CN113627651A - Method for predicting average service life of logistics storage equipment of tobacco enterprise - Google Patents

Method for predicting average service life of logistics storage equipment of tobacco enterprise Download PDF

Info

Publication number
CN113627651A
CN113627651A CN202110781009.0A CN202110781009A CN113627651A CN 113627651 A CN113627651 A CN 113627651A CN 202110781009 A CN202110781009 A CN 202110781009A CN 113627651 A CN113627651 A CN 113627651A
Authority
CN
China
Prior art keywords
storage equipment
logistics storage
function
data
failure
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110781009.0A
Other languages
Chinese (zh)
Inventor
孟祝
何杰明
何振峰
贾宁
刘晓迪
朱宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Tobacco Huizhou Co ltd
Original Assignee
Guangdong Tobacco Huizhou Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong Tobacco Huizhou Co ltd filed Critical Guangdong Tobacco Huizhou Co ltd
Priority to CN202110781009.0A priority Critical patent/CN113627651A/en
Publication of CN113627651A publication Critical patent/CN113627651A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/15—Correlation function computation including computation of convolution operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/04—Ageing analysis or optimisation against ageing

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • General Engineering & Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Mathematical Physics (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Mathematical Analysis (AREA)
  • Data Mining & Analysis (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Algebra (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a method for predicting the average service life of logistics storage equipment of a tobacco enterprise, which solves the problem of efficiently and accurately predicting the average service life of the logistics storage equipment. In addition, the unknown parameter estimation of the mixed Weibull distribution is carried out by using the nonlinear least square method based on the L-M algorithm, the calculation is faster and more efficient compared with the common maximum likelihood estimation method in the parameter estimation, the high-quality estimation can be carried out on the parameters of the mixed Weibull distribution, the principle is simple, the calculation is simple, and the popularization is easy.

Description

Method for predicting average service life of logistics storage equipment of tobacco enterprise
Technical Field
The invention relates to the technical field of equipment average life prediction, in particular to a method for predicting the average life of logistics storage equipment of a tobacco enterprise.
Background
In the recent development process of the tobacco industry in China, in order to improve the tobacco sale and transportation efficiency and reduce the logistics cost of tobacco enterprises, the logistics distribution centers are established in a dispute, and the establishment of the logistics centers can play a positive role in improving the levels of raw material transportation, cigarette storage, sale distribution and the like of the tobacco industry.
The reliability of the logistics storage equipment greatly influences the safe operation of the logistics distribution center, the service life data and maintenance data related to the equipment are important components for an equipment engineer to master the safety condition of the device and reasonably plan a maintenance plan, and for the equipment which works for a period of time, the service life of the equipment can be evaluated according to the working information and the working state of the equipment obtained when the equipment works; for a device in a working state, two types of data can be acquired: the failure data and the performance degradation data are based on the failure data, and the probability distribution of the failure rate and the reliability of the equipment can be obtained. The general reliability life models are exponential distribution, normal distribution, Weibull distribution and the like, wherein the Weibull distribution is most widely applied. The failure data of the logistics storage equipment of the tobacco enterprise is complex, on one hand, the service life distribution of different equipment is different due to the difference of the process, the raw materials and the like; in addition, the service life distribution of each device is different due to different conditions of the devices; in addition, the logistics storage equipment is provided with equipment such as a vehicle penetrating device and a tray transfer vehicle, the mechanical structure is complex, and the failure reason can be caused under the combined action of multiple mechanisms. Therefore, the problem of low prediction accuracy occurs when the standard Weibull distribution is used for predicting the service life of the logistics storage equipment of the tobacco enterprise. In addition, other methods, such as a prediction method based on machine learning, generally require more types of data to achieve prediction accuracy, require larger data amount, and are complex to calculate; the mixed Weibull distribution contains more parameters, can fit more complex data and can fit various distributions of the service life of equipment, but the mixed Weibull distribution has more parameters, and the solving complexity of the parameters is relatively increased; at present, a graph estimation method in common parameter estimation methods is simple and convenient, but has large human influence factors; the nonlinear least square method has high accuracy, but is generally difficult to directly calculate an accurate solution, and needs to be solved by a nonlinear algorithm.
In 11/27/2020, Chinese invention patent (publication No. CN112001077A) discloses a method for evaluating the life of petrochemical safety key equipment based on incomplete data, aiming at the conditions of incomplete equipment data record, various deletion data inclusion and small subsample data with little field test data, fusing the existing failure condition data of the equipment, estimating the initial value of the failure rate of the given equipment through maximum likelihood, iterating the condition maximum likelihood estimation value of the failure rate of the equipment according to the condition EM algorithm, randomly filling a virtual complete sample with the residual life distribution of the equipment at the ending time, avoiding the problem of low prediction accuracy when the standard Weibull distribution is directly used for predicting the life of the logistics storage equipment of tobacco enterprises, ensuring that the life evaluation result is more accurate, but establishing the accuracy of the maximum likelihood estimation method under the premise of correct parameter distribution prediction, and is computationally inefficient.
Disclosure of Invention
In order to solve the problem of how to efficiently and accurately predict the average service life of the logistics storage equipment, the invention provides a method for predicting the average service life of the logistics storage equipment of a tobacco enterprise, which has the advantages of less required data types, high efficiency and quickness in calculation and easiness in popularization.
In order to achieve the technical effects, the technical scheme of the invention is as follows:
a method for predicting average life of logistics storage equipment of tobacco enterprises, the method at least comprises:
s1, acquiring failure time sample data of logistics storage equipment,
s2, preliminarily determining a reliability function curve of the logistics storage equipment according to the failure time sample data of the logistics storage equipment;
s3, acquiring fault cause data of the logistics storage equipment, and determining the number of times n of the mixed Weibull distribution function and a probability density function corresponding to the n-times mixed Weibull distribution function according to the fault cause data;
s4, defining unknown parameters to be estimated in the probability density function, and estimating the unknown parameters based on a nonlinear least square method of an L-M algorithm by combining the reliability values on the reliability function curve of the logistics storage equipment preliminarily determined in the step S2 to obtain parameter estimation values;
and S5, introducing a mean life calculation formula, and predicting the mean life of the logistics storage equipment by combining the parameter estimation value.
Preferably, the failure time sample data of the lifetime of the logistics storage equipment in step S1 includes the data of the number of the same type of failed equipment and the lifetime time data of the logistics storage equipment.
Preferably, the process of preliminarily determining the reliability function curve of the logistics storage equipment in step S2 is as follows:
acquiring failure time sample data of m logistics storage equipment in total, and arranging the failure time sample data of the m logistics storage equipment from small to large according to failure time, namely t1≤t2≤…≤tm；
Calculating an observed value of the cumulative failure probability, wherein the expression is as follows:
Figure BDA0003156974340000031
wherein, i is 1, 2, …, m, which represents the order of data in m data;
Figure BDA0003156974340000032
an observed value representing a cumulative probability of failure for the ith data;
and calculating the reliability value, wherein the expression is as follows:
Figure BDA0003156974340000033
wherein the content of the first and second substances,
Figure BDA0003156974340000034
representing the reliability value of the ith data to obtain an observation sample data set:
Figure BDA0003156974340000035
Figure BDA0003156974340000036
with tiIs shown as the abscissa of the graph,
Figure BDA0003156974340000037
and drawing a reliability function curve by using the data in the observation sample data set as a vertical coordinate.
Preferably, in step S3, when determining the number of repetitions n of the mixed weibull distribution function according to the failure cause data, first determining the failure cause of the logistics storage equipment with an occurrence probability greater than E from the failure cause data of the logistics storage equipment, where E is an occurrence probability threshold, and taking the failure cause of the logistics storage equipment with an occurrence probability greater than E as the main failure cause of the logistics storage equipment, and then determining the number n of main failure causes, and taking the number n of main failure causes as the number of repetitions of the mixed weibull distribution function.
Preferably, the expression of the probability density function corresponding to the mixed weibull distribution function is:
Figure BDA0003156974340000038
wherein f is(t) represents a probability density function; n represents the multiplicity of the mixed Weibull distribution function; w is aiAre weighted, satisfy
Figure BDA0003156974340000039
wi、βi、ηiAll are parameters of n-fold mixed Weibull distribution functions and are also unknown parameters to be estimated in the probability density function.
Preferably, after determining the probability density function corresponding to the n-fold mixed weibull distribution function, the reliability function corresponding to the n-fold mixed weibull distribution function is obtained according to the probability density function corresponding to the n-fold mixed weibull distribution function, and the expression is:
Figure BDA00031569743400000310
preferably, in step S4, the process of estimating the unknown parameters by using the L-M algorithm is as follows:
the unknown parameter [ wi，βi，ηi]The set of (a) is denoted as vector θ';
the unknown parameters were transformed as follows:
wi＝(sinai)2，ηi＝exp(bi)，βi＝exp(ci) At this time, the unknown parameters to be estimated are converted into vectors: theta ═ ai，ci，bi]；
According to the observation sample data set
Figure BDA0003156974340000041
And the principle of a nonlinear least square method, and setting an optimization model objective function of parameter estimation as follows:
Figure BDA0003156974340000042
wherein the content of the first and second substances,
Figure BDA0003156974340000043
j is the jth reliability value in the reliability function corresponding to the probability density function, wherein j is 1, 2, …, m; executeThe method comprises the following steps:
s41: introducing a parameter l, making l equal to 0, and giving initial values theta of all unknown parameters to be estimatedlSetting error precision epsilon > 0, setting initial step length lambdalWhen the value is 0.01, an initial objective function value f (θ) is calculatedl)；
S42: calculate the residual vector for the l-th iteration: r (theta)l)＝(r1(θl)，r2(θl)，…，rm(θl))；
Note the book
Figure BDA0003156974340000044
Is R at the first iterationjThe value of (1), then in the residual vector
Figure BDA0003156974340000045
Calculating the objective function at thetalThe Jacobian matrix of (a), denoted as J (θ)l)；
S43: according to J (theta)l) And r (theta)l) Calculating a matrix A, wherein the calculation expression is as follows: a ═ J (θ)l)]TJ(θl) And calculating a matrix B, wherein the calculation expression is as follows: b ═ J (θ)l)]Tr(θl)；
S44: calculating an iteration increment Δl＝-[A+λlI]-1B, calculating the next iteration value according to the iteration increment, wherein the iteration formula is thetal+1＝θl+Δl；
S45: and determining an estimation value of the unknown parameter according to the iteration condition.
Preferably, the specific process of step S45 is:
s451. if | | | DeltalStopping iteration and obtaining the optimal solution if | | < epsilon
Figure BDA0003156974340000046
Outputting an estimation value result of the unknown parameters; otherwise, go to step S452;
s452, calculating f (theta)l+1) And f (theta)l) If f (θ)l+1)-f(θl) If < 0, let λl+1＝0.1×λlAnd l +1, return to performing step S42; otherwise, let λl＝10×λlThe flow returns to step S44.
Preferably, after step S4 and before step S5, the method further comprises: according to the parameter estimation value, a curve of the reliability function R (t) corresponding to the n-fold mixed Weibull distribution function is determined, and the curve is compared with the reliability function curve of the logistics storage equipment preliminarily determined in the step S2, so that the effectiveness of the reliability function fitting by adopting the n-fold mixed Weibull distribution function provided by the technical scheme can be verified.
Preferably, the average life calculation formula is:
Figure BDA0003156974340000047
wherein, wig、ηig、βigRespectively the original unknown parameters wi、βi、ηiIs the gamma function, and L is the average life of the logistics storage equipment.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention provides a method for predicting the average service life of logistics storage equipment of a tobacco enterprise, which comprises the steps of firstly obtaining failure time sample data of the logistics storage equipment, and only under the condition that historical failure data of the logistics storage equipment is known, utilizing mixed Weibull distribution function fitting, wherein the required data types are few, and the reliability modeling of the logistics storage equipment of the tobacco enterprise by using mixed Weibull distribution is more practical and accurate compared with a standard Weibull distribution function. In addition, the unknown parameter estimation of the mixed Weibull distribution is carried out by using the nonlinear least square method based on the L-M algorithm, the calculation is faster and more efficient compared with the common maximum likelihood estimation method in the parameter estimation, the high-quality estimation can be carried out on the parameters of the mixed Weibull distribution, the principle is simple, the calculation is simple, and the popularization is easy.
Drawings
FIG. 1 is a flow chart illustrating a method for predicting the average lifetime of logistics storage equipment of a tobacco enterprise according to an embodiment of the present invention;
fig. 2 is a graph showing a reliability function curve corresponding to the dual mixed weibull distribution function proposed in the embodiment of the present invention, and a comparison graph of the reliability function curve of the logistics storage equipment which is preliminarily determined.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for better illustration of the present embodiment, certain parts of the drawings may be omitted, enlarged or reduced, and do not represent actual dimensions;
it will be understood by those skilled in the art that certain well-known descriptions of the figures may be omitted.
The positional relationships depicted in the drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
the technical scheme of the invention is further explained by combining the drawings and the embodiment;
the positional relationships depicted in the drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
examples
The tobacco enterprise logistics storage equipment mainly comprises a goods shelf, a stacking vehicle, a carrying vehicle, an entry and exit conveying device, a sorting device, a lifting machine, a case unpacking machine, a case sealing machine, a carrying robot and a computer management and monitoring system. These devices may constitute automated, semi-automated, mechanized commercial warehouses to stack, access, and sort carrier items. In terms of classification, the logistics storage equipment can be divided into two major categories: logistics equipment and warehousing equipment.
The reliability of the logistics storage equipment greatly affects the safe operation of the logistics distribution center, and for predicting the average life of the logistics storage equipment, the embodiment provides a method for predicting the average life of the logistics storage equipment of the tobacco enterprise, the flow chart of the method is shown in fig. 1, and referring to fig. 1, the method includes:
s1, acquiring failure time sample data of the logistics storage equipment, wherein the sample data comprises the data of the number of the same type of failure equipment and the data of service life time of the logistics storage equipment and represents known historical failure data of the logistics storage equipment.
S2, preliminarily determining a reliability function curve of the logistics storage equipment according to the failure time sample data of the logistics storage equipment;
in this embodiment, a case unpacking machine of tobacco enterprise logistics storage equipment is taken as a specific equipment object, a median rank is calculated according to sample data, and a reliability function curve of the case unpacking machine is determined by adopting an average rank method;
specifically, the method comprises the following steps:
acquiring failure time sample data of m logistics storage equipment in total, and arranging the failure time sample data of the m logistics storage equipment from small to large according to failure time, namely t1≤t2≤…≤tm；
Calculating an observed value of the cumulative failure probability, wherein the expression is as follows:
Figure BDA0003156974340000061
wherein, i is 1, 2, …, m, which represents the order of data in m data;
Figure BDA0003156974340000062
an observed value representing a cumulative probability of failure for the ith data;
and calculating the reliability value, wherein the expression is as follows:
Figure BDA0003156974340000063
wherein the content of the first and second substances,
Figure BDA0003156974340000064
representing the reliability value of the ith data to obtain an observation sample data set:
Figure BDA0003156974340000065
Figure BDA0003156974340000066
with tiIs shown as the abscissa of the graph,
Figure BDA0003156974340000067
and drawing a reliability function curve by using the data in the observation sample data set as a vertical coordinate.
S3, acquiring fault cause data of the logistics storage equipment, and determining the number of times n of the mixed Weibull distribution function and a probability density function corresponding to the n-times mixed Weibull distribution function according to the fault cause data;
in this embodiment, when determining the multiple n of the mixed weibull distribution function according to the failure cause data, first determine the failure cause of the logistics storage equipment with an occurrence probability greater than E from the failure cause data of the logistics storage equipment, where E is an occurrence probability threshold, take the failure cause of the logistics storage equipment with an occurrence probability greater than E as the main failure cause of the logistics storage equipment, then determine the number n of the main failure causes, and take the number n of the main failure causes as the multiple of the mixed weibull distribution function. Where E is a threshold used to define a differential screening of the primary failure cause from the other failure cause data.
The expression of the probability density function corresponding to the mixed weibull distribution function is:
Figure BDA0003156974340000071
wherein f (t) represents a probability density function; n represents the multiplicity of the mixed Weibull distribution function; w is aiAre weighted, satisfy
Figure BDA0003156974340000072
wi、βi、ηiAll are parameters of n-fold mixed Weibull distribution functions and are also unknown parameters to be estimated in the probability density function.
After the probability density function corresponding to the n-fold mixed Weibull distribution function is determined, the reliability function corresponding to the n-fold mixed Weibull distribution function is obtained according to the probability density function corresponding to the n-fold mixed Weibull distribution function, and the expression is as follows:
Figure BDA0003156974340000073
in this embodiment, the main failure cause of the box unpacking machine is analyzed to be two failures, namely, a cylinder failure and a motor failure, so that the dual weibull distribution function can be determined to be used, and then the failure probability density function of the box unpacking machine is determined to be according to the expression of the dual weibull distribution function:
Figure BDA0003156974340000074
reliability function of
Figure BDA0003156974340000075
S4, defining unknown parameters to be estimated in the probability density function, and estimating the unknown parameters based on a nonlinear least square method of an L-M algorithm by combining the reliability values on the reliability function curve of the logistics storage equipment preliminarily determined in the step S2 to obtain parameter estimation values;
the process of estimating the unknown parameters by using the L-M algorithm comprises the following steps:
the unknown parameter [ wi，βi，ηi]The set of (a) is denoted as vector θ';
the unknown parameters were transformed as follows:
wi＝(sinai)2，ηi＝exp(bi)，βi＝exp(ci) At this time, the unknown parameters to be estimated are converted into vectors: theta ═ ai，ci，bi]；
According to the observation sample data set
Figure BDA0003156974340000076
And principle of non-linear least square methodSetting an optimization model objective function of parameter estimation as follows:
Figure BDA0003156974340000077
wherein the content of the first and second substances,
Figure BDA0003156974340000078
j is the jth reliability value in the reliability function corresponding to the probability density function, wherein j is 1, 2, …, m; the following steps are carried out:
s41: introducing a parameter l, making l equal to 0, and giving initial values theta of all unknown parameters to be estimatedlSetting error precision epsilon > 0, setting initial step length lambdalWhen the value is 0.01, an initial objective function value f (θ) is calculatedl)；
S42: calculate the residual vector for the l-th iteration: r (theta)l)＝(r1(θl)，r2(θl)，…，rm(θl))；
Note the book
Figure BDA0003156974340000081
Is R at the first iterationjThe value of (1), then in the residual vector
Figure BDA0003156974340000082
Calculating the objective function at thetalThe Jacobian matrix of (a), denoted as J (θ)l)；
S43: according to J (theta)l) And r (theta)l) Calculating a matrix A, wherein the calculation expression is as follows: a ═ J (θ)l)]TJ(θl) And calculating a matrix B, wherein the calculation expression is as follows: b ═ J (θ)l)]Tr(θl)；
S44: calculating an iteration increment Δl＝-[A+λlI]-1B, calculating the next iteration value according to the iteration increment, wherein the iteration formula is thetal+1＝θl+Δl；
S45: and determining an estimation value of the unknown parameter according to the iteration condition. The specific process of step S45 is:
s451. if | | | DeltalStopping iteration and optimizing if | < epsilonSolution (II)
Figure BDA0003156974340000084
Outputting an estimation value result of the unknown parameters; otherwise, go to step S452;
s452, calculating f (theta)l+1) And f (theta)l) If f (θ)l+1)-f(θl) If < 0, let λl+1＝0.1×λlAnd l +1, return to performing step S42; otherwise, let λl＝10×λlThe flow returns to step S44.
In this embodiment, fitting is performed by fitting a double weibull distribution function to sample data, and based on the double weibull distribution function, there are 5 unknown parameters to be estimated, w each1，β1，β2，η1，η2Record vector θ' ═ w1，β1，β2，η1，η2]Wherein, in the step (A),
Figure BDA0003156974340000083
finally, according to the parameter estimation value, determining a curve of the reliability function r (t) corresponding to the dual hybrid weibull distribution function, and comparing the curve with the reliability function curve of the logistics storage equipment preliminarily determined in step S2, where the comparison result is shown in fig. 2, where the reliability function curve of the logistics storage equipment preliminarily determined in step S2 is also an actual reliability curve based on actual data, and corresponds to a solid line in fig. 2; the curve of the reliability function r (t) corresponding to the dual mixed weibull distribution function corresponds to the solid line in fig. 2, and it can be seen from fig. 2 that the accuracy of the method of fitting by using the mixed weibull distribution function in the present invention is high, and further, the step S5 is executed:
and S5, introducing a mean life calculation formula, and predicting the mean life of the logistics storage equipment by combining the parameter estimation value. The average life calculation formula is as follows:
Figure BDA0003156974340000091
wherein, wig、ηig、βigRespectively the original unknown parameters wi、βi、ηiIs the gamma function, and L is the average life of the logistics storage equipment.
In this embodiment, under the dual weibull distribution function, in combination with the parameter estimation value, the average life prediction formula of the case unpacking machine is:
Figure BDA0003156974340000092
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.

Claims (10)

1. A method for predicting the average service life of logistics storage equipment of a tobacco enterprise is characterized by at least comprising the following steps:
s1, acquiring failure time sample data of logistics storage equipment,
s2, preliminarily determining a reliability function curve of the logistics storage equipment according to the failure time sample data of the logistics storage equipment;
s3, acquiring fault cause data of the logistics storage equipment, and determining the number of times n of the mixed Weibull distribution function and a probability density function corresponding to the n-times mixed Weibull distribution function according to the fault cause data;
s4, defining unknown parameters to be estimated in the probability density function, and estimating the unknown parameters based on a nonlinear least square method of an L-M algorithm by combining the reliability values on the reliability function curve of the logistics storage equipment preliminarily determined in the step S2 to obtain parameter estimation values;
and S5, introducing a mean life calculation formula, and predicting the mean life of the logistics storage equipment by combining the parameter estimation value.
2. The method for predicting the average life of the logistics storage equipment of the tobacco enterprise as claimed in claim 1, wherein the failure time sample data of the life of the logistics storage equipment in step S1 comprises the same type of failure equipment quantity data and life time data of the logistics storage equipment.
3. The method for predicting the average lifetime of logistics storage equipment of tobacco enterprises as claimed in claim 1, wherein the step S2 of preliminarily determining the reliability function curve of logistics storage equipment comprises:
acquiring failure time sample data of m logistics storage equipment in total, and arranging the failure time sample data of the m logistics storage equipment from small to large according to failure time, namely t1≤t2≤…≤tm；
Calculating an observed value of the cumulative failure probability, wherein the expression is as follows:
Figure FDA0003156974330000011
wherein, i is 1, 2, …, m, which represents the order of data in m data;
Figure FDA0003156974330000012
an observed value representing a cumulative probability of failure for the ith data;
and calculating the reliability value, wherein the expression is as follows:
Figure FDA0003156974330000013
wherein the content of the first and second substances,
Figure FDA0003156974330000014
representing the reliability value of the ith data to obtain an observation sample data set:
Figure FDA0003156974330000015
Figure FDA0003156974330000021
with tiIs shown as the abscissa of the graph,
Figure FDA0003156974330000022
and drawing a reliability function curve by using the data in the observation sample data set as a vertical coordinate.
4. The method for predicting the average lifetime of logistics storage equipment of a tobacco enterprise as claimed in claim 3, wherein in step S3, when determining the multiple n of the mixed Weibull distribution function according to the failure cause data, first determining the failure cause of the logistics storage equipment with the occurrence probability greater than E from the failure cause data of the logistics storage equipment, where E is an occurrence probability threshold, and using the failure cause of the logistics storage equipment with the occurrence probability greater than E as the main failure cause of the logistics storage equipment, then determining the number n of the main failure causes, and using the number n of the main failure causes as the multiple of the mixed Weibull distribution function.
5. The method for predicting the average service life of the logistics storage equipment of the tobacco enterprise as claimed in claim 4, wherein the expression of the probability density function corresponding to the n-fold mixed Weibull distribution function is as follows:
Figure FDA0003156974330000023
wherein f (t) represents a probability density function; n represents the multiplicity of the mixed Weibull distribution function; w is aiAre weighted, satisfy
Figure FDA0003156974330000024
wi、βi、ηiAll are parameters of n-fold mixed Weibull distribution functions and are also unknown parameters to be estimated in the probability density function.
6. The method for predicting the average service life of the logistics storage equipment of the tobacco enterprise as claimed in claim 5, wherein after the probability density function corresponding to the n-fold mixed Weibull distribution function is determined, the corresponding reliability function is obtained according to the probability density function corresponding to the n-fold mixed Weibull distribution function, and the expression is as follows:
Figure FDA0003156974330000025
7. the method for predicting the average life of the logistics storage equipment of the tobacco enterprise as claimed in claim 6, wherein the estimation of the unknown parameters by the L-M algorithm in step S4 comprises the following steps:
the unknown parameter [ wi，βi，ηi]The set of (a) is denoted as vector θ';
the unknown parameters were transformed as follows:
wi＝(sin ai)2，ηi＝exp(bi)，βi＝exp(ci) At this time, the unknown parameters to be estimated are converted into vectors: theta ═ ai，ci，bi]；
According to the observation sample data set
Figure FDA0003156974330000026
And the principle of a nonlinear least square method, and setting an optimization model objective function of parameter estimation as follows:
Figure FDA0003156974330000027
wherein the content of the first and second substances,
Figure FDA0003156974330000028
j is the jth reliability value in the reliability function corresponding to the probability density function, wherein j is 1, 2, …, m; the following steps are carried out:
s41: introducing a parameter l, making l equal to 0, and giving initial values theta of all unknown parameters to be estimatedlSetting error precision epsilon > 0, setting initial step length lambdalWhen the value is 0.01, an initial objective function value f (θ) is calculatedl)；
S42: calculate the residual vector for the l-th iteration: r (theta)l)＝(r1(θl)，r2(θl)，…，rm(θl))；
Note the book
Figure FDA0003156974330000031
Is R at the first iterationjThe value of (1), then in the residual vector
Figure FDA0003156974330000032
Calculating the objective function at thetalThe Jacobian matrix of (a), denoted as J (θ)l)；
S43: according to J (theta)l) And r (theta)l) Calculating a matrix A, wherein the calculation expression is as follows: a ═ J (θ)l)]TJ(θl) And calculating a matrix B, wherein the calculation expression is as follows: b ═ J (θ)l)]Tr(θl)；
S44: calculating an iteration increment Δl＝-[A+λlI]-1B, calculating the next iteration value according to the iteration increment, wherein the iteration formula is thetal+1＝θl+Δl；
S45: and determining an estimation value of the unknown parameter according to the iteration condition.
8. The method for predicting the average life of the logistics storage equipment of the tobacco enterprise as claimed in claim 7, wherein the specific process of the step S45 is as follows:
s451. if | | | DeltalStopping iteration and obtaining the optimal solution if | | < epsilon
Figure FDA0003156974330000033
Outputting an estimation value result of the unknown parameters; otherwise, go to step S452;
s452, calculating f (theta)l+1) And f (theta)l) If f (θ)l+1)-f(θl) If < 0, let λl+1＝0.1×λlAnd l +1, return to performing step S42; otherwise, let λl＝10×λlThe flow returns to step S44.
9. The method for predicting the average lifetime of logistics storage equipment of a tobacco enterprise as claimed in claim 8, wherein after step S4 and before step S5, further comprising: and according to the parameter estimation value, determining a curve of the reliability function R (t) corresponding to the n-fold mixed Weibull distribution function, and comparing the curve with the reliability function curve of the logistics storage equipment preliminarily determined in the step S2.
10. The method for predicting the average life of the logistics storage equipment of the tobacco enterprise as claimed in claim 9, wherein the average life calculation formula is as follows:
Figure FDA0003156974330000034
wherein, wig、ηig、βigRespectively the original unknown parameters wi、βi、ηiIs the gamma function, and L is the average life of the logistics storage equipment.
CN202110781009.0A 2021-07-09 2021-07-09 Method for predicting average service life of logistics storage equipment of tobacco enterprise Pending CN113627651A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110781009.0A CN113627651A (en) 2021-07-09 2021-07-09 Method for predicting average service life of logistics storage equipment of tobacco enterprise

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110781009.0A CN113627651A (en) 2021-07-09 2021-07-09 Method for predicting average service life of logistics storage equipment of tobacco enterprise

Publications (1)

Publication Number Publication Date
CN113627651A true CN113627651A (en) 2021-11-09

Family

ID=78379361

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110781009.0A Pending CN113627651A (en) 2021-07-09 2021-07-09 Method for predicting average service life of logistics storage equipment of tobacco enterprise

Country Status (1)

Country Link
CN (1) CN113627651A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107148081A (en) * 2017-06-02 2017-09-08 重庆邮电大学 Mono-station location method based on nonlinear restriction least square
CN109933924A (en) * 2019-03-19 2019-06-25 云南电网有限责任公司玉溪供电局 A kind of protective relaying device life-span prediction method based on LM method parameter Estimation
CN110728008A (en) * 2018-06-27 2020-01-24 中国电力科学研究院有限公司 Method and device for determining expected life of smart electric meter
CN113792939A (en) * 2021-09-30 2021-12-14 国网湖南省电力有限公司 Electric energy meter reliability prediction method and device based on mixed Weibull distribution

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107148081A (en) * 2017-06-02 2017-09-08 重庆邮电大学 Mono-station location method based on nonlinear restriction least square
CN110728008A (en) * 2018-06-27 2020-01-24 中国电力科学研究院有限公司 Method and device for determining expected life of smart electric meter
CN109933924A (en) * 2019-03-19 2019-06-25 云南电网有限责任公司玉溪供电局 A kind of protective relaying device life-span prediction method based on LM method parameter Estimation
CN113792939A (en) * 2021-09-30 2021-12-14 国网湖南省电力有限公司 Electric energy meter reliability prediction method and device based on mixed Weibull distribution

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
滕永兴 等: "基于混合威布尔分布的电能表寿命预测研究", 《电气传动》, vol. 51, no. 1, pages 61 - 66 *
王冕: "水稻插秧机可靠性评价研究", 《中国优秀硕士学位论文全文数据库 农业科技辑》, no. 12, pages 60 *
金少华 等: "利用贝叶斯理论对电工产品进行可靠性评估", 《低压电器》, no. 01, pages 8 - 10 *
陆祖建 等: "关于威布尔分布的若干问题", 《质量与可靠性》, no. 6, pages 9 - 15 *

Similar Documents

Publication Publication Date Title
US6625589B1 (en) Method for adaptive threshold computation for time and frequency based anomalous feature identification in fault log data
CN113011530B (en) Intelligent ammeter fault prediction method based on multi-classifier fusion
CN112083244B (en) Integrated intelligent diagnosis system for faults of avionic equipment
CN106570778A (en) Big data-based data integration and line loss analysis and calculation method
CN107967485A (en) Electro-metering equipment fault analysis method and device
CN111191897A (en) Service flow online compliance prediction method and system based on bidirectional GRU neural network
CN111915026A (en) Fault processing method and device, electronic equipment and storage medium
CN116467674B (en) Intelligent fault processing fusion updating system and method for power distribution network
CN109634790B (en) Disk fault prediction method based on cyclic neural network
CN112685910A (en) Complex equipment power pack fault prediction method based on hybrid prediction model
CN114266289A (en) Complex equipment health state assessment method
CN104699614A (en) Software defect component predicting method
Xiang Using ARIMA‐GARCH Model to Analyze Fluctuation Law of International Oil Price
CN113065675A (en) Optimal equipment maintenance method based on residual life prediction
CN111260127A (en) Fault prediction system and method based on full-machine digital twin model
CN109359742B (en) Method for generating preventive maintenance period of subway subsystem
CN114036647A (en) Power battery safety risk assessment method based on real vehicle data
Guo et al. Towards practical and synthetical modelling of repairable systems
CN113627651A (en) Method for predicting average service life of logistics storage equipment of tobacco enterprise
CN115460061B (en) Health evaluation method and device based on intelligent operation and maintenance scene
CN115687330A (en) Automobile fault code analysis method, storage medium and electronic equipment
CN109978038B (en) Cluster abnormity judgment method and device
CN113065683A (en) Price prediction method, device, equipment and storage medium for vehicle pledge
CN113064883A (en) Method for constructing logistics wind control model, computer equipment and storage medium
Sharma et al. Ranking of reliability models based on accurate estimation and weighted function

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination