allocation. The results of multiple experts in identifying human risk factors and their corresponding weights are combined to obtain the human risk factors and their weight ratios.

8. The method for identifying risk factors in complex systems based on the PRA framework according to claim 5, characterized in that: When performing fault tree analysis on the split subsystems: Draw a fault tree diagram for the subsystem; for the bottom events in the subsystem fault tree diagram, define them as risk factors of the system; intermediate events are identified as risk events of the system; the top event is the risk consequence of the system; for a complex system composed of multiple subsystems, the top event of the subsystem is the risk event of the overall system, the top event of the system is the risk consequence of the system, and the sum of the risk factors of the subsystems is the internal risk factor of all complex systems.

9. The method for identifying risk factors in complex systems based on the PRA framework according to claim 6, characterized in that: The system's external risk factors are identified, and the external risk events caused by these external factors are determined as the initial events of the system's risk events. For multiple initial events, the related cyclic events are identified to complete the system risk event chain. The atomic events at the end of the final system risk event tree are considered as external risk factors affecting system risk.

10. A method for identifying risk factors in complex systems based on the PRA framework according to claim 7, characterized in that: For the results of different experts, the weights are combined to determine the human risk factors assessed by the experts and their weight ratios. Risk items that meet the weight ratio requirements are selected as the human risk factors of the system.

Images