CN115719013B - Multistage maintenance decision modeling method and device for intelligent manufacturing production line - Google Patents

Abstract

The invention discloses a multi-level maintenance decision-making modeling method and device for an intelligent manufacturing production line. The method includes: using historical fault data of components to obtain the fault time distribution of functional faults and degradation faults of components; , to construct the functional life model and degradation life model of the component; according to the functional life model and degradation life model of the component, construct the functional availability model, performance availability model and performance availability penalty model of the equipment; use the functional availability model of the equipment, The performance availability model and the performance availability penalty model calculate the loss of the production line due to functional failures and degradation failures, and construct the production line benefit model; according to the constraints required by the production line, optimize and solve the production line benefit model to obtain Maintenance decisions that maximize line efficiency. The present invention can construct a multi-level maintenance decision-making model of an intelligent manufacturing production line, reduce unplanned downtime, and maximize production line benefits.

Description

A multi-level maintenance decision-making modeling method and device for an intelligent manufacturing production line

technical field

The invention belongs to the technical field of intelligent manufacturing, and in particular relates to a multi-level maintenance decision-making modeling method and device for an intelligent manufacturing production line, computer equipment, and a computer-readable storage medium.

Background technique

The intelligent manufacturing production line is composed of multiple manufacturing equipment, and is a manufacturing system that performs interconnection communication and collaborative operation between multiple equipment through information systems such as MES (Manufacturing Execution System). During the production and operation of the production line, functional failures or degradation failures will occur randomly. Random functional failures or hidden degradation failures may cause unexpected shutdown of the production line, which in turn will cause serious economic and social benefit losses. In the flow operation of the production line, the failure of any equipment will cause the loss of the production line to varying degrees. Also, different devices have multiple critical components, and different components have multiple failure modes. The maintenance decisions of most existing manufacturing lines only consider the fault prediction at the equipment level, without considering the impact of underlying components or even different faults, and cannot provide sufficient technical support for the active operation and maintenance of intelligent manufacturing lines.

Contents of the invention

The purpose of the present invention is to provide a multi-level maintenance decision-making modeling method and device, computer equipment, and computer-readable storage medium for an intelligent manufacturing production line, which can build a multi-level maintenance decision-making model for an intelligent manufacturing production line, and effectively support intelligent manufacturing production. Analysis and formulation of line maintenance plans to reduce unplanned downtime and maximize production line benefits.

One aspect of the present invention provides a multi-level maintenance decision modeling method for an intelligent manufacturing production line, the intelligent manufacturing production line includes m equipment E _i , i=1,..., m, each equipment E _i has n _i components C _ij , j=1,...,n _i , the method includes:

Obtaining the fault time distribution step: using the historical fault data of the component C _ij to obtain the fault time distribution of the functional fault and the fault time distribution of the degraded fault of the component C _ij ;

根据部件C_ij的相互影响的多个退化故障的故障时间分布，构建部件C_ij的基于多退化故障随机依赖性的退化寿命模型

The component life model construction step, according to the failure time distribution of all functional failures of the component C _ij , constructs the functional life model of the component C _ij based on the random dependence of the multifunctional faults

According to the failure time distribution of multiple degraded faults of components C _ij interacting with each other, a degradation life model based on the random dependence of multiple degraded faults of components C _ij is constructed

构建设备E_i的基于多部件功能结构依赖性的功能可用度模型A_i(t)，根据部件C_ij的退化寿命模型

构建设备E_i的基于多部件性能结构依赖性的性能可用度模型B_i(t)和性能可用度惩罚模型AB_i(t)，其中，t为时刻；Equipment availability model construction steps, according to the functional life model of component C _ij

Construct the functional availability model A _i (t) of the equipment E _i based on the multi-component functional structure dependence, according to the degradation life model of the component C _ij

Construct the performance availability model B _i (t) and the performance availability penalty model AB _i (t) based on multi-component performance structure dependence of the equipment E _i , where t is the moment;

The production line benefit model construction step uses the functional availability model A _i (t) of the equipment E _i to calculate the loss of the production line due to functional failure; uses the performance availability model B _i (t) of the equipment E _i and the performance availability Penalty model AB _i (t), calculates the loss of the production line due to degraded faults, and constructs a production line benefit model based on the economic dependence of multiple devices according to the losses caused by functional faults and degraded faults of the production line;

The maintenance decision obtaining step is to optimize and solve the production line benefit model according to the constraints required by the production line, and obtain a maintenance decision that maximizes the production line benefit.

融合为设备E_i的功能模型

计算设备在时刻t的预期功能故障时长

得到设备E_i的功能可用度

Preferably, in the step of constructing the equipment availability model, the functional life model of the component C _ij

fused as a functional model of device E _i

Calculate the expected functional failure time of the equipment at time t

Get the functional availability of equipment E _i

得到设备E_i的性能故障时间

从而得到设备E_i的性能可用度模型

Preferably, in the step of constructing the equipment availability model, the expected degradation failure duration of the component C _ij within Γ(t) is calculated

Get the performance failure time of equipment E _i

Thus, the performance availability model of equipment E _i is obtained

Wherein, Γ(t) is the expected function normal time of the equipment E _i , Γ(t)=t−α _i (t).

Preferably, in the equipment availability model construction step, the performance availability penalty model AB _i (t) of the equipment E _i is constructed as follows:

Among them, ω _ij is the influence factor of the performance of component C _ij on the production quality of equipment E _i .

Preferably, in the step of constructing the production line benefit model, the expected failure time point is obtained according to the function availability model A _i (t) and the performance availability model B _i (t), according to the series-parallel relationship between each device and The performance availability penalty model AB _i (t), calculates the penalty in the time period between each expected failure time point.

其中，z为蒙特卡洛抽样次数，z＞10000。Preferably, in the step of constructing the component life model, Monte Carlo sampling is performed on the failure time distributions of all functional failures, and the earliest failure component statistics X _ij =min in the wth sampling sample of each distribution are taken {x _{ij, 1} , ..., x _{ij, w} , ..., x _{ij, z} }, by fitting this statistic, the functional life model is obtained

Among them, z is the number of Monte Carlo sampling, z>10000.

作为退化寿命模型，其中，

为部件C_ij的退化故障数，σ_lk为协方差矩阵，衡量多个退化故障的随机依赖性程度。Preferably, in the component life model construction step, a multivariate joint distribution function is constructed for multiple degradation faults affecting each other

As a degenerate lifetime model, where,

is the number of degradation faults of component C _ij , and σ _lk is the covariance matrix, which measures the degree of random dependence of multiple degradation faults.

Preferably, in the component life model construction step, based on the random dependence between functional faults and degraded faults, according to the occurrence of functional faults with a certain probability and a certain proportion, the mean value of the fault distribution time of degraded faults is simultaneously increased Revise the degenerate life model in the way of large variance

Another aspect of the present invention provides a multi-level maintenance decision modeling device for an intelligent manufacturing production line, the intelligent manufacturing production line includes m equipment E _i , i=1,...,m, each equipment E _i With n _i components C _ij , j=1, . . . , n _i , the device comprises:

The failure time distribution obtaining module: use the historical failure data of the component C _ij to obtain the failure time distribution of the functional failure and the failure time distribution of the degradation failure of the component C _ij ;

根据部件C_ij的相互影响的多个退化故障的故障时间分布，构建部件C_ij的退化寿命模型

设备可用度模型构建模块，根据部件C_ij的功能寿命模型

构建设备E_i的功能可用度模型A_i(t)，根据部件C_ij的退化寿命模型

构建设备E_i的性能可用度模型B_i(t)和性能可用度惩罚模型AB_i(t)，其中，t为时刻；The component life model building block, according to the failure time distribution of all functional failures of the component C _ij , constructs the functional life model of the component C _ij

According to the failure time distribution of multiple degradation faults of components C _ij interacting with each other, the degradation life model of components C _ij is constructed

Equipment Availability Model Building Block, Functional Lifetime Model for Components C _ij

Construct the functional availability model A _i (t) of the equipment E _i , according to the degradation life model of the component C _ij

Build the performance availability model B _i (t) and the performance availability penalty model AB _i (t) of the equipment E _i , where t is the moment;

Production line benefit model building block, using the functional availability model A _i (t) of equipment E _i to calculate the loss of the production line due to functional failure; _{using the performance availability model B i (} t) and performance availability of equipment E i Penalty model AB _i (t), calculates the loss of the production line due to degraded faults, and constructs a production line benefit model based on the losses caused by functional faults and degraded faults of the production line;

The maintenance decision obtaining module optimizes and solves the production line benefit model according to the constraints required by the production line, and obtains a maintenance decision that maximizes the production line benefit.

Another aspect of the present invention provides a computer device, including a memory and a processor, the memory stores a computer program, and it is characterized in that, when the processor executes the computer program, the steps of the above method are implemented.

Another aspect of the present invention provides a computer-readable storage medium on which a computer program is stored, wherein the computer program implements the steps of the above-mentioned method when executed by a processor.

According to the multi-level maintenance decision-making modeling method and device, computer equipment, and computer-readable storage medium of the intelligent manufacturing production line according to the above aspects of the present invention, a multi-level maintenance model of the intelligent manufacturing production line can be constructed to effectively support the maintenance of the intelligent manufacturing production line The analysis and formulation of the plan can reduce unplanned downtime and maximize the efficiency of the production line.

Description of drawings

In order to illustrate the technical solution of the present invention more clearly, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative labor: Fig. 1 is the flow chart of the multi-level maintenance decision-making modeling method of the intelligent manufacturing production line in the embodiment of the present invention picture.

Fig. 2 is an explanatory diagram of the principle of a production line efficiency model according to an embodiment of the present invention.

Fig. 3 is a structural diagram of a multi-level maintenance decision modeling device for an intelligent manufacturing production line according to an embodiment of the present invention.

FIG. 4 is a structural diagram of a computer device according to an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Apparently, the described embodiments are only some of the embodiments of the present invention, not all of them. the embodiment. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Embodiments of the present invention provide a multi-level maintenance decision modeling method for an intelligent manufacturing production line. An intelligent manufacturing line usually includes multiple devices, different devices have multiple key components, and different components have multiple failure modes. Taking a small production line as an example, the production line includes 1 robot and 4 machine tools and other equipment. The robot has key components such as slide rails, servo motors, and reducers. Different components have multiple failure modes, and their impacts are also different. For example For servo motors, the encoder is abnormal due to mechanical shock and cannot work; there is a short circuit of inter-turn insulation, bearing wear, etc., resulting in a decrease in system performance, but it can still work.

For convenience, the following assumptions are made:

An intelligent manufacturing production line includes m devices, and the devices are represented by E _i , i=1,...,m;

Each device E _i has n _i components, the components are denoted by C _ij , j=1,...,n _i ;

个，表示为

性能退化类故障共

个，表示为

满足

Each component C _ij has p _ij faults, among which, the loss of function faults have a total of

one, expressed as

Total performance degradation faults

one, expressed as

satisfy

Among them, the failure of function loss, referred to as functional failure, once it occurs, the components cannot work, and the production line will be shut down, such as mechanical fracture, structural jamming, electrical short circuit, etc., which need to be shut down for overhaul or replacement; If it happens, the components can still work, but the quality of the work is degraded, which leads to an increase in the rate of defective products in the production line, and when it reaches the threshold line, it needs to be shut down for maintenance.

Different from the fault prediction and maintenance decision-making based on equipment level in the prior art, the multi-level maintenance decision-making modeling method of the intelligent manufacturing production line in the embodiment of the present invention considers multiple faults and their random dependencies, multiple components and their structural dependencies, Multi-equipment and its economic dependence, according to the idea of modularization, comprehensive multi-level dependency modeling, bottom-up analysis and construction of the first-level fault distribution model, the second-level component life model, the third-level equipment availability model and the fourth-level The production line benefit model provides technical support for enterprises to build maintenance decision models.

In order to apply the multi-level maintenance decision-making modeling method of the embodiment of the present invention, firstly, by means of FMEA (Failure Mode and Effect Analysis, failure mode and impact analysis), FTA (Fault Tree Analysis, fault tree analysis) and other means, carry out from top to bottom Production line→equipment→component→decomposition of failure modes; for key failure modes (hereinafter referred to as failures), they are divided into two types: degradation failures and functional failures. The former can be directly or indirectly detected through equipment sensor monitoring or maintenance detection "Discover" the degeneration fault process and degree; the faults that cannot be "discovered" by the process are classified as functional faults. For example: inter-turn insulation short circuit of robot servo motor, bearing wear, etc., are all degenerate faults in mechanism, but need to install current/vibration sensors online, or use special offline device tests regularly during maintenance to find out. The equipment is equipped with online monitoring Or regular maintenance testing is classified as a degradation failure, otherwise it is classified as a functional failure.

Fig. 1 is a flowchart of a multi-level maintenance decision modeling method for an intelligent manufacturing production line according to an embodiment of the present invention. As shown in FIG. 1 , the multi-level maintenance decision modeling method for an intelligent manufacturing production line according to an embodiment of the present invention includes steps S1-S5.

Step S1: Obtaining step of failure time distribution

In this step, using the historical fault data of the component C _ij , the fault time distribution of the functional fault and the fault time distribution of the degenerative fault of the component C _ij are obtained.

In one embodiment, the "fault distribution model" of key faults, that is, the fault time distribution, is obtained by using the historical fault data of components such as on-line monitoring of equipment or periodic testing of maintenance, through life prediction algorithms, such as for functional faults, By modeling the stochastic process at the occurrence time of similar faults in history, the fault time distribution function PDF ¹ of functional faults is constructed; for degenerate faults, a mechanism-based or data-driven method is used to construct a degradation model to obtain the fault time distribution function PDF ² of degenerate faults . Using the fault distribution model, the fault probability at different moments can be obtained, and the fault consequences (such as severity) can be integrated to calculate the corresponding fault risk.

Step S2: Construction steps of component life model

根据部件C_ij的相互影响的多个退化故障的故障时间分布，构建部件C_ij的基于多退化故障随机依赖性的退化寿命模型

In this step, according to the failure time distribution of all functional failures of components C _ij , a functional life model based on the random dependence of multifunctional failures of components C _ij is constructed

According to the failure time distribution of multiple degraded faults of components C _ij interacting with each other, a degradation life model based on the random dependence of multiple degraded faults of components C _ij is constructed

In one embodiment, a "component life model" based on multiple fault stochastic dependencies is constructed for key components (such as functional components such as sliding guide rails of robots, reducers, and servo motors).

为部件C_ij的功能故障数，通过故障时间分布获得步骤S1获得

为

的故障时间分布，对所有功能故障的故障时间分布分别进行蒙特卡洛抽样

假如对每个功能故障的故障时间分布进行z次蒙特卡洛抽样，一般地z＞10000。取每个分布的第w次抽样样本x_ijk，w，考虑功能故障的竞争失效关系(即任一功能故障发生，设备停机)，最早发生(抽样时间最小)的故障

组成统计量X_ij＝min{x_ij，1，...，x_ij，w，....，x_ij，z}，用合适的分布进行拟合，获得部件功能寿命模型

In the construction of component life model, the random dependence between multiple functional failures is first considered, for functional failure

is the number of functional failures of component C _ij , obtained through step S1 of obtaining the distribution of failure time

for

The failure time distribution of all functional failures, Monte Carlo sampling is performed on the failure time distribution of all functional failures

If the failure time distribution of each functional failure is subjected to z Monte Carlo sampling, generally z>10000. Take the wth sampling sample x _{ijk, w} of each distribution, consider the competitive failure relationship of functional failures (that is, any functional failure occurs, and the equipment stops), the failure that occurs earliest (minimum sampling time)

Composition statistic X _ij = min{x _{ij, 1} ,..., x _{ij, w} ,..., x _{ij, z} }, fit with a suitable distribution to obtain the component functional life model

和

其中

为部件C_ij的退化故障数，通过步骤S1获得退化故障

和

的故障时间分布

和

表示其联合分布，σ_lk为协方差矩阵，衡量随机依赖性程度。依次类推2个以上故障情形，构建部件退化寿命模型。In the construction of the component life model, the stochastic dependence among multiple degradation faults is secondly considered. Considering that performance degradation is not a single degradation, it is often an overall degradation, and most of the time the degradation process of the same component is correlated (mostly positive correlation). For example, for the servo motor of a robot, the more the motor grease is lost, the bearing wear will increase, and the vibration of the motor will increase. The performance of the bearing will lead to a decrease in efficiency, an increase in heat generation, and an increase in temperature, resulting in other deformation and insulation problems, etc. , where the two monitoring quantities of vibration and temperature rise are positive correlation degradation faults. Since the more serious the degradation, the higher the risk of failure, the influence relationship of the degradation quantity is described by the failure probability, then two or more degraded failures that affect each other can construct their multivariate joint distribution function, through the covariance matrix or correlation coefficient, etc. Measure the size of the correlation. In one example, for 2 degenerate faults affecting each other

and

in

is the number of degenerate faults of component C _ij , the degenerate faults are obtained through step S1

and

The failure time distribution of

and

Represents its joint distribution, σ _lk is the covariance matrix, which measures the degree of random dependence. By analogy with more than two fault situations, a component degradation life model is constructed.

In the construction of component life model, the stochastic dependence between functional failure and degradation failure is further considered. The occurrence of functional failures will increase the risk of dependent degenerate failures with a certain probability. For example, the abnormal power supply of servo motors occurs randomly and is a functional failure, and the abnormal power supply will cause the risk of degradation failures such as insulation. Regarding the random dependence of functional faults and degenerate faults, considering functional faults does not change the distribution type of degenerate faults (such as normal distribution), but the occurrence of functional faults will reduce the mean value of degenerate fault time with a certain probability and increase at the same time Its variance, thus obtaining the modified degradation life model

Step S3: Construction steps of equipment availability model

构建设备E_i的基于多部件功能结构依赖性的功能可用度模型A_i(t)，根据部件C_ij的退化寿命模型

构建设备E_i的基于多部件性能结构依赖性的性能可用度模型B_i(t)和性能可用度惩罚模型AB_i(t)。In this step, according to the functional lifetime model of component C _ij

Construct the functional availability model A _i (t) of the equipment E _i based on the multi-component functional structure dependence, according to the degradation life model of the component C _ij

Construct the performance availability model B _i (t) and the performance availability penalty model AB _i (t) of the equipment E _i based on the multi-component performance structure dependence.

In one embodiment, for key equipment (such as robots and CNC machine tools in small production lines), a three-level equipment availability model based on multi-component structure dependencies is constructed.

Specifically, in the construction of the equipment availability model, at time t:

融合为单个设备的功能模型

计算设备在时刻t的预期功能故障时长为

设备功能正常工作时间预期为Γ(t)＝t-α_i(t)，设备功能可用度模型

在0～Γ(t)内，根据各部件的退化寿命模型，计算各部件C_ij在Γ(t)内的预期退化故障时长

设备E_i的性能故障时间为

设备性能可用度模型

在Γ(t)-β_i(t)～Γ(t)之间，设备“带病运行”，考虑部件性能影响设备的生产质量，引入ω_ij作为部件C_ij的性能对设备生产质量的影响因子，退化越严重对生产质量的影响越大，ω_ij是t的增函数，随

的增大而增大；因此，随着时间推进，第一个部件到达退化寿命，引入第一个ω_ij，若不维修，第二个部件退化超差，引入第二个ω_ij，多部件的影响累加(如线性累加)，依次类推，得到设备性能可用度惩罚模型，设备性能可用度惩罚模型AB_i(t)是一个分段函数：For example, the function life model of each component can be calculated by the above-mentioned Monte Carlo sampling and taking a small method

Functional models fused into a single device

The expected functional failure duration of computing equipment at time t is

The normal working time of equipment function is expected to be Γ(t)=t-α _i (t), and the equipment function availability model

Within 0～Γ(t), according to the degradation life model of each component, calculate the expected degradation failure time of each component C _ij within Γ(t)

The performance failure time of equipment E _i is

Equipment Performance Availability Model

Between Γ(t) _-βi (t)～Γ(t), the equipment "operates with a disease", considering that the performance of components affects the production quality of the equipment, and introduces ω _ij as the impact of the performance of the component C _ij on the production quality of the equipment Factor, the more serious the degradation, the greater the impact on production quality, ω _ij is an increasing function of t, with

Therefore, as time progresses, the first component reaches the degradation life, and the first ω _ij is introduced. If it is not repaired, the degradation of the second component is out of tolerance, and the second ω _ij is introduced, and the multi-component The impact accumulation (such as linear accumulation), and so on, to obtain the equipment performance availability penalty model, the equipment performance availability penalty model AB _i (t) is a piecewise function:

的顺序统计量，即从大到小排序的第k个，

即为β_i(1)(t)。Here, (k) in β _i(k) (t) means

The order statistics of , that is, the kth sorted from large to small,

That is, β _i(1) (t).

According to the performance availability penalty model AB _i (t) of the equipment E _i , the penalty factor of the equipment E _i at a certain time t can be obtained.

Step S4: Construction steps of production line benefit model

In this step, use the functional availability model A _i (t) of equipment E _i to calculate _{the loss of the production line due to functional failure; use the performance availability model B i (} t) and performance availability penalty model of equipment E i AB _i (t), calculate the loss caused by the degradation failure of the production line, and construct the production line benefit model based on the economic dependence of multiple equipment according to the loss caused by the functional failure and degradation failure of the production line.

In the construction step of the production line benefit model, the expected failure time point is obtained according to the function availability model A _i (t) and the performance availability model B _i (t), and according to the series-parallel relationship between each device and the performance availability Penalize the model AB _i (t), calculating the loss in the time period between each expected failure time point.

In the intelligent manufacturing production line, the connection relationship between each device is different, and the loss of the production line caused by equipment failure is different. For example, in the aforementioned example of a small production line, the production line includes 1 robot and 4 machine tools 1-4, the relationship between the robot and the 4 machine tools is in series, and the 4 machine tools are connected in series and then in parallel. For example, machine tool 1 and Machine tool 2 is connected in series, machine tool 3 is connected in series with machine tool 4, and the two series branches are connected in parallel. Under this connection relationship, if the function of the robot fails, the entire production line will be shut down, and the effective working time of all equipment with availability less than the robot is the same as that of the robot; if the machine tool 1 is shut down, the machine tool 2 will be shut down, and the production line capacity will be halved. Through the availability of each equipment and the series-parallel relationship between each equipment, the production line capacity analysis is carried out.

Fig. 2 is an explanatory diagram of the principle of a production line efficiency model according to an embodiment of the present invention. As shown in Figure 2, all equipment is normal at the beginning _, and the availability and production capacity are _both 100%. penalty factor of ; at time t2, if machine tool 1 fails, the production capacity of machine tool 1 is 0, and machine tool 2 is in series with machine tool 1, so the production capacity of machine tool 1 is also 0%; the production capacity of the robot is halved; at t3 and t4, machine tool 3 and Machine tool 4 has degraded faults respectively, and their production capacity decreases, and the penalty factor of the branch composed of machine tool 3 and machine tool 4 connected in series is accumulated. In the end, the function of the t5 robot failed, and the entire line was discontinued. Thus, the device function availability and performance availability in each time interval are obtained. In the maintenance decision optimization process, time points such as t1, t2, t3, t4, and t5 are the expected failure time points obtained according to the functional availability model A _i (t) and the performance availability model B _i (t).

In the construction step of the production line benefit model, the functional availability model A _i (t) of the equipment E _i is used to calculate the output loss of each expected failure time period, and the output loss is the difference between the production line availability and the rated The product of productivity X, at time t2-t5 in Fig. 3, yield loss CA _2-5 due to functional failure is 50%*(t5-t2)*X, loss after t5 is 100%; utilizing equipment E _i performance availability Model B _i (t) and performance availability penalty model AB _i (t), calculate the loss caused by the production of substandard products CB _1-2 = 0.5X*AB ₂ , 0.5X means that the parallel branch is only responsible for half of the production capacity; similarly CB _4-5 =0.5X*(AB ₃ +AB ₄ ).

In the production line benefit, the economic dependence of multiple devices can be obtained statistically, that is, each maintenance will generate a fixed cost C ₀ , and the fixed cost does not change with the number of maintenance objects, but depends on fixed parts such as personnel/scheduling and waiting Downtime loss for maintenance, etc. Generally, the waiting time for after-event maintenance is much longer than that for predictive maintenance, so the fixed cost of after-event maintenance is much higher than that of preventive maintenance.

In production line benefits, secondly consider the production loss C ₁ under multi-equipment combined maintenance, where C ₁ is the maximum time for multi-equipment parallel maintenance*equipment productivity; and consider replacement of specific components or maintenance of specific failure modes of spare parts and maintenance costs C ₂ .

Production line benefit t*X under normal operation, after deducting the above-mentioned loss due to functional failure and degradation failure, fixed cost C ₀ , production loss C ₁ and maintenance cost C ₂ , etc., the production line benefit is obtained.

Step S5: Maintenance Decision Obtaining Step

In this step, according to the constraints required by the production line, the optimization and solution of the production line benefit model is carried out to obtain a maintenance decision that maximizes the production line benefit.

Specifically, on the basis of building the production line benefit model, according to safety production requirements such as major functional failures (for example, fatal failures are always less than the specified value), quality control requirements for defective product rate (for example, the defective product rate at any time is lower than the specified value ), monthly output requirements (such as maintaining basic output benefits every month), key equipment availability requirements (key equipment function availability must be greater than the specified value) and other production line requirements to add relevant constraints; according to the production line scheduling situation, select the appropriate Time period, using maintenance time, maintenance failures or maintenance parts as optimization variables, using optimization algorithms such as dynamic solving, to optimize and solve the production line benefit model, and obtain the maintenance time and maintenance objects that maximize the production line benefit under constraint conditions, etc. maintenance decisions.

In summary, the multi-level maintenance decision-making modeling method of the intelligent manufacturing production line in the embodiment of the present invention can build a multi-level maintenance model of the intelligent manufacturing production line, provide technical support for the active operation and maintenance of the production line in the intelligent manufacturing environment, and reduce Unplanned downtime to maximize production line efficiency.

Embodiments of the present invention also provide a multi-level maintenance decision-making modeling device for an intelligent manufacturing production line. Fig. 3 is a structural diagram of a multi-level maintenance decision modeling device for an intelligent manufacturing production line according to an embodiment of the present invention. As shown in Figure 3, the multi-level maintenance decision-making modeling device of the intelligent manufacturing production line in this embodiment includes:

Failure time distribution obtaining module 101: use the historical failure data of component C _ij to obtain the failure time distribution of functional failure and the failure time distribution of degraded failure of component C _ij ;

根据部件C_ij的相互影响的多个退化故障的故障时间分布，构建部件C_ij的基于多退化故障随机依赖性的退化寿命模型

The component life model construction module 102, according to the failure time distribution of all functional failures of the component C _ij , constructs the functional life model of the component C _ij based on the random dependency of the multifunctional failure

According to the failure time distribution of multiple degraded faults of components C _ij interacting with each other, a degradation life model based on the random dependence of multiple degraded faults of components C _ij is constructed

构建设备E_i的基于多部件功能结构依赖性的功能可用度模型A_i(t)，根据部件C_ij的退化寿命模型

构建设备E_i的基于多部件性能结构依赖性的性能可用度模型B_i(t)和性能可用度惩罚模型AB_i(t)，其中，t为时刻；Equipment availability model building block 103, according to the functional life model of component C _ij

Construct the functional availability model A _i (t) of the equipment E _i based on the multi-component functional structure dependence, according to the degradation life model of the component C _ij

Construct the performance availability model B _i (t) and the performance availability penalty model AB _i (t) based on multi-component performance structure dependence of the equipment E _i , where t is the moment;

The production line benefit model building block 104 uses the functional availability model A _i (t) of the equipment E _i to calculate the loss of the production line due to functional failure; uses the performance availability model B _{i (} t) and performance availability of the equipment E i Degree penalty model AB _i (t), calculate the loss caused by the degradation fault of the production line, and construct the production line benefit model based on the economic dependence of multiple equipment according to the loss caused by the functional fault and degradation fault of the production line;

The maintenance decision obtaining module 105 optimizes and solves the production line benefit model according to the constraint conditions required by the production line, and obtains a maintenance decision that maximizes the production line benefit.

For specific examples of the multi-level maintenance decision modeling device for an intelligent manufacturing production line in this embodiment, refer to the above definition of the multi-level maintenance decision modeling method for an intelligent manufacturing production line, and details will not be repeated here. Each module in the above-mentioned multi-level maintenance decision-making modeling device of the intelligent manufacturing production line can be fully or partially realized by software, hardware and combinations thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.

Embodiments of the present invention also provide a computer device, which may be a server, and its internal structure may be as shown in FIG. 4 . The computer device includes a processor, memory and a network interface connected by a system bus. Wherein, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs and databases. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store the operating parameter data of each frame. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps of the multi-level maintenance decision-making modeling method of the intelligent manufacturing production line in this embodiment are realized.

Those skilled in the art can understand that the structure shown in Figure 4 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation to the computer equipment on which the solution of the application is applied. The specific computer equipment can be More or fewer components than shown in the figures may be included, or some components may be combined, or have a different arrangement of components.

The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the multi-level maintenance decision of the intelligent manufacturing production line according to the embodiment of the present invention is realized Steps in the modeling method.

Certain exemplary embodiments of the present invention have been described above only by way of illustration, and it goes without saying that those skilled in the art can use various methods without departing from the spirit and scope of the present invention. The described embodiments are modified. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

Claims (10)

1. A multi-level maintenance decision modeling method for an intelligent manufacturing production line, the intelligent manufacturing production line includes m equipment E _i , i=1,..., m, each equipment E _{i has} n _{i parts} C _ij , j=1,...,n _i , characterized in that the method includes: Obtaining the fault time distribution step: using the historical fault data of the component C _ij to obtain the fault time distribution of the functional fault and the fault time distribution of the degraded fault of the component C _ij ; The component life model construction step, according to the failure time distribution of all functional failures of the component C _ij , constructs the functional life model of the component C _ij based on the random dependence of the multifunctional faults

According to the failure time distribution of multiple degraded faults of components C _ij interacting with each other, a degradation life model based on the random dependence of multiple degraded faults of components C _ij is constructed

Equipment availability model construction steps, according to the functional life model of component C _ij

Construct the functional availability model A _i (t) of the equipment E _i based on the multi-component functional structure dependence, according to the degradation life model of the component C _ij

Construct the performance availability model B _i (t) and the performance availability penalty model AB _i (t) based on multi-component performance structure dependence of the equipment E _i , where t is the moment; The production line benefit model construction step uses the functional availability model A _i (t) of the equipment E _i to calculate the loss of the production line due to functional failure; uses the performance availability model B _i (t) of the equipment E _i and the performance availability The penalty model AB _i (t) calculates the loss of the production line due to the degradation fault, and constructs the production line benefit model based on the economic dependence of multiple equipment according to the loss of the production line due to the functional fault and the degradation fault; The constraint conditions required by the production line are optimized and solved for the production line benefit model, and the maintenance decision that maximizes the production line benefit is obtained. In the construction step of the equipment availability model, the functional life model of the component C _ij

fused as a functional model of device E _i

Calculate the expected functional failure time of the equipment at time t

Get the functional availability of equipment E _i

2. The method as claimed in claim 1, characterized in that, in the equipment availability model building step, the expected degradation failure duration of the calculation component _Cij in Γ(t)

Get the performance failure time of equipment E _i

Thus, the performance availability model of equipment E _i is obtained

Among them, Γ(t) is the expected functional normal time of equipment E _i , Γ(t)=t−α _i (t). 3. The method according to claim 2, characterized in that, in the equipment availability model building step, the performance availability penalty model AB _i (t) of the equipment E _i is constructed as follows:

Among them, ω _ij is the influence factor of the performance of component C _ij on the production quality of equipment E _i . 4. The method according to any one of claims 1-3, wherein, in the construction step of the production line benefit model, according to the function availability model A _i (t) and the performance availability model B _i ( t) Obtain the expected failure time point, and calculate the loss within the time period between each expected failure time point according to the series-parallel connection relationship between each device and the performance availability penalty model AB _i (t). 5. The method according to any one of claims 1-3, characterized in that, in the component life model building step, Monte Carlo sampling is carried out respectively to the time-to-failure distributions of all functional failures, and each distribution is taken The earliest fault composition statistics in the wth sampling sample of X _ij ＝min{x _{ij, 1} ,...,X _{ij, w} ,..., x _{ij, z} }, by fitting this statistic, the functional life model is obtained

Among them, z is the number of Monte Carlo sampling, z>10000. 6. The method according to any one of claims 1-3, wherein, in the component life model building step, a multivariate joint distribution function is constructed for multiple degradation faults that influence each other

As a degenerate lifetime model, where,

is the number of degradation faults of component C _ij , and σ _lk is the covariance matrix, which measures the degree of random dependence of multiple degradation faults. 7. The method according to any one of claims 1-3, characterized in that, in the component life model construction step, based on the random dependence between functional failures and degradation failures, according to the occurrence of functional failures and With a certain probability and a certain proportion, the mean value of the fault distribution time of the degraded fault is reduced while increasing its variance, and the degraded life model is corrected

8. A multi-level maintenance decision modeling device for an intelligent manufacturing production line, the intelligent manufacturing production line includes m equipment E _i , i=1,..., m, and each equipment E _i has n _i components C _ij , j=1,...,n _i , characterized in that the device includes: The failure time distribution obtaining module: use the historical failure data of the component C _ij to obtain the failure time distribution of the functional failure and the failure time distribution of the degradation failure of the component C _ij ; The component life model building block, according to the failure time distribution of all functional failures of the component C _ij , constructs the functional life model of the component C _ij based on the random dependence of the multifunctional failures

According to the failure time distribution of multiple degraded faults of components C _ij interacting with each other, a degradation life model based on the random dependence of multiple degraded faults of components C _ij is constructed

Equipment Availability Model Building Block, Functional Lifetime Model for Components C _ij

Construct the functional availability model A _i (t) of the equipment E _i based on the multi-component functional structure dependence, according to the degradation life model of the component C _ij

Construct the performance availability model B _i (t) and the performance availability penalty model AB _i (t) based on multi-component performance structure dependence of the equipment E _i , where t is the moment; Production line benefit model building block, using the functional availability model A _i (t) of equipment E _i to calculate the loss of the production line due to functional failure; _{using the performance availability model B i (} t) and performance availability of equipment E i Penalty model AB _i (t), calculates the loss of the production line due to degraded faults, and constructs a production line benefit model based on the economic dependence of multiple equipment according to the losses caused by functional faults and degraded faults of the production line; the maintenance decision acquisition module, according to The constraint conditions required by the production line are optimized and solved for the production line benefit model, and the maintenance decision that maximizes the production line benefit is obtained. The equipment availability model building block models the functional lifetime of the component C _ij

fused as a functional model of device E _i

Calculate the expected functional failure time of the equipment at time t

Get the functional availability of equipment E _i

9. A computer device comprising a memory and a processor, the memory stores a computer program, wherein the processor implements the method according to any one of claims 1-7 when executing the computer program step. 10. A computer-readable storage medium, on which a computer program is stored, wherein, when the computer program is executed by a processor, the steps of the method according to any one of claims 1-7 are implemented.

Images