A system reliability prediction method and system based on interpolation iteration algorithm
The system reliability is calculated through the interpolation iterative algorithm, and the problems of complex probability distribution and low-efficiency simulation calculation are solved, and efficient and accurate system reliability prediction is achieved, which is suitable for multi-state systems and non-exponential random distributions.
Patent Information
- Application Number
- CN202210318939.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-29
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-03-29
AI Technical Summary
When calculating the reliability of the system, it is difficult to solve the problem of difficult to solve the inverse function of complex probability distribution and low simulation calculation efficiency, especially when the probability distribution form is complex and the probability of random events is small, it is difficult to achieve high precision.
The system reliability prediction method based on the interpolation iteration algorithm is adopted. By setting the total number of system simulations and the prediction cut-off time, the cumulative probability distribution function and probability density function of component state transition are determined, and the state transition time is calculated using the interpolation iteration algorithm, and the state transition time is calculated based on the random number. and probability density function updates component state and calculates system reliability.
This method improves computational efficiency and accuracy, is suitable for multi-state systems and non-exponential random distribution, expands the scope of application of reliability prediction, overcomes the shortcomings of the inverse function method, and provides higher computational efficiency and better compatibility .
Smart Images

Figure CN114676576B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of system engineering technology, and more specifically, to a system reliability prediction method and system based on an interpolation iterative algorithm. Background Art
[0002] The simulation-based calculation method is an important technical approach in the field of modern numerical calculation. It has become one of the emerging mainstream technologies for engineering calculations such as system reliability analysis, and is particularly suitable for situations where variables obey complex random distributions. In view of the complexity of random distributions and the efficiency requirements of simulation calculations, specific numerical calculation implementation technologies should be designed.
[0003] Reliability prediction based on simulation calculation methods requires simulating a large number of random system state transitions to calculate system reliability. The random number calculation method based on the inverse function is to substitute the random number uniformly distributed in the range of 0-1 into the inverse function of the probability distribution to obtain the random number of the corresponding distribution. This method is suitable for situations where the inverse function of probability distributions such as exponential distribution is easy to solve. When the random variables obey complex probability distributions, it is usually difficult to analytically find the inverse function of the probability distribution, which leads to the inability to obtain the random number of the corresponding distribution. For example, when the calculation formula of the indicator function in the system reliability analysis is complex, or when the variables in the indicator function obey the probability distribution that is difficult to obtain an analytical formula for the inverse function, the random number calculation method based on the inverse function is difficult to apply. Although common methods such as the rejection method can overcome the defect that the inverse function is difficult to solve, it is difficult to achieve high accuracy with a small number of simulation calculations when the probability distribution form is complex and the probability of random events is small.
[0004] In view of the defects and shortcomings of common methods, reliability prediction of complex engineering systems requires a random number generation algorithm with high computational efficiency and applicable to those for which the inverse function is difficult to obtain. This algorithm is not limited by the type of probability distribution function and has a wider range of applications and higher computational efficiency than common methods such as the inverse function method and the rejection method. Summary of the invention
[0005] In response to at least one defect or improvement need in the prior art, the present invention provides a system reliability prediction method and system based on an interpolation iterative algorithm, which has high computational efficiency and controllable computational accuracy, and can be widely applied to system reliability prediction of multi-state system components and system component failures that obey exponential distribution or non-exponential general random distribution.
[0006] To achieve the above object, according to one aspect of the present invention, a system reliability prediction method based on an interpolation iterative algorithm is provided, comprising:
[0007] Set the total number of system simulations and prediction deadline;
[0008] According to the system component composition structure, determine the system state corresponding to the combination of different component states;
[0009] Determine the cumulative probability distribution function and probability density function of the state transition of each component of the system;
[0010] According to the cumulative working time of the components, determine the residual cumulative probability distribution function and the residual cumulative probability density function of the state transition of each component;
[0011] Calculate the residual cumulative probability distribution function of system state transfer according to the residual cumulative probability distribution function of each component state transfer;
[0012] An interpolation iterative algorithm is used to determine the system state transfer time;
[0013] Generate a random number, determine the component for state transfer based on the random number and the residual cumulative probability density function of each component state transfer, and update the state of the component;
[0014] The system state is determined based on the updated component state combination, the cumulative predicted failure times of the system within the prediction deadline are calculated, and the system reliability is calculated based on the cumulative predicted failure times.
[0015] Further, the determining of the system status corresponding to the combination of different component statuses includes:
[0016] The total number of system components is K, and the state variable of component j is x j , 1≤j≤K, Γ is the set of components that cause system failure, and the system state function is recorded as Φ, then the following relationship is satisfied:
[0017]
[0018] Furthermore, the cumulative working time of component j is recorded as The cumulative probability distribution function of component j’s state transition is denoted as F j (t), whose probability density function is denoted by f j (t), the residual cumulative probability distribution function of component j state transition is recorded as Its residual cumulative probability density function is recorded as and The calculation formula is:
[0019]
[0020]
[0021] Furthermore, the residual cumulative probability distribution function of the system state transition is recorded as The calculation formula is:
[0022]
[0023] Furthermore, the adopting of an interpolation iterative algorithm to determine the system state transfer time includes:
[0024] Step 6.1: Generate a random number ξ from a uniform distribution in the range 0-1;
[0025] Step 6.2: For the residual cumulative distribution function Given the initial time range of interpolation [t1,t m ], and satisfy the following relations:
[0026]
[0027] In the interval [t1,t m ], insert (m-2) interpolation points at equal intervals, and change t1, t m and the interpolation points are arranged in ascending order to form a vector t, according to the cumulative probability distribution function Calculate the corresponding probability value vector p, This gives the interpolation data point vector combination:
[0028]
[0029] Step 6.3: According to the residual cumulative probability distribution function of the system state transition, determine the ranking position of ξ in the probability value vector p:
[0030]
[0031] In the formula, represents the neighboring probability of the random number ξ;
[0032] Step 6.4: The time corresponding to the probability of the random number ξ As the time range for the next interpolation, let When the preset condition is met, execute step 6.5, otherwise, return to step 6.2. The preset condition is expressed as:
[0033]
[0034] Among them, α is the preset value for the termination of interpolation iteration;
[0035] Step 6.5: Calculate the system state transition time t (n) :
[0036]
[0037] Further, the component for determining the state transition according to the random number and the residual cumulative probability density function of the state transition of each component includes:
[0038] The random number is recorded as δ, and the component j of the state transition is determined * , so that the formula is satisfied:
[0039]
[0040] Furthermore, the calculation of the cumulative predicted number of failures and the reliability includes:
[0041] Step 8.1: Accumulate the random numbers of the generated system state transition time to obtain the current cumulative working time T of the system. The current simulation round is recorded as N. sim , the total number of simulations is recorded as N max ;
[0042] Step 8.2: When T ≤ T Z And Φ(x1,x2,…,x K )=0, N sim Add 1, the cumulative number of system failures N increases by 1, and execute step 8.5;
[0043] Step 8.3: When T>T Z When N sim Add 1, then execute step 8.5;
[0044] Step 8.4: If the conditions of steps 8.2 and 8.3 are not met, the random number of the generated system state transition time is added to the cumulative working time of the component, and then the system state corresponding to the different component state combinations is re-determined, and the subsequent steps are repeated;
[0045] Step 8.5: When N sim >N max , calculate the system reliability R S :
[0046]
[0047] According to a second aspect of the present invention, there is provided a system reliability prediction system based on an interpolation iterative algorithm, comprising:
[0048] The setting module is used to set the total number of system simulations and the prediction deadline;
[0049] A state determination module is used to determine the system state corresponding to different component state combinations according to the system component composition structure;
[0050] A function calculation module is used to determine the cumulative probability distribution function and probability density function of the state transfer of each component of the system, determine the residual cumulative probability distribution function and residual cumulative probability density function of the state transfer of each component according to the cumulative working time of the component, and calculate the residual cumulative probability distribution function of the system state transfer according to the residual cumulative probability distribution function of the state transfer of each component;
[0051] An interpolation iteration module is used to determine the system state transfer time by using an interpolation iteration algorithm;
[0052] A state transition component determination module is used to generate a random number, determine the component for state transition according to the random number and the residual cumulative probability density function of each component state transition, and update the state of the component;
[0053] The reliability calculation module is used to determine the system state according to the updated component state combination, calculate the cumulative predicted failure times of the system within the prediction deadline, and calculate the system reliability according to the cumulative predicted failure times.
[0054] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:
[0055] 1) Wide range of applications. The present invention is applicable to reliability prediction of faults that obey exponential distribution and non-exponential general random distribution. As long as the cumulative probability distribution function of the target distribution can be determined (or the approximate distribution of the target distribution is given), the algorithm can be used to calculate random numbers that obey complex distributions, which is particularly suitable for system reliability prediction based on simulation calculation methods. The present invention has no special requirements for the random distribution type of faults and has a wide range of applications.
[0056] 2) High computational efficiency. The present invention overcomes the defects and shortcomings of the inverse function method. While the inverse function of the cumulative probability distribution function of the target distribution cannot be derived, it also inherits the advantage of the inverse function method in terms of high computational efficiency. Compared with other commonly used methods such as the selection method, the interpolation iteration algorithm provided by the present invention does not require repeated selection, thereby improving computational efficiency.
[0057] 3) The calculation accuracy is controllable. The accuracy of the algorithm of the present invention is mainly affected by the selection of interpolation data points. The more interpolation iterations there are, the higher the accuracy. The number of interpolation iterations can be controlled according to the interpolation iteration termination coefficient to strike a balance between the amount of calculation and the calculation accuracy.
[0058] 4) Good expansion compatibility. In reliability prediction based on simulation calculation methods, the interpolation iteration algorithm provided by the present invention has good method compatibility. For example, the interpolation iteration algorithm of the present invention can efficiently provide state transition time that obeys complex distribution, and combined with other accelerated simulation strategies, it can further improve the calculation efficiency of reliability prediction of high-reliability systems. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0060] Figure 1 is a schematic diagram of a selection process of providing an interpolation point sequence according to an embodiment of the present invention;
[0061] Figure 2 is a schematic diagram of the interpolation result and its distribution test of an embodiment of the present invention;
[0062] Figure 3 Schematic diagram of the effect of the number of interpolation points on interpolation accuracy in an embodiment of the present invention;
[0063] Figure 4 It is a system composition block diagram of an embodiment of the present invention. DETAILED DESCRIPTION
[0064] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.
[0065] In the specification and claims of the present application, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.
[0066] A system reliability prediction method based on an interpolation iterative algorithm according to an embodiment of the present invention comprises the following steps:
[0067] Step 1: Set the number of simulations and the deadline for reliability estimation.
[0068] Step 1.1: Set the system failure count N = 0, the simulation count variable N sim =0, and the maximum number of simulations is N max .
[0069] Step 1.2: Set the current time T = 0 and the deadline T for determining the reliability estimate Z .
[0070] Step 2: According to the system component composition structure, determine the system state corresponding to the component state combination.
[0071] Assume that the system consists of K components, and the state variable of component j (1≤j≤K) is xj Γ is the set of components that cause system failure, and the system state function is recorded as Φ, which should satisfy the following relationship:
[0072]
[0073] Step 3: Determine the cumulative probability distribution function and probability density function of the state transition of each component of the system.
[0074] Regardless of whether the system is an unrepairable system or a repairable system, the failure and repair of components are regarded as component state transitions. Let the cumulative probability distribution function of component state transition be F j (t), the probability density function is f j (t), and F j (t) shall meet the following requirements:
[0075] 0≤F j (t)≤1,t≥0 (2)
[0076] Step 4: According to the cumulative working time of the component, calculate the residual cumulative probability distribution function and the residual probability density function of the component state transition.
[0077] Component j cumulative working time According to the cumulative probability distribution function F of component state transition j (t) and the probability density function f j (t), calculate the residual cumulative probability distribution function of state transitions that obey a non-exponential general random distribution Its probability density function
[0078]
[0079]
[0080] Step 5: Calculate the residual cumulative probability distribution function of the system state transition
[0081] The system contains K components, and the residual cumulative distribution function of the system state transition probability is for:
[0082]
[0083] Step 6: Execute the interpolation iteration algorithm to calculate the nth system state transition time t (n) .
[0084] Step 6.1: Generate a random number ξ from a uniform distribution in the range 0-1.
[0085] Step 6.2: For the residual cumulative distribution function Given the initial time range of interpolation [t1,t m ], and satisfy the following relations:
[0086]
[0087] In the interval [t1,t m ], insert (m-2) interpolation points at equal intervals. m and the interpolation points are arranged in ascending order to form a vector t. According to the cumulative probability distribution function Calculate the corresponding probability value vector This gives the interpolation data point vector combination:
[0088]
[0089] Step 6.3: According to the residual cumulative probability distribution function of the system state transition, determine the ranking position of ξ in the probability value vector p:
[0090]
[0091] represents the neighboring probability of a random number ξ.
[0092] Step 6.4: The time corresponding to the probability of the random number ξ As the time range for the next interpolation. When the condition of formula (9) is met, execute step 6.5; otherwise, return to step 6.2.
[0093]
[0094] Wherein, α is the preset value for terminating the interpolation iteration. The smaller the value of α is, the higher the calculation accuracy is, but a relatively longer calculation time is required. The value of α can be adjusted according to the accuracy requirement, for example, α = 0.01.
[0095] Step 6.5: Use linear interpolation to calculate the system state transition time:
[0096]
[0097] Step 7: Generate a random number, determine the component for state transfer based on the random number and the residual cumulative probability density function of each component state transfer, and update the state of the component.
[0098] According to the Monte Carlo random selection mechanism, a random number δ uniformly distributed in the range of 0-1 is generated to determine the component j of the state transition * , so that the formula holds:
[0099]
[0100] For multi-state components, the transfer state of the component can be further determined by the same logic.
[0101] Step 8: Determine the system state based on the updated component state combination, calculate the cumulative predicted failure times of the system within the prediction deadline, and calculate the system reliability based on the cumulative predicted failure times.
[0102] Step 8.1: Update the current accumulated working time T:
[0103] T←T+t (n) (12)
[0104] Depending on different conditions, it can be executed in the following three ways:
[0105] Step 8.2: When T ≤ T Z And Φ(x1,x2,…,x K )=0, N sim ←N sim +1, N←N+1, execute step 8.5;
[0106] Step 8.3: When T>T Z When N sim ←N sim +1, then proceed to step 8.5;
[0107] Step 8.4: When the above conditions are not met, Then proceed to step 2.
[0108] Step 8.5: When N sim >N max , calculate the system reliability R S ; Otherwise, proceed to step 1.2.
[0109]
[0110] For a single-component unrepairable system, assume that the component failure time follows the Weibull distribution with parameters η = 100.00 and b = 2.00. According to the interpolation iteration algorithm of the present invention, the interpolation point generation and selection process is as follows: Figure 1 As shown, the interpolation result distribution test is as follows Figure 2 Fewer interpolation data points will result in low calculation accuracy, while more data points will increase the amount of calculation and reduce the calculation efficiency. Figure 3As shown in the figure. The interpolation point sequence is generated through multiple iterations, and the interval range of its probability value p is continuously reduced, which can gradually approach the random number ξ. The advantage of interpolation iteration is that it can ensure the unidirectional improvement of interpolation accuracy and avoid the efficiency defect of excessively increasing the number of interpolation points to improve accuracy. Since the steps of generating the interpolation point sequence and interpolation calculation are all substitution calculations, which do not involve solving equations, the calculation speed is fast and the execution efficiency is high.
[0111] The reliability prediction method based on the accelerated simulation strategy of the embodiment of the present invention can be applied to various probability distributions. Figure 4 The multi-component system shown is taken as an example. Assuming that the failure time and maintenance time of all components in the system obey the Weibull distribution, the reliability prediction method based on the accelerated simulation strategy of the embodiment of the present invention includes the following steps:
[0112] Step 1: Set the number of simulations and the deadline for reliability estimation.
[0113] Step 1.1: Set the system failure count N = 0, the simulation count variable N sim =0, and the maximum number of simulations is N max =10 4 .
[0114] Step 1.2: Set the current time T = 0 and the deadline for reliability estimation T Z =100h.
[0115] Step 2: Figure 4 The system shown in the figure consists of three components, and the state variable of component j is x j The system status is shown in Table 1. Assume that all components are in good initial state, that is, x1=x2=x3=1.
[0116] Table 1 System status table
[0117] <![CDATA[x1]]> <![CDATA[x2]]> <![CDATA[x3]]> <![CDATA[Φ(x1,x2,x3)]]> 1 1 1 1 0 1 1 1 1 0 1 1 1 1 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0
[0118] Step 3: Assume the cumulative probability distribution function F of component state transition j (t) and the probability density function f j (t). For a repairable system, if the state transition of a component follows a Weibull distribution, the shape parameter b and scale parameter η of its failure and repair are shown in Table 2. For an unrepairable system, the repair transfer parameters of the components in Table 2 are omitted.
[0119] Table 2 Component parameters
[0120]
[0121] Taking the failover of component 1 as an example, the probability distribution function of its failover is:
[0122]
[0123] The probability density function of its failover is:
[0124]
[0125] Step 4: Accumulate working time by component Calculate the residual cumulative probability distribution function and residual probability density function of component state transition. The initial cumulative working time of component j Taking the fault transfer of component 1 as an example, the residual cumulative probability distribution function of its fault state transfer is and probability density They are:
[0126]
[0127]
[0128] Step 5: Calculate the residual cumulative probability distribution function of the system state transition. The system contains 3 components, and the residual cumulative distribution function of the system state transition probability is for:
[0129]
[0130] Step 6: Follow the steps below to calculate the nth system state transition time t( 1 ), n=1 when executed for the first time.
[0131] Step 6.1: Generate a random number ξ from a uniform distribution in the range 0-1.
[0132] Step 6.2: Given the initial interpolation time range, generate an interpolation sequence within the interpolation time range. m ], and satisfy the following relations:
[0133]
[0134] Assume random number ξ=0.38, take t1=0.00, t m =200.00, m=11. According to the formula, we can get The relationship that satisfies the formula.
[0135] In the interval [0,200], 9 interpolation points are inserted at equal intervals to form a vector t = [0.00, 20.00, 40.00, ..., 200.00]. According to the cumulative probability distribution function By calculating the corresponding probability value vector p, we can get the interpolation point vector combination, as shown in Table 3.
[0136] Table 3 Interpolation point combinations
[0137] <![CDATA[t i / h]]> 0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00 160.00 180.00 200.00 <![CDATA[p i ]]> 0.000 0.144 0.297 0.434 0.550 0.647 0.725 0.788 0.837 0.876 0.906
[0138] Step 6.3: Determine the ranking position of ξ in the probability value vector p:
[0139] 0<...<0.297<ξ<0.434<...<0.906 (20)
[0140] Step 6.4: Use the time corresponding to the probability of the random number ξ as the time range for the next interpolation. That is, t1 = 40.00, t m =60.00, when the conditions of the formula are met, execute step 6.5; otherwise, return to step 6.2.
[0141]
[0142] The smaller the value of α is, the higher the calculation accuracy is, for example, α=0.01.
[0143] Step 6.5: Use linear interpolation to calculate the system state transition time t( 1 ):
[0144]
[0145] Step 7: Generate a random number δ uniformly distributed in the range of 0-1 according to the Monte Carlo random selection mechanism to determine the component j of the state transition * For example, the current component state combination is (1,1,1). According to the formula, the system state transition time t( 1 )=51.807. According to the Monte Carlo random selection mechanism, the residual probability density of the component's current state transition is calculated according to the formula
[0146]
[0147]
[0148]
[0149] The proportions of different component state transfers are:
[0150]
[0151]
[0152]
[0153] For example, when the random number δ=0.5, according to the formula, component 2 is selected for failover:
[0154] 0.42017<ζ<0.42017+0.42017 (29)
[0155] Step 8: Update the current accumulated working time T:
[0156] T←T+t (1) (30)
[0157] According to the different execution conditions of step 8, the current condition is T<T Z And Φ(x1,x2,x3)=1, step 2 should be executed.
[0158] Step 9: When N sim >N max , calculate the system reliability R S ; Otherwise, proceed to step 1.2.
[0159]
[0160] According to the current specific implementation steps, N sim <N max , step 1.2 should be performed.
[0161] Since the simulation calculation needs to be executed several times in a loop, it will not be described here due to space limitations.
[0162] In order to reflect the effectiveness of the method of the present invention, the results are compared here in combination with the theoretical values of the analytical method. When all parameters b in Table 2 are set to 1.00 (Weibull distribution degenerates into exponential distribution), the calculation results are shown in Table 4. According to the parameter settings in Table 2, the calculation results are shown in Table 5.
[0163] Table 4 Comparison between the method of the present invention and the theoretical value
[0164] System Type Cut-off time T / h Theoretical value <![CDATA[Number of simulations N max > <![CDATA[System reliability R s > Relative error / % Repairable system 100.00 0.9879 <![CDATA[10 4 ]]> 0.9877 0.0202 Unrepairable system 100.00 0.6920 <![CDATA[10 4 ]]> 0.6909 0.1590
[0165] Table 5 Reliability prediction results based on the method of the present invention
[0166]
[0167]
[0168] It should be noted that although the operations of the method of the embodiments of this specification are described in a specific order in the above embodiments, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results.
[0169] A system reliability prediction system based on an interpolation iteration algorithm according to an embodiment of the present invention includes:
[0170] The setting module is used to set the total number of system simulations and the prediction deadline;
[0171] A state determination module is used to determine the system state corresponding to different component state combinations according to the system component composition structure;
[0172] A function calculation module is used to determine the cumulative probability distribution function and probability density function of the state transfer of each component of the system, determine the residual cumulative probability distribution function and residual cumulative probability density function of the state transfer of each component according to the cumulative working time of the component, and calculate the residual cumulative probability distribution function of the system state transfer according to the residual cumulative probability distribution function of the state transfer of each component;
[0173] An interpolation iteration module is used to determine the system state transfer time by using an interpolation iteration algorithm;
[0174] A state transition component determination module is used to generate a random number, determine the component for state transition according to the random number and the residual cumulative probability density function of each component state transition, and update the state of the component;
[0175] The reliability calculation module is used to determine the system state according to the updated component state combination, calculate the cumulative predicted failure times of the system within the prediction deadline, and calculate the system reliability according to the cumulative predicted failure times.
[0176] For the specific definition of the system reliability prediction system, please refer to the definition of the system reliability prediction method above, which will not be repeated here. Each module in the above system reliability prediction system can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0177] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0178] 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 system reliability prediction method based on interpolation iteration algorithm, characterized in that: include: Set the total number of system simulations and prediction deadline; According to the system component composition structure, determine the system state corresponding to the combination of different component states; Determine the cumulative probability distribution function and probability density function of the state transition of each component of the system; According to the cumulative working time of the components, determine the residual cumulative probability distribution function and the residual cumulative probability density function of the state transition of each component; Calculate the residual cumulative probability distribution function of system state transfer according to the residual cumulative probability distribution function of each component state transfer; The interpolation iteration algorithm is used to determine the system state transfer time including: Step 6.1: Generate a random number from a uniform distribution in the range 0-1 ; Step 6.2: For the residual cumulative probability distribution function , giving the initial time range for interpolation , and satisfy the following relations: ; In the interval Insert at equal intervals m -2 interpolation points, , and the interpolation points are arranged in ascending order to form a vector , according to the residual cumulative probability distribution function , calculate the corresponding probability value vector , , thus we can get the interpolation data point vector combination: ; Step 6.3: According to the residual cumulative probability distribution function of the system state transition, determine In the probability vector Sorting position in: ; In the formula, , Represents a random number The neighboring probability of Step 6.4: Random Numbers The time corresponding to the neighboring probability , As the time range for the next interpolation, let , , when the preset condition is met, execute step 6.5, otherwise, return to step 6.
2. The preset condition is expressed as: ; in, The preset value for terminating the interpolation iteration; Step 6.5: Calculate system state transition time : ; Generate a random number, determine the component for state transfer based on the random number and the residual cumulative probability density function of each component state transfer, and update the state of the component; The system state is determined based on the updated component state combination, the cumulative predicted failure times of the system within the prediction deadline are calculated, and the system reliability is calculated based on the cumulative predicted failure times.
2. A system reliability prediction method based on interpolation iteration algorithm as claimed in claim 1, characterized in that: Determining the system status corresponding to different component status combinations includes: The total number of system components is recorded as , components The state variable is , , is the set of components that cause system failure, and the system state function is recorded as , then the following relationship is satisfied: 。 3. A system reliability prediction method based on interpolation iteration algorithm as claimed in claim 2, characterized in that: The components The accumulated working time is recorded as , components The cumulative probability distribution function of state transition is recorded as , and its probability density function is recorded as , the components The residual cumulative probability distribution function of state transition is recorded as , and its residual cumulative probability density function is recorded as , and The calculation formula is: 。 4. A system reliability prediction method based on interpolation iteration algorithm as claimed in claim 3, characterized in that: The residual cumulative probability distribution function of the system state transition is recorded as , The calculation formula is: 。 5. A system reliability prediction method based on interpolation iteration algorithm as claimed in claim 3, characterized in that: The component for determining the state transition according to the random number and the residual cumulative probability density function of the state transition of each component includes: Let this random number be , determine the components of state transition , so that the formula is satisfied: 。 6. A system reliability prediction method based on interpolation iteration algorithm as claimed in claim 2, characterized in that: The calculation of the cumulative predicted number of failures and the reliability includes: Step 8.1: Add up the random numbers of the generated system state transition time to get the current cumulative working time of the system , the current simulation round is recorded as The total number of simulations is recorded as ; Step 8.2: When and hour, Add 1, the cumulative number of system failures Add 1 and execute step 8.5; Step 8.3: When hour, Add 1, then execute step 8.5; Step 8.4: If the conditions of steps 8.2 and 8.3 are not met, the random number of the generated system state transition time is added to the cumulative working time of the component, and then the system state corresponding to the different component state combinations is re-determined, and the subsequent steps are repeated; Step 8.5: When , calculate the system reliability : 。 7. A system reliability prediction system based on an interpolation iterative algorithm, used to implement the steps of the method as claimed in any one of claims 1 to 6, characterized in that: include: The setting module is used to set the total number of system simulations and the prediction deadline; A state determination module is used to determine the system state corresponding to different component state combinations according to the system component composition structure; A function calculation module is used to determine the cumulative probability distribution function and probability density function of the state transfer of each component of the system, determine the residual cumulative probability distribution function and residual cumulative probability density function of the state transfer of each component according to the cumulative working time of the component, and calculate the residual cumulative probability distribution function of the system state transfer according to the residual cumulative probability distribution function of the state transfer of each component; An interpolation iteration module is used to determine the system state transfer time by using an interpolation iteration algorithm; A state transition component determination module is used to generate a random number, determine the component for state transition according to the random number and the residual cumulative probability density function of each component state transition, and update the state of the component; The reliability calculation module is used to determine the system state according to the updated component state combination, calculate the cumulative predicted failure times of the system within the prediction deadline, and calculate the system reliability according to the cumulative predicted failure times.
