Ship system reliability calculation method and system based on discrete event and Monte Carlo simulation

Abstract

The invention belongs to the related technical field of reliability evaluation, and discloses a ship system reliability calculation method and system based on discrete events and Monte Carlo simulation. The method comprises the following steps: (a) constructing a simulation model according to a reliability block diagram of a to-be-processed ship system; (b) setting simulation test times N according to a total task of the ship system to be processed; (c) for the ith simulation test, setting simulation time and token generation time intervals, generating a plurality of tokens of which the tags are 0 at the reliability entrance, transmitting each token backwards along the directed connecting line to which the token belongs, updating the tags in the transmission process, and taking the corresponding tag as a final tag when the token reaches the reliability exit; (d) if i is equal to i + 1, repeating the step (c) until igt; and N, obtaining the final labels of all the tokens in all the simulation tests, and calculating the reliability of the simulation model by using the final labels of all the tokens. According to the invention, the problem that the reliability of a complex system is difficult to calculate is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field related to reliability assessment, and more specifically, relates to a ship system reliability calculation method and system based on discrete events and Monte Carlo simulation. Background Art

[0002] Traditional ship system reliability assessment methods are mainly based on empirical statistics and theoretical analysis, relying on a large amount of historical data and strict mathematical models. However, the ship system structure is complex, containing a large number of components and subsystems. These systems are both in series and parallel, and there are complex dependencies and influences between them, which leads to many limitations of traditional methods in dealing with such complex systems. On the one hand, the acquisition cost of historical data is high and not necessarily comprehensive. On the other hand, strict mathematical models often cannot fully describe the dynamic characteristics and inherent complexity of the actual system.

[0003] Discrete event simulation, as an effective tool for dealing with complex system behaviors, has been widely used in many fields. Its main advantage is that it can simulate the dynamic change process of the system under different states, and reflect the interaction between the system and the environment by constructing an event schedule based on the passage of time. For the reliability assessment of ship systems, discrete event simulation can provide a more flexible and detailed analysis method, which can take into account various random and uncertain factors in the operation of the system. However, despite the above advantages of discrete event simulation, how to accurately integrate the reliability requirements and performance indicators of the system into the simulation model, and how to efficiently extract and calculate reliability data from the simulation process are still one of the main challenges currently faced. In addition, due to the complexity of the system, a single discrete event simulation may be difficult to achieve a comprehensive assessment of the system reliability. Therefore, there is an urgent need for a method to solve the reliability data of the above discrete events. Summary of the invention

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a ship system reliability calculation method and system based on discrete events and Monte Carlo simulation, which solves the problem that the reliability of complex systems is difficult to calculate.

[0005] To achieve the above object, according to one aspect of the present invention, a method for calculating the reliability of a ship system based on discrete events and Monte Carlo simulation is provided, the method comprising the following steps:

[0006] (a) constructing a simulation model according to the reliability block diagram of the ship system to be processed, the simulation model comprising a reliability object, a directed connection line, a reliability inlet and a reliability outlet, the reliability inlet being arranged at the initial position of the simulation model, the reliability outlet being arranged at the end of the simulation model, and the directed connection line being used to connect the reliability object;

[0007] (b) setting the number of simulation tests N according to the total tasks of the ship system to be processed;

[0008] (c) For the ith simulation test, set the simulation time T and the time interval Δt for token generation. Within the simulation time T, generate multiple tokens with a label of 0 at the reliability entrance according to the time interval Δt. Each token is transmitted backward along the corresponding directed connection line. The label is updated during the transmission process. When the token arrives at the reliability exit, the corresponding label is the final label. In this way, the final labels of all tokens in this simulation test are obtained.

[0009] (d) i=i+1, repeat step (b) until i>N, obtain the final labels of all tokens in all simulation tests, and use the final labels of all tokens to calculate the reliability of the simulation model, where N is the number of simulation tests set according to the total task of the ship system to be processed.

[0010] Further preferably, the reliability object includes a reliability device, a relationship component and a logic component, the reliability device is a device object in the reliability block diagram, the relationship component is arranged at the front end of a plurality of parallel reliability devices, and the logic component is arranged at the end of a plurality of parallel reliability devices;

[0011] Further preferably, the total task includes multiple simulation tests, one simulation test includes one or more task sequences, one task sequence includes one or more task profiles, one task profile includes one or more task stages, the task profile sets all the task stages included in the simulation test, and the task stage is set with the start time and end time of the current task stage.

[0012] Further preferably, when setting the reliability device in the simulation model, it is necessary to set the task sequence, task profile, task phase, distribution type and parameters of failure and repair time of the reliability device in the simulation test, and the set task sequence, task profile, task phase, distribution type and parameters of failure and repair time are used to determine whether the reliability device is in a normal state, a failed state or an unplanned state at the current moment;

[0013] When setting the relationship component, it is necessary to set the task sequence, task profile, task phase and component relationship between the reliability devices involved in the simulation test of the relationship component; the set task sequence, task profile and task phase are used to determine whether the relationship component is in a planned state or an unplanned state at the current moment; the component relationship between the reliability devices is preset as hot backup or cold backup, hot backup means that each directed connection line from the relationship component to the subsequent reliability device is feasible, and cold backup means that only one of all the directed connection lines from the relationship component to the subsequent reliability device is feasible;

[0014] The logic component is set according to the relationship between the device objects set in the reliability block diagram. When the relationship between the device objects in the reliability block diagram is parallel, the logic component is set to or; when the relationship between the device objects in the reliability block diagram is &, the logic component is set to and; when the relationship between the device objects in the reliability block diagram is r / n, the logic component is set to r / n.

[0015] Further preferably, the label of the unplanned state is 2, the planned state includes normal and failure states, the label of the normal state is 0, and the label of the failure state is 1.

[0016] Further preferably, in step (b), updating the label during the transfer process is performed in the following manner:

[0017] When the token is passed to the successor reliability object A, the current token is removed, m tokens are created, the labels of each token are set according to the preset rules, and the created m tokens are passed to the successor reliability object B, where m is the number of successor reliability objects B of the successor reliability object A, until the token is passed to the last reliability object in the simulation model, and the token and label in the last reliability object are directly passed to the reliability exit, thus completing a token transfer cycle.

[0018] Further preferably, the preset rules are performed as follows:

[0019] When the successor reliability object A is a relationship component, according to whether the successor reliability objects B of the successor reliability object A are all unplanned, if all are unplanned, the label of the token of the successor reliability object A is set to 2; otherwise, it is determined whether the relationship between the successor reliability devices is a cold backup or a hot backup. When it is a cold backup, the label of at most one reliability device among all the successor reliability devices is 0, and the labels of the rest are 2. When it is a hot backup, the labels of the successor reliability devices are all flag0, and flag0 is the label of the token currently passed to the successor reliability object A;

[0020] When the successor reliability object A is a reliability device, the state of the current reliability device is determined according to the task sequence, task profile, task phase, distribution type and parameters of failure and repair time set for the reliability device, and the label of the reliability device is set according to the state;

[0021] When the successor reliability object A is a logic component, it waits for Num tokens of Num predecessor objects directly connected to the logic component to arrive at the logic component, sets the label of the token on the logic component according to the logic type set on the logic component, and the predecessor object is a reliability component or a logic component.

[0022] Further preferably, when the successor reliability object A is a reliability device, the state of the reliability device is normal, faulty or unplanned. When it is unplanned, the token label is 2; when it is faulty, the token label is max(1, flag0); when it is normal, the label is flag0.

[0023] Further preferably, when the successor reliability object A is a logic component, when the logic component is or, first determine whether there is 0 in the labels of Num tokens, if so, set the label of the current logic component to 0; if there is no 0 in the labels of Num tokens, and there is 1 in the label, set the label of the current logic component to 1; in other cases, the label of the current logic component is set to 2;

[0024] When the logic component is AND, first determine whether there is 1 in the labels of Num tokens. If so, set the label of the current logic component to 1; if there is no 1 in the labels of Num tokens and there is 0 in the labels, set the label of the current logic component to 0; in other cases, the label of the current logic component is set to 2;

[0025] When the logical component is r / n, if the number of 0s in the labels of Num tokens is greater than or equal to r, the label of the current logical component is set to 0; if r is greater than Num or the labels of Num tokens are not all 2, the label of the current logical component is set to 1; in other cases, the label of the current logical component is set to 2.

[0026] According to another aspect of the present invention, a ship system reliability calculation system based on discrete events and Monte Carlo simulation is provided, the system comprising an actuator for executing the above-mentioned ship system reliability calculation method based on discrete events and Monte Carlo simulation.

[0027] In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:

[0028] 1. The present invention constructs multiple simulation tests and calculates the reliability of the simulation model by counting the output token labels of all simulation tests. It is a Monte Carlo simulation method based on probability statistical numerical calculation method. Through random sampling and statistical analysis, the reliability of the ship system under different working conditions is more accurately evaluated and calculated. The value is getting closer and closer to the theoretical value with the increase of simulation times, which effectively solves the shortcoming that the reliability of complex systems is difficult to calculate, and provides a new idea for analyzing the reliability of complex systems and improving efficiency;

[0029] 2. The present invention converts the reliability block diagram into a simulation model, which can simulate the dynamic behavior and failure mode of each component of the system. Through this conversion, the simulation model can not only specifically describe the dependencies between components, but also reflect the response of the system under different failure conditions. This process provides accurate dynamic input data for reliability evaluation, effectively supports the reliability analysis of large-scale complex systems, and simplifies the modeling process of complex systems;

[0030] 3. The present invention updates the tags of tokens according to the set rules, mainly by capturing discrete events during the operation of the system, dynamically adjusting the token status, reflecting the changes of various components in the system, and by designing specific update rules, the simulation can automatically determine the timing of the event, and adjust the tag status of each token according to random input, flexibly simulating the dynamic changes and fault occurrence process of the system; this can not only refine the state transition of each step, but also combine the randomness of Monte Carlo simulation, so that each simulation run has a variety of results, and finally improve the accuracy of system reliability calculation by summarizing the results of a large number of simulation tests;

[0031] 4. The present invention can not only make full use of the advantages of discrete event simulation in processing the dynamic characteristics of the system, but also make use of the powerful ability of Monte Carlo simulation in dealing with uncertainty problems, to provide a more comprehensive and accurate analysis tool for ship system reliability assessment. This method is expected to effectively improve the accuracy and efficiency of ship system reliability assessment, which is of great practical significance for ensuring the safe and stable operation of the ship system. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a schematic diagram of a modeling object according to a preferred embodiment of the present invention;

[0033] Figure 2 It is a schematic diagram of a topological structure formed by combining various components according to a preferred embodiment of the present invention;

[0034] Figure 3 is a reliability block diagram of a ship system according to a preferred embodiment of the present invention;

[0035] Figure 4It is a reliability model of a certain ship system constructed based on a discrete event simulation model according to a preferred embodiment of the present invention;

[0036] Figure 5 It is the experimental result calculated based on discrete event and Monte Carlo simulation method according to the preferred embodiment of the present invention;

[0037] Figure 6 is a schematic diagram of a task profile definition according to a preferred embodiment of the present invention;

[0038] Figure 7 is a schematic diagram of reliability equipment definition according to a preferred embodiment of the present invention, wherein (a) is a schematic diagram of failure characteristic definition of reliability equipment, including failure time interval distribution function and parameters, maintenance duration distribution function and parameters, etc., (b) is a schematic diagram of reliability equipment start-stop cycle definition, which is used to define the work task sequence of reliability equipment;

[0039] Figure 8 is a schematic diagram of relationship component and logic component definition according to a preferred embodiment of the present invention, wherein (a) is a schematic diagram of relationship type definition of relationship component, and (b) is a schematic diagram of logic type definition of logic component;

[0040] Fig. 9 It is a reliability block diagram constructed according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0042] As a numerical calculation method based on probability statistics, Monte Carlo simulation has also been widely used in system engineering and reliability engineering. Its basic idea is to estimate the performance indicators of the system through a large number of random sampling methods, which is suitable for dealing with uncertainty problems in complex systems. Combining the Monte Carlo simulation method with discrete event simulation can make up for the shortcomings of a single method, and more accurately evaluate and calculate the reliability of ship systems under different working conditions through random sampling and statistical analysis.

[0043] A ship system reliability modeling and calculation method based on discrete event and Monte Carlo simulation includes the following steps:

[0044] Step 1: Establish a discrete event simulation model during ship operation. The specific steps are as follows:

[0045] Step 101: Establish a reliability block diagram of the system according to the operating principle of the ship system as a reference object for modeling the discrete event simulation model.

[0046] Step 102: According to the overall mission cycle of the system, define the corresponding mission sequence, mission profile, and mission phase calendar in the discrete event simulation model. The mission sequence is mainly used to describe the temporal relationship of each typical mission profile of the ship equipment in a certain mission sequence process, mainly including the mission sequence name, mission profile name, mission profile order, relative start interval time between mission profiles, and allowed delay time of mission profile; the mission profile is mainly to analyze the events and environment experienced by the main equipment during the specified mission period, mainly including the mission profile name, mission phase name, mission phase order, mission phase duration, and allowed delay time of mission phase; the mission phase is mainly the basic component element of a mission profile, including the mission phase name, mission phase maintenance type, required equipment name, required equipment quantity, and required equipment relationship (series, parallel, voting).

[0047] Step 103: According to the equipment objects in the reliability block diagram, establish the corresponding "reliability equipment" object in the discrete event simulation model, specifically set its failure and maintenance time distribution type (exponential distribution, normal distribution, Weibull distribution) and parameters, and set the task sequence, task profile, and task stage of the reliability equipment in the start-stop cycle.

[0048] The set task sequence, task profile, task phase, distribution type and parameters of failure and repair time are used to determine whether the reliability equipment is in normal state, failure state or unplanned state at the current moment.

[0049] Step 104: According to the relationship between the devices in the reliability block diagram, a corresponding "relationship component" object is established in the discrete event simulation model, and its relationship type is specifically set, including task sequence, task profile, task phase, component relationship (hot backup / cold backup) and the reliability devices involved;

[0050] The component relationship between reliability devices can be preset as hot backup or cold backup. Hot backup means that every directed connection line from the relationship component to the subsequent reliability device is feasible, and cold backup means that only one of all directed connection lines from the relationship component to the subsequent reliability device is feasible.

[0051] The set task sequence, task profile and task phase are used to determine whether the relationship component is in a planned state or an unplanned state at the current moment;

[0052] Step 105: According to the nodes in the reliability block diagram, a corresponding "logic component" object is established in the discrete event simulation model, and its logic type ("or", "and", "r / n") and logic switching rules are specifically set. "or" is the parallel connection of the device objects in the reliability block diagram, and "and" is the relationship of the device objects given in the reliability block diagram as "&". In this embodiment, Fig. 9 The “&” shown in the figure; “r / n” is given by the reliability block diagram. Figure 3 The "r / n" given in is "2 / 3", where r = 2, r represents the number of normal configurable objects required at the node, and n = 3, n represents the total number of predecessor reliability objects;

[0053] Step 106: According to the connection lines of the reliability block diagram, a "connection line" object is established in the discrete event simulation model to connect the above three components.

[0054] Step 107: Create a "reliability entry" object to generate a token object: the token object is a temporary object in the simulation model, generated by the reliability entry object, driven by the discrete event simulation engine, and guided by the directed line segment. The time the Token object passes through each component is 0, and its attributes include a flag parameter, 0 for normal, 1 for failure, and 2 for unplanned status. Then connect to the starting position of the ship system through a connecting line object to define its sampling time interval; create an "exit" object to collect the token object after a single sampling.

[0055] Step 108: Create a "simulation test" object and define the simulation test: including input parameters, output parameters, test definition, and concurrency. The input parameter is defined as the test index, which takes a value of 1-N, where N is the total number of Monte Carlo simulations. The output parameter is the reliability of the system. The test definition is a full permutation of the input parameter category, and the output parameter generates a full factorial test. The concurrency number is the number of simultaneous simulation tests that can be customized based on the hardware conditions of the computing device to speed up the simulation test.

[0056] Step 2: Simulate and record the state based on discrete events. The specific steps are as follows:

[0057] Step 201: running a discrete event-based ship system simulation model;

[0058] Step 202: According to the token generation time interval Δt defined by the reliability entry, the simulation engine advances the simulation clock from T=0 to generate a Token object every Δt, and initializes the flag state to 0;

[0059] Step 203: The token object moves along the directed line segment to the successor component. When the token reaches the component, the current token state is first obtained, recorded as flag0, and then the token is removed, and n new tokens are created at the current position to associate with n objects of the successor devices, and further judgment is made:

[0060] If the current component type is a relationship component object, based on whether the successor reliability object B of the successor reliability object A is all unplanned, if all are unplanned, the label of the token of the successor reliability object A is set to 2; otherwise, determine whether the relationship between the successor reliability devices is cold backup or hot backup. When it is a cold backup, the label of at most one reliability device among all the successor reliability devices is 0, and the labels of the rest are 2. When it is a hot backup, the labels of the successor reliability devices are all flag0, and flag0 is the label of the token currently passed to the successor reliability object A.

[0061] If the current component type is a reliability device object, the current device state (normal, faulty, unplanned) is determined based on the task sequence defined by the object (whether it is in a planned or unplanned state when judging) and the random failure function (whether it is in a normal or faulty state when judging). If it is an unplanned state, the flag of the new Token is 2 (unplanned); if it is a faulty state, the flag of the new Token is max (1, flag0); if it is a normal state, the flag of the new Token is flag0, and the model-related custom script is triggered.

[0062] If the current component type is a logical component object, first determine whether the number of tokens contained in the current component is equal to the number of component predecessor nodes. If not, wait; if they are equal, determine the logic type, and then determine the logic switching time period corresponding to the current simulation time, so as to obtain the corresponding logic type and set the flag state of the component:

[0063] If it is "or", first determine whether there is 0 in the labels of Num tokens. If so, set the label of the current logic component to 0; if there is no 0 in the labels of Num tokens and there is 1 in the label, set the label of the current logic component to 1; in other cases, the label of the current logic component is set to 2;

[0064] If it is "and", first determine whether there is 1 in the labels of Num tokens. If so, set the label of the current logic component to 1; if there is no 1 in the labels of Num tokens and there is 0 in the labels, set the label of the current logic component to 0; in other cases, the label of the current logic component is set to 2;

[0065] If it is "r / n", it corresponds to a redundant system. A redundant system means that there are multiple identical components or subsystems in a system. As long as a certain number of components or subsystems are working properly, the entire system can continue to operate. If the number of flag status values ​​​​in the current token is greater than or equal to r, the component flag value is 0; otherwise, if r is greater than the current number of tokens or not all tokens are in the state of 2 (unplanned), the component flag value is 1 (fault), and the rest is 2 (unplanned).

[0066] Step 204: If there are multiple token objects in the system at time t, traverse all reliability devices, relationship components, and logic components, and execute step 3 to update and move the token.

[0067] Step 205: If the token reaches the exit or the number of subsequent reliability objects of the current component is 0, it means that the system has reached the end. At this time, the token will no longer be deleted. The flag status value of the token indicates the status of the system corresponding to this sampling (normal, faulty, unplanned).

[0068] Step 206: If the simulation end time is reached, the simulation stops automatically, and the reliability of the system can be obtained based on the proportion of the token values ​​collected at the exit, and the single simulation ends.

[0069] Step 3: Based on Monte Carlo simulation test and data statistical analysis, the specific steps are as follows:

[0070] Step 301: After the simulation model is established, define a temporary variable index, with a value of 1-N and a step size of 1 to perform multiple simulation tests. In test 1, index is 1, and the test result is the reliability value R1; in test 2, index is 2, and the test result is the reliability value R2; in test N, index is N, and the test result is the reliability value R n ; In N experiments, only the index value is different, and other variables (fault, maintenance distribution and parameters, sampling interval, etc.) are the same.

[0071] Step 302: According to the basic principle of Monte Carlo simulation, the average value of system reliability obtained after multiple simulations gradually approaches the accurate value as the number of simulations increases. The Monte Carlo simulation results are statistically analyzed to calculate the system reliability, mean time between failures (MTBF), failure probability and other indicators in different time periods, thereby calculating the system reliability. After N tests are completed, the final mean time between failures and mean repair time of the system are obtained.

[0072] The present invention will be further described below with reference to specific embodiments.

[0073] The modeling objects and corresponding icons required by the present invention are as follows: Figure 1 Shown: reliability entry, reliability equipment, relationship components, logical components, directed connection lines, and reliability exit.

[0074] To further illustrate the integrity of the modeling object, Figure 2 The various logical topological structures formed by the combination of various components are shown: series connection, parallel connection, bypass connection, voting, and bridge connection. The errors between the simulation test results of various topological structures and the mathematical theoretical calculation values ​​are all within an acceptable range, and the test process is omitted due to space reasons.

[0075] like Figure 3 The figure shows an example of a reliability block diagram of a ship system. The system consists of a cooling water system, a condensate feed water system, a water making system, a lubricating oil system, a shaft cooling water system, and an electric power system. Each system is connected in series, and each system has its own logical topology. The equipment failures, maintenance time distribution, and start-stop cycle task sequences within the system are all different, forming a complex system.

[0076] Based on the above system reliability block diagram example and related data, the ship system reliability block diagram modeling method based on discrete event simulation described in the present invention is further explained:

[0077] Step 1: Establish the reliability block diagram of the system according to the operating principle of the ship system. Figure 3 , as a reference object for modeling discrete event simulation models.

[0078] Step 2: Establish a discrete event simulation model based on the operation process of the ship's equipment, subsystems and systems. Preferably, this example uses discrete event simulation software FactorySimulation to implement:

[0079] According to the overall task cycle of the system, the corresponding task sequence, task profile, and task phase calendar are defined in the discrete event simulation model. In this example, a total task sequence lasting 20 days is defined, and the total task sequence is divided into four task profiles: low-condition phase 1 (0-1:0:0:0), high-condition phase 1 (1:0:0:0-10:0:0:0), low-condition phase 2 (10:0:0:0-11:0:0:0), and high-condition phase 2 (11:0:0:0-20:0:0:0). Figure 6 shown.

[0080] According to the equipment objects in the reliability block diagram, create "reliability equipment" objects, including the cooling water system, which consists of the following equipment: 2 steam turbine circulating water pumps, 3 fresh water pumps, 2 seawater centralized cooling devices, 2 Type I seawater pumps, and 2 Type II seawater pumps; the condensate feed water system, which consists of the following equipment: 1 condensate module, 1 condensate delivery source, 2 steam turbine feed water pump systems, and 2 auxiliary electric feed water pump units; the water making system, which consists of the following equipment: 2 reciprocating salt removal pumps, and 2 combined water transmission devices; the lubricating oil system, which consists of the following equipment: 3 lubricating oil pumps; the shaft cooling water system, which consists of the following equipment: 2 shaft cooling water pumps; the power system, which consists of the following equipment: 1 steam turbine generator set, 2 small steam turbine generator sets, 1 step-up DC conversion device, 1 step-up rectifier device, 1 medium-voltage DC switchboard, 2 variable frequency speed regulators, and 1 propulsion motor. The topological connection relationship of each device is shown in the reliability block diagram. For each device, set its failure time distribution parameters and maintenance time distribution parameters. Set the task sequence, task profile, and task phase of the reliability device in the start-stop cycle. Figure 7 shown.

[0081] According to the relationship between the devices in the reliability diagram, create a "relationship component" object, set its relationship type according to the relationship of the subsequent components, including task sequence, task profile, task phase (custom), component relationship and the reliability components involved, such as Figure 8 shown.

[0082] According to the nodes in the reliability diagram, create a "Logical Component" object, and set its logic type and logic switching rules as follows: Figure 8 or means that one of the multiple devices is working properly, and means that multiple devices are working properly at the same time, and r / n means it is set according to the value given in the original reliability diagram.

[0083] According to the connection lines of the reliability block diagram, a "connection line" object is established in the discrete event simulation model, and the above three components are connected according to the reliability block diagram.

[0084] Create a "Reliability Entry" object and connect it to the starting point of the ship system through a connecting line object, and define its sampling time interval as 20 minutes.

[0085] Create a "Simulation Test" object, define the input parameter index value as 1-10000, the step size as 1, the output parameter as the number of tokens under the three states (normal, faulty, and unplanned) at the exit, and finally initialize the experiment definition (a total of 10,000 simulations for the full factorial experiment).

[0086] Step 3: Run simulation experiments, conduct multiple experiments and statistical analysis of data;

[0087] According to the Token generation time, a Token is first generated at T=0, that is, the first sampling time is 0, and then a new Token is generated every 20 minutes.

[0088] The token object moves along the directed line segment to the subsequent component. When the token reaches the component, the current token is first removed and n new tokens are created at the current position, where n is the number of subsequent reliability objects of the current device, and then further judgment is made. Take the ship system as an example:

[0089] The token object moves along the directed connection line from the reliability entrance to the relationship component, and determines that its successor components are turbine circulating water pump 1 and turbine circulating water pump 2. The relationship is hot backup mode, and the task sequence is 0-20:0:0:0 including the current time. Therefore, Token is created separately, and its state is set to the current state of the component (normal). Then it moves along the connection line to the two successor components and removes the current Token from the simulation model.

[0090] When the token Token of steam turbine circulating water pump 1 enters, the current status of the equipment is first obtained based on the equipment's fault interval and maintenance duration, that is, the equipment start-stop cycle: if it is an unplanned state (state 2), the flag value of the Token is updated to 2; if it is a faulty state, it is necessary to further determine whether the subsequent reliability object logic component of the steam turbine circulating water pump is in an unplanned state. The judgment logic component object knows that its state is normal, so the flag value of the Token is updated to 1 (fault); if it is a normal state, the flag state value is updated to 0 (normal), and the steam turbine circulating water pump 2 performs the same judgment process.

[0091] When the Token values ​​on the above two devices are updated, they will flow along the connection line to the logic component object. The logic type of the logic component is "or", so as long as one of the flag values ​​of the two Tokens is 0 (normal), it is normal. After the judgment is completed, a new Token is generated on the logic component, and its state value is updated to the state just obtained. It flows along the connection line to the next component and removes the current two Tokens.

[0092] If there are multiple token objects at time t, all reliability devices, relationship components, and logic components are traversed and step 3 is executed to update and move the token.

[0093] When the token passes through the cooling water system, condensate feed water system, water making system, lubricating oil system, shaft cooling water system and power system, it finally reaches the exit. At this time, the flag value of the Token object at the exit is the reliability of the system during this sampling process (normal, faulty, unplanned).

[0094] The above process is repeated until the simulation end time (20:0:0:0) is reached, the simulation stops automatically, all tokens at the exit are counted, and the proportion of tokens with a flag value of 0 is calculated, which is the average reliability of the system during this simulation.

[0095] Repeat the above process until all tests are completed. After the test, check the detailed test data to get the approximate reliability of the system: 99.744310%. According to the Monte Carlo simulation principle, this value is very close to the theoretical reliability of the system, and it gradually approaches as the number of simulation tests increases.

[0096] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A ship system reliability calculation method based on discrete event and Monte Carlo simulation, characterized in that: The method comprises the following steps: (a) constructing a simulation model according to the reliability block diagram of the ship system to be processed, the simulation model comprising a reliability object, a directed connection line, a reliability inlet and a reliability outlet, the reliability inlet being arranged at the initial position of the simulation model, the reliability outlet being arranged at the end of the simulation model, and the directed connection line being used to connect the reliability object; (b) For the i-th simulation test, set the simulation time T and the time interval Δt for token generation. Within the simulation time T, generate multiple tokens with a label of 0 at the reliability entrance according to the time interval Δt. Each token is transmitted backward along the corresponding directed connection line. The label is updated during the transmission process. When the token arrives at the reliability exit, the corresponding label is the final label. In this way, the final labels of all tokens in this simulation test are obtained. (c) i=i+1, repeat step (b) until i>N, obtain the final labels of all tokens in all simulation tests, and use the final labels of all tokens to calculate the reliability of the simulation model, where N is the number of simulation tests set according to the total task of the ship system to be processed.

2. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 1, characterized in that: The reliability object includes a reliability device, a relationship component and a logic component. The reliability device is a device object in the reliability block diagram. The relationship component is arranged at the front end of a plurality of parallel reliability devices. The logic component is arranged at the end of a plurality of parallel reliability devices.

3. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 1 or 2, characterized in that: The total task includes multiple simulation tests, one simulation test includes one or more task sequences, one task sequence includes one or more task profiles, one task profile includes one or more task stages, the task profile sets all the task stages included in the simulation test, and the task stage is set with the start time and end time of the current task stage.

4. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 3, characterized in that: When setting the reliability device in the simulation model, it is necessary to set the task sequence, task profile, task phase, distribution type and parameters of failure and repair time of the reliability device in the simulation test. The set task sequence, task profile, task phase, distribution type and parameters of failure and repair time are used to determine whether the reliability device is in a normal state, a failed state or an unplanned state at the current moment; When setting the relationship component, it is necessary to set the task sequence, task profile, task phase and component relationship between the reliability devices involved in the simulation test of the relationship component; the set task sequence, task profile and task phase are used to determine whether the relationship component is in a planned state or an unplanned state at the current moment; the component relationship between the reliability devices is preset as hot backup or cold backup, hot backup means that each directed connection line from the relationship component to the subsequent reliability device is feasible, and cold backup means that only one of all the directed connection lines from the relationship component to the subsequent reliability device is feasible; The logic component is set according to the relationship between the device objects set in the reliability block diagram. When the relationship between the device objects in the reliability block diagram is parallel, the logic component is set to or; when the relationship between the device objects in the reliability block diagram is &, the logic component is set to and; when the relationship between the device objects in the reliability block diagram is r / n, the logic component is set to r / n, where r is the number of normal device objects required at the node, and n is the total number of predecessor reliability objects.

5. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 4, characterized in that: The label of the unplanned state is 2, and the planned state includes normal and failure states. The label of the normal state is 0, and the label of the failure state is 1.

6. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 5, characterized in that: In step (b), updating the label during the transfer process is performed as follows: When the token is passed to the successor reliability object A, the current token is removed, m tokens are created, the labels of each token are set according to the preset rules, and the created m tokens are passed to the successor reliability object B, where m is the number of successor reliability objects B of the successor reliability object A, until the token is passed to the last reliability object in the simulation model, and the token and label in the last reliability object are directly passed to the reliability exit, thus completing a token transfer cycle.

7. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 6, characterized in that: The preset rules are as follows: When the successor reliability object A is a relationship component, according to whether the successor reliability objects B of the successor reliability object A are all unplanned, if all are unplanned, the label of the token of the successor reliability object A is set to 2; otherwise, it is determined whether the relationship between the successor reliability devices is a cold backup or a hot backup. When it is a cold backup, the label of at most one reliability device among all the successor reliability devices is 0, and the labels of the rest are 2. When it is a hot backup, the labels of the successor reliability devices are all flag0, and flag0 is the label of the token currently passed to the successor reliability object A; When the successor reliability object A is a reliability device, the state of the current reliability device is determined according to the task sequence, task profile, task phase, distribution type and parameters of failure and repair time set for the reliability device, and the label of the reliability device is set according to the state; When the successor reliability object A is a logic component, it waits for Num tokens of Num predecessor objects directly connected to the logic component to arrive at the logic component, sets the label of the token on the logic component according to the logic type set on the logic component, and the predecessor object is a reliability component or a logic component.

8. A ship system reliability calculation method based on discrete event and Monte Carlo simulation as claimed in claim 6, characterized in that: When the successor reliability object A is a reliability device, the status of the reliability device is normal, faulty or unplanned. When it is unplanned, the token label is 2; when it is faulty, the token label is max(1, flag0); when it is normal, the label is flag0.

9. The method for calculating the reliability of a ship system based on discrete events and Monte Carlo simulation as claimed in claim 6, characterized in that: When the successor reliability object A is a logic component, when the logic component is or, first determine whether there is 0 in the labels of Num tokens. If so, set the label of the current logic component to 0; if there is no 0 in the labels of Num tokens and there is 1 in the label, set the label of the current logic component to 1; in other cases, the label of the current logic component is set to 2; When the logic component is AND, first determine whether there is 1 in the labels of Num tokens. If so, set the label of the current logic component to 1; if there is no 1 in the labels of Num tokens and there is 0 in the labels, set the label of the current logic component to 0; in other cases, the label of the current logic component is set to 2; When the logical component is r / n, if the number of 0s in the labels of Num tokens is greater than or equal to r, the label of the current logical component is set to 0; if r is greater than Num or the labels of Num tokens are not all 2, the label of the current logical component is set to 1; in other cases, the label of the current logical component is set to 2.

10. A ship system reliability calculation system based on discrete events and Monte Carlo simulation, characterized in that: The system comprises an actuator, which is used to execute the ship system reliability calculation method based on discrete events and Monte Carlo simulation as described in any one of claims 1-9.

Images