Health analysis method for complex system

Abstract

The invention provides a health analysis method for a complex system, relates to the technical field of system health analysis, and aims to introduce a prediction capability for system operation reliability in health analysis, and the method comprises the following steps: calculating the reliability of each device; calculating a health value of each device according to the reliability; the overall reliability of the system is comprehensively calculated based on the serial and parallel connection relation of equipment; a system health value is calculated. The method has the advantage of providing more effective and comprehensive health analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of system health analysis, and in particular to a health analysis method for a complex system. Background Art

[0002] With the rapid development of modern science and technology, especially information technology, the complexity and integration of a large number of complex systems are constantly increasing, and the cost of their maintenance and guarantee is becoming increasingly high.

[0003] As complex systems are put into use, the probability of failure and functional failure gradually increases due to the increase in components and influencing factors of complex systems. Considering the reliability, safety and economy of complex systems, how to effectively use various monitoring indicators and historical data of equipment to comprehensively analyze the health status of equipment, plan maintenance according to the health status to reduce maintenance and guarantee costs, and improve the safety, integrity and mission success of the system. Gradually, health analysis technology with prediction technology as the core has gained more and more attention and application. However, how to conduct an analysis with all-round considerations and how to save computing resources as much as possible are difficult to balance. Existing technologies usually lack the reliability of the system by establishing a model for the life characteristics of equipment in complex systems and the comprehensive dynamic evaluation of equipment correlation relationships, which greatly reduces the accuracy of health analysis. In addition, this single analysis method lacks robustness.

[0004] Therefore, it is necessary to introduce an accurate and convenient calculation method for system operation reliability in health analysis, realize the transformation from condition monitoring to health analysis, identify and manage the occurrence of system failures with the help of health analysis results and plan maintenance, so as to improve the working health level of complex systems and maximize the value of system use with less maintenance investment. Summary of the invention

[0005] The purpose of the present invention is to provide a health analysis method for a complex system, which can introduce the prediction capability of system operation reliability into the health analysis.

[0006] The present invention is achieved through the following technical solutions:

[0007] A health analysis method for a complex system includes the following steps:

[0008] Calculate the reliability R of each device i (t i ), R i (t i ) indicates that the i-th device has been working for a period of time t i reliability;

[0009] Calculate the health value H of each device according to the reliability i , H iIndicates the current health value of the i-th device;

[0010] Comprehensively calculate the overall reliability R of the system based on the series and parallel relationship of the equipment;

[0011] Calculate the system health value H.

[0012] Preferably, the method for calculating the reliability of each device is:

[0013] Calculate the lifetime factor α of the device i (t i ), α i (t i ) is the life factor of the i-th device;

[0014] Calculate the fault factor β of the equipment according to the fault alarm i , β i is the failure factor of the ith device;

[0015] Calculate the maintenance factor γ of the equipment i , γ i is the maintenance factor of the i-th equipment;

[0016] The reliability of the equipment is calculated based on the life factor, the failure factor and the maintenance factor.

[0017] Preferably, the method for calculating the life factor of the device is:

[0018]

[0019] Among them, e is a natural constant, t i is the working life of the i-th equipment, θ i is the maximum service life of the ith device;

[0020] Calculate the fault factor β of the equipment according to the fault alarm i The method is:

[0021]

[0022] Fr i =∑ n w in ×Fa in ;

[0023]

[0024] Among them, Fr i is the overall failure rate of the ith device, w in is the severity weight of the fault of the ith device at the nth level, Fa inis the failure rate of the ith device based on the nth level fault, t ip is the duration of the pth fault of the ith device in the nth level fault, t total,i is the total running time of the i-th device;

[0025] The method for calculating the maintenance factor of the equipment is:

[0026]

[0027] Among them, w im is the event severity coefficient corresponding to the mth maintenance of the i-th device, S im w im is the single maintenance factor of the mth maintenance of the ith equipment, Mit and Pit represent the actual duration and planned duration of the mth maintenance of the ith equipment, respectively.

[0028] Preferably, the method for calculating the reliability of the device is:

[0029] R i (t i )=γ i β i α i (t i )

[0030] Preferably, the method for calculating the health value of each device is:

[0031] calculating a health baseline for the device;

[0032] A current health value of the device is calculated based on the health baseline.

[0033] Preferably, the method for calculating the health baseline for the device is:

[0034]

[0035] Among them, Y i is the health baseline of the ith device, x ij is the value of the jth indicator detection item of the i-th device, a ij is the weight of the j-th indicator detection item of the i-th device;

[0036] The method for calculating the current health value of the device based on the health baseline is:

[0037] H i =R i (t i )Y i ;

[0038] Among them, H iis the current health value of the i-th device.

[0039] Preferably, the method for comprehensively calculating the overall reliability of the system based on the series and parallel relationship of the devices is:

[0040] If it is a series system, the reliability R of the system is:

[0041]

[0042] If it is a parallel system, the reliability R of the system is:

[0043]

[0044] Wherein, p is the total number of the devices.

[0045] Preferably, the method for calculating the system health value H is:

[0046]

[0047] Among them, π i is the weight of the ith device.

[0048] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0049] The present invention can accurately reflect the health status of each device, helping operation and maintenance personnel to promptly identify devices that may have problems and perform preventive maintenance or adjustments;

[0050] The present invention performs reliability analysis while performing analysis, making fuller and more comprehensive use of the collected and calculated data, and improving the comprehensiveness, accuracy and effectiveness of health analysis;

[0051] When calculating the reliability of each device, the present invention combines the three dimensions of life factor, failure factor and maintenance factor to more comprehensively and accurately reflect the actual operating status of the device and improve the accuracy of the evaluation.

[0052] The present invention specifically calculates the overall reliability of the system based on the series and parallel relationship of the devices in the system. Since the connection relationship of the system is taken into consideration, a more accurate system-level reliability evaluation can be provided.

[0053] The system health value calculated by the present invention provides a global and quantifiable indicator, which helps to intuitively understand the overall health status of the entire system, has important reference significance for the long-term stability and reliability of the system, and facilitates global optimization management and maintenance;

[0054] The invention has reasonable design, simple calculation and high cost performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of a flow chart of a health analysis method for a complex system provided in Example 1 of the present invention;

[0056] Figure 2 A schematic diagram of a failure distribution model of a device during its service life provided in Example 1 of the present invention;

[0057] Figure 3 This is a system structure diagram of the case provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0059] Example 1

[0060] This embodiment provides a health analysis method for a complex system. Figure 1 , including the following steps:

[0061] Calculate the reliability R of each device i (t i ), R i (t i ) represents the life span of the i-th device at its lifetime t i reliability;

[0062] Calculate the health value H of each device according to the reliability i , H i Indicates the current health value of the i-th device;

[0063] Comprehensively calculate the overall reliability R of the system based on the series and parallel relationship of the equipment;

[0064] Calculate the system health value H.

[0065] Since a complex system is composed of multiple devices, the design content of this embodiment 1 mainly consists of four parts: first, the reliability of the equipment is calculated based on the equipment's life, maintenance, alarm, etc.; second, a health baseline for the equipment is established, and the health value of the equipment is calculated based on the reliability of the equipment; third, the overall reliability of the complex system is comprehensively calculated based on the series and parallel relationships between the devices; fourth, the overall health value of the system is comprehensively calculated based on the overall reliability of the system and the health value of each device.

[0066] In this embodiment, the method for calculating the reliability of each device is:

[0067] Calculate the lifetime factor α of the device i (t i ), α i (t i ) is the life factor of the i-th device;

[0068] Calculate the fault factor β of the equipment according to the fault alarm i , β i is the failure factor of the ith device;

[0069] Calculate the maintenance factor γ of the equipment i , γ i is the maintenance factor of the i-th equipment;

[0070] The reliability of the equipment is calculated based on the life factor, the failure factor and the maintenance factor.

[0071] Furthermore, the method for calculating the life factor of the device is:

[0072]

[0073] Among them, e is a natural constant, t i is the working life of the i-th equipment, θ i is the maximum service life of the ith device;

[0074] The setting principle of the above life factor is as follows:

[0075] The failure distribution model of equipment during its service life can be found in Figure 2 , the probability that the equipment works normally under the specified conditions is the reliability of the equipment, which is represented by R(t). T represents the random variable of product failure time, t represents the specified equipment working time, and the reliability of the equipment at the specified time t is expressed as follows:

[0076] R(t)=P(T>t);

[0077] The cumulative failure distribution function of the equipment at a specified time t is represented by F(t), which is complementary to R(t), namely:

[0078] F(t)=P(T <t);

[0079] R(t)+F(t)=1;

[0080] Based on the classical reliability theory, the life characteristics of the equipment follow the Weibull distribution, and its Weibull distribution probability density function is expressed as follows:

[0081]

[0082] Where: θ and γ are characteristic life and shape parameters respectively. When γ = 1, the Weibull distribution becomes an exponential distribution; when γ = 2, the Weibull distribution becomes a Rayleigh distribution;

[0083] The Weibull distribution function F(t) is expressed as follows:

[0084]

[0085] Combining the above, we can get the reliability function α i (t):

[0086]

[0087] Since the device failure characteristics follow Figure 2 The bathtub failure model is used, so γ = 2 is used, and Rayleigh distribution is used. The life factor function according to the service life change is:

[0088]

[0089] Next, the fault factor β of the device is calculated based on the fault alarm. i The method is:

[0090]

[0091] Fr i =∑ n w in ×Fa in ;

[0092]

[0093] Among them, Fr i is the overall failure rate of the ith device, w in is the severity weight of the fault of the ith device at the nth level, Fa in is the failure rate of the ith device based on the nth level fault, t ip is the duration of the pth fault of the ith device in the nth level fault, t total,i is the total running time of the i-th device;

[0094] Furthermore, the method for calculating the maintenance factor of the equipment is:

[0095]

[0096] Among them, w im is the event severity coefficient corresponding to the mth maintenance of the i-th device, S im w imis the single maintenance factor of the mth maintenance of the ith equipment, Mit and Pit represent the actual duration and planned duration of the mth maintenance of the ith equipment, respectively.

[0097] Based on this, the current reliability of the equipment can be calculated by combining the maintenance factor, failure factor, and life factor of the equipment. The method for calculating the reliability of the equipment is:

[0098] R i (t i )=γ i β i α i (t i )

[0099] On the other hand, the method for calculating the health value of each device is:

[0100] calculating a health baseline for the device;

[0101] A current health value of the device is calculated based on the health baseline.

[0102] Furthermore, the method for calculating the health baseline for the device is:

[0103]

[0104] Among them, Y i is the health baseline of the ith device, x ij is the value of the jth indicator detection item of the i-th device, a ij is the weight of the j-th indicator detection item of the i-th device;

[0105] When establishing a health baseline, you can define the health index Y of the equipment under m different working conditions, such as: (y1, y2, ..., y m ) T ,y1,y2,…,y m ∈R 1 , each health index is calculated by n important indicators of the device;

[0106] Suppose that the important indicator detection item X under m working conditions is in the form of: (x1, x2,…, x m ) T ,x1,x2,…,x m ∈R n ;

[0107] Assume the weight value of the indicator is A, such as: (a1, a2, ..., a m ) T ,a1,a2,…,a m ∈R 1 ;

[0108] The following formula can be established:

[0109] y1=x 11 a1+x 12 a2+…+x 1n a n

[0110] y2=x 21 a1+x 22 a2+…+x 2n a n

[0111]

[0112] y m =x m1 a1+x m2 a2+…+x mn a b

[0113] It can be written as the following linear transformation:

[0114]

[0115] It can be further expressed as X m×b A b×1 =Y m×1 , since m≠n, linear regression is used to solve the equation, let J=min||XA-Y|| 2 :

[0116] Take the derivative to find the extreme value:

[0117] Then we can conclude that: X T XA=X T Y;

[0118] The equipment health baseline weight matrix A is derived as: A = (X T X) -1 X T Y;

[0119] Based on the health baseline weight matrix, the current theoretical health value of the device can be calculated through the current relevant monitoring indicators, that is, the health baseline calculated for the device in this embodiment.

[0120] On this basis, the method for calculating the current health value of the device based on the health baseline is:

[0121] H i =R i (t i )Y i ;

[0122] Among them, H i is the current health value of the i-th device.

[0123] As a preferred solution of this embodiment, the method for comprehensively calculating the overall reliability of the system based on the series and parallel relationships of the devices is:

[0124] If it is a series system, the reliability R of the system is:

[0125]

[0126] If it is a parallel system, the reliability R of the system is:

[0127]

[0128] Wherein, p is the total number of the devices.

[0129] It should be noted that if it is a more complex parallel-series combination system, the overall system can be divided into multiple subsystems. Each subsystem is a pure series or pure parallel subsystem. The reliability of each subsystem is calculated separately, and then the overall reliability is calculated according to the connection relationship between the subsystems in the same way.

[0130] In addition, the method for calculating the system health value H is:

[0131]

[0132] Among them, π i is the weight of the ith device.

[0133] The following is a specific calculation example provided by this embodiment. For the system structure diagram of this embodiment, please refer to Figure 3 .

[0134] First, calculate the reliability of the equipment. Refer to Table 1 for equipment life data.

[0135] Table 1 Year data

[0136] equipment Maximum working years Years of working experience Allocator 10 3 Transmitter (main) 15 4 Transmitter (equipment) 15 3 Tower 15 5

[0137] According to the calculation method of life factor Calculate the life factors of the distributor, main transmitter, backup transmitter and tower respectively, and the corresponding θ i and t i Substitute the maximum working life and the working life respectively, for example: the life reliability of the distributor is Similarly, the life reliability of the transmitter (main), transmitter (backup) and tower are 0.9316, 0.9608 and 0.8948 respectively.

[0138] Then the fault factors within the last 24 hours are calculated. Table 2 shows the statistics of fault warning duration.

[0139] Table 2 Statistics of fault warning duration

[0140]

[0141] The fault level weight distribution is: severe fault is 5, major fault is 3, and general fault is 1.

[0142] For example, the failure rate of the distributor is The failure factor within 24 hours is e -0.00119863×24 =0.9716. Similarly, the failure factors of the transmitter (main), transmitter (backup) and tower within 24 hours are 0.9877, 0.9877 and 1 respectively.

[0143] Then the maintenance factor of the equipment is calculated. Table 3 shows the maintenance event duration statistics.

[0144] Table 3 Maintenance event duration statistics

[0145]

[0146] The weight distribution of event severity levels is as follows: severe events are 0.6, major events are 0.25, and general events are 0.15, so that the sum of the weights is 1.

[0147] Taking the distributor as an example, assuming that the distributor number is 1, according to the maintenance time and type in the table, the single maintenance factor is calculated according to the method of this embodiment:

[0148] S 13 =1, S 11 =1;

[0149] Then calculate γ1:

[0150]

[0151] Therefore, the maintenance factors of the transmitter (main), transmitter (backup) and tower are 1, 1, and 1 respectively.

[0152] Finally, the reliability of the equipment can be calculated. For example, the reliability of the distributor is 0.9167×0.9716×0.9138=0.8139. The reliability of the transmitter (main), transmitter (backup) and tower are 0.9201, 0.9490 and 0.8948 respectively.

[0153] Assuming that the equipment passes through the health baseline matrix measured under various working conditions before leaving the factory, the theoretical health value, i.e. the health baseline, is calculated in combination with the current relevant indicators of the equipment. Refer to Table 4:

[0154] Table 4 Theoretical health values

[0155] equipment Current health theoretical value Allocator 0.999 Transmitter (main) 0.999 Transmitter (equipment) 0.999 Tower 0.999

[0156] The current health value of the distributor is the product of the reliability and the theoretical health value, that is, 0.8130×0.999=0.812187. Similarly, the current health values ​​of the transmitter (main), transmitter (backup) and tower are 0.9191799, 0.948051 and 0.8939052 respectively.

[0157] According to the connection relationship of the system, we can know that the overall reliability of the system is:

[0158] R=R 分配器 ×[1-(1-R 发射机主 )×(1-R 发射机备 )]×R 铁塔

[0159] =0.8139×[1-(1-0.9201)×(1-0.9490)]×0.8948=0.7253;

[0160] The health value of the system is based on weight, and the weight distribution of devices is shown in Table 5.

[0161] Table 5 Weight distribution of devices

[0162] equipment Weight Allocator 0.8 Transmitter (main) 0.9 Transmitter (equipment) 0.9 Tower 0.95

[0163] Therefore, the health value of the system is:

[0164]

[0165] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A health analysis method for a complex system, characterized in that: The following steps are involved: Calculate the reliability R of each device i (t i ), R i (t i ) indicates that the i-th device has been working for a period of time t i reliability; Calculate the health value H of each device according to the reliability i , H i Indicates the current health value of the i-th device; Comprehensively calculate the overall reliability R of the system based on the series and parallel relationship of the equipment; Calculate the system health value H.

2. A complex system health analysis method according to claim 1, characterized in that: The method for calculating the reliability of each device is: Calculate the lifetime factor α of the device i (t i ), α i (t i ) is the life factor of the i-th device; Calculate the fault factor β of the equipment according to the fault alarm i , β i is the failure factor of the ith device; Calculate the maintenance factor γ of the equipment i , γ i is the maintenance factor of the i-th equipment; The reliability of the equipment is calculated based on the life factor, the failure factor and the maintenance factor.

3. A complex system health analysis method according to claim 2, characterized in that: The method for calculating the life factor of the equipment is: Among them, e is a natural constant, t i is the working life of the i-th equipment, θ i is the maximum service life of the ith device; Calculate the fault factor β of the equipment according to the fault alarm i The method is: Fri i =∑ n w in ×Fa in ; Among them, Fr i is the overall failure rate of the ith device, w in is the severity weight of the fault of the ith device at the nth level, Fa in is the failure rate of the ith device based on the nth level fault, t ip is the duration of the pth fault of the ith device in the nth level fault, t total,i is the total running time of the i-th device; The method for calculating the maintenance factor of the equipment is: Among them, w im is the event severity coefficient corresponding to the mth maintenance of the i-th device, S im w im is the single maintenance factor of the mth maintenance of the ith equipment, Mit and Pit represent the actual duration and planned duration of the mth maintenance of the ith equipment, respectively.

4. A complex system health analysis method according to claim 3, characterized in that: The method for calculating the reliability of the equipment is: R i (t i )=γ i b i a i (t i ) 5. A complex system health analysis method according to claim 1, characterized in that: The method for calculating the health value of each device is: calculating a health baseline for the device; A current health value of the device is calculated based on the health baseline.

6. A complex system health analysis method according to claim 5, characterized in that: The method for calculating the health baseline for the device is: Among them, Y i is the health baseline of the ith device, x ij is the value of the jth indicator detection item of the i-th device, a ij is the weight of the j-th indicator detection item of the i-th device; The method for calculating the current health value of the device based on the health baseline is: H i =R i (t i )Y i ; Among them, H i is the current health value of the i-th device.

7. A complex system health analysis method according to claim 1, characterized in that: The method for comprehensively calculating the overall reliability of the system based on the series and parallel relationship of the equipment is: If it is a series system, the reliability R of the system is: If it is a parallel system, the reliability R of the system is: Wherein, p is the total number of the devices.

8. The health analysis method for a complex system according to claim 1, characterized in that: The method for calculating the system health value H is: Among them, π i is the weight of the ith device.

Images