A reliability prediction and maintenance strategy optimization method for train door systems

Abstract

The present invention provides a reliability prediction and maintenance strategy optimization method for a train door system. The method comprises: classifying and counting the actual fault data of the door system according to different components, and calculating the average fault rate of each component; constructing the initial state transfer matrix of each component according to the average fault rate of each component, and calculating the one-step state transfer vector of the faulty component of the door system; calculating the universal generation function of each subsystem and the door system according to the one-step state transfer vector of each faulty component of the door system; performing reliability fitting according to the universal generation function of the door system to obtain the reliability curve of the door system, inputting the reliability parameters of the door system into the maintenance cycle optimization model, and obtaining the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system. The method of the present invention fully considers the multi-state characteristics of the train door system, and provides a more accurate reliability analysis method for the train door system.

Description

Technical Field

[0001] The present invention relates to the technical field of reliability prediction of train door systems, and in particular to a reliability prediction and maintenance strategy optimization method for train door systems. Background Art

[0002] For existing reliability prediction methods, many of them determine the reliability distribution of the system by fitting the fault distribution according to the interval of fault occurrence time. This method not only requires a large amount of data for analysis, but also has a large deviation in the fitting results when there is less system fault data. In practice, the door system is a complex multi-state system. The various states of the system will shift due to faults and repairs. The reliability of the system can be easily calculated based on the transition between these states. Moreover, in the fault data, the average failure rate of each component of the system will not have a large deviation due to the amount of data. Therefore, the reliability analysis of the system has a high accuracy.

[0003] In actual operation, train operating units generally perform periodic maintenance according to the maintenance procedures specified by the manufacturer at the time of leaving the factory. For example, the door system is cleaned every 60 days and lubricated every 90 days. However, due to the different actual operating conditions of each train, under-repair or over-repair often occurs according to the original maintenance procedures. Therefore, it is a relatively new idea and method to establish a maintenance model based on system reliability to optimize the maintenance strategy, which is of great significance to reducing maintenance costs and reasonably setting maintenance procedures.

[0004] The door system is a relatively complex integrated system, and its functional block diagram is a series-parallel mixed structure. The series-parallel mixed system is generally divided into a single series or single parallel subsystem.

[0005] The shortcomings of a train door system reliability prediction and maintenance strategy optimization scheme in the prior art include: most of the methods use the fault time interval as a parameter to perform reliability fitting; only a small part of them conduct reliability analysis based on the logical relationship between the failure rate of the faulty component and the system structure. Reliability fitting only through the system time interval not only has high requirements on the number of fault data, but also has low accuracy in predicting the results when there are few fault data, and reliability analysis only through the interval of fault data also lacks certain rationality. Therefore, the practicality of the maintenance cycle obtained by optimizing the maintenance strategy through the reliability analysis results obtained by this method is relatively low, and it cannot provide practical maintenance suggestions to the maintenance department. Summary of the invention

[0006] The embodiment of the present invention provides a reliability prediction and maintenance strategy optimization method for a train door system, so as to provide an accurate reliability analysis method for the train door system.

[0007] In order to achieve the above object, the present invention adopts the following technical scheme.

[0008] A reliability prediction and maintenance strategy optimization method for a train door system, comprising:

[0009] Collect actual fault data of the door system of the train, classify and count the collected actual fault data according to different components, and calculate the average failure rate of each component;

[0010] Each component of the door system is regarded as a two-state component, and the initial state transfer matrix of each component is constructed according to the average failure rate of each component. The one-step state transfer vector of the faulty component of the door system is calculated according to the initial state transfer matrix of each component.

[0011] The door system is divided into multiple subsystems, and the universal generating function of each subsystem is calculated according to the one-step state transfer vector of the faulty component of each door system and the series-parallel relationship of the component, and then the universal generating function of the door system is calculated according to the universal generating function of each subsystem;

[0012] The reliability curve of the door system is obtained by reliability fitting based on the universal generating function of the door system, and the reliability parameters of the door system are obtained based on the reliability curve. The reliability, unreliability, reliability threshold, replacement cost and preventive repair cost parameters of the door system are input into the maintenance cycle optimization model to obtain the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system.

[0013] Preferably, the actual fault data of the door system of the train is collected, and the collected actual fault data is classified and counted according to different components to calculate the average failure rate of each component, including:

[0014] Collect the actual failure data of the door system of the train, classify and count these failure data according to different components, and calculate the average failure rate of different components according to the average failure rate formula shown in formula 18. The average failure rate represents the ratio of the total number of product failures to the total number of experiments within the statistical time;

[0015]

[0016] Where λ(t) is the average failure rate; n f (t) is the number of failed products; Δn f (t) is the number of failures within the time interval;

[0017] ns (t) is the number of products at time t.

[0018] Preferably, each component constituting the vehicle door system is regarded as a two-state component, an initial state transfer matrix of each component is constructed according to the average failure rate of each component, and a one-step state transfer vector of the faulty component of the vehicle door system is calculated according to the initial state transfer matrix of each component, including:

[0019] Each component of the door system is regarded as a two-state component. The initial state transfer matrix of each component is calculated using formula (1) according to the average failure rate of each component, and the one-step state transfer vector of the door system fault component is calculated using formula (2).

[0020] Assume that the state space of a system is X and the time space is T. The set expression of the two is X∈

[0021] {0,1,2,3,…}; T∈{0,1,2,3,…}, the discrete-time Markov chain {X(n)|T} should always satisfy equation (1);

[0022]

[0023] In the formula, when n is 0, it represents the initial time sequence origin of the discrete-time Markov chain, and x 0 is the initial state;

[0024] After n steps of transition, the Markov chain changes from state to state i Change to x j The probability of ij The probability of an n-step transition of (n) is mathematically described as:

[0025] p ij (n) = Pr {X (m + n) = x j |X(m)=x i},0≤m≤n (2)

[0026] Further explanation of discrete Markov chain, the transition probability p ij (m,n) can be uniquely determined by the difference between m and n.

[0027] Preferably, the initial probability transfer matrix table of the door system components is shown in Table 3:

[0028] Table 3

[0029]

[0030]

[0031] The one-step state transfer vector table of the door system fault component is shown in Table 4:

[0032] Table 4

[0033]

[0034] Preferably, the door system is divided into a plurality of subsystems, a universal generating function of each subsystem is calculated according to the one-step state transfer vector of each door system fault component and the series-parallel relationship of the components, and then a universal generating function of the door system is calculated according to the universal generating function of each subsystem, including:

[0035] According to the functional block diagram, the entire door system is divided into multiple subsystems connected in series or parallel. According to the initial state transfer matrix of each component, the one-step state transfer vector of the door system faulty component and the series-parallel relationship of the components, the universal generating function of each subsystem for one year is calculated according to formula (5). Because the universal generating function of the system for 10 years is to be calculated, multiple iterations are performed according to formulas (6), (7) and (8) to obtain the two-step, three-step... state transfer probability vectors of each door system faulty component), and then the universal generating function of each subsystem under different years is calculated;

[0036] The series-parallel relationship of each subsystem and the universal generating function of each subsystem are synthesized to obtain the universal generating function of the door system;

[0037] For all i,j∈X, 0≤p ij ≤1, and the sum of each row of P is 1. In this case, the matrix P is a random matrix. The random value X(0) represents the initial state of the Markov chain, and its probability distribution is called the initial probability vector. The mathematical expression of the probability vector is:

[0038] p(0)=[p 0 (0),p 1 (0),…,p m (0)] (5)

[0039] The Chapman-Kolmogorov equation is used to calculate the n-step transition probability matrix P(n), and its mathematical expression is:

[0040] P(n)=P·P(n-1)=P n (6)

[0041] Where P is the one-step probability transfer matrix of the Markov chain. The n-step probability transfer matrix is ​​the n-th power of the one-step probability transfer matrix;

[0042] State probability p j The value of (n) depends on the initial state probability when n=0 and the number of steps of subsequent state probability transfer, and its mathematical expression is:

[0043]

[0044] Transforming Equation 7 into a matrix form, the matrix expression of the n-step state probability vector is obtained as follows:

[0045] p(n)=p(0)·P n (8)

[0046] Where p(0) is the row vector of the initial state probability (when n=0), and p(n) is the n-step state probability vector obtained after n-step transfer.

[0047] Preferably, the door system is divided into a plurality of subsystems, a universal generating function of each subsystem is calculated according to the one-step state transfer vector of each door system fault component and the series-parallel relationship of the components, and then a universal generating function of the door system is calculated according to the universal generating function of each subsystem, including:

[0048] The door system is divided into subsystem 1, subsystem 2 and subsystem 3. Subsystem 2 and subsystem 3 are connected in parallel and then in series with subsystem 1. Each subsystem is formed by components in series. Subsystem 1 includes door control button / ATD and door controller. Subsystem 2 includes drive motor, screw rod, nut assembly, long and short guide column, door frame, guide rail / lower swing arm, slide, door leaf, 98 limit switch, main lock, auxiliary lock and door lock travel switch. Subsystem 3 includes door leaf, slide, guide rail / lower swing arm, door frame, long and short guide column, nut assembly, screw rod, drive motor, main lock and auxiliary lock.

[0049] The universal generating function of each subsystem is calculated based on the one-step state transfer vector of each door system fault component and the series-parallel relationship of the components. The universal generating function of subsystem 1 is: U 1 (1) = 0.0104z 0 +0.9896z 1 ;

[0050] The general generating function of subsystem 2 is:

[0051] U 2 =0.2067z 0 +0.7933z 1 ;

[0052] The general generating function of subsystem 3 is: 3 (1) = 0.1785z 0 +0.8215z 1 ;

[0053] According to the series-parallel relationship of the three subsystems that make up the door system, the general generating function of the door system is:

[0054] U 系 =0.0491z 0 +0.3098z 0.5 +0.6483z 1 .

[0055] Preferably, the reliability curve of the door system is obtained by performing reliability fitting according to the universal generating function of the door system, and the reliability, unreliability, reliability threshold, replacement cost and preventive repair cost parameters in the reliability curve of the door system are input into the maintenance cycle optimization model to obtain the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system, including:

[0056] The general generating function model of the door system is a z function model, in which the index of z represents the performance state of the system, and the coefficient of z represents the probability of different performance states. The door system has three states in each year. According to the general generating function model of the door system, the probabilities corresponding to the performance of different states of the door system and the change of the probability with the number of transfer steps are obtained. Then, the reliability curve of the door system is obtained through reliability fitting, and the reliability parameters of the door system are obtained according to the reliability curve.

[0057] Input the reliability, unreliability, reliability threshold, replacement cost and preventive repair cost parameters of the door system into the maintenance cycle optimization model shown in formula (3) to obtain the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system;

[0058]

[0059] Where C(T) represents the maintenance cost required for a single maintenance cycle of the system; c p The cost of a regular maintenance; f The cost required for a fault repair; R(t) is the reliability function of the system, T is the maintenance cycle, which is a variable; R(T) is the reliability of the system at time T; F(T) is the unreliability of the system at time T; The time during which the system operates reliably within a single maintenance cycle; R e is the reliability constraint value. The significance of reliability constraint is that the system needs to meet the minimum reliability during the entire maintenance cycle.

[0060] It can be seen from the technical solutions provided by the above-mentioned embodiments of the present invention that the method of the present invention fully considers the multi-state characteristics of the train door system, and also considers the impact of different faults on the system, thereby providing a more accurate reliability analysis method for the train door system.

[0061] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0063] Figure 1 A schematic flow chart of a method for predicting reliability and optimizing maintenance strategies for a vehicle door system provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of a functional block diagram of a vehicle door system provided by an embodiment of the present invention;

[0065] Figure 3 A structural diagram of a door subsystem provided by an embodiment of the present invention;

[0066] Figure 4 A reliability curve diagram of a door system provided by an embodiment of the present invention;

[0067] Figure 5 A diagram showing the relationship between the maintenance cycle and maintenance cost of a vehicle door system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0069] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0070] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0071] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0072] The embodiment of the present invention proposes a train door system reliability prediction and maintenance strategy optimization method based on the Markov process-general generating function method, so as to enable the door system to formulate a reasonable maintenance strategy according to the reliability degradation of the system and provide operation and maintenance support for the vehicle maintenance department. The processing flow of the method is as follows Figure 1 As shown, the processing steps include the following:

[0073] Step 1: Data processing and calculation of average failure rate.

[0074] The actual fault data of the door system of the train are collected, and these fault data are classified and counted according to different components. The results are shown in Table 1. The average failure rate of different components is calculated according to the average failure rate formula as shown in Table 2. The calculation formula of the average failure rate is shown in Formula 19

[0075]

[0076] Where λ(t) is the average failure rate; n f (t) is the number of failed products; Δn f (t) is the number of failures within the time interval;

[0077] n s (t) is the number of products at time t. The average failure rate represents the ratio of the total number of product failures to the total number of experiments within the statistical time.

[0078] Table 1: Door system fault data table

[0079]

[0080]

[0081] Table 2: Failure rate of main components of door system

[0082]

[0083] Step 2: Construct the one-step state transition matrix and state probability vector of the component

[0084] After calculating the average failure rate of each component, the one-step state transfer matrix of the system is constructed according to the average failure rate of each component. Each component of the door system is regarded as a two-state component. According to formula (1), the initial state transfer matrix of each component can be obtained, as shown in Table 3. After obtaining the initial state transfer matrix of the component, the one-step state transfer vector of the door system fault component can be calculated according to formula (2), as shown in Table 4.

[0085] The initial state transfer matrix of the component is the basis for the calculation of the universal generating function method. When calculating the universal generating function of the subsystem and the system in step 3, we must first calculate the one-step state transfer vector of the component based on the initial state transfer matrix and the initial state probability vector of the component, calculate the universal generating function of the subsystem based on the one-step state transfer vector of each component and the series-parallel relationship of the components, and then calculate the universal generating function of the system based on the universal generating function of the subsystem (according to formula 10-16).

[0086] Assume that the state space of a system is X and the time space is T. The set expression of the two is X∈

[0087] {0,1,2,3,…}; T∈{0,1,2,3,…}, the discrete-time Markov chain {X(n)|T} should always satisfy Equation 1.

[0088]

[0089] Where n is 0, which represents the initial time sequence origin of the discrete-time Markov chain; x 0 is the initial state. This formula means that the current state of the Markov chain is only related to the current state and has nothing to do with its past state. This effect is also called the no aftereffect of the Markov process.

[0090] The characterization of discrete-time Markov chains often requires consideration of the state transition probability p ij (n), that is, after the Markov chain has undergone n-step transitions, the state value x i Change to x j The probability of ij The probability of an n-step transition of (n) is mathematically described as:

[0091] p ij (n) = Pr{X(m+n) = x j |X(m)=x i},0≤m≤n (2)

[0092] Further explanation of discrete Markov chain, the transition probability p ij (m,n) can be uniquely determined by the difference between m and n. The mathematical description of the discrete-time Markov chain is:

[0093] Table 3: Initial probability transfer matrix of door system components

[0094]

[0095] Table 4: One-step state transition vector table of door system fault components

[0096]

[0097]

[0098] Step 3: Construct the system functional block diagram and system general generation function model.

[0099] After constructing the state transfer matrix and one-step state transfer vector of the components and constructing the functional block diagram of the system in step 3, the entire system can be divided into a single series or parallel subsystem according to the functional block diagram. At this time, it is necessary to calculate the universal generating function of the subsystem for one year based on the state transfer matrix and one-step state vector in step 2. Because the universal generating function of the system for 10 years needs to be calculated, it is necessary to perform multiple iterations according to formula 6 to obtain (the state transfer probability vector of two steps, three steps...), and then obtain the universal generating function of the system under different years according to the above calculation method.

[0100] Before modeling the system using the general generating function method, it is necessary to construct a functional block diagram of the system to clarify the series and parallel structure of the system. The functional block diagram for executing the door opening and closing function of the door system is as follows: Figure 2 As shown in the figure. According to the functional block diagram of the door system, the door system is a complex series-parallel hybrid system, and the entire system needs to be partially divided into a single series-parallel system, such as Figure 3As shown. The entire door system is formed by connecting subsystem 2 and subsystem 3 in parallel and then in series with subsystem 1, and each subsystem is formed by connecting components in series. Subsystem 1 includes door control button / ATO and door controller. Subsystem 2 includes drive motor, screw rod, nut assembly, long and short guide column, door carrier, guide rail / lower swing arm, slide, door leaf, 98 limit switch, main lock, auxiliary lock and door lock travel switch. Subsystem 3 includes door leaf, slide, guide rail / lower swing arm, door carrier, long and short guide column, nut assembly, screw rod, drive motor, main lock and auxiliary lock.

[0101] Before constructing the system universal function model, it is necessary to establish the universal generation function of the components, as shown in Table 5. The universal generation functions of the three subsystems are established based on the universal generation functions of the components. According to Formula 13, the universal generation function of subsystem 1 is: 1 (1) = 0.0104z 0 +0.9896z 1 ;

[0102] The general generating function of subsystem 2 is:

[0103] U 2 =0.2067z 0 +0.7933z 1 ;

[0104] The general generating function of subsystem 3 is: 3 (1) = 0.1785z 0 +0.8215z 1 .

[0105] According to the series-parallel relationship of the three subsystems that make up the door system, the general generating function of the door system is:

[0106] U 系 =0.0491z 0 +0.3098z 0.5 +0.6483z 1

[0107] Here, the universal generating function of the door system is obtained by taking one year as a node. After one transition, the universal generating function of the door system for the next year is obtained. According to this method, the universal generating function of the system in the next 10 years can be calculated, as shown in Table 6.

[0108] Table 5 z function table of faulty components of door system

[0109]

[0110] Table 6: Door system z function table

[0111]

[0112] Step 4: System reliability prediction and maintenance strategy optimization

[0113] After obtaining the general generating function model of the door system, the probabilities corresponding to the different state performances of the door system can be obtained, as well as the changes of these probabilities with the number of transfer steps.

[0114] The general generating function model of the door system is a z function model, where the index of z represents the performance state of the system, and the coefficient of z represents the probability of different performance states. Table 6 shows that this is the general generating function model of the door system for 10 years. In each year, the door system has three states. In the next step, we need to set the performance threshold and then fit the reliability curve.

[0115] When the system required performance w = 1, the system is reliable only when the performance state is ≥ 1. According to Table 6, the probability of the door system having a performance state ≥ 1 under different years can be obtained. The reliability curve of the system can be obtained by reliability fitting with these probabilities (reliability) as the ordinate and the corresponding year as the abscissa. Figure 4 shown.

[0116] After obtaining the reliability curve of the door system, the reliability R, unreliability F (F = 1-R), and reliability threshold R e , Replacement cost f , preventive repair costs p The maintenance cycle optimization model shown in formula (3) is input with the same parameters.

[0117] According to the reliability curve, the reliability R(T) and unreliability F(T) (F=1-R) of the system at different times can be obtained. The reliability threshold can be formulated according to the actual situation of the target by the operating company engineers. The replacement cost and preventive maintenance cost can be obtained based on the maintenance ledger.

[0118]

[0119] Where C(T) represents the maintenance cost required for a single maintenance cycle of the system; c p The cost of a regular maintenance; f The cost required for a fault repair; R(t) is the reliability function of the system, T is the maintenance cycle, which is a variable; R(T) is the reliability of the system at time T; F(T) is the unreliability of the system at time T; The time during which the system operates reliably within a single maintenance cycle; R e is the reliability constraint value. The significance of reliability constraint is that the system needs to meet the minimum reliability during the entire maintenance cycle.

[0120] The maintenance cycle optimization model output is as follows: Figure 5 The relationship between the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system shown. The maintenance cost data for the inspection and cleaning of the passenger compartment side door is: c p = 400 yuan per time, c f = 1500 yuan /

[0121] time. The maintenance cost data for the lubrication of the passenger compartment side door is: c p = 600 yuan per time, c f = 1500 yuan per time. Since the door system is one of the most frequently used systems in the train, its reliability should not be less than 0.9 when it reaches the inspection and cleaning cycle, and its reliability should not be less than 0.85 when it is lubricated.

[0122] According to Figure 5 it can be known that the inspection and cleaning work cycle of the optimized passenger compartment side door is 90 days, and the lubrication work cycle of the passenger compartment side door is 120 days. According to Table 7, the inspection and cleaning work cycle of the passenger compartment side door in the current situation of the door system is 60 days, and the current situation of the lubrication work cycle of the passenger compartment side door is 90 days. The preventive maintenance cost for one year using the current maintenance cycle is 4800 yuan, and the preventive maintenance cost for one year using the optimized maintenance cycle is 3400 yuan, saving 1400 yuan per year, and the maintenance cost is reduced by 29.17%.

[0123] Table 7 Current situation of the maintenance cycle of the door system

[0124]

[0125] In summary, the reliability analysis by the method described in the present invention is that the reliability accuracy is only related to the failure rate of components and the logical structure of the system, and is not affected by the sample size. When using this method for reliability analysis, the accuracy of reliability prediction is improved, so the practicability of the maintenance strategy optimization based on reliability is also improved. In addition, this method is simple to calculate, has a high degree of generality, is convenient for programming, and can provide theoretical support for the maintenance strategy optimization of the door system in actual engineering.

[0126] The present invention does not require fitting the failure data according to the failure time interval, and only needs the failure rate of each component and the functional block diagram of the system to predict the system reliability, and can also optimize the maintenance strategy according to the reliability distribution law, providing targeted maintenance and maintenance suggestions for the door system for the maintenance department, thereby reducing time and economic costs and improving the reliability of urban rail vehicles.

[0127] The method of the present invention is easy to program the algorithm of the method using relevant software, has a fast calculation speed, a high degree of programming, and is convenient and practical. Therefore, this method has certain economic and social benefits.

[0128] The present invention aims at the problem of optimizing maintenance strategies for the reliability of the current door system, and proposes a method for predicting the reliability and optimizing the maintenance strategies of the train door system based on the Markov process-universal generating function method. It considers how to accurately analyze the reliability of a complex multi-state system such as the door system, and optimizes the maintenance strategy according to the reliability analysis results. The method of the present invention only needs to calculate the failure rate of each component of the system and the logical relationship between the components of the system based on the historical fault data, and then the reliability analysis of the system can be performed. The analysis results will not be affected by the number of samples, and the accuracy of the reliability analysis has been greatly improved compared with the traditional method. The maintenance strategy of the system can also be optimized through the law of changes in the reliability of the system, and targeted door maintenance suggestions can be provided to the maintenance department, thereby reducing economic and time costs. The algorithm of the present invention is simple to calculate, has strong versatility, and is convenient and practical. Therefore, the invention has certain economic and social benefits.

[0129] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0130] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0131] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0132] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A reliability prediction and maintenance strategy optimization method for train door systems. It is characterized in that include: Collect actual fault data of the train door system, classify and count the collected actual fault data according to different components, and calculate the average failure rate of each component; Each component of the door system is regarded as a two-state component, and the initial state transfer matrix of each component is constructed according to the average failure rate of each component. The one-step state transfer vector of the faulty component of the door system is calculated according to the initial state transfer matrix of each component. The door system is divided into multiple subsystems, and the universal generating function of each subsystem is calculated according to the one-step state transfer vector of each faulty component of the door system and the series-parallel relationship of the components, and then the universal generating function of the door system is calculated according to the universal generating function of each subsystem and the series-parallel relationship between the subsystems; The reliability curve of the door system is obtained by performing reliability fitting according to the universal generating function of the door system, and the reliability parameters of the door system are obtained according to the reliability curve. The reliability, unreliability, reliability threshold, replacement cost and preventive repair cost parameters of the door system are input into the maintenance cycle optimization model to obtain the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system; The method of dividing the door system into a plurality of subsystems, calculating the universal generating function of each subsystem according to the one-step state transfer vector of each door system fault component and the series-parallel relationship of the components, and then calculating the universal generating function of the door system according to the universal generating function of each subsystem includes: According to the functional block diagram, the entire door system is divided into multiple subsystems connected in series or parallel. According to the initial state transfer matrix of each component, the one-step state transfer vector of the door system faulty component and the series-parallel relationship of the components, the universal generating function of each subsystem for one year is calculated according to formula (5). According to formulas (6), (7) and (8), multiple iterations are performed to obtain the two-step, three-step, etc. state transfer probability vectors of each door system faulty component, and then the universal generating function of each subsystem under different years is calculated. According to the series-parallel relationship of each subsystem, the universal generating function of each subsystem is synthesized to obtain the universal generating function of the door system; For all i,j∈X, X is the state space of the system, 0≤p ij ≤1, and the sum of each row of P is 1. In this case, the matrix P is a random matrix. The random value X(0) represents the initial state of the Markov chain. Its probability distribution is called the initial probability vector. The mathematical expression of the probability vector is: p(0)=[p 0 (0),p 1 (0),…,p m (0)] (5) The Chapman-Kolmogorov equation is used to calculate the n-step transition probability matrix P(n), and its mathematical expression is: P(n)=P·P(n-1)=P n (6) Where P is the one-step probability transfer matrix of the Markov chain, and the n-step probability transfer matrix is ​​the nth power of the one-step probability transfer matrix; State probability p j The value of (n) depends on the initial state probability when n=0 and the number of steps of subsequent state probability transfer, and its mathematical expression is: Transforming Equation 7 into a matrix form, the matrix expression of the n-step state probability vector is obtained as follows: p(n)=p(0)·P n (8) Where p(0) is the row vector of the initial state probability, and when n=0, p(n) is the n-step state probability vector obtained after n-step transfer.

2. The method according to claim 1, It is characterized in that The actual fault data of the door system of the train is collected, and the collected actual fault data is classified and counted according to different components, and the average failure rate of each component is calculated, including: The actual fault data of the door system of the train is collected, and these fault data are classified and counted according to different components. The average failure rate of different components is calculated according to the average failure rate formula shown in formula 19. The average failure rate represents the ratio of the total number of product failures to the total number of products within the statistical time period; In the formula, is the average failure rate; n f (t) is the number of failed products; Δn f (t) is the number of failure faults within the time interval; n s (t) is the number of products at time t.

3. The method according to claim 2, It is characterized in that The method of treating each component of the door system as a two-state component, constructing an initial state transfer matrix of each component according to the average failure rate of each component, and calculating a one-step state transfer vector of the faulty component of the door system according to the initial state transfer matrix of each component includes: Each component of the door system is regarded as a two-state component. The initial state transfer matrix of each component is calculated using formula (1) according to the average failure rate of each component, and the one-step state transfer vector of the door system fault component is calculated using formula (2). The time series space is T, and the set expression of the two is X∈{0,1,2,3,…}; T∈{0,1,2,3,…}, and the discrete-time Markov chain {X(n)|T} should always satisfy equation (1); In the formula, when n is 0, it represents the initial time sequence origin of the discrete-time Markov chain, and x 0 is the initial state; After n steps of transition, the Markov chain changes from state to state i Change to x j The probability of ij The probability of an n-step transition of (n) is mathematically described as: p ij (n)=Pr{X(m+n)=x j |X(m)=x i },0≤m≤n (2) Further explanation of discrete Markov chain, the transition probability p ij (m,n) can be uniquely determined by the difference between m and n.

4. The method according to claim 1, It is characterized in that The method of dividing the door system into a plurality of subsystems, calculating the universal generating function of each subsystem according to the one-step state transfer vector of each door system fault component and the series-parallel relationship of the components, and then calculating the universal generating function of the door system according to the universal generating function of each subsystem includes: The door system is divided into subsystem 1, subsystem 2 and subsystem 3. Subsystem 2 and subsystem 3 are connected in parallel and then in series with subsystem 1. Each subsystem is formed by components in series. Subsystem 1 includes door control button / ATO and door controller. Subsystem 2 includes drive motor, screw rod, nut assembly, long and short guide column, door frame, guide rail / lower swing arm, slide, door leaf, 98 limit switch, main lock, auxiliary lock and door lock travel switch. Subsystem 3 includes door leaf, slide, guide rail / lower swing arm, door frame, long and short guide column, nut assembly, screw rod, drive motor, main lock and auxiliary lock. The universal generating function of each subsystem is calculated based on the one-step state transfer vector of each door system fault component and the series-parallel relationship of the components. The universal generating function of subsystem 1 is: U 1 (1) = 0.0104z 0 +0.9896z 1 ; The general generating function of subsystem 2 is: 2 =0.2067z 0 +0.7933z 1 ; The general generating function of subsystem 3 is: 3 (1) = 0.1785z 0 +0.8215z 1 ; According to the series-parallel relationship of the three subsystems that make up the door system, the general generating function of the door system is: U 系 =0.0491z 0 +0.3098z 0.5 +0.6483z 1 The exponent of z represents the performance state of the system, and the coefficient of z represents the probability that different performance states may occur.

5. The method according to claim 4, It is characterized in that The reliability curve of the door system is obtained by performing reliability fitting according to the universal generating function of the door system, and the reliability parameters of the door system are obtained according to the reliability curve. The reliability, unreliability, reliability threshold, replacement cost and preventive repair cost parameters of the door system are input into the maintenance cycle optimization model to obtain the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system, including: The general generating function model of the door system is a z-function model. The door system has three states in each year. According to the general generating function model of the door system, the probabilities corresponding to the performance of the door system in different states and the change of the probability with the number of transfer steps are obtained. Then, the reliability curve of the door system is obtained through reliability fitting. Input the reliability, unreliability, reliability threshold, replacement cost and preventive repair cost parameters in the reliability curve of the door system into the maintenance cycle optimization model shown in formula (3) to obtain the optimal maintenance cycle and maintenance cost of different maintenance contents of the door system; Where C(M) represents the maintenance cost required for a single maintenance cycle of the system; p The cost of a regular maintenance; f The cost required for a fault repair; R(t) is the reliability function of the system, M is the maintenance cycle, which is a variable; R(M) is the reliability of the system at time T; F(M) is the unreliability of the system at time T; The time during which the system operates reliably within a single maintenance cycle; R e is the reliability constraint value. The significance of reliability constraint is the minimum reliability that the system needs to meet during the entire maintenance cycle.

Images