A newly developed maintenance method and system for degraded equipment

By building a target optimization model to optimize the preventive replacement time of the equipment, the problem of insufficient consideration of equipment reliability in existing maintenance strategies is solved, the reliability and safety of the equipment is improved, and the maintenance costs are reduced.

CN111428356BActive Publication Date: 2024-11-08ROCKET FORCE UNIV OF ENG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010196262.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-03-19
Publication Date
2024-11-08
Estimated Expiration
2040-03-19

AI Technical Summary

Technical Problem

The existing maintenance strategies based on prediction information lack consideration of equipment reliability when maintaining equipment, resulting in too large or too small detection intervals, making it difficult to ensure the reliability and safety of equipment, and frequent inspections will accelerate equipment degradation and increase costs.

Method used

By constructing a target optimization model with the objective function of the minimum long-term average cost and the maximum long-term average availability, the optimal preventive replacement time of the device is determined, and the detection interval is dynamically adjusted to optimize maintenance strategies to ensure the reliability and safety of the device.

Benefits of technology

Improve the reliability and safety of the equipment, and avoid unnecessary losses by determining the optimal preventive replacement time at each monitoring time point.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111428356B_ABST
    Figure CN111428356B_ABST
Patent Text Reader

Abstract

The present invention relates to a maintenance method and system for newly developed degraded equipment. The method comprises: based on the degradation process of the equipment, obtaining the prediction information of the remaining life of the equipment corresponding to the current monitoring time point, including probability density distribution information, cumulative distribution information and reliability information; based on the prediction information, constructing a target optimization model with the minimum long-term average cost and the maximum long-term average availability as the objective function, and the preventive replacement time being greater than the monitoring time point as the constraint condition; solving the target optimization model to determine the optimal preventive replacement time; judging whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time; if so, determining the optimal preventive replacement time as the monitoring time point of the next monitoring; if not, determining the monitoring time point as the execution replacement time; the equipment performs preventive replacement activities at the execution replacement time. The present invention can improve the reliability and safety of the equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of reliability engineering, and in particular to a maintenance method and system for newly developed degraded equipment. Background Art

[0002] Under the framework of prognostics and health management (PHM) technology, the remaining life of degraded equipment is first predicted, and then maintenance activities are scientifically and reasonably arranged based on the predictive information, such as inspection and maintenance, replacement of key parts, and spare parts ordering, so as to reduce equipment maintenance costs and improve equipment safety and reliability as much as possible. At present, the maintenance strategy based on predictive information makes full use of the degradation data monitored in real time. Through the established degradation model and decision optimization model, dynamic maintenance is achieved as the health status of the equipment changes, which transforms the maintenance activities from the past "passive" to "active", overcoming many drawbacks of traditional maintenance strategies (such as failure maintenance and planned maintenance). Therefore, the CBM strategy based on predictive information has developed rapidly in recent years.

[0003] At present, maintenance strategies based on predictive information mainly focus on making maintenance decisions under the condition of equal interval inspections, and usually only use the lowest cost or expense as a single indicator, which brings two problems:

[0004] 1) For many key engineering equipment, ensuring its reliability is the top priority, and maintenance costs are only one aspect of optimization. However, if the maintenance strategy only focuses on maintenance costs, it will inevitably be difficult to apply the research results to actual projects. For example, for special equipment such as weapons and missile systems, it is more important to ensure the reliability of the weapons and missile systems and ensure that the system has a long enough normal operation time to meet the strategic requirements in actual projects and complete the mission. However, most of the existing maintenance strategies based on predictive information only focus on costs and lack consideration for improving equipment reliability.

[0005] 2) The premise of executing the maintenance strategy is to be able to obtain degradation data that reflects the health status of the equipment. In practice, the detection interval of degradation data is generally periodic. However, there is a contradiction: when the detection interval is too large, the health status of the degraded equipment cannot be well grasped, the best replacement time is delayed, and the risk of equipment failure between two inspections increases, making it difficult to ensure the reliability and safety of the equipment; when the detection interval is too small, although it is easy to grasp the health status of the degraded equipment, frequent status detection will accelerate the degradation of the equipment, reduce the remaining life of the equipment, increase the detection cost, and ultimately cause unnecessary losses. Summary of the invention

[0006] The purpose of the present invention is to provide a maintenance method and system for newly developed degraded equipment to improve the reliability and safety of the equipment.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A newly developed maintenance method for degraded equipment, comprising:

[0009] Based on the degradation process of the equipment, obtaining prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring; the prediction information of the remaining life includes probability density distribution information, cumulative distribution information and reliability information;

[0010] Based on the predicted information of the remaining life, a target optimization model is constructed with the minimum long-term average cost and the maximum long-term average availability as the objective function and the preventive replacement time being greater than the monitoring time point as the constraint condition;

[0011] Solving the target optimization model to determine the optimal preventive replacement time for the current monitoring; the optimal preventive replacement time is the optimal replacement time for the device predicted by the current monitoring;

[0012] Determining whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time;

[0013] When the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, the optimal preventive replacement time is determined as the monitoring time point for the next monitoring, and the number of monitoring times is increased by 1, and the step of returning to the degradation process of the equipment to obtain the prediction information step of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring;

[0014] When the cost of the replacement strategy corresponding to the monitoring time point is not greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, the monitoring time point is determined as the execution replacement time; and the device executes the preventive replacement activity at the execution replacement time.

[0015] Optionally, the acquiring, based on the degradation process of the equipment, prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring specifically includes:

[0016] The degradation process of the device is obtained based on the Wiener process; the degradation process of the device is {X(t), t≥0}, X(t)=X(0)+λt+σ B B(t); where λ is the drift coefficient, σ B is the diffusion coefficient, X(t) is the degradation data at time t, B(t) is the standard Brownian motion describing the random dynamics of the degradation process, X(0) is the initial state of the device;

[0017] Obtain the life of the device; the life of the device is T = inf{t:X(t)≥ω|X(0)<ω}; wherein inf represents the infimum of the set, and ω is the failure threshold;

[0018] Get the remaining life of the device at the monitoring time point of the current monitoring; the remaining life is L k = inf{l k >0:X(l k +t k )≥ω}; where t k is the monitoring time point of the current monitoring, l k is the random variable L k The realized value of

[0019] According to the remaining life of the device at the monitoring time point of the current monitoring, the prediction information of the remaining life of the device is determined; the prediction information of the remaining life of the device is:

[0020]

[0021]

[0022]

[0023] in, is the probability density distribution information; is the cumulative distribution information, is the reliability information; x k Indicates that at t k Degradation data monitored at all times, x k =X(t k );X 1:k Indicates t1 to t k All degradation data at time, X 1:k ={x1,x2,…,x k}.

[0024] Optionally, the target optimization model is:

[0025]

[0026] Where C(T) is the long-term average cost; A(T) is the long-term average availability; C p Cost of performing preventive replacement; C f Cost of performing failure replacement; T p The average length of time required to complete preventive replacement activities; T fis the average length of time required to complete failure replacement activities; T is the preventive replacement time to be optimized; R k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k Reliability information, F k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k The cumulative distribution information, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the device is k (τ|X 1:k ) is the monitoring time point t k The probability density function of the remaining life of the equipment after a period of time τ, τ is the monitoring time point t k The time after which the device is running.

[0027] Optionally, solving the target optimization model to determine the optimal preventive replacement time for the current monitoring specifically includes:

[0028] Determining an availability threshold; the availability threshold being a minimum level of the long-term average availability;

[0029] Based on the availability threshold, an equivalent model of the target optimization model is obtained; the equivalent model is Among them, ζ is the availability threshold;

[0030] Based on the equivalent model, the decision range of the preventive replacement time is determined; the decision range of the preventive replacement time is [T min ,T max ], the decision scope is the decision scope that satisfies the constraint conditions in the hierarchical model;

[0031] Within the decision range, the objective function in the equivalent model is optimized to obtain the optimal preventive replacement time; the optimal preventive replacement time is

[0032] Optionally, the determining whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time further includes:

[0033] Using the formula Determine the cost of the replacement strategy corresponding to the monitoring time point; where t k To monitor the time point, is the cost of the replacement strategy corresponding to the monitoring time point, C m is the cost of each monitoring, k is the number of monitoring times, C p Costs of carrying out preventive replacements;

[0034] Using the formula Determine the cost of the replacement strategy corresponding to the optimal preventive replacement time; wherein, is the cost of the replacement strategy corresponding to the optimal preventive replacement time, C f The cost of performing failure replacement, R k (T * |X 1:k ) is the reliability information of the optimal preventive replacement time, F k (T * |X 1:k ) is the cumulative distribution information of the optimal preventive replacement time, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the equipment is obtained, where τ is the monitoring time point t k The time after which the device is running.

[0035] The present invention also provides a maintenance system for a newly developed degraded device, comprising:

[0036] A module for acquiring prediction information of the remaining life, used to acquire prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring based on the degradation process of the equipment; the prediction information of the remaining life includes probability density distribution information, cumulative distribution information and reliability information;

[0037] A target optimization model building module is used to build a target optimization model based on the prediction information of the remaining life, with the minimum long-term average cost and the maximum long-term average availability as the objective function, and with the preventive replacement time being greater than the monitoring time point as the constraint condition;

[0038] A solution module, used for solving the target optimization model to determine the optimal preventive replacement time of the current monitoring; the optimal preventive replacement time is the optimal time for replacing the device predicted by the current monitoring;

[0039] A judgment module, used to judge whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time;

[0040] A module for determining the monitoring time point for the next monitoring, which is used to determine the optimal preventive replacement time as the monitoring time point for the next monitoring when the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, and increase the number of monitoring times by 1, and return to the step of obtaining prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring based on the degradation process of the equipment;

[0041] The module for determining the execution replacement time is used to determine the monitoring time point as the execution replacement time when the cost of the replacement strategy corresponding to the monitoring time point is not greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time; the device performs preventive replacement activities at the execution replacement time.

[0042] Optionally, the remaining life prediction information acquisition module specifically includes:

[0043] A degradation process acquisition unit, used to acquire the degradation process of the device based on the Wiener process; the degradation process of the device is {X(t), t≥0}, X(t)=X(0)+λt+σ B B(t); where λ is the drift coefficient, σ B is the diffusion coefficient, X(t) is the degradation data at time t, B(t) is the standard Brownian motion describing the random dynamics of the degradation process, X(0) is the initial state of the device;

[0044] A life acquisition unit, used to acquire the life of the device; the life of the device is T = inf{t:X(t)≥ω|X(0)<ω}; wherein inf represents the infimum of the set, and ω is the failure threshold;

[0045] The remaining life acquisition unit is used to obtain the remaining life of the device at the monitoring time point of the current monitoring; the remaining life is L k = inf{l k >0:X(l k +t k )≥ω}; where t k is the monitoring time point of the current monitoring, l k is the random variable L k The realized value of

[0046] A prediction information determination unit is used to determine the prediction information of the remaining life of the device according to the remaining life of the device at the monitoring time point of the current monitoring; the prediction information of the remaining life of the device is:

[0047]

[0048]

[0049]

[0050] in, is the probability density distribution information; is the cumulative distribution information, is the reliability information; x k Indicates that at t k Degradation data monitored at all times, x k =X(t k );X 1:k Indicates t1 to t k All degradation data at time, X 1:k ={x1,x2,…,x k}.

[0051] Optionally, the target optimization model constructed by the target optimization model construction module is:

[0052]

[0053] Where C(T) is the long-term average cost; A(T) is the long-term average availability; C p Cost of performing preventive replacement; C f Cost of performing failure replacement; T p The average length of time required to complete preventive replacement activities; T f is the average length of time required to complete failure replacement activities; T is the preventive replacement time to be optimized; R k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k Reliability information, F k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k The cumulative distribution information, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the device is k (τ|X 1:k ) is the monitoring time point t k The probability density function of the remaining life of the equipment after a period of time τ, τ is the monitoring time point t k The time after which the device is running.

[0054] Optionally, the solution module specifically includes:

[0055] an availability threshold determination unit, configured to determine an availability threshold; the availability threshold being the lowest level of the long-term average availability;

[0056] An equivalent model acquisition unit is used to acquire an equivalent model of the target optimization model based on the availability threshold; the equivalent model is Among them, ζ is the availability threshold;

[0057] A decision range determination unit is used to determine the decision range of the preventive replacement time based on the equivalent model; the decision range of the preventive replacement time is [T min ,T max ], the decision scope is the decision scope that satisfies the constraint conditions in the hierarchical model;

[0058] The optimization unit is used to optimize the objective function in the equivalent model within the decision range to obtain the optimal preventive replacement time; the optimal preventive replacement time is

[0059] Optionally, also include:

[0060] The cost determination module of the replacement strategy corresponding to the monitoring time point is used to determine whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, using the formula Determine the cost of the replacement strategy corresponding to the monitoring time point; where t k To monitor the time point, is the cost of the replacement strategy corresponding to the monitoring time point, C m is the cost of each monitoring, k is the number of monitoring times, C p Costs of carrying out preventive replacements;

[0061] The cost determination module of the replacement strategy corresponding to the optimal preventive replacement time is used to use the formula Determine the cost of the replacement strategy corresponding to the optimal preventive replacement time; wherein, is the cost of the replacement strategy corresponding to the optimal preventive replacement time, C f The cost of performing failure replacement, R k (T * |X 1:k ) is the reliability information of the optimal preventive replacement time, F k (T * |X 1:k ) is the cumulative distribution information of the optimal preventive replacement time, R k (τ|X 1:k ) is the monitoring time point t kAfter a period of time τ, the reliability information of the equipment is obtained, where τ is the monitoring time point t k The time after which the device is running.

[0062] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0063] The present invention makes maintenance prediction based on the reliability and cost of the equipment, determines the corresponding optimal preventive replacement time at each monitoring time point, and then compares the cost of the current replacement strategy with the cost of the replacement strategy corresponding to the optimal preventive replacement time. When the cost of the current replacement strategy is low, preventive replacement is performed at the current time; when the cost of the current replacement strategy is high, a dynamic monitoring interval can be obtained to determine the monitoring time point for the next monitoring, thereby improving the reliability and safety of the equipment during the maintenance process. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0065] Figure 1 It is a flow chart of the maintenance method of the newly developed degraded equipment of the present invention;

[0066] Figure 2 This is a schematic diagram of the structure of the maintenance system for the newly developed degradation equipment of the present invention;

[0067] Figure 3 It is a flow chart of a specific implementation case of the present invention;

[0068] Figure 4 is the degradation data of the gyroscope in the simulation example of the present invention;

[0069] Figure 5 It is a decision result analysis diagram in the simulation example of the present invention;

[0070] Figure 6 This is an example of the maintenance strategy in the simulation example of the present invention. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0073] Figure 1 FIG. 1 is a flow chart of the maintenance method for the newly developed degraded equipment of the present invention. Figure 1 As shown, the maintenance method of the newly developed degraded equipment of the present invention comprises the following steps:

[0074] Step 100: Based on the degradation process of the equipment, obtain the prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring. The prediction information of the remaining life includes probability density distribution information, cumulative distribution information and reliability information. The specific process is as follows:

[0075] According to existing work, the degradation process of the equipment is obtained based on the Wiener process {X(t), t≥0}, X(t)=X(0)+λt+σ B B(t). Where λ is the drift coefficient, σ B is the diffusion coefficient, B(t) is the standard Brownian motion describing the random dynamics of the degradation process, X(t) is the degradation data at time t driven by the standard Brownian motion B(t), X(0) is the initial state of the device. Generally, the initial state X(0)=0.

[0076] On this basis, it is assumed that the degradation data obtained are at discrete time points t1<t2<…<t k Monitored, let x k =X(t k ) indicates that at t k The degradation data monitored at all times can be k All the degradation data at the moment are expressed as a set, namely X 1:k ={x1,x2,…,x k}. According to the concept of first hitting time (FHT), it is believed that once the degradation state {X(t), t≥0} that represents the health level of the degraded equipment reaches the set failure threshold ω for the first time, the life of the degraded equipment ends. Therefore, the life of the degraded equipment T=inf{t:X(t)≥ω|X(0)<ω}; where inf represents the lower bound of the set, that is, the first hitting time; ω is the failure threshold, which is generally determined by industrial standards.

[0077] Get the remaining life of the device at the current monitoring time point. The formula is:

[0078] L k = inf{l k >0:X(l k +tk )≥ω};

[0079] Among them, t k is the monitoring time point of the current monitoring, l k is the random variable L k The realized value of

[0080] According to the remaining life of the device at the monitoring time point of the current monitoring, the corresponding prediction information of the remaining life of the device is determined as follows:

[0081]

[0082]

[0083]

[0084] in, is the probability density distribution information; is the cumulative distribution information, is the reliability information; x k Indicates that at t k Degradation data monitored at all times, x k =X(t k );X 1:k Indicates t1 to t k All degradation data at time, X 1:k ={x1,x2,…,x k}.

[0085] It can be seen that the above formula fully utilizes the monitoring data X of the degraded equipment. 1:k , and can be adaptively updated with new degradation data. Among them, the parameter estimation part can be estimated by the expectation maximization algorithm to achieve adaptive update of the remaining life distribution. Of course, the heuristic algorithm search can also be used. Moreover, the method of the present invention is not only applicable to the prediction information given here, but also to the prediction information obtained from other relevant literature. In other words, as long as the three functions related to the remaining life are obtained, the method can be applied. Therefore, the adaptive maintenance method proposed in the present invention has certain versatility.

[0086] Step 200: Based on the prediction information of the remaining life, a target optimization model is constructed with the minimum long-term average cost and the maximum long-term average availability as the objective function, and the preventive replacement time greater than the monitoring time point as the constraint condition. In order to make the maintenance strategy better applied to actual projects, it is necessary to determine the optimal preventive replacement time T while minimizing the long-term average cost and maximizing the long-term average availability. * Therefore, this will be transformed into a multi-objective optimization problem. Based on this, at the monitoring time point t k, a multi-objective optimization model can be constructed by the following objective function and constraints:

[0087]

[0088] Where C(T) is the long-term average cost; A(T) is the long-term average availability; C p Cost of performing preventive replacement; C f Cost of performing failure replacement; T p The average length of time required to complete preventive replacement activities; T f is the average length of time required to complete failure replacement activities; T is the preventive replacement time to be optimized; R k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k Reliability information, F k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k The cumulative distribution information, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the device is k (τ|X 1:k ) is the monitoring time point t k The probability density function of the remaining life of the equipment after a period of time τ, τ is the monitoring time point t k The time after which the device is running.

[0089] The multi-objective optimization model established in the present invention makes full use of the prediction information and is based on the observed degradation data X 1:k Adaptive updating of C(T) and A(T) can be achieved at each monitoring time point.

[0090] Step 300: Solve the target optimization model to determine the optimal preventive replacement time for the current monitoring. The optimal preventive replacement time is the optimal replacement time for the device predicted by the current monitoring.

[0091] In order to determine the solution corresponding to the multi-objective optimization model, the present invention constructs a decision boundary, the purpose of which is to solve the trade-off between availability and cost, thereby helping managers make more reasonable replacement decisions. Specifically, under the constraint that the long-term average availability A(T) at least meets the minimum level ζ, the long-term average cost C(T) is minimized. Therefore, the multi-objective optimization model can be rewritten as follows:

[0092]

[0093] The above formula is defined as an equivalent model of the multi-objective optimization model, where: ζ is the availability threshold, which is generally set according to actual needs. Obviously, this equivalent model ensures that the maintenance decision made by the manager must minimize the cost under the premise of meeting the availability threshold, so it is in line with actual needs.

[0094] In order to satisfy the constraints in the equivalent model, the decision range of the preventive replacement time T can be obtained, that is, T∈[T min ,T max Based on this, the objective function can be optimized within the decision range, and the optimal preventive replacement time T * It can be calculated by the following formula:

[0095]

[0096] At this point, the optimal preventive replacement time at the current monitoring time point is obtained through the established multi-objective optimization model, which not only ensures the availability of degraded equipment but also minimizes the cost.

[0097] Step 400: Determine whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time. If yes, execute step 500; if not, execute step 600.

[0098] Before judging, first use the formula Determine the cost of the replacement strategy corresponding to the monitoring time point, using the formula Determine the cost of the replacement strategy corresponding to the optimal preventive replacement time. k To monitor the time point, is the cost of the replacement strategy corresponding to the monitoring time point, C m is the cost of each monitoring, k is the number of monitoring times, C p Costs of carrying out preventive replacements; is the cost of the replacement strategy corresponding to the optimal preventive replacement time, C f The cost of performing failure replacement, R k (T * |X 1:k ) is the reliability information of the optimal preventive replacement time, F k (T * |X 1:k ) is the cumulative distribution information of the optimal preventive replacement time, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the equipment is obtained, where τ is the monitoring time point tk The time after which the device is running.

[0099] Then compare the cost of the replacement strategy corresponding to the monitoring time point with the cost of the replacement strategy corresponding to the optimal preventive replacement time. If the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, it means that the cost of the replacement strategy at the current monitoring time point is higher, and the better strategy should be to postpone the preventive replacement activity to the next monitoring time point; if the cost of the replacement strategy corresponding to the monitoring time point is not greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, it means that the cost of the replacement strategy at the current monitoring time point is lower.

[0100] Step 500: Determine the optimal preventive replacement time as the monitoring time point of the next monitoring. The number of monitoring times is increased by 1, and return to step 100. Repeat the above process at each monitoring time point until the replacement execution time is determined.

[0101] Step 600: Determine the monitoring time point as the execution replacement time. The device executes the preventive replacement activity at the execution replacement time.

[0102] Figure 2 This is a schematic diagram of the structure of the maintenance system for the newly developed degradation equipment of the present invention. Figure 2 As shown, the maintenance system of the newly developed degradation equipment of the present invention includes the following structure:

[0103] The remaining life prediction information acquisition module 201 is used to obtain the prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring based on the degradation process of the equipment; the remaining life prediction information includes probability density distribution information, cumulative distribution information and reliability information.

[0104] The target optimization model construction module 202 is used to construct a target optimization model based on the remaining life prediction information, with the minimum long-term average cost and the maximum long-term average availability as the objective function, and the preventive replacement time being greater than the monitoring time point as the constraint condition.

[0105] The solution module 203 is used to solve the target optimization model to determine the optimal preventive replacement time for the current monitoring; the optimal preventive replacement time is the optimal time to replace the device predicted by the current monitoring.

[0106] The judgment module 204 is used to judge whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time.

[0107] The module 205 for determining the time point for next monitoring is used to determine the optimal preventive replacement time as the monitoring time point for next monitoring when the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, and increase the number of monitoring times by 1, and return to the step of obtaining the prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring based on the degradation process of the equipment.

[0108] The execution replacement time determination module 206 is used to determine the monitoring time point as the execution replacement time when the cost of the replacement strategy corresponding to the monitoring time point is not greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time; the device performs a preventive replacement activity at the execution replacement time.

[0109] As another embodiment, the remaining life prediction information acquisition module 201 of the maintenance system for degraded equipment newly developed by the present invention specifically includes:

[0110] A degradation process acquisition unit, used to acquire the degradation process of the device based on the Wiener process; the degradation process of the device is {X(t), t≥0}, X(t)=X(0)+λt+σ B B(t); where λ is the drift coefficient, σ B is the diffusion coefficient, X(t) is the degradation data at time t, B(t) is the standard Brownian motion describing the random dynamics of the degradation process, X(0) is the initial state of the device.

[0111] The life acquisition unit is used to acquire the life of the device; the life of the device is T=inf{t:X(t)≥ω|X(0)<ω}; wherein inf represents the infimum of the set, and ω is the failure threshold.

[0112] The remaining life acquisition unit is used to obtain the remaining life of the device at the monitoring time point of the current monitoring; the remaining life is L k = inf{l k >0:X(l k +t k )≥ω}; where t k is the monitoring time point of the current monitoring, l k is the random variable L k The realized value of .

[0113] A prediction information determination unit is used to determine the prediction information of the remaining life of the device according to the remaining life of the device at the monitoring time point of the current monitoring; the prediction information of the remaining life of the device is:

[0114]

[0115]

[0116]

[0117] in, is the probability density distribution information; is the cumulative distribution information, is the reliability information; x k Indicates that at t k Degradation data monitored at all times, x k =X(t k );X 1:k Indicates t1 to t k All degradation data at time, X 1:k ={x1,x2,…,x k}.

[0118] As another embodiment, the target optimization model constructed by the target optimization model construction module 202 of the maintenance system for the newly developed degraded equipment of the present invention is:

[0119]

[0120] Where C(T) is the long-term average cost; A(T) is the long-term average availability; C p Cost of performing preventive replacement; C f Cost of performing failure replacement; T p The average length of time required to complete preventive replacement activities; T f is the average length of time required to complete failure replacement activities; T is the preventive replacement time to be optimized; R k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k Reliability information, F k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k The cumulative distribution information, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the device is k (τ|X 1:k ) is the monitoring time point t k The probability density function of the remaining life of the equipment after a period of time τ, τ is the monitoring time point t k The time after which the device is running.

[0121] As another embodiment, the solution module 203 of the maintenance system for degraded equipment newly developed by the present invention specifically includes:

[0122] The availability threshold determination unit is used to determine an availability threshold; the availability threshold is the lowest level of the long-term average availability.

[0123] An equivalent model acquisition unit is used to acquire an equivalent model of the target optimization model based on the availability threshold; the equivalent model is Where ζ is the availability threshold.

[0124] A decision range determination unit is used to determine the decision range of the preventive replacement time based on the equivalent model; the decision range of the preventive replacement time is [T min ,T max ], the decision scope is the decision scope that satisfies the constraints in the hierarchical model.

[0125] The optimization unit is used to optimize the objective function in the equivalent model within the decision range to obtain the optimal preventive replacement time; the optimal preventive replacement time is

[0126] As another embodiment, the maintenance system of the newly developed degradation equipment of the present invention further includes:

[0127] The cost determination module of the replacement strategy corresponding to the monitoring time point is used to determine whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, using the formula Determine the cost of the replacement strategy corresponding to the monitoring time point; where t k To monitor the time point, is the cost of the replacement strategy corresponding to the monitoring time point, C m is the cost of each monitoring, k is the number of monitoring times, C p To cover the costs of performing preventive replacements.

[0128] The cost determination module of the replacement strategy corresponding to the optimal preventive replacement time is used to use the formula Determine the cost of the replacement strategy corresponding to the optimal preventive replacement time; wherein, is the cost of the replacement strategy corresponding to the optimal preventive replacement time, C f The cost of performing failure replacement, R k (T * |X 1:k ) is the reliability information of the optimal preventive replacement time, F k (T * |X 1:k) is the cumulative distribution information of the optimal preventive replacement time, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the equipment is obtained, where τ is the monitoring time point t k The time after which the device is running.

[0129] A specific implementation case is provided below to further illustrate the present invention. Figure 1 and Figure 2 The scheme shown.

[0130] Figure 3 Schematic diagram of a specific implementation example of the present invention. Figure 3 As shown, this specific implementation case includes the following steps:

[0131] Firstly, based on the existing remaining life prediction work, the random degradation model is used to obtain the prediction information associated with the remaining life, the probability density function, the cumulative distribution function and the reliability function.

[0132] Then, the optimal preventive replacement time at the current monitoring time point is calculated through a multi-objective optimization model.

[0133] Finally, construct the decision criteria Determine whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time. k+1 is the multi-objective optimization model at the monitoring time point t k The predicted next monitoring time point, i.e. t k+1 At the monitoring time point t k The optimal preventive replacement time is obtained, so there is The decision criterion K C (t k ) indicates the monitoring time point t k The replacement strategy is postponed to the next monitoring time point t k+1 The cost ratio between the replacement strategies. If K C (t k )>1, indicating that the replacement strategy at the current monitoring time point is more expensive, and a better strategy should be to postpone the preventive replacement activity to the next monitoring time point; if K C (t k )≤1, then the preventive replacement activity is performed at the current monitoring time point. The above process is repeated at each predicted monitoring time point until the preventive replacement activity is performed, thereby obtaining the optimal preventive replacement time and dynamic detection interval.

[0134] A simulation example is provided below to verify the effectiveness and superiority of the present invention.

[0135] The simulation example of the present invention uses the actual degradation data of the gyroscope in the inertial navigation system. During the simulation process, according to the technical indicators of the gyroscope, the failure threshold of the drift coefficient is set to ω = 0.37 (° / hour), and the health status of the gyroscope is monitored every 2.5 hours. When the gyroscope runs for 180.5 hours, the drift coefficient reaches the failure threshold for the first time, and the life is considered to have ended.

[0136] Figure 4 is the degradation data of the gyroscope in the simulation example of the present invention, such as Figure 4 As shown in Figure 2, a total of 73 degradation data points were collected during the monitoring period. In addition, the parameters in the multi-objective optimization model were set as C p =5000 yuan, C f = 15000 yuan, C m =100 yuan, T p = 5 hours, T f =200 hours, availability threshold ζ = 0.969.

[0137] Specifically, firstly, taking the current monitoring time point 135 hours as an example, it is assumed that the long-term average availability must be maintained at least above 0.969, that is, the availability threshold ζ=0.969. Figure 5 This is a decision result analysis diagram in the simulation example of the present invention, such as Figure 5 As shown, Figure 5 The decision range under the condition of meeting the availability threshold is given, that is, the preventive replacement time T∈[141,147]. Obviously, through the formula It can be obtained that the optimal preventive replacement time at the current monitoring time point of 135 hours is the 147th hour. Therefore, the trade-off problem between cost and availability is well solved, and the cost is minimized while meeting the availability requirement.

[0138] According to the previous analysis, it can be obtained that the optimal preventive replacement time at the monitoring time point of 135 hours is the 147th hour. However, this replacement time is not the optimal preventive replacement time for the degraded equipment in the entire life cycle, because an optimal preventive replacement time corresponding to it will be obtained at each monitoring time point. Therefore, the present invention proposes an adaptive decision criterion to determine the optimal preventive replacement time and the optimal dynamic detection interval of the degraded equipment in the entire life cycle. For the adaptive maintenance strategy proposed in the present invention, Figure 6 An illustrative example of this type of gyroscope is given, Figure 6 This is an example of the maintenance strategy in the simulation example of the present invention. Figure 6As shown in Figure 1, using the obtained prediction information, the degradation model has been identified when the gyroscope is monitored for 40 hours. Therefore, the first monitoring time point to enter the adaptive maintenance phase is t1 = 40 hours. Then, based on the degradation data monitored from the initial time to t1 and the proposed decision criterion, if K C (t1) < 1, then the preventive replacement activity is performed at the monitoring time point t1, otherwise, the next monitoring time point is scheduled to t2 = T1 * , where T1 * is the optimal preventive replacement time obtained by the multi-objective optimization model at the monitoring time point t1. k , repeat the above process until K C (t k )<1. As can be seen from the figure, a total of 4 monitorings were carried out during the adaptive maintenance phase, and preventive replacement activities were performed at the 4th monitoring time point. Therefore, the optimal preventive replacement time and the optimal status detection interval of the equipment during its entire life cycle can be obtained, and each monitoring time point obtained is determined by a multi-objective optimization model. Therefore, the present invention will achieve the optimization of equipment health management to a certain extent, and can guide and help managers to scientifically and reasonably arrange performance testing and maintenance activities, and has certain application value in engineering practice.

[0139] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0140] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A newly developed maintenance method for degraded equipment, characterized in that: include: Based on the degradation process of the equipment, obtaining prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring; The prediction information of the remaining life includes probability density distribution information, cumulative distribution information and reliability information; Based on the predicted information of the remaining life, a target optimization model is constructed with the minimum long-term average cost and the maximum long-term average availability as the objective function and the preventive replacement time being greater than the monitoring time point as the constraint condition; Solving the target optimization model to determine the optimal preventive replacement time for the current monitoring; The optimal preventive replacement time is the optimal time for replacing the device predicted by the current monitoring; Determining whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time; When the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, the optimal preventive replacement time is determined as the monitoring time point for the next monitoring, and the number of monitoring times is increased by 1, and the step of returning to the degradation process of the equipment to obtain the prediction information step of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring; When the cost of the replacement strategy corresponding to the monitoring time point is not greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, the monitoring time point is determined as the execution replacement time; the device performs the preventive replacement activity at the execution replacement time; The target optimization model is: Where C(T) is the long-term average cost; A(T) is the long-term average availability; C p Cost of performing preventive replacement; C f Cost of performing failure replacement; T p The average length of time required to complete preventive replacement activities; T f is the average length of time required to complete failure replacement activities; T is the preventive replacement time to be optimized; R k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k Reliability information, F k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k The cumulative distribution information, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the device is k (τ|X 1:k ) is the monitoring time point t k The probability density function of the remaining life of the equipment after a period of time τ, τ is the monitoring time point t k The time after which the device is running.

2. The maintenance method of the newly developed degraded equipment according to claim 1 is characterized in that: The obtaining, based on the degradation process of the equipment, prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring specifically includes: The degradation process of the device is obtained based on the Wiener process; the degradation process of the device is {X(t), t≥0}, X(t)=X(0)+λt+σ B B(t); where λ is the drift coefficient, σ B is the diffusion coefficient, X(t) is the degradation data at time t, B(t) is the standard Brownian motion describing the random dynamics of the degradation process, X(0) is the initial state of the device; Obtain the life of the device; the life of the device is T = inf{t:X(t)≥ω|X(0)<ω}; wherein inf represents the infimum of the set, and ω is the failure threshold; Get the remaining life of the device at the monitoring time point of the current monitoring; the remaining life is L k = inf{l k >0:X(l k +t k )≥ω}; where t k is the monitoring time point of the current monitoring, l k is the random variable L k The realized value of According to the remaining life of the device at the monitoring time point of the current monitoring, the prediction information of the remaining life of the device is determined; the prediction information of the remaining life of the device is: in, is the probability density distribution information; is the cumulative distribution information, is the reliability information; x k Indicates that at t k Degradation data monitored at all times, x k =X(t k );X 1:k Indicates t1 to t k All degradation data at time, X 1:k ={x1,x2,...,x k }.

3. The maintenance method of the newly developed degraded equipment according to claim 1 is characterized in that: Solving the target optimization model to determine the optimal preventive replacement time for the current monitoring specifically includes: Determining an availability threshold; the availability threshold being a minimum level of the long-term average availability; Based on the availability threshold, an equivalent model of the target optimization model is obtained; the equivalent model is Among them, ζ is the availability threshold; Based on the equivalent model, the decision range of the preventive replacement time is determined; the decision range of the preventive replacement time is [T min ,T max ], the decision range is the decision range that satisfies the constraint conditions in the equivalent model; Within the decision range, the objective function in the equivalent model is optimized to obtain the optimal preventive replacement time; the optimal preventive replacement time is 4. The maintenance method of the newly developed degraded equipment according to claim 1 is characterized in that: The determining whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time also includes: Using the formula Determine the cost of the replacement strategy corresponding to the monitoring time point; where t k To monitor the time point, is the cost of the replacement strategy corresponding to the monitoring time point, C m is the cost of each monitoring, k is the number of monitoring times, C p Costs of carrying out preventive replacements; Using the formula Determine the cost of the replacement strategy corresponding to the optimal preventive replacement time; wherein, is the cost of the replacement strategy corresponding to the optimal preventive replacement time, C f The cost of performing failure replacement, R k (T * |X 1:k ) is the reliability information of the optimal preventive replacement time, F k (T * |X 1:k ) is the cumulative distribution information of the optimal preventive replacement time, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the equipment is obtained, where τ is the monitoring time point t k The time after which the device is running.

5. A newly developed maintenance system for degraded equipment, characterized in that: include: A module for acquiring prediction information of the remaining life, used to acquire prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring based on the degradation process of the equipment; the prediction information of the remaining life includes probability density distribution information, cumulative distribution information and reliability information; A target optimization model building module is used to build a target optimization model based on the prediction information of the remaining life, with the minimum long-term average cost and the maximum long-term average availability as the objective function, and with the preventive replacement time being greater than the monitoring time point as the constraint condition; A solution module, used for solving the target optimization model to determine the optimal preventive replacement time for the current monitoring; The optimal preventive replacement time is the optimal time for replacing the device predicted by the current monitoring; A judgment module, used to judge whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time; A module for determining the monitoring time point for the next monitoring, which is used to determine the optimal preventive replacement time as the monitoring time point for the next monitoring when the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, and increase the number of monitoring times by 1, and return to the step of obtaining prediction information of the remaining life of the equipment corresponding to the monitoring time point of the current monitoring based on the degradation process of the equipment; a replacement execution time determination module, configured to determine the monitoring time point as the replacement execution time when the cost of the replacement strategy corresponding to the monitoring time point is not greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time; and the device executes the preventive replacement activity at the replacement execution time; The target optimization model constructed by the target optimization model construction module is: Where C(T) is the long-term average cost; A(T) is the long-term average availability; C p Cost of performing preventive replacement; C f Cost of performing failure replacement; T p The average length of time required to complete preventive replacement activities; T f is the average length of time required to complete failure replacement activities; T is the preventive replacement time to be optimized; R k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k Reliability information, F k (Tt k |X 1:k ) is the monitoring time point t k The corresponding time Tt k The cumulative distribution information, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the device is k (τ|X 1:k ) is the monitoring time point t k The probability density function of the remaining life of the equipment after a period of time τ, τ is the monitoring time point t k The time after which the device is running.

6. The maintenance system for the newly developed degraded equipment according to claim 5, characterized in that: The remaining life prediction information acquisition module specifically includes: A degradation process acquisition unit, used to acquire the degradation process of the device based on the Wiener process; the degradation process of the device is {X(t), t≥0}, X(t)=X(0)+λt+σ B B(t); where λ is the drift coefficient, σ B is the diffusion coefficient, X(t) is the degradation data at time t, B(t) is the standard Brownian motion describing the random dynamics of the degradation process, X(0) is the initial state of the device; A life acquisition unit, used to acquire the life of the device; the life of the device is T = inf{t:X(t)≥ω|X(0)<ω}; wherein inf represents the infimum of the set, and ω is the failure threshold; The remaining life acquisition unit is used to obtain the remaining life of the device at the monitoring time point of the current monitoring; the remaining life is L k = inf{l k >0:X(l k +t k )≥ω}; where t k is the monitoring time point of the current monitoring, l k is the random variable L k The realized value of A prediction information determination unit is used to determine the prediction information of the remaining life of the device according to the remaining life of the device at the monitoring time point of the current monitoring; the prediction information of the remaining life of the device is: in, is the probability density distribution information; is the cumulative distribution information, is the reliability information; x k Indicates that at t k Degradation data monitored at all times, x k =X(t k );X 1:k Indicates t1 to t k All degradation data at time, X 1:k ={x1,x2,...,x k }.

7. The maintenance system for the newly developed degraded equipment according to claim 5, characterized in that: The solution module specifically includes: an availability threshold determination unit, configured to determine an availability threshold; the availability threshold being the lowest level of the long-term average availability; An equivalent model acquisition unit is used to acquire an equivalent model of the target optimization model based on the availability threshold; the equivalent model is Among them, ζ is the availability threshold; A decision range determination unit is used to determine the decision range of the preventive replacement time based on the equivalent model; the decision range of the preventive replacement time is [T min ,T max ], the decision range is the decision range that satisfies the constraint conditions in the equivalent model; The optimization unit is used to optimize the objective function in the equivalent model within the decision range to obtain the optimal preventive replacement time; the optimal preventive replacement time is 8. The maintenance system for newly developed degraded equipment according to claim 5, characterized in that: Also includes: The cost determination module of the replacement strategy corresponding to the monitoring time point is used to determine whether the cost of the replacement strategy corresponding to the monitoring time point is greater than the cost of the replacement strategy corresponding to the optimal preventive replacement time, using the formula Determine the cost of the replacement strategy corresponding to the monitoring time point; where t k To monitor the time point, is the cost of the replacement strategy corresponding to the monitoring time point, C m is the cost of each monitoring, k is the number of monitoring times, C p Costs of carrying out preventive replacements; The cost determination module of the replacement strategy corresponding to the optimal preventive replacement time is used to use the formula Determine the cost of the replacement strategy corresponding to the optimal preventive replacement time; wherein, is the cost of the replacement strategy corresponding to the optimal preventive replacement time, C f The cost of performing failure replacement, R k (T * |X 1:k ) is the reliability information of the optimal preventive replacement time, F k (T * |X 1:k ) is the cumulative distribution information of the optimal preventive replacement time, R k (τ|X 1:k ) is the monitoring time point t k After a period of time τ, the reliability information of the equipment is obtained, where τ is the monitoring time point t k The time after which the device is running.