Residual life prediction method and system of KooN hot standby redundancy safety control system

By constructing a three-parameter Weibull distribution model and estimating parameters using the maximum likelihood method, the life prediction problem of KooN type safety control system under right-cut data is solved, and accurate residual life prediction and data demand are achieved.

CN118938637APending Publication Date: 2024-11-12CHINA PETROLEUM & CHEMICAL CORP +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310532874.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-11
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the case of including right-cut data, it is difficult to obtain the accurate prediction value of the remaining life of the KooN type security control system, and the existing methods have low accuracy and high data demand.

Method used

The right-cut data of multiple devices was obtained through the life test, and a three-parameter Weibull distribution model of a single device was constructed. The model parameters were estimated using the maximum likelihood method and optimization method, and the reliability model was established, which was then predicted to predict the remaining KooN type safety control system. Service life.

Benefits of technology

The precise life prediction of the KooN type safety control system is realized when the right-cut data is included, which reduces the demand for data, improves the prediction accuracy, and reflects the failure form of the actual equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938637A_ABST
    Figure CN118938637A_ABST
Patent Text Reader

Abstract

The invention provides a residual life prediction method and system for a KooN hot standby redundancy safety control system, and belongs to the technical field of modeling calculation. The method comprises the following steps: acquiring the service life data of a plurality of devices containing right truncation data through a service life test, and reducing the demanded quantity of the data. And constructing a three-parameter Weibull distribution model of a single device to describe the reliability of the device, thereby further reflecting the fault form of the actual device. And estimating parameters of the three-parameter Weibull distribution model based on a maximum likelihood method and an optimization method according to the service life data of the plurality of devices so as to obtain the three-parameter Weibull distribution model obeyed by the devices. And predicting the residual service life of the KooN type safety control system through the three-parameter Weibull distribution model obeyed by the equipment. Therefore, the purpose of obtaining the accurate prediction value of the residual life of the KooN type safety control system under the condition of containing the right truncation data is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of modeling and computing, and in particular to a method for predicting the remaining life of a KooN hot standby redundant safety control system and a system for predicting the remaining life of a KooN hot standby redundant safety control system. Background Art

[0002] Reliability is an important indicator to describe whether equipment can work properly. In key fields such as petrochemical, aerospace, and defense industries, once equipment fails, huge safety accidents often occur, causing a large number of casualties and economic losses. Therefore, the reliability requirements of equipment are more stringent. Modeling and calculating the reliability of equipment can obtain key information such as failure rate, reliability, and remaining service life of the equipment. The remaining service life can guide the maintenance and replacement of equipment, reduce unnecessary maintenance, and reduce economic costs. Redundancy is often used in engineering to improve the reliability of equipment, extend its life, and reduce the frequency of overall failures. Hot standby redundant systems are widely used in safety control systems and their components. Common redundancies include 1oo2 systems, 2oo3 systems, 2oo4 and 3oo5 systems, and the more general is the KooN system, that is, N identical devices are running at the same time. As long as K or more devices among them can work normally, the system can work normally. Therefore, for KooN type safety control systems, it is meaningless to predict the remaining service life of a single device in the system, and it is necessary to predict the remaining service life of the entire system.

[0003] The mainstream life prediction methods currently include empirical methods, machine learning methods, and statistical methods. The empirical methods rely on expert experience and are often less accurate; machine learning methods require a large amount of equipment degradation data, which is often difficult to obtain; statistical methods only require equipment life data to obtain more accurate prediction results, so they are highly valued. Commonly used statistical models include exponential distribution models and two-parameter and three-parameter Weibull distribution models. The probability density functions of the three models are:

[0004]

[0005]

[0006]

[0007] Among them, t is the running time of the equipment, β, γ, and η are the shape, location, and scale parameters of the distribution respectively. It can be seen that the exponential distribution and the two-parameter Weibull distribution can be regarded as simplified forms of the three-parameter Weibull distribution. When γ = 0, the two-parameter Weibull distribution is obtained, and when γ = 0 and β = 1, the exponential distribution is obtained. Therefore, the three-parameter Weibull distribution is more versatile than the previous two.

[0008] In existing research, some researchers use exponential distribution to predict the life information of the system. However, many literatures point out that the life of most electronic, mechanical, and electromechanical products can be considered to obey the Weibull distribution [Reference: Yin Wenhui. Reliability assessment of dedicated safety facilities for 650MW nuclear power units and its software development [D]. Southeast University, 2019.], so some researchers have proposed using the two-parameter Weibull distribution to predict the life of the system. It is worth pointing out that the two-parameter Weibull distribution assumes that the location parameter γ of the device is 0, which will lead to the failure of the obtained equipment reliability model. At any point in the interval (0, +∞), in fact, for most of the equipment that has been applied, the failure will not occur in the early stage of the equipment operation. Therefore, using the two-parameter Weibull model to describe the reliability of the equipment is imperfect. In addition to the mainstream statistical model-based methods, some researchers use machine learning methods such as neural networks to predict the life of the system. However, since most neural networks do not have a relatively complete mathematical foundation, the prediction is highly random, and the network training has high requirements for the quantity and quality of data, so this method is less practical. In addition, due to the high reliability of some equipment, it is difficult to obtain a large amount of complete life data, and it is often necessary to use right-truncated data for life prediction. Therefore, how to obtain an accurate prediction value of the remaining life of the KooN type safety control system when including right-truncated data is still a difficult problem. Summary of the invention

[0009] The purpose of the embodiments of the present invention is to provide a method and system for predicting the remaining life of a KooN hot standby redundant safety control system, so as to at least solve the above-mentioned problem that it is difficult to obtain an accurate prediction value of the remaining life of a KooN type safety control system when right-truncated data is included.

[0010] In order to achieve the above object, the first aspect of the present invention provides a method for predicting the remaining life of a KooN hot standby redundant safety control system, comprising:

[0011] Obtain life data of multiple devices including right-censored data through life tests;

[0012] Construct a three-parameter Weibull distribution model for a single device;

[0013] According to the life data of multiple equipment, the parameters of the three-parameter Weibull distribution model are estimated based on the maximum likelihood method and optimization method to obtain the three-parameter Weibull distribution model that the equipment obeys;

[0014] The remaining useful life of the KooN type safety control system is predicted by the three-parameter Weibull distribution model obeyed by the equipment.

[0015] Optionally, the above method estimates the parameters of the three-parameter Weibull distribution model based on the maximum likelihood method and the optimization method according to the life data of multiple devices to obtain the three-parameter Weibull distribution model obeyed by the equipment, including:

[0016] Based on the life data of multiple equipments, the likelihood function of the three-parameter Weibull distribution model is obtained;

[0017] Use optimization methods to find the maximum value of the likelihood function and obtain the best estimate of the shape parameter Best estimate of the location parameter and the best estimate of the scale parameter

[0018] Optionally, the three-parameter Weibull distribution model of a single device is constructed as follows:

[0019] Construct the probability density function, cumulative distribution function and reliability function of the three-parameter Weibull distribution, where the probability density function of the three-parameter Weibull distribution is: The cumulative distribution function is The reliability function is t is the equipment operation time, γ ≥ 0 is the location parameter, β > 0 is the shape parameter, and η > 0 is the scale parameter.

[0020] Optionally, the likelihood function of the three-parameter Weibull distribution model is obtained based on the life data of multiple devices, including:

[0021] When there are n life span data, r complete life span data, and nr right-truncated data, r is recorded as t j , let nr be t k , then the likelihood function is expressed as

[0022] t i Denoted as including t j and t k Life data, i = 1, 2, ..., n, to simplify the likelihood function, we get Taking the logarithm, we get the log-likelihood function:

[0023] Optionally, the above optimization method is used to solve the maximum value of the likelihood function and obtain the best estimate of the shape parameter Best estimate of the location parameter and the best estimate of the scale parameter include:

[0024] Set β∈(1,10] and use the optimization method to solve the optimization objective function get

[0025] Optionally, the above optimization method includes an interior point method.

[0026] Optionally, the above-mentioned prediction of the remaining useful life of the KooN type safety control system by the three-parameter Weibull distribution model obeyed by the equipment includes:

[0027] According to the working principle of the KooN type safety control system, it is determined that when at least N-K+1 devices in the KooN type safety control system fail, the KooN type safety control system is in a system failure state, and the distribution function of the system failure F s (t) is Among them, N represents that there are N identical devices running simultaneously in the KooN type safety control system, and K represents that there are K devices running normally among N identical devices;

[0028] Distribution function F s (t) Take the derivative and get the probability density function f of system failure s (t) is Among them, F s (t) and f s (t) all satisfy The condition when When F s (t) and f s (t) are all 0;

[0029] The running time of the current KooN type safety control system is recorded as δ ≥ 0. According to the conditional probability formula, the probability density function f of the remaining service life T of the KooN type safety control system is obtained: T (T) is

[0030]

[0031] For the probability density function f T (T) Calculate the expectation and obtain the expected value of the remaining service life of the KooN type safety control system after being put into use for δ time:

[0032] According to f s(t), when the running time of the KooN type safety control system When 1-F s (δ) = 1, Then the expected value of remaining life E(T) is expressed as

[0033] Optionally, the life data of multiple devices including right-truncated data obtained through life testing may include:

[0034] The timed truncation test or the constant truncation test is used to obtain the life data of multiple devices including right-truncation data.

[0035] Optionally, the life of any device follows a three-parameter Weibull distribution, a two-parameter Weibull distribution or an exponential distribution.

[0036] A second aspect of the present invention provides a remaining life prediction system for a KooN hot standby redundant safety control system, comprising:

[0037] An equipment life data acquisition module, used for acquiring life data of multiple equipments including right-truncated data through life tests;

[0038] Distribution model building module, used to build a three-parameter Weibull distribution model for a single device;

[0039] The distribution model parameter estimation module is used to estimate the parameters of the three-parameter Weibull distribution model based on the life data of multiple devices, based on the maximum likelihood method and optimization method, so as to obtain the three-parameter Weibull distribution model that the equipment obeys;

[0040] The remaining service life prediction module is used to predict the remaining service life of the KooN type safety control system through the three-parameter Weibull distribution model obeyed by the equipment.

[0041] In a third aspect of the present invention, a machine-readable storage medium is provided, on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the remaining life prediction method of the KooN hot standby redundant safety control system.

[0042] In a fourth aspect of the present invention, an electronic device is provided, the electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the remaining life prediction method of the KooN hot standby redundant safety control system when executing the computer program.

[0043] Through the above technical scheme, a method and system for predicting the remaining life of a KooN hot standby redundant safety control system is provided. The life data of the equipment is obtained through life tests. When the equipment reliability is high and the complete life data is difficult to obtain, the method and system can obtain the life data of multiple equipments including right-truncated data, thereby reducing the demand for data, thereby effectively avoiding the problem of low estimation accuracy due to the difficulty in obtaining equipment life data. A three-parameter Weibull distribution model is constructed to describe the reliability of the equipment, thereby further reflecting the failure form of the actual equipment. Then, based on the life data of multiple equipments, a likelihood function is established to more effectively utilize the life data of the equipment. Then, the maximum likelihood method and optimization method are used to estimate the parameters of the three-parameter Weibull distribution model to obtain the three-parameter Weibull distribution model that the equipment obeys, that is, a reliability model is established, and then the remaining service life of the KooN type safety control system is calculated through the reliability model, and a relatively accurate life prediction result for the KooN type safety control system is obtained. Thus, the purpose of obtaining an accurate prediction value of the remaining life of the KooN type safety control system in the case of right-truncated data is achieved.

[0044] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0046] Figure 1 It is a flow chart of a method for predicting the remaining life of a KooN hot standby redundant safety control system provided by an embodiment of the present invention;

[0047] Figure 2 It is a probability density function form actually obeyed by the device in the 2oo4 system provided by an embodiment of the present invention and an estimated probability density function;

[0048] Figure 3 It is a distribution function form actually obeyed by the device in the 2oo4 system and an estimated distribution function provided by an embodiment of the present invention;

[0049] Figure 4 is an actual value and a predicted value of the remaining useful life of a 2oo4 system provided by an embodiment of the present invention;

[0050] Figure 5 is an actual value and a predicted value of the remaining useful life of a 2oo3 system provided by an embodiment of the present invention;

[0051] Figure 6 is an actual value and a predicted value of the remaining useful life of a 1oo2 system provided by an embodiment of the present invention;

[0052] Figure 7 is the actual value and predicted value of the remaining useful life of a 3oo5 system provided by an embodiment of the present invention;

[0053] Figure 8 It is a structural block diagram of a remaining life prediction system of a KooN hot standby redundant safety control system provided by an embodiment of the present invention;

[0054] Fig. 9 It is a schematic diagram of the structure of an electronic device provided by a preferred embodiment of the present invention.

[0055] Description of Reference Numerals

[0056] 10 - electronic device, 100 - processor, 101 - memory, 102 - computer program. DETAILED DESCRIPTION

[0057] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0058] Example 1

[0059] Figure 1 FIG. 1 is a flowchart of a method for predicting the remaining life of a KooN hot standby redundant safety control system provided by an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a method for predicting the remaining life of a KooN hot standby redundant safety control system, the method comprising:

[0060] S110: Acquire life data of a plurality of devices including right-truncated data through life tests;

[0061] In some implementations of this embodiment, the above-mentioned obtaining the life data of multiple devices including right-truncated data through life testing includes: using a timed truncation test or a constant truncation test to obtain the life data of multiple devices including right-truncated data.

[0062] Specifically, a life test is performed on the equipment in the KooN type safety control system to obtain a series of life data. When the equipment reliability is high and the complete life data is difficult to obtain, a timed truncation test or a constant truncation test can be used to obtain life data including right truncation data. This method only needs to obtain the life data of the equipment, that is, the failure time, through life tests, and the demand for data is relatively small.

[0063] Among them, the timed truncation test means that a test time is set in advance, and the test is stopped when the specified time is reached. The fixed number truncation test means that the test is stopped when a preset number of failures occurs in the experimental sample.

[0064] S120: Construct a three-parameter Weibull distribution model for a single device;

[0065] In some implementations of this embodiment, the three-parameter Weibull distribution model of a single device is constructed as follows: constructing a three-parameter Weibull distribution probability density function, a cumulative distribution function, and a reliability function, wherein the three-parameter Weibull distribution probability density function is: The cumulative distribution function is The reliability function is t is the equipment operation time, γ ≥ 0 is the location parameter, β > 0 is the shape parameter, and η > 0 is the scale parameter.

[0066] Specifically, the three-parameter Weibull distribution model is a generalized form of the exponential distribution and the two-parameter Weibull distribution model. Compared with the exponential distribution and the two-parameter Weibull distribution, the three-parameter Weibull distribution model is used to describe the reliability of the equipment and can better reflect the failure mode of the actual equipment.

[0067] S130: according to the life data of multiple devices, the parameters of the three-parameter Weibull distribution model are estimated based on the maximum likelihood method and the optimization method to obtain the three-parameter Weibull distribution model obeyed by the equipment;

[0068] In some implementations of this embodiment, the above-mentioned method estimates the parameters of the three-parameter Weibull distribution model based on the maximum likelihood method and the optimization method according to the life data of multiple devices to obtain the three-parameter Weibull distribution model obeyed by the device, including: obtaining the likelihood function of the three-parameter Weibull distribution model based on the life data of multiple devices. Using the optimization method to solve the maximum value of the likelihood function to obtain the best estimate of the shape parameter Best estimate of the location parameter and the best estimate of the scale parameter

[0069] Specifically, the likelihood function of the three-parameter Weibull distribution is obtained based on the life data, and then the model parameters of the Weibull distribution are estimated based on the maximum likelihood method. The establishment of the above likelihood function can utilize the right-truncated life data of the equipment, which is particularly effective when the equipment data is less. When the equipment reliability is high and the complete life data is less, the life data of the equipment can be more effectively utilized through S130. And the maximum value of the likelihood function is solved by the optimization method, which can simplify the process of solving the likelihood function. Therefore, the parameters of the distribution model are estimated by combining the maximum likelihood method with the optimization method, and a reliability model is established to calculate the remaining service life of the KooN type safety control system.

[0070] Among them, since the location parameter is introduced into the three-parameter Weibull distribution model obeyed by the equipment, the equipment failure occurs after the location parameter, that is, there is a threshold as the starting point of the possible occurrence time of the failure, which is more in line with the actual equipment.

[0071] S140: Predict the remaining useful life of the KooN type safety control system through the three-parameter Weibull distribution model obeyed by the equipment.

[0072] In the above implementation process, the method obtains the life data of the equipment through life tests, and when the equipment reliability is high and the complete life data is difficult to obtain, the method can obtain the life data of multiple equipment including right-truncated data, thereby reducing the demand for data, thereby effectively avoiding the problem of low estimation accuracy due to the difficulty in obtaining equipment life data. And a three-parameter Weibull distribution model is constructed to describe the reliability of the equipment, thereby further reflecting the failure form of the actual equipment. Then, based on the life data of multiple equipment, a likelihood function is established to more effectively utilize the life data of the equipment. Then, the maximum likelihood method and optimization method are used to estimate the parameters of the three-parameter Weibull distribution model to obtain the three-parameter Weibull distribution model that the equipment obeys, that is, a reliability model is established, and then the remaining service life of the KooN type safety control system is calculated through the reliability model, and a relatively accurate life prediction result of the KooN type safety control system is obtained. Thereby, the purpose of obtaining an accurate prediction value of the remaining life of the KooN type safety control system in the case of right-truncated data is achieved.

[0073] It should be noted that this method can not only make predictions for all devices in the entire safety control system, but also make predictions for all hardware of a certain device.

[0074] Optionally, the likelihood function of the three-parameter Weibull distribution model is obtained based on the life data of multiple devices, including:

[0075] When there are n life span data, r complete life span data, and nr right-truncated data, r is recorded as t j , let nr be t k , then the likelihood function is expressed as

[0076] t i Denoted as including t j and t k Life data, i = 1, 2, ..., n, to simplify the likelihood function, we get Taking the logarithm, we get the log-likelihood function:

[0077] Optionally, the above optimization method is used to solve the maximum value of the likelihood function and obtain the best estimate of the shape parameter Best estimate of the location parameter and the best estimate of the scale parameter include:

[0078] Set β∈(1,10] and use the optimization method to solve the optimization objective function get

[0079] Specifically, in order to find the optimal Three parameters, this method uses the optimization method to iteratively search for the best value and obtain the parameter estimate that maximizes the log-likelihood function. According to the physical meaning of the parameters, we take γ∈[0,min(t i )], β∈(0,10]. However, when β<1, the equipment failure rate λ(t) decreases with the increase of running time, which is usually used to describe the situation of new product testing. When β=1, the failure rate is constant, and when β>1, the failure rate decreases with the increase of running time, which are often used to describe the random failure and aging failure of equipment respectively. Therefore, when this method is used to describe the reliability of the KooN system, β∈(1,10]. The optimization objective function of this method is:

[0080]

[0081]

[0082] Use optimization methods such as interior point method to solve the objective function and obtain the optimal parameter estimate

[0083] Optionally, the above optimization method includes an interior point method.

[0084] Specifically, the calculation steps of the interior point method are as follows:

[0085] (1) The original constrained optimization problem is:

[0086]

[0087]

[0088] Among them, θ = (β, γ, η), the penalty function can be constructed:

[0089] g(θ)=ln(β-1)+ln(10-β)+ln(min(t i )-γ)+ln(γ)+ln(η)

[0090] (2) Take μ (1) is the initial penalty factor, usually μ (1) =1. Then the original constrained optimization problem can be transformed into an unconstrained optimization problem:

[0091]

[0092] (3) Take the initial solution θ in the feasible region D (0) =(β (0) ,γ (0) ,η (0) ), let k = 1, ε>0 is the set error limit, usually a very small positive number, such as 10-5.

[0093] (4) From θ * (μ (k-1) ) to solve the above optimization problem using the unconstrained optimization method, we can get φ(θ,μ (k) ) of the extreme point θ * (μ (k) ), when k = 1, θ * (μ (0) )=θ (0) .

[0094] (5) Check the iteration termination criteria. If ||θ is satisfied, * (μ (k) )-θ * (μ (k-1) )||≤ε, the iterative calculation is stopped and θ * (μ (k) ) is the optimal solution of the original objective function θ * ; Otherwise, go to the next step.

[0095] (6) Let μ (k+1) =cμ (k) , k=k+1, where c is the decreasing coefficient, usually 0.1, and is transferred to (4).

[0096] Optionally, the above-mentioned prediction of the remaining useful life of the KooN type safety control system by the three-parameter Weibull distribution model obeyed by the equipment includes:

[0097] According to the working principle of the KooN type safety control system, it is determined that when at least N-K+1 devices in the KooN type safety control system fail, the KooN type safety control system is in a system failure state, and the distribution function of the system failure F s (t) is Among them, N represents that there are N identical devices running simultaneously in the KooN type safety control system, and K represents that there are K devices running normally among N identical devices;

[0098] Distribution function F s (t) Take the derivative and get the probability density function f of system failure s (t) is Among them, F s (t) and f s (t) all satisfy The condition when When F s (t) and f s (t) are all 0;

[0099] The running time of the current KooN type safety control system is recorded as δ≥0, and the probability density function f of the remaining service life T of the KooN type safety control system is obtained according to the conditional probability formula: T (T) is

[0100]

[0101] Where T is the remaining life, δ is the current running time, and t is the total running time. p(t-δ=T|t≥δ) represents the probability that the system is not damaged at time δ, that is, when t≥δ, the remaining life is T. The second line in the formula is the expansion of the first line according to the conditional probability formula. The numerator is the probability that the system is not damaged at time δ and the remaining life is T at the same time. Since T≥0, when p(t-δ=T) occurs, p(t≥δ) must occur, so the numerator is transformed into the numerator of the third line, that is, the probability that the system fails at t=T+δ. The denominator of the second line in the formula is the probability that the total running time is greater than the current running time, which is equivalently replaced by the denominator of the third line, that is, the probability that the system is not broken at the current time.

[0102] For the probability density function f T (T) Calculate the expectation and obtain the expected value of the remaining service life of the KooN type safety control system after being put into use for δ time:

[0103] According to fs (t), when the running time of the KooN type safety control system hour, Then the expected value of remaining life E(T) is expressed as

[0104] in, is the expected life of the KooN safety control system from the time of commissioning to failure. It can be seen that the remaining service life of the system is When , it decreases linearly. Therefore, when the running time is short, it is only necessary to calculate the system's expected life span once and then subtract it from the current running time to obtain the remaining service life. It is not necessary to calculate the integral all the time to obtain the current remaining service life, which reduces the calculation complexity. When , the remaining service life of the system presents a nonlinear form, and the accurate remaining service life needs to be obtained by calculating the above formula. Since the calculation of the integral is very complicated, this method uses the computer to calculate its numerical solution using the adaptive integral method, which does not require relevant professional knowledge of the system and is easy to implement.

[0105] Specifically, according to the definition of the distribution function, the distribution function of the KooN type safety control system failure, that is, the failure of the N-K+1th device, is obtained, thereby obtaining the probability density function of the system failure. According to the definition of the remaining life and the conditional probability formula, the probability density function of the remaining service life of the system is obtained, thereby obtaining the expected value of the remaining service life. According to the formula of the expected value, the remaining life is segmented, thereby further simplifying the calculation of the remaining life. This method is based on the conditional probability calculation of the remaining service life when it runs to a certain moment, which can better describe the nonlinear relationship of the equipment life, and divide the remaining life into piecewise functions, simplifying the complexity of the remaining life calculation and improving the accuracy of the equipment remaining life prediction. The remaining service life derived based on conditional probability has a nonlinear relationship with the running time and is closer to the actual system life.

[0106] In the above implementation process, this method can be widely used in the life prediction of safety control systems with KooN structures. This method can model the reliability of the system and calculate its expected remaining life in real time, guide the maintenance and replacement of equipment, and prevent cost expenditure caused by excessive maintenance and safety accidents caused by untimely maintenance.

[0107] Optionally, the life of any device follows a three-parameter Weibull distribution, a two-parameter Weibull distribution or an exponential distribution.

[0108] It should be noted that the remaining life prediction method of the KooN hot standby redundant safety control system is applicable only if the equipment life obeys the Weibull distribution, whether it is a three-parameter Weibull distribution, a two-parameter Weibull distribution or an exponential distribution, in order to achieve the best prediction effect. And if the equipment life does not obey the Weibull distribution, the remaining life prediction method of the KooN hot standby redundant safety control system is also applicable.

[0109] Example 2

[0110] This embodiment uses a KooN hot standby redundant safety control system remaining life prediction method to predict the remaining service life of the 2oo4 type safety control system. The specific process is as follows:

[0111] Since the failure data of the KooN system is difficult to obtain, the data that conforms to the Weibull distribution can be generated by combining computer simulation as life data for experiments to verify the accuracy of this method in predicting the remaining service life of the KooN system.

[0112] Step 1: Life Test Obtain the equipment life which may include right-censored data.

[0113] Assume that a certain model of equipment obeys a three-parameter Weibull distribution (lifetime unit: hours) with shape parameter β = 5, location parameter γ = 3000, and scale parameter η = 20000. Based on Matlab 2020b software, generate 100 life data that obey this distribution, and assume that the redundant system is a 2oo4 system, that is, N = 4, K = 2. The generated 100 data are sorted as shown in Table 1:

[0114] Table 1 Lifespan data of 100 devices generated (unit: h)

[0115]

[0116] It is worth pointing out that the lifespans of the 90th to 100th devices in the table are the same. This is because it is difficult to obtain complete life data for some high-reliability devices, and it is often necessary to conduct timed or constant life tests to obtain right-truncated data. In order to simulate this situation, the 10 largest devices among the 100 generated device life data are changed to be the same as the 90th device, indicating that the life test was truncated at this moment, that is, in this life test, the first 90 devices ran into a fault state, and the corresponding data was a complete life, while the last 10 devices had not yet failed at the end of the test, and were right-truncated data.

[0117] Step 2: Construct a three-parameter Weibull distribution model for a single device.

[0118] The probability density function f(t), distribution function F(t) and reliability function R(t) of the life of a single device in the 2oo4 system are: t is the equipment operation time, γ ≥ 0 is the location parameter, β > 0 is the shape parameter, and η > 0 is the scale parameter.

[0119] Step 3: Solve the three parameters of the equipment distribution based on the maximum likelihood method and optimization method.

[0120] Among the 100 life span data, the complete life span data is t j ,j=1,2,…,90, right truncated data are denoted as t k ,k=1,2,…,10, then its likelihood function can be expressed as:

[0121]

[0122] Note i To include t j and t k The life data of 100 devices, i = 1, 2, ..., 100, can be simplified to obtain:

[0123]

[0124] Taking the logarithm we get:

[0125]

[0126] In order to find the optimal parameters β, γ, and η, an optimization method is used to iteratively search for the best value and obtain the parameter estimation value that maximizes the log-likelihood function. For this embodiment, the optimization objective function is:

[0127]

[0128]

[0129] Use the interior point method to solve the objective function and obtain the optimal parameter estimate Therefore, the probability density function f(t), distribution function F(t) and reliability function R(t) of the life of a single device in the 2oo4 system are as follows:

[0130]

[0131]

[0132]

[0133] The probability density function form actually obeyed by the equipment in the 2oo4 system and the estimated probability density function are as follows: Figure 2 As shown in the figure, the actual cumulative distribution function and the estimated cumulative distribution function in the 2oo4 system are as follows Figure 3 As shown, we can see that although the three parameters β, γ, and η of the Weibull distribution actually obeyed by the device are different from the estimated parameters There are some differences, but the distribution functions of the two are almost the same, and similar distribution functions mean that the two have similar expected values, that is, similar life expectancy.

[0134] Step 4: Calculate the expected remaining useful life of the 2oo4 system.

[0135] According to the working principle of 2oo4 system, the system fails when at least 3 devices in the system fail. Therefore, the distribution function of system failure is F s (t) is:

[0136]

[0137] Derivative the above equation and substitute the estimated parameters The probability density function of system failure can be obtained as f s (t) is:

[0138]

[0139] The above F s (t), f s (t) are all under the condition of t≥8902.28, when 0< <8902.28,

[0140] F s (t), f s (t) is 0.

[0141] Let the current system running time be δ ≥ 0, and the probability density function f of the remaining service life T of the system is T (T), the expected value of remaining life E(T) can be expressed as form.

[0142] The remaining useful life of the system is When , it decreases linearly. Therefore, when the running time is short, it is only necessary to calculate the system's expected life span once and then subtract it from the current running time to obtain the expected life span. It is not necessary to calculate the integral all the time to obtain the current remaining service life, which reduces the calculation complexity. When the remaining service life of the system is nonlinear, the accurate remaining service life needs to be calculated by the formula Because the calculation of the integral is very complicated, this embodiment uses an adaptive integral method based on a computer to calculate its numerical solution, and obtains the actual remaining life mean value and the predicted life expectancy value at some moments of the equipment operation as shown in Table 2 below:

[0143] Table 2 2oo4 system life prediction results (unit: h)

[0144]

[0145] It can be seen that the life results predicted by the scheme of this method are relatively close to the actual values, and the absolute value of the maximum percentage error is about 5.9%.

[0146] Set the system's running time δ∈[5000,26000], calculate the remaining life expectancy of the 2oo4 system, and obtain the remaining life corresponding to each moment of δ as follows Figure 4 As shown, Figure 4 It is the actual value and predicted value of the remaining service life of the 2oo4 system provided by an embodiment of the present invention. And it is compared with the actual remaining service life of the system at the corresponding time δ, where the actual value is obtained by generating 10,000 sets of the same 2oo4 system and calculating the mean value of the remaining service life of the system whose actual service life is greater than δ. It can be seen that the error between the predicted value and the actual value of the remaining service life is small, and the form of both is consistent with the inference of the remaining service life of the system by this method, that is, when δ is small, the remaining service life decreases linearly with the increase of the operating time, and when δ is large, the nonlinearity between the two becomes more serious.

[0147] Example 3

[0148] This embodiment uses a KooN hot standby redundant safety control system remaining life prediction method to predict the remaining service life of the 2oo3 type safety control system. The specific process is as follows:

[0149] Assuming that the life of a certain model of equipment obeys a three-parameter Weibull distribution (life unit: day) with shape parameter β = 3, location parameter γ = 700, and scale parameter η = 4000, 100 life data that obey this distribution are generated based on Matlab 2020b software, and assuming that the redundant system is a 2oo3 system, that is, N = 3, K = 2. The 100 generated data are sorted as shown in Table 3:

[0150] Table 3 Lifespan data of 100 devices generated (unit: d)

[0151]

[0152] According to the technical solution and steps 1-4 in Example 1, the remaining life of the 2oo3 hot standby redundant system composed of the equipment is predicted using the life data in the above table, and the actual remaining life mean and the predicted life expectancy at some moments of equipment operation are obtained as shown in Table 4 below:

[0153] Table 4 2oo3 system life prediction results (unit: d)

[0154]

[0155] It can be seen that the life results predicted by the scheme of this method are relatively close to the actual values, and the absolute value of the maximum percentage error is about 3.2%.

[0156] Set the system's running time δ∈[500,6000], calculate the remaining life expectancy of the 2oo3 system, obtain the remaining life at each moment corresponding to δ and compare it with the actual remaining life of the system at the corresponding moment δ, as follows: Figure 5 As shown, Figure 5 It is an actual value and a predicted value of the remaining useful life of a 2oo3 system provided by an embodiment of the present invention.

[0157] Example 4

[0158] This embodiment uses a KooN hot standby redundant safety control system remaining life prediction method to predict the remaining service life of the 1oo2 type safety control system. The specific process is as follows:

[0159] Assume that the life of a certain model of equipment obeys a three-parameter Weibull distribution (life unit: hours) with shape parameter β = 6, location parameter γ = 10000, and scale parameter η = 30000. Based on Matlab 2020b software, generate 100 life data that obey this distribution, and assume that the redundant system is a 1oo2 system, that is, N = 2, K = 1. Among the 100 life data generated, there are 10 right-truncated data, as shown in Table 5 below:

[0160] Table 5 Lifespan data of 100 devices generated (unit: h)

[0161]

[0162] According to the technical solution and steps 1-4 in Example 1, the remaining life of the 1oo2 hot standby redundant system composed of the equipment of this type is predicted using the 100 pieces of life data generated, and the actual remaining life mean and the predicted life expectancy at some moments of equipment operation are obtained as shown in Table 6 below:

[0163] Table 6 1oo2 system life prediction results (unit: h)

[0164]

[0165] It can be seen that the life results predicted by the scheme of this method are relatively close to the actual values, and the maximum absolute value of the percentage error is about 4.8%.

[0166] Set the system's running time δ∈[10000,45000], calculate the expected remaining life of the 1oo2 system, obtain the remaining life at each moment of δ and compare it with the actual remaining life of the system at the corresponding moment of δ, as follows: Figure 6 As shown, Figure 6 It is an actual value and a predicted value of the remaining useful life of a 1oo2 system provided by an embodiment of the present invention.

[0167] Example 5

[0168] This embodiment uses a KooN hot standby redundant safety control system remaining life prediction method to predict the remaining service life of the 3oo5 type safety control system. The specific process is as follows:

[0169] Assuming that the life of a certain type of equipment obeys a three-parameter Weibull distribution (life unit: hours) with shape parameter β = 4, location parameter γ = 5000, and scale parameter η = 25000, 100 life data that obey this distribution are generated based on Matlab 2020b software, and assuming that the redundant system is a 3oo5 system, that is, N = 3, K = 1. The 100 generated life data contain 10 right-truncated data, which are sorted as shown in Table 7:

[0170] Table 7 Lifespan data of 100 devices generated (unit: h)

[0171]

[0172] According to the technical solution and steps 1-4 in Example 1, the remaining life of the 3oo5 hot standby redundant system composed of this type of equipment is predicted using the life data in the above table, and the actual remaining life mean and predicted life expectancy at some moments of equipment operation are obtained as shown in Table 8 below:

[0173] Table 8 3oo5 system life prediction results (unit: h)

[0174]

[0175] It can be seen that the life results predicted according to the scheme of this method are relatively close to the actual values, and the absolute value of the maximum percentage error is about 0.4%.

[0176] Set the system's running time δ∈[10000,35000], calculate the expected remaining life of the 3oo5 system, obtain the remaining life at each moment of δ and compare it with the actual remaining life of the system at the corresponding moment of δ, as follows: Figure 7 As shown, Figure 7 It is the actual value and predicted value of the remaining useful life of the 3oo5 system provided by an embodiment of the present invention.

[0177] Example 6

[0178] Figure 8 FIG. 1 is a structural block diagram of a remaining life prediction system of a KooN hot standby redundant safety control system provided by an embodiment of the present invention. Figure 8 As shown, an embodiment of the present invention provides a remaining life prediction system for a KooN hot standby redundant safety control system, comprising:

[0179] An equipment life data acquisition module, used for acquiring life data of multiple equipments including right-truncated data through life tests;

[0180] Distribution model building module, used to build a three-parameter Weibull distribution model for a single device;

[0181] The distribution model parameter estimation module is used to estimate the parameters of the three-parameter Weibull distribution model based on the life data of multiple devices, based on the maximum likelihood method and optimization method, so as to obtain the three-parameter Weibull distribution model that the equipment obeys;

[0182] The remaining service life prediction module is used to predict the remaining service life of the KooN type safety control system through the three-parameter Weibull distribution model obeyed by the equipment.

[0183] In the above implementation process, the system obtains the life data of the equipment through life tests, and when the equipment reliability is high and the complete life data is difficult to obtain, the system can obtain the life data of multiple equipment including right-truncated data, thereby reducing the demand for data, thereby effectively avoiding the problem of low estimation accuracy due to the difficulty in obtaining equipment life data. And a three-parameter Weibull distribution model is constructed to describe the reliability of the equipment, thereby further reflecting the failure form of the actual equipment. Then, based on the life data of multiple equipment, a likelihood function is established to more effectively utilize the life data of the equipment. Then, the maximum likelihood method and optimization method are used to estimate the parameters of the three-parameter Weibull distribution model to obtain the three-parameter Weibull distribution model that the equipment obeys, that is, a reliability model is established, and then the remaining service life of the KooN type safety control system is calculated through the reliability model, and a relatively accurate life prediction result for the KooN type safety control system is obtained. Thereby, the purpose of obtaining an accurate prediction value of the remaining life of the KooN type safety control system is achieved when the right-truncated data is included.

[0184] Example 7

[0185] An embodiment of the present invention provides a machine-readable storage medium having instructions stored thereon. When the instructions are executed by a processor, the processor is configured to execute the above-mentioned remaining life prediction method for the KooN hot standby redundant safety control system.

[0186] Machine-readable storage media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0187] An embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned remaining life prediction method of the KooN hot standby redundant safety control system is implemented.

[0188] like Fig. 9 FIG. 1 is a schematic diagram of an electronic device provided by an embodiment of the present invention. Fig. 9 As shown, the electronic device 10 of this embodiment includes: a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the steps in the above method embodiment are implemented. Alternatively, when the processor 100 executes the computer program 102, the functions of each module / unit in the above device embodiment are implemented.

[0189] Exemplarily, the computer program 102 may be divided into one or more modules / units, one or more modules / units are stored in the memory 101, and are executed by the processor 100 to complete the present invention. One or more modules / units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 102 in the terminal device 10. For example, the computer program 102 may be divided into an equipment life data acquisition module, a distribution model construction module, a distribution model parameter estimation module, and a remaining service life prediction module.

[0190] The electronic device 10 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 10 may include, but is not limited to, a processor 100 and a memory 101. Those skilled in the art will appreciate that Fig. 9 It is only an example of the electronic device 10 and does not constitute a limitation of the electronic device 10. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0191] The processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0192] The memory 101 may be an internal storage unit of the electronic device 10, such as a hard disk or memory of the electronic device 10. The memory 101 may also be an external storage device of the electronic device 10, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 10. Further, the memory 101 may also include both an internal storage unit of the electronic device 10 and an external storage device. The memory 101 is used to store computer programs and other programs and data required by the electronic device 10. The memory 101 may also be used to temporarily store data that has been output or is to be output.

[0193] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0194] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0195] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0196] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0197] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0198] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0199] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for predicting the remaining life of a KooN hot standby redundant safety control system. It is characterized in that include: Obtain life data of multiple devices including right-censored data through life tests; Construct a three-parameter Weibull distribution model for a single device; According to the life data of the plurality of devices, the parameters of the three-parameter Weibull distribution model are estimated based on the maximum likelihood method and the optimization method to obtain the three-parameter Weibull distribution model obeyed by the devices; The remaining useful life of the KooN type safety control system is predicted by a three-parameter Weibull distribution model obeyed by the equipment.

2. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 1, It is characterized in that The estimating the parameters of the three-parameter Weibull distribution model based on the maximum likelihood method and the optimization method according to the life data of the plurality of devices to obtain the three-parameter Weibull distribution model obeyed by the devices includes: Based on the life data of the plurality of devices, a likelihood function of a three-parameter Weibull distribution model is obtained; Use optimization methods to solve the maximum value of the likelihood function and obtain the best estimate of the shape parameter Best estimate of the location parameter and the best estimate of the scale parameter 3. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 1, It is characterized in that The three-parameter Weibull distribution model of a single device is constructed, including: Construct a three-parameter Weibull distribution probability density function, cumulative distribution function and reliability function, wherein the three-parameter Weibull distribution probability density function is: The cumulative distribution function is The reliability function is t is the equipment operation time, γ ≥ 0 is the location parameter, β > 0 is the shape parameter, and η > 0 is the scale parameter.

4. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 2, It is characterized in that The method of obtaining a likelihood function of a three-parameter Weibull distribution model based on the life data of the plurality of devices includes: When there are n life span data, r complete life span data, and nr right-truncated data, r is recorded as t j , let nr be t k , then the likelihood function is expressed as t i Denoted as including t j and t k Life data, i = 1, 2, ..., n, to simplify the likelihood function, we get Taking the logarithm, we get the log-likelihood function:

5. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 4, It is characterized in that The optimization method is used to solve the maximum value of the likelihood function to obtain the best estimate of the shape parameter Best estimate of the location parameter and the best estimate of the scale parameter include: Set β∈(1,10] and use the optimization method to solve the optimization objective function get 6. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 5, It is characterized in that The optimization method includes an interior point method.

7. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 2, It is characterized in that The method of predicting the remaining useful life of the KooN type safety control system by using the three-parameter Weibull distribution model to which the device is subjected comprises: According to the working principle of the KooN type safety control system, it is determined that when at least N-K+1 devices in the KooN type safety control system fail, the KooN type safety control system is in a system failure state, and the distribution function of the system failure F s (t) is Among them, N represents that there are N identical devices running simultaneously in the KooN type safety control system, and K represents that there are K devices running normally among N identical devices; Distribution function F s (t) Take the derivative and get the probability density function f of system failure s (t) is Among them, F s (t) and f s (t) all satisfy The condition when When F s (t) and f s (t) are all 0; The running time of the current KooN type safety control system is recorded as δ≥0, and the probability density function f of the remaining service life T of the KooN type safety control system is obtained according to the conditional probability formula: T (T) is For the probability density function f T (T) Calculate the expectation and obtain the expected value of the remaining service life of the KooN type safety control system after being put into use for δ time: According to f s (t), when the running time of the KooN type safety control system When 1-F s (δ) = 1, Then the expected value of remaining life E(T) is expressed as 8. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 1, It is characterized in that The method of obtaining the life data of a plurality of devices including right-truncated data through life testing includes: The timed truncation test or the constant truncation test is used to obtain the life data of multiple devices including right-truncation data.

9. The remaining life prediction method of the KooN hot standby redundant safety control system according to claim 1, It is characterized in that The life of any equipment follows a three-parameter Weibull distribution, a two-parameter Weibull distribution or an exponential distribution.

10. A remaining life prediction system for a KooN hot standby redundant safety control system. It is characterized in that include: An equipment life data acquisition module, used for acquiring life data of multiple equipments including right-truncated data through life tests; Distribution model building module, used to build a three-parameter Weibull distribution model for a single device; A distribution model parameter estimation module, used to estimate the parameters of the three-parameter Weibull distribution model based on the maximum likelihood method and the optimization method according to the life data of the plurality of devices, so as to obtain the three-parameter Weibull distribution model obeyed by the device; The remaining service life prediction module is used to predict the remaining service life of the KooN type safety control system through a three-parameter Weibull distribution model obeyed by the device.

11. A machine-readable storage medium having instructions stored thereon, It is characterized in that When the instruction is executed by a processor, the processor is configured to execute the remaining life prediction method of the KooN hot standby redundant safety control system as claimed in any one of claims 1 to 9.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the remaining life prediction method of the KooN hot standby redundant safety control system according to any one of claims 1 to 9 is implemented.