Reliability modeling and evaluation method for multi-phase system task considering complex dynamic characteristics

Abstract

The application discloses a kind of multi-stage system task reliability modeling evaluation methods considering complex dynamic characteristics, belong to system reliability modeling and evaluation technical field, including: according to the structure of system component unit, function principle and task flow, establish system function model and stage model;Based on the finite state machine with constraint, the state-transition model of each structure component unit is established respectively;According to the task activity of each stage and the success and failure criterion of task, the task flow model of system in the corresponding stage is constructed;Integrate the above steps, obtain the reliability model of multi-stage task, carry out task reliability simulation, calculate the occurrence probability of the structure component unit state or structure component unit state combination concerned.This application guarantees the efficient and stable operation of complex dynamic multi-stage system.

Description

Technical Field

[0001] This application relates to a multi-stage system task reliability modeling and evaluation method that considers complex dynamic characteristics, belonging to the field of system reliability modeling and evaluation technology. Background Technology

[0002] Traditional multi-stage system task modeling methods often establish system reliability models based on the mapping relationship between unit states and system states, such as reliability block diagrams, multi-stage fault trees, binary decision diagrams, and Bayesian networks. These methods only establish the structural relationships between the system and its units, neglecting the behavior of the system and its units. Even models based on dynamic fault trees, Petri Nets, Markov models, and dynamic Bayesian networks only describe the time-series failure relationships or the dynamic characteristics of partially correlated failures. They cannot characterize the system's state change processes, time-varying relationships, time delay effects, and dynamic control processes. To analyze the aforementioned dynamic characteristics of the system, it is necessary to explicitly establish models expressing these dynamic characteristics and develop analytical methods for solving these models. Summary of the Invention

[0003] According to one aspect of this application, a method for modeling and evaluating the reliability of multi-stage system tasks that considers complex dynamic characteristics is provided, which ensures the efficient and stable operation of complex dynamic multi-stage systems.

[0004] A multi-stage system task reliability modeling and evaluation method considering complex dynamic characteristics is characterized by the following steps:

[0005] S1: Based on the system's structural components, functional principles, and task flow, establish the system's functional model and stage model;

[0006] S2: Based on constrained finite state machines, establish state-transition models for each structural component;

[0007] S3: Based on the tasks and activities at each stage and the criteria for success and failure, construct the task flow model of the system at the corresponding stage;

[0008] S4: Integrate the above steps to obtain a multi-stage task reliability model, conduct task reliability simulation, and calculate the probability of occurrence of the state of the structural unit of interest or the combination of states of the structural unit of interest.

[0009] Furthermore, S1 includes:

[0010] S11: Identify the structural components of the system, clarify the functional relationships between the structural components in the system, establish the connection relationships between the structural components, and establish the influence relationship between the external environment and the structural components.

[0011] S12: Based on the system operating principle and the success and failure criteria of the task, divide the task into stages and establish a stage model based on finite state machine representation.

[0012] Furthermore, S2 includes:

[0013] S21: Conduct failure mode and effect analysis on the system and its structural components to identify the key failure modes of each structural component.

[0014] S22: Based on the analysis results, obtain the complete set of states of the structural components, determine the transition relationships of each state, and clarify the events associated with the transitions and the constraints on the occurrence of the transitions.

[0015] Furthermore, S3 includes:

[0016] Construct task flow models for each stage, including task success flow models and task failure flow models.

[0017] Furthermore, S4 includes:

[0018] S41: Establish the relationship between the phase model, state-transition model and task flow model to obtain a multi-stage task reliability model;

[0019] S42: Set the simulation input parameters, including the number of simulations N and the state of the structural unit of interest C. T Or K structural unit states (C1, C2, ..., C K )combination;

[0020] S43: Read the multi-stage task reliability model and conduct task reliability simulation based on the Monte Carlo method;

[0021] S44: Statistical simulation data to calculate the probability of occurrence of structural unit states or combinations of structural unit states.

[0022] Furthermore, in step S43, the mission reliability simulation based on the Monte Carlo method includes the following steps:

[0023] (1) Initialize parameters, current simulation number i=0, the state of the structural unit or combination of structural unit states of interest;

[0024] (2) Determine whether the simulation number N has been reached. If yes, end the simulation. If no, proceed to step (3).

[0025] (3) Determine whether the stage model has been fully traversed. If so, increment the simulation number by 1 and proceed to step (2).

[0026] If not, access the current stage model and proceed to step (4).

[0027] (4) Access the task flow model associated with the phase model, and proceed to step (5).

[0028] (5) Proceed from the starting point of the task flow model to the next node of the task flow, and determine whether the current node is the termination node. If yes, exit the task flow model and go to step (3); otherwise, go to step (6).

[0029] (6) Determine whether the current node is associated with the state-transition model. If not, proceed to step (5). If yes, obtain the output of the state-transition model and proceed to step (7).

[0030] (7) Determine whether the state of the structural component is one of the states of the structural component or one of the combinations of states of the structural component. If yes, count and record the state and proceed to step (5). If no, proceed directly to step (5).

[0031] Furthermore, the formula for calculating the probability of occurrence of a structural component state or a combination of structural component states is as follows:

[0032] The state C of the structural unit that is of interest during the simulation process T The number of times it appears is If the total number of simulations is N, then the probability of the occurrence of the state of the structural unit of interest is:

[0033] ;

[0034] For the state combination C of the structural constituent unit of interest G =(C1,C2,…,C K The number of occurrences can be obtained from the AND logic of these states. Let the number of occurrences of the state combinations of the structural components be... If the total number of simulations is N, then the probability of the occurrence of the state combination of the structural unit of interest is:

[0035] .

[0036] The beneficial effects that this application can produce include:

[0037] The reliability modeling and evaluation method for multi-stage system tasks considering complex dynamic characteristics provided in this application captures the structural composition, functional principles, and unit state evolution laws of complex dynamic multi-stage systems by constructing system functional models, stage models, and state-transition models for each unit. This effectively solves the pain point of traditional modeling methods being unable to adapt to the dynamic characteristics and multi-stage coupling relationships of systems. The process model constructed by combining stage task activities and success / failure criteria achieves accurate replication of task scenarios at each stage, providing a basic framework for reliability analysis that fits the actual tasks. By integrating multi-view models to conduct simulation evaluation and calculate the probability of occurrence of target unit states and state combinations, it not only improves the comprehensiveness, accuracy, and pertinence of reliability modeling, but also provides quantitative support for system task optimization, fault prediction, and reliability improvement decisions, helping to reduce task execution risks and ensure the efficient and stable operation of complex multi-stage systems. This solves the problem that previous multi-stage system reliability analysis methods could not characterize system delay behavior or express system dynamic processes. Moreover, the combination of various modeling methods reduces the difficulty of model construction, and the use of Monte Carlo simulation evaluation overcomes the constraints of analytical methods in practical applications, making it more suitable for real industrial systems. Attached Figure Description

[0038] Figure 1 This is a flowchart illustrating a reliability modeling and evaluation method for a multi-stage task system considering complex dynamic characteristics, as described in this application.

[0039] Figure 2 This is a flowchart of the simulation evaluation involved in the reliability modeling and evaluation method for a multi-stage task system considering complex dynamic characteristics in this application;

[0040] Figure 3 This is a functional model of a forced termination system in one embodiment of this application;

[0041] Figure 4 This is a phase model of a forced termination system in one embodiment of this application;

[0042] Figure 5 This application provides a state-transition model of an external system in one embodiment.

[0043] Figure 6 This application provides a state-transition model of the control system in one embodiment.

[0044] Figure 7 This application provides a state-transition model for a signal receiver in one embodiment.

[0045] Figure 8 This application provides a state-transition model for a limit switch in one embodiment.

[0046] Figure 9 This application provides a state-transition model of a battery in one embodiment.

[0047] Figure 10 This is a state-transition model of the running processing terminal in one embodiment of this application;

[0048] Figure 11 This application provides a state-transition model for the release device in one embodiment.

[0049] Figure 12 This is a system behavior model at startup time in one embodiment of this application;

[0050] Figure 13 This is a process model for forcibly terminating the system task when the external system fails to start in stage 1 of one embodiment of this application;

[0051] Figure 14 This is a process model for forcibly terminating system tasks in stage 1 of one embodiment of this application when an external system computational anomaly occurs;

[0052] Figure 15 This is a process flow model for forcibly terminating the system task in stage 2 of one embodiment of this application when the external system computation is abnormal;

[0053] Figure 16 This is a process model for forcibly terminating the system task when the external system calculation times out in stage 2 of one embodiment of this application. Detailed Implementation

[0054] The present application is described in detail below with reference to the embodiments, but the present application is not limited to these embodiments.

[0055] See Figure 1-16 ,like Figure 1 As shown, a multi-stage system task reliability modeling and evaluation method considering complex dynamic characteristics includes the following steps:

[0056] S1: Based on the system's structural components, functional principles, and task flow, establish the system's functional model and stage model;

[0057] S2: Based on constrained finite state machines, establish state-transition models for each structural component;

[0058] S3: Based on the tasks and activities at each stage and the criteria for success and failure, construct the task flow model of the system at the corresponding stage;

[0059] S4: Integrate the above steps to obtain a multi-stage task reliability model, conduct task reliability simulation, and calculate the probability of occurrence of the state of the structural unit of interest or the combination of states of the structural unit of interest.

[0060] Specifically, based on the physical structure, core functional principles, and actual task execution flow of the multi-stage system, the functional relationships, information interaction, and energy transfer relationships of each component unit are analyzed. A structured modeling method is used to establish a system functional model, clarifying the role and function of each unit in the overall system function realization. At the same time, based on the characteristics of task time sequence, working condition switching, and target nodes, combined with the logical hierarchy and stage boundary conditions of system task execution, the task stages are reasonably divided, a system stage model is established, clarifying the task objectives, execution duration, working condition parameters, and switching rules between stages for each stage, thus achieving a clear deconstruction of the multi-stage task process. To address the operational characteristics, fault modes, and dynamic behaviors of each component unit constrained by system conditions and stages, constrained finite state machine theory is introduced. A state-transition model is independently established for each component unit. The model clearly defines all effective states of the unit, including normal working state, different degrees of degradation state, various fault states, and repair state. Transition triggering conditions between states are defined, such as changes in operating conditions, time accumulation, external stimuli, fault occurrence, and repair completion. State transition constraint rules that match system stages and overall constraints are added, such as prohibited state transitions of the unit under specific stages and forced state transitions of the unit during stage switching. This accurately depicts the dynamic state evolution of a single unit in a multi-stage task. Based on the phase model established in S1, for each task phase, specific task activities, execution logic, and resource scheduling methods are defined, and the core criteria for success and failure of system tasks in each phase are clarified, such as the realization of key unit functions, achievement of task indicators, and timeliness of phase task completion. Based on the unit state-transition model established in S2, the set of units participating in the task in each phase, the collaborative relationships between units, and the state coupling effect are sorted out. A phased system task process model is constructed using process modeling methods, clarifying the impact mechanism of unit state on task process execution in each phase, the triggering and termination conditions of process nodes, and the judgment logic of task failure in each phase, so as to realize the correlation mapping between system state and task execution in a single phase task process. By organically integrating the system function model and stage model of S1, the state-transition model of S2, and the phased task flow model of S3, a multi-stage system task reliability overall model considering complex dynamic characteristics is constructed. The information interaction interface between the modules of the model, the model parameter transmission rules during stage switching, and the calculation logic of unit state coupling are clarified. Based on this overall model, multi-stage system task reliability simulation analysis is carried out. By setting the simulation scenario, initial state, and simulation step size, the entire process of multi-stage task execution is simulated, and the state evolution trajectory of each component unit in each stage of the task is tracked. Finally, the probability of occurrence of a single state of the component unit of interest and the combination of states of multiple component units within the entire task cycle is calculated. Based on the probability results, a quantitative evaluation of the reliability of multi-stage system tasks is achieved, providing data support for system reliability optimization and task planning.

[0061] S1 includes:

[0062] S11: Identify the structural components of the system, clarify the functional relationships between the structural components in the system, establish the connection relationships between the structural components, and establish the influence relationship between the external environment and the structural components.

[0063] S12: Based on the system operating principle and the success and failure criteria of the task, divide the task into stages and establish a stage model based on finite state machine representation.

[0064] Specifically, this study comprehensively identifies all core and auxiliary components of the multi-stage system, clarifies the core functions, performance indicators, and functional boundaries of each component, and analyzes the functional coupling, information interaction, physical connection, and resource scheduling relationships among the components, establishing a model of inter-component relationships. Simultaneously, it analyzes the influence of the external environment on the system as a whole and its components, establishing a model of the influence relationship between the external environment and the analyzed system, clarifying the objects, degrees of influence, and forms of action of environmental factors. Based on the system's core operating principles, the execution sequence and logical hierarchy of multi-stage tasks, and combining the success criteria, failure criteria, and termination conditions of the tasks at each stage and the overall system, a clear stage division of the entire process is established, clarifying the boundary characteristics, stage objectives, execution duration, operating parameters, and transition trigger conditions for each stage. Based on finite state machine theory, each task stage is defined as a state of the finite state machine, the transition rules between stages are defined as state transition conditions, and the task execution requirements and constraints within a stage are defined as state attributes, establishing a system stage model based on finite state machine representation to achieve a formal description of the multi-stage task process.

[0065] In one embodiment, S1 includes:

[0066] S11: Identify all the components of the forced termination system, clarify the functional relationships between the components in the system, establish the connection relationships between the components, and at the same time establish the influence relationship between the external environment and the system.

[0067] Specifically, the forced termination system mainly consists of a control system, a remote control center, an operation processing terminal, a battery, a release device A, a release device B, a signal receiver, and limit switches.

[0068] The forced termination system implements forced termination measures in the event of external system malfunction, employing a combination of automatic equipment termination and remote control. The control system identifies external system faults and sends a termination command to the operation processing terminal. Upon receiving the termination command, the operation processing terminal issues execution commands to each release device. If the control system is unable to issue a termination command due to a fault, the remote control center issues the termination command, which is transmitted to the operation processing terminal via a signal receiver. Upon receiving the termination command, the operation processing terminal issues execution commands to each release device. Based on the above functional principle analysis, the following can be obtained: Figure 3 The forced system function model shown.

[0069] S12: Based on the system operating principle and the success and failure criteria of the task, divide the task into stages and establish a stage model based on state machine representation.

[0070] Specifically, the forced termination of the system can be divided into three phases: startup, phase 1, and phase 2. Phase 1 lasts for 30 seconds, and phase 2 lasts for 70 seconds. The phase model is shown in the attached figure. Figure 4 As shown.

[0071] S2 includes:

[0072] S21: Conduct failure mode and effect analysis on the system and its structural components to identify the key failure modes of each structural component.

[0073] S22: Based on the analysis results, obtain the complete set of states of the structural components, determine the transition relationships of each state, and clarify the events associated with the transitions and the constraints on the occurrence of the transitions.

[0074] Specifically, a comprehensive Failure Mode and Effects Analysis (FMEA) is conducted on the system as a whole and its constituent units. By combining the system's operating conditions, task phase characteristics, and external environmental influences, potential failure modes of each constituent unit are identified. The causes, impact levels, detection methods, and degree of impact on the unit's own functions, local system functions, and even the overall task execution of each failure mode are clarified, and the key failure modes of each constituent unit are screened out. Based on the failure analysis results and the normal operating characteristics of each component, a complete set of states for each component is constructed. This set includes normal operating states, performance degradation states of varying degrees, fault states corresponding to each critical fault mode, repair / recovery states, standby / dormancy states, and all other valid operating states. Based on the unit's operating mechanism and fault evolution laws, direct and indirect transition relationships between states are determined, and the associated events triggering state transitions are identified, including time accumulation, changes in operating parameters, external environmental stimuli, fault occurrence, repair completion, stage switching, and human intervention. Simultaneously, considering overall system constraints, task stage requirements, environmental adaptation conditions, and inter-unit coupling relationships, the constraints for each state transition are clarified, including transition prohibition rules under specific stages, transition threshold restrictions under extreme environments, transition linkage conditions under multi-unit collaboration, and transition recovery rules after fault repair. This completes the construction of the constrained finite state machine state-transition model for each component.

[0075] In one embodiment, S2 includes:

[0076] S21: Conduct FMEA analysis on the system and its components to identify the failure modes of each component;

[0077] Specifically, FMEA analysis was conducted on the system and its constituent units to obtain the key failure modes of each unit, as shown in Table 1.

[0078] Table 1

[0079]

[0080] In addition, the forced termination system accepts input from external systems. There are three failure modes of external systems: startup failure, computational error, and task timeout.

[0081] S22: Based on the results of the failure analysis, obtain the complete set of unit states, determine the transition relationships of each state, and clarify the events associated with the transitions and the constraints on the occurrence of the transitions.

[0082] Specifically, the external system may experience two failure modes in Phase 1: startup failure and computational anomaly. In Phase 2, it may experience two failure modes: computational anomaly and task timeout. That is, the transition of the external system from its normal operating state to these states (startup failure, computational anomaly, and task timeout) is time-constrained, and therefore, the following can be obtained: Figure 5 The external system state-transition model is defined, where T is the global simulation time variable. When the global simulation time is less than or equal to the cumulative time of stage 1, the external system state may transition from "normal operation" to "startup failure". When the global simulation time is less than or equal to the cumulative time of stage 1 and stage 2, the external system state may transition from "normal operation" to "computational anomaly". In other words, in stage 1, the system may transition from "normal operation" to "startup failure" or "computational anomaly", and the specific state to which it transitions is determined by the events set on the transition.

[0083] The main failure mode of the control system is the inability to issue a termination command, which can occur throughout the entire mission phase. Therefore, the following can be obtained: Figure 6 The control system state-transition model is shown. The system state-transition models for the signal receiver, limit switch, and battery can be obtained using a similar approach; their state-transition models are attached. Figure 7 -Appendix Figure 9 As shown.

[0084] There are two failure modes for the processing terminal: failure to execute functions correctly and erroneous execution of functions. These can occur throughout the entire task phase, and therefore, the following can be obtained: Figure 10 The state-transition model of the operating processing terminal is shown. The state-transition model of the release device can be obtained in the same way, and the state-transition model of the release device is attached. Figure 11 As shown.

[0085] S3 includes:

[0086] Construct task flow models for each stage, including task success flow models and task failure flow models.

[0087] Specifically, when constructing the task success process model, the core guidance is the task success criteria for this stage. All core activities, execution sequences, unit participation sets, and resource allocation requirements required to complete the established task within the stage are sorted out. Based on the normal working status and compliance status transition rules of each participating unit, the triggering conditions, unit collaboration logic, information / resource transmission paths, and process advancement judgment criteria for each node in the success process are clarified. The correlation between the evolution of the normal state of the unit and the advancement of the success process nodes is established, the boundary conditions for the smooth completion of the process are defined, and a verifiable task success process model is formed.

[0088] When constructing the task failure process model, the core guidance is the task failure criteria for this stage. Combined with the identified key unit failure modes and state anomaly transition patterns, all failure scenarios, failure propagation paths, unit abnormal state combinations, and process interruption nodes that may lead to task failure in this stage are identified. The triggering conditions for each failure scenario are clarified, including single unit failure, multi-unit coupled failure, unit state anomalies caused by environmental limits, and process stagnation caused by mismatch between stage operating conditions. The correlation between the evolution of unit abnormal states and the triggering of failure process nodes is established, the judgment criteria for process interruption and task failure are defined, and a task failure process model covering all failure scenarios is formed.

[0089] At the same time, the boundary thresholds and switching logic of successful and failed processes are clearly defined, and the coupling of the two types of process models with the unit state-transition model and the stage model is realized to ensure that the process model can accurately reflect the impact of the dynamic changes of unit state on the execution results of stage tasks.

[0090] In one embodiment, S3 includes:

[0091] Construct system workflow models for each stage. The constructed workflow models include both workflow models for successful task execution and workflow models for task failure.

[0092] Specifically, the task process for each stage is as follows:

[0093] At startup, two parallel limit switches provide a start signal, the battery provides full power, and the system is forced to terminate and enter standby mode. The constructed workflow model is attached. Figure 12 As shown, it is denoted as model M-0.

[0094] In Phase 1, the task process is as follows:

[0095] When the external system is fault-free, the system should be forcibly terminated and kept in standby mode. If the external system experiences one of two random faults—either a startup failure or a computational anomaly—then the system should be terminated.

[0096] a) When an external system fails to start, the control system should issue an automatic equipment termination command. After receiving the command, the operation processing terminal sends an execution command to the release device A, and the release device A performs the action. The above is the task success process. If any system or device fails during the above task process, the task will fail.

[0097] (b) When an external system experiences a computational malfunction, the control system should issue an automatic equipment termination command. Upon receiving the command, the operation processing terminal sends an execution command to the release device A, which then executes the command. After a 10-second delay, the remote control center sends a termination command to the signal receiver. Upon receiving the command, the operation processing terminal sends an execution command to the release device B, which then executes the command. If the control system fails to issue a termination command, the remote control center sends a remote control termination command after a 15-second delay. Release devices A and B then execute the commands. The task terminates when the release devices successfully execute their commands. In the above task flow, any process that correctly reaches the release device's execution action is considered a successful task; other processes are considered task failures.

[0098] Based on the workflow analyzed above, the task flow for this stage is constructed as follows: Figure 13 Appendix Figure 14 As shown, they are denoted as M-1-a and M-1-b, respectively.

[0099] In Phase 2, the task process is as follows:

[0100] When the external system is fault-free, the system should be forcibly terminated and kept in standby mode. If the external system experiences either a random computational error or a task timeout, the system should be forced to terminate and the termination function should be executed.

[0101] a) When an external system experiences a computational malfunction, the control system should issue an automatic equipment termination command. After receiving the command, the operation processing terminal waits for the signal receiver to issue a termination command. The remote control center issues a termination command to the signal receiver after a 10-second delay. After receiving the command, the signal receiver sends it to the operation processing terminal. After receiving the command, the operation processing terminal issues an execution command to the release device B. If the control system fails to issue a termination command, the remote control center issues a remote control termination command after a 15-second delay. After being sent sequentially by the signal receiver and the operation processing terminal, the release device B executes the command. The task ends when the release device successfully executes the command. In the above task flow, the process that correctly reaches the release device to execute the action is a successful process, and the other processes are task failures.

[0102] b) When an external system experiences a task timeout failure, the control system should issue an automatic equipment termination command. Upon receiving the command, the operation processing terminal sends an execution command to the release device B. If the control system fails to issue a termination command, the remote control center issues a remote control termination command after a 15-second delay. After being sent sequentially by the signal receiver and the operation processing terminal, the release device B executes the command. The task terminates simultaneously when the release device successfully executes the command. In the above task flow, the flow that correctly reaches the release device to execute the action is the successful flow of the task, while the other flows are the failed flows of the task.

[0103] Based on the above workflow analysis, the task flow model for Phase 2 is as follows: Figure 15 Appendix Figure 16 As shown, they are denoted as M-2-a and M-2-b, respectively.

[0104] S4 includes:

[0105] S41: Establish the relationship between the phase model, state-transition model and task flow model to obtain a multi-stage task reliability model;

[0106] S42: Set simulation input parameters, including the number of simulations N and the state of the constituent units of interest C. T Or K constituent unit states (C1, C2, ..., C K )combination;

[0107] S43: Read the multi-stage task reliability model and conduct task reliability simulation based on the Monte Carlo method;

[0108] S44: Statistical simulation data to calculate the probability of occurrence of the constituent unit state or combination of constituent unit states.

[0109] Specifically, a bidirectional relationship is established among the stage model, the unit state-transition model, and the task flow model. The triggering rules for stage switching events in the stage model on unit state transitions are clarified; the driving logic of state evolution in the unit state-transition model on node advancement / interruption in the task flow model is defined; and the feedback mechanism of process execution results in the task flow model on stage continuation / termination in the stage model is established. By unifying the model data interface, defining parameter transfer rules, and clarifying coupling logic, the three types of models are organically integrated to obtain a multi-stage task reliability overall model that can be directly used for simulation analysis. Based on reliability assessment requirements, core simulation input parameters are set, including the number of simulations N and the state C of the single component unit requiring key attention. T Or, a combination of states of K constituent units (C1, C2, ..., C K Simultaneously, auxiliary simulation parameters such as simulation step size, system initial state, external environment parameter sequence, and unit failure probability distribution can be set to ensure the pertinence and accuracy of simulation analysis. The system reads the constructed multi-stage task reliability overall model, imports all set simulation input parameters, and independently conducts multi-stage task full-process simulation for each simulation sample based on the Monte Carlo random sampling principle. During the simulation, trigger events and occurrence times of unit state transitions are randomly generated, and the stage switching process, unit state evolution trajectory, and execution status of task success / failure processes are tracked in real time. The state C of the constituent units of interest in each simulation is recorded. T Or a combination of states (C1, C2, ..., C KThe system collects data from N independent simulations, classifies and statistically analyzes the data from the N Monte Carlo simulations, and counts the effective number of occurrences of a single component state of interest, or the effective number of simultaneous occurrences of a combination of K component states. Based on the principle of frequency-based probability estimation, the system calculates the corresponding probability of occurrence, and performs error analysis and validity verification on the simulation data to ensure the reliability of the probability calculation results.

[0110] In step S43, the mission reliability simulation based on the Monte Carlo method is carried out, including the following steps:

[0111] (1) Initialize parameters, current simulation number i=0, the state of the structural unit or combination of structural unit states of interest;

[0112] (2) Determine whether the simulation number N has been reached. If yes, end the simulation. If no, proceed to step (3).

[0113] (3) Determine whether the stage model has been fully traversed. If so, increment the simulation number by 1 and proceed to step (2).

[0114] If not, access the current stage model and proceed to step (4).

[0115] (4) Access the task flow model associated with the phase model, and proceed to step (5).

[0116] (5) Proceed from the starting point of the task flow model to the next node of the task flow, and determine whether the current node is the termination node. If yes, exit the task flow model and go to step (3); otherwise, go to step (6).

[0117] (6) Determine whether the current node is associated with the state-transition model. If not, proceed to step (5). If yes, obtain the output of the state-transition model and proceed to step (7).

[0118] (7) Determine whether the state of the structural component is one of the states of the structural component or one of the combinations of states of the structural component. If yes, count and record the state and proceed to step (5). If no, proceed directly to step (5).

[0119] Specifically, the simulation parameters are initialized, the current simulation number i is set to 0, and the states C of the constituent units of interest that need to be monitored in this simulation are calibrated. T Or a combination of states of K constituent units (C1, C2, ..., C K), complete the initialization configuration of the simulation counting and status recording modules; determine whether the current simulation number i has reached the preset number of simulations N. If yes, immediately end the current Monte Carlo simulation process and output the full simulation process data; if no, go to step (3) to continue the simulation; call the stage model and determine whether the current simulation sample has completed the full process traversal of the multi-stage task. If yes, increment the current simulation number i by 1 and return to step (2) to start the next simulation; if no, access the current stage model node to be executed, obtain the working conditions, constraints and associated process model information of the stage, and go to step (4); according to the association relationship of the current stage model, access the matching task process model, obtain the process node advancement rules, node association logic and termination judgment conditions, and go to step (4). To step (5); proceed from the starting point of the current task flow model to the next node of the task flow according to the preset logic, determine whether the current node is the process termination node, if yes, exit the current task flow model and return to step (3) to continue the stage traversal judgment; if no, proceed to step (6); determine whether the current process node is bound / associated with the unit state-transition model, if not associated, directly continue to advance the node according to the process logic and return to step (5); if associated, call the corresponding unit state-transition model, generate the unit state transition result based on Monte Carlo random sampling, obtain the real-time output data of the state-transition model, and proceed to step (7); combine the real-time unit state output by the state-transition model with the preset focus to form unit state C. T Or, the constituent units focus on the combination of states (C1, C2, ..., C K If a match is found, the count of the state / state combination of interest is incremented and the relevant information such as the current simulation stage, process node, and state occurrence time is recorded simultaneously. Then, the process node is returned to step (5) to continue. If the match fails, the process is returned directly to step (5).

[0120] The formula for calculating the probability of occurrence of a structural unit state or a combination of structural unit states is as follows:

[0121] The state C of the structural unit that is of interest during the simulation process T The number of times it appears is If the total number of simulations is N, then the probability of the occurrence of the state of the structural unit of interest is:

[0122] ;

[0123] For the state combination C of the structural constituent unit of interest G =(C1,C2,…,C K The number of occurrences can be obtained from the AND logic of these states. Let the number of occurrences of the state combinations of the structural components be... If the total number of simulations is N, then the probability of the occurrence of the state combination of the structural unit of interest is:

[0124] .

[0125] Specifically, the full data of N Monte Carlo simulations are classified, deduplicated, and validated. The occurrence frequency of each combination of states of interest is counted, and the corresponding probability of occurrence is calculated using quantitative formulas. After the probability calculation is completed, error analysis, confidence interval calculation, and validity verification can be performed on the simulation results to ensure the credibility and reference value of the probability calculation results.

[0126] In one embodiment, S4 includes:

[0127] S41: Establish the relationship between the stage model, the system task flow model, and the unit state-transition model to finally obtain the multi-stage task reliability model.

[0128] Specifically, at each stage, the system task flow model related to that stage is associated. In the system task flow model, each action to obtain the unit state is associated with the corresponding unit state-transition model. The detailed association is shown in Table 2.

[0129] Table 2

[0130]

[0131] S42: Set simulation input parameters. The input parameters include the number of simulations, the state of the element of interest, or a combination of element states.

[0132] Specifically, the number of simulations is set to N=100000, and a combination of two states is selected: the external system is normal and the forced termination system does not execute the termination function, and the external system fails and the forced termination system executes the termination function normally.

[0133] S43: Read the multi-stage task reliability model and conduct task reliability simulation based on the Monte Carlo method;

[0134] Specifically, initialize the parameters, set the current simulation number i=0, and set the combinations of states of interest to C. G1 =(External system is normal, forced system termination did not execute the termination function), C G2=(External system failure, forced termination of normal system execution termination function). Starting from the startup time model of the stage model, access its associated task flow model M-0. This stage does not involve external systems and begins with obtaining the battery status. Obtaining the battery status requires accessing the battery's state-transition model. If the battery status is normal, power can be supplied, and the process proceeds to the next node. Next, the status of the limit switch is obtained. At this point, the state transition model of the limit switch needs to be accessed. If the status is normal, a start signal is sent, completing the normal startup process, and subsequent stage models are continuously accessed. If, during this process, the battery or limit switch status is faulty, the simulation round ends directly, and the next simulation round begins.

[0135] S44: Statistically analyze simulation data to obtain the probability of occurrence of the state of the unit of interest or the combination of unit states;

[0136] Specifically, in N simulations, the external system operated normally 92,000 times, and failed 8,000 times. When the external system was normal, the forced termination failed 91,998 times; when the external system failed, the forced termination executed normally 7,991 times. Therefore, the state (external system normal, forced termination failed) occurred 91,998 times, and the state (external system failure, forced termination executed normally) occurred 7,991 times. The probabilities of these two state combinations are calculated as follows:

[0137] ;

[0138] .

[0139] It is worth noting that this application accurately captures the structural, functional, and unit state evolution laws of complex dynamic multi-stage systems by adaptively constructing multi-dimensional models, thus solving the pain point of insufficient adaptability of traditional modeling. Furthermore, it replicates actual task scenarios by combining stage task characteristics and builds a practical reliability analysis framework. Finally, by integrating model simulation evaluation and probability calculation, it not only improves the comprehensiveness, accuracy, and pertinence of reliability modeling, but also provides quantitative decision support for system task optimization, fault prediction, and reliability improvement, effectively reducing task execution risks and ensuring the efficient and stable operation of complex dynamic multi-stage systems.

[0140] The above description is merely a few embodiments of this application and is not intended to limit this application in any way. Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any changes or modifications made by those skilled in the art without departing from the scope of the technical solution of this application using the disclosed technical content are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A multi-stage system task reliability modeling and evaluation method considering complex dynamic characteristics, characterized in that, Includes the following steps: S1: Based on the system's structural components, functional principles, and task flow, establish the system's functional model and stage model; S1 includes: S11: Identify the structural components of the system, clarify the functional relationships between the structural components in the system, establish the connection relationships between the structural components, and establish the influence relationship between the external environment and the structural components. S12: Based on the system operating principle and the success and failure criteria of the task, divide the task into stages and establish a stage model based on finite state machine representation; S2: Based on constrained finite state machines, establish state-transition models for each structural component; S3: Based on the tasks and activities at each stage and the criteria for success and failure, construct the task flow model of the system at the corresponding stage; S4: Integrate the above steps to obtain a multi-stage task reliability model, conduct task reliability simulation, and calculate the probability of occurrence of the state of the structural unit of interest or the combination of states of the structural unit of interest. S3 includes: Construct task flow models for each stage, including task success flow models and task failure flow models; S4 includes: S41: Establish the relationship between the phase model, state-transition model and task flow model to obtain a multi-stage task reliability model; S42: Set the simulation input parameters, including the number of simulations N and the state of the structural unit of interest C. T Or K structural unit states (C1, C2, ..., C K )combination; S43: Read the multi-stage task reliability model and conduct task reliability simulation based on the Monte Carlo method; S44: Statistical simulation data to calculate the probability of occurrence of the state of structural components or the combination of states of structural components; The formula for calculating the probability of occurrence of a structural unit state or a combination of structural unit states is as follows: The state C of the structural unit that is of interest during the simulation process T The number of times it appears is If the total number of simulations is N, then the probability of the occurrence of the state of the structural unit of interest is: ； For the state combination C of the structural constituent unit of interest G =(C1,C2,…,C K The number of occurrences can be obtained from the AND logic of these states. Let the number of occurrences of the state combinations of the structural components be... If the total number of simulations is N, then the probability of the occurrence of the state combination of the structural unit of interest is: 。 2. The multi-stage system task reliability modeling and evaluation method considering complex dynamic characteristics according to claim 1, characterized in that, S2 includes: S21: Conduct failure mode and effect analysis on the system and its structural components to identify the key failure modes of each structural component. S22: Based on the analysis results, obtain the complete set of states of the structural components, determine the transition relationships of each state, and clarify the events associated with the transitions and the constraints on the occurrence of the transitions.

3. The multi-stage system task reliability modeling and evaluation method considering complex dynamic characteristics according to claim 1, characterized in that, In step S43, the mission reliability simulation based on the Monte Carlo method is carried out, including the following steps: (1) Initialize parameters, current simulation number i=0, the state of the structural unit or combination of structural unit states of interest; (2) Determine whether the simulation number N has been reached. If yes, end the simulation. If no, proceed to step (3). (3) Determine whether the stage model has been fully traversed. If so, increment the simulation number by 1 and proceed to step (2). If not, access the current stage model and proceed to step (4). (4) Access the task flow model associated with the phase model, and proceed to step (5). (5) Proceed from the starting point of the task flow model to the next node of the task flow, and determine whether the current node is the termination node. If yes, exit the task flow model and go to step (3); otherwise, go to step (6). (6) Determine whether the current node is associated with the state-transition model. If not, proceed to step (5). If yes, obtain the output of the state-transition model and proceed to step (7). (7) Determine whether the state of the structural component is one of the states of the structural component or one of the combinations of states of the structural component. If yes, count and record the state and proceed to step (5). If no, proceed directly to step (5).

Images