Equipment maintenance strategy determination method and device, equipment and medium

Abstract

The invention relates to the technical field of prediction and health management, and particularly provides an equipment maintenance strategy determination method and device, equipment and a medium, and the method comprises the steps: determining a plurality of target model parameter intervals of a to-be-solved life prediction model constructed through a degradation model algorithm based on a sample degradation data set of target equipment, and obtaining a to-be-solved life prediction model; determining a basic probability distribution value corresponding to each target model parameter interval; converting a plurality of target model parameter intervals described by evidence variables into target model parameters described by random variables; updating the to-be-solved life prediction model through the target model parameters to obtain a target residual life prediction model associated with the target equipment; optimizing the unit time expected cost model in combination with a residual life value, predicted based on the target residual life prediction model, of the target equipment at the target time to obtain a maintenance strategy of the target equipment at the target time; the reliability of the determined equipment maintenance strategy can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of prediction and health management, and particularly to a method, device, equipment, and medium for determining an equipment maintenance strategy. Background Art

[0002] For equipment with high safety and reliability requirements such as aircraft engines, preventive maintenance can preemptively eliminate potential failures, reduce the risk of equipment operation, and thus reduce significant economic losses caused by sudden equipment failures.

[0003] In related technologies, a proportional accelerated degradation model considering the proportional relationship between the drift coefficient and the diffusion coefficient can be established based on the Wiener process as a remaining useful life (RUL) prediction model to predict the remaining life of equipment, and an equipment maintenance strategy can be determined according to the predicted remaining life of the equipment by the remaining life prediction model of the equipment.

[0004] However, in the process of determining the remaining life prediction model and maintenance strategy of equipment based on the proportional accelerated degradation model, a large amount of sample data is usually required, resulting in the inability to construct a highly reliable remaining life prediction model under the condition of a small sample size based on the solutions provided in related technologies, thereby affecting the reliability of the determined equipment maintenance strategy. Summary of the Invention

[0005] In view of the above problems, the present disclosure is proposed. The present disclosure provides a method, device, equipment, and medium for determining an equipment maintenance strategy, which can improve the reliability of the determined equipment maintenance strategy.

[0006] According to one aspect of the present disclosure, a method for determining an equipment maintenance strategy is provided, including:

[0007] Based on a sample degradation data set of a target device, determine multiple target model parameter intervals of a life prediction model to be solved constructed by a degradation model algorithm, and determine a basic probability assignment value corresponding to each target model parameter interval;

[0008] Based on the multiple target model parameter intervals and the basic probability assignment value corresponding to each target model parameter interval, convert the multiple target model parameter intervals described by evidence variables into target model parameters described by random variables;

[0009] Update the life prediction model to be solved with the target model parameters to obtain a target remaining life prediction model associated with the target device;

[0010] Optimize the expected cost model per unit time by combining the remaining health value of the target device at the target time predicted based on the target remaining life prediction model, to obtain the maintenance strategy of the target device at the target time.

[0011] According to another aspect of the present disclosure, there is provided a device maintenance strategy determination device, including:

[0012] A determination module, configured to determine multiple target model parameter intervals of a life prediction model to be solved constructed based on a degradation model algorithm based on a sample degradation data set of a target device, and determine a basic probability assignment value corresponding to each of the target model parameter intervals;

[0013] A conversion module, configured to convert the multiple target model parameter intervals described by evidence variables into target model parameters described by random variables based on the multiple target model parameter intervals and the basic probability assignment values corresponding to each of the target model parameter intervals;

[0014] An update module, configured to update the life prediction model to be solved through the target model parameters to obtain a target remaining life prediction model associated with the target device;

[0015] An optimization module, configured to optimize the expected cost model per unit time by combining the remaining health value of the target device predicted based on the target remaining life prediction model, to obtain the maintenance strategy of the target device at the target time.

[0016] According to yet another aspect of the present disclosure, there is provided an electronic device, including a memory, a processor, and a computer program stored on the memory, where the processor executes the computer program to implement the above method.

[0017] According to yet another aspect of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented.

[0018] The device maintenance strategy determination method, device, equipment and medium provided by the present disclosure can, when the sample size in the sample degradation dataset is small, introduce the evidence theory to determine multiple target model parameter intervals in the to-be-solved life prediction model constructed based on the degradation model algorithm, and combine the basic probability assignment values corresponding to each of the target model parameter intervals, convert the multiple target model parameter intervals into target model parameters described by random variables as evidence variables, so as to obtain a target remaining life prediction model associated with the target device, perform remaining life prediction on the target device, and determine the maintenance strategy through the remaining life prediction result, which not only improves the accuracy of the model parameters of the determined remaining life prediction model and the accuracy of the determined remaining life prediction result under the condition of a small sample size, but also improves the reliability of the device maintenance strategy determined under the small sample condition.

[0019] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further explanation of the claimed technology. Brief Description of the Drawings

[0020] By describing the embodiments of the present disclosure in more detail in conjunction with the drawings, the above and other objects, features and advantages of the present disclosure will become more obvious. The drawings are used to provide a further understanding of the embodiments of the present disclosure, and constitute a part of the specification, and are used to explain the present disclosure together with the embodiments of the present disclosure, and do not constitute a limitation to the present disclosure. In the drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 is a flowchart illustrating a method for determining a device maintenance strategy according to an embodiment of the present disclosure.

[0022] Figure 2 is a schematic diagram showing the relationship between the maintenance cost and time of a device according to an embodiment of the present disclosure.

[0023] Figure 3 is a schematic diagram of a sample degradation dataset according to an embodiment of the present disclosure.

[0024] Figure 4 is a comparison chart of the predicted return oil flow of a piston pump and the sample return oil flow according to an embodiment of the present disclosure.

[0025] Figure 5 is an error chart of a remaining life prediction result according to an embodiment of the present disclosure.

[0026] Figure 6 is a schematic diagram of the average error of remaining life prediction according to an embodiment of the present disclosure.

[0027] Figure 7The RUL intervals of a piston pump according to an embodiment of the present disclosure, and the legal relationship diagram of the BPA corresponding to each RUL interval.

[0028] Figure 8 The relationship diagram between the probability density function of the target model parameters described by random variables and the RUL according to an embodiment of the present disclosure.

[0029] Figure 9 The relationship diagram between the expected maintenance cost per unit time and the ordering time and the maintenance time respectively according to an embodiment of the present disclosure.

[0030] Figure 10 The schematic diagram of the expected maintenance cost per unit time of a piston pump according to an embodiment of the present disclosure.

[0031] Figure 11 The schematic diagram of the saving rate of a sample piston pump at different observation times according to an embodiment of the present disclosure.

[0032] Figure 12 The block diagram of a device maintenance strategy determination device according to an embodiment of the present disclosure.

[0033] Figure 13 The schematic diagram of a computer program product according to an embodiment of the present disclosure.

[0034] Figure 14 The hardware block diagram of an electronic device according to an embodiment of the present disclosure. Detailed implementation manners

[0035] In order to make the purpose, technical solutions and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure. It should be understood that the present disclosure is not limited by the exemplary embodiments described herein.

[0036] To solve the above problems, an embodiment of the present disclosure provides a device maintenance strategy determination method. The device maintenance strategy determination method can be applied to a terminal device, and the terminal device can be an electronic device such as a computer, a notebook or a server, such as Figure 1 As shown, the device maintenance strategy determination method includes:

[0037] Step S101, based on the sample degradation data set of the target device, determine multiple target model parameter intervals of the life prediction model to be solved constructed by the degradation model algorithm, and determine the basic probability assignment value corresponding to each target model parameter interval;

[0038] Step S102: Based on multiple target model parameter intervals and the basic probability assignment values corresponding to each target model parameter interval, convert the multiple target model parameter intervals described by evidence variables into target model parameters described by random variables.

[0039] Step S103: Update the life prediction model to be solved by using the target model parameters to obtain a target remaining life prediction model associated with the target device.

[0040] Step S104: Combine the remaining life value of the target device at the target time predicted based on the target remaining life prediction model to optimize the expected cost model per unit time, and obtain a maintenance strategy for the target device at the target time.

[0041] In summary, the device maintenance strategy determination method provided by the disclosed example can, when the sample size in the sample degradation dataset is small, introduce the evidence theory to determine multiple target model parameter intervals in the life prediction model to be solved constructed based on the degradation model algorithm, and combine the basic probability assignment values corresponding to each target model parameter interval to convert the multiple target model parameter intervals as evidence variables into target model parameters described by random variables, so as to obtain a target remaining life prediction model associated with the target device, perform remaining life prediction on the target device, and determine the maintenance strategy through the remaining life prediction result. This not only improves the accuracy of the model parameters of the determined remaining life prediction model and the accuracy of the determined remaining life prediction result under the condition of a small sample size, but also improves the reliability of the device maintenance strategy determined under the small sample condition.

[0042] The following Figure 1 elaborates in detail on the specific implementation manners of each step in the

[0043] In step S101, the terminal device determines multiple target model parameter intervals of the life prediction model to be solved constructed by using the degradation model algorithm based on the sample degradation dataset of the target device, and determines the basic probability assignment values corresponding to each target model parameter interval.

[0044] In the embodiments of the present disclosure, the target device is a device for which a device maintenance strategy needs to be formulated. The sample degradation data set includes the degradation parameter values of the target device at multiple time points. Among them, the degradation parameter values are determined based on the functions of the device, and the embodiments of the present disclosure do not limit this. For example, when the target device is a piston pump, the degradation parameter value can be the oil return flow; the degradation model algorithm can be determined based on actual needs, and the embodiments of the present disclosure do not limit this. For example, the degradation model algorithm is the Wiener algorithm, or the inverse Gaussian (IG) algorithm, etc.; the basic probability assignment value is the function value of the Basic Probability Assignment (BPA) function.

[0045] In an alternative embodiment, the degradation model algorithm is the Wiener algorithm. The process by which the terminal device determines multiple target model parameter intervals of the remaining life prediction model to be solved constructed based on the degradation model algorithm based on the sample degradation data set of the target device includes: constructing the remaining life prediction model to be solved based on the Wiener algorithm, and determining the probability density function of the remaining life; then, based on the sample degradation data set and the probability density function, determining the likelihood function of the model parameters to be solved in the remaining life prediction model to be solved, where the model parameters to be solved are the drift coefficient and the diffusion coefficient; further, solving the likelihood function to obtain multiple target model parameter intervals of the remaining life prediction model to be solved. Using the Wiener algorithm that fits the remaining life prediction model more accurately under small sample conditions, constructing the remaining life prediction model to be solved of the target device and multiple target model parameter intervals, so as to improve the accuracy and reliability of the degradation state of the target device fitted by the determined target remaining life prediction model associated with the target device.

[0046] Among them, the remaining life prediction model to be solved is:

[0047] X(t) = X(0) + τΛ(t) + σB(Λ(t)); (Formula 1)

[0048] In Formula 1, X(t) is the degradation value of the target device at time t, X(0) is the initial degradation value, τ is the drift coefficient, σ is the diffusion coefficient, B(·) is the standard Brownian function, Λ(t) is a monotonically increasing function of time, and Λ(t) follows a normal distribution with a mean of τΔΛ(t) and a variance of σ 2 ΔΛ(t). ΔX(t) ~ N(τΔΛ(t), σ 2 ΔΛ(t)), and the increment of Λ(t) is ΔΛ(t) = Λ(t + Δt) - Λ(t).

[0049] Then the probability density function of the remaining life is:

[0050]

[0051] Among them, based on the sample degradation dataset and the probability density function, the likelihood function of the model parameters to be solved in the remaining life prediction model to be solved is as follows:

[0052]

[0053] In Formula 3, X is the degradation parameter value in the sample degradation dataset, T is the time point corresponding to the degradation parameter value in the sample degradation dataset, δ=(τ,σ), τ>0, σ>0, M is the data volume of the sample data associated with each sample device in the degradation dataset, the sample data includes the degradation parameter value and the time point corresponding to the degradation parameter value; N is the number of sample devices in the degradation dataset; it can be understood that m is the m-th sample data among M sample data, and n is the n-th sample device among N sample devices.

[0054] Optionally, the terminal device can determine multiple target model parameter intervals of the remaining life prediction model to be solved based on the EM algorithm or maximum likelihood estimation. Among them, the number of multiple target model parameter intervals can be determined according to actual needs, and this disclosure embodiment does not limit this. The multiple target model parameter intervals determined based on the likelihood function can be defined as:

[0055]

[0056] In Formula 4, λ=(δ d ,δ u are multiple target model parameter intervals, is the multiple minimum parameter vectors in the multiple target model parameter intervals, is the multiple maximum vectors in the multiple target model parameter intervals, m n is the numerical value of the number of multiple target parameter intervals.

[0057] In an optional implementation manner, the process for the terminal device to determine the basic probability assignment value corresponding to each target model parameter interval includes: determining the multiple target model parameter intervals as evidence variables based on the evidence theory and constructing a credibility assignment function for the evidence variables; then, solving the credibility assignment function to obtain the basic probability assignment value corresponding to each target model parameter interval. The multiple target model parameter intervals can be determined as evidence variables based on the evidence theory, and a credibility assignment function for the evidence variables can be constructed to determine the basic probability assignment value of each target model parameter interval, so as to determine the target remaining life prediction model associated with the target device by converting the multiple target model parameters from evidence variables to random variables.

[0058] Among them, the credibility assignment function is as follows:

[0059] m[λ i |T,X] = Me[λ i |T,X]; (Formula 5)

[0060] In Formula 5, m is the credibility assignment function, λ i is the i-th target parameter interval, and Me is:

[0061]

[0062] In Formula 6, φ is the distribution range of the evidence variable, that is, the identification framework of the evidence variable.

[0063] In an alternative implementation, before the terminal device converts the multiple target model parameter intervals described by the evidence variable into target model parameters described by a random variable, it can also: update the basic probability assignment value corresponding to each of the target model parameter intervals based on the Bayesian algorithm to obtain an updated basic probability assignment value corresponding to each of the target model parameter intervals. Updating the basic probability assignment value corresponding to each target model parameter interval in combination with the Bayesian algorithm can further improve the accuracy of the determined basic probability assignment value of each target model parameter interval.

[0064] Among them, the process of updating the basic probability assignment value corresponding to each of the target model parameter intervals based on the Bayesian algorithm to obtain an updated basic probability assignment value corresponding to each of the target model parameter intervals can be implemented based on the basic probability assignment value update formula, where the basic probability assignment value update formula is:

[0065]

[0066] In Formula 7, m(λ i |X po ,T po ) is the prior of the basic probability assignment value corresponding to the target model parameter interval; m(λ i |X pr ,T pr ) is the updated basic probability assignment value corresponding to the target model parameter interval.

[0067] In step S102, the terminal device converts the multiple target model parameter intervals described by the evidence variable into target model parameters described by a random variable based on the multiple target model parameter intervals and the basic probability assignment value corresponding to each of the target model parameter intervals.

[0068] In an embodiment of the present disclosure, the terminal device may convert the model parameters described by evidence variables into model parameters described by random variables based on a homogenization formula.

[0069] In an alternative embodiment, the process by which the terminal device converts the multiple target model parameter intervals described by evidence variables into target model parameters described by random variables based on the multiple target model parameter intervals and the basic probability assignment values corresponding to each of the target model parameter intervals includes: inputting the multiple target model parameter intervals and the basic probability assignment values corresponding to each of the target model parameter intervals into an evidence variable homogenization formula to obtain a probability density function of the target model parameters described by random variables; then, solving the probability density function of the target model parameters to obtain the target model parameters described by random variables. Based on the homogenization formula, the model parameters described by evidence variables can be quickly and efficiently converted into model parameters described by random variables, improving data processing efficiency.

[0070] Wherein, the homogenization formula is:

[0071]

[0072] In Equation (8), f δ (δ′) is the probability density function of the model parameter δ′ described by random variables obtained by homogenizing the model parameter interval δ described by evidence variables, ζ i (·) is an indicator function, m(δ i ) is the basic probability assignment value of the i-th target parameter interval, is the maximum value of the i-th target parameter interval, δ i is the minimum value of the i-th target parameter interval, and K(δ) is the number of focal elements of the model parameter interval δ described by evidence variables.

[0073] Wherein, and,

[0074] It should be noted that in an embodiment of the present disclosure, given the degradation parameter threshold D, the probability density function of the target model parameters described by random variables is:

[0075]

[0076] Then, given the degradation parameter threshold D, the probability density function of the target model parameters described by random variables at any time t is:

[0077]

[0078] In Equation (10), dt is the degradation parameter value at time t.

[0079] In step S103, the terminal device updates the to-be-solved life prediction model with the target model parameters to obtain a target remaining life prediction model associated with the target device.

[0080] In step S104, the terminal device combines the remaining life value of the target device at the target time predicted based on the target remaining life prediction model to optimize the expected cost model per unit time, and obtains the maintenance strategy of the target device at the target time.

[0081] In the embodiments of the present disclosure, the target time is the time when a maintenance strategy needs to be formulated for the target device. Specifically, it can be determined based on actual needs, and the embodiments of the present disclosure do not limit this.

[0082] In an alternative embodiment, the terminal device may further: construct a maintenance cost model of the device regarding the cost of each failure replacement and repair, the cost of each preventive maintenance, and the maintenance cost of the target remaining life value; and construct a spare part cost model of the device regarding the spare part storage cost and spare part shortage cost per unit time, and the spare part shortage time and storage time; and construct an expected total operating life model regarding the current operating time, the expected time and storage time of spare part shortage, and the remaining life value; further, based on the maintenance cost model, the spare part cost model of the device and the expected total operating life model, construct the expected cost model per unit time, where the target remaining life value is the remaining life value of the device at the maintenance time. The expected cost model per unit time of the device can be constructed by combining the information of the maintenance cost model, the spare part cost model of the device and the expected total operating life in three dimensions, improving the reliability of the constructed expected cost model per unit time, and further improving the accuracy of the determined maintenance strategy.

[0083] It should be noted that, in the embodiments of the present disclosure, the expected cost model per unit time may be:

[0084]

[0085] In formula 11, f(t m , t or ) is the expected cost per unit time within the life cycle, is the optimal maintenance time; is the optimal ordering time; t ar is the delivery time; t m is the maintenance time; t or is the ordering time; t f is the system failure time; tb It is the time from order to delivery.

[0086] Among them, according to the order time t or , the delivery time t ar , the system failure time t f and the maintenance time t m Based on the relationship among them, the maintenance cost includes after-sales maintenance cost (t f ≤t ar =t m ) and preventive maintenance cost (t f >t m >t ar ), as Figure 2 shown, Figure 2 shows a schematic diagram of the relationship between the maintenance cost and time of a device provided by an embodiment of the present disclosure. Among them, t0 is the current running time. If the failure time is in the blue area, it means the device fails before the spare part is delivered, and it is regarded as post-failure repair; if the failure time is in the yellow area, it means the device fails after the spare part is delivered, and it is regarded as preventive maintenance.

[0087] Among them, the maintenance cost model is:

[0088]

[0089] In Formula 12, f m (t m ) is the maintenance cost at the maintenance time t m , c r is the cost of replacement repair for each failure; c m is the cost of each preventive maintenance; is the target remaining life value.

[0090] The device spare part cost model is:

[0091]

[0092] In Formula 13, f s (t or ) is the device spare part cost at the order time t or , c q is the spare part shortage cost per unit time; c c is the spare part storage cost per unit time, is the time of spare part shortage, and, is the storage time. Among them, the time of spare part shortage is:

[0093]

[0094] In Formula 14, tar is the delivery time, t f is the failure time.

[0095] The storage time is:[[]]

[0096]

[0097] The expected total operating life model is:[[]]

[0098]

[0099] In Equation (16), t0 is the current operating time.

[0100] The expected cost model per unit time is:[[]]

[0101]

[0102] In an alternative embodiment, the process of the terminal device optimizing the expected cost model per unit time to obtain the maintenance strategy of the target device at the target time may include: based on the genetic optimization algorithm, with the lowest expected cost per unit time as the goal, optimizing the expected cost model per unit time to obtain the target ordering time and target maintenance time of the target device; and determining the target ordering time and target maintenance time of the target device as the maintenance strategy of the target device at the target time. The expected cost model per unit time can be optimized based on the genetic optimization algorithm, with the lowest expected cost per unit time as the goal, to quickly obtain the target ordering time and target maintenance time of the target device.

[0103] Exemplarily, in this disclosure, taking the target device as a piston pump as an example, the advantages of the device maintenance strategy determination scheme provided in this application are verified based on the Wiener algorithm, where Figure 3 shows a schematic diagram of the sample degradation data set of the piston pump, including the sample oil return flow rates of 3 sample piston pumps at multiple time points. The given interval values of the drift coefficient and the diffusion coefficient are shown in Table 1.

[0104] Table 1

[0105]

[0106] And, since the data in the piston pump test data set is a random variable, after converting the random variable into an evidence variable, Λ(t) is:[[]]

[0107] Λ(t) = exp(α p (t / ν p ) - ω p (t / ν p ) -1 ); (Equation (18))

[0108] In Formula 18, α p = 0.00122; ω p = 0.6527; v p = 1.663.

[0109] Based on the above basic conditions, the process of determining the target remaining life prediction model in the embodiments of the present disclosure is executed to obtain the target remaining life prediction models of three sample piston pumps, and the oil return flow rates at the last six time points in the 3 sample piston pumps are predicted. As Figure 4 shown, Figure 4 shows the predicted values of the oil return flow rates at the last six time points of the 3 sample piston pumps predicted by the target remaining life prediction model determined based on the embodiments of the present disclosure, and the comparison chart of the sample oil return flow rates at the last six time points in the 3 sample piston pumps. Among them, the predicted value of the oil return flow rate determined by the target remaining life prediction model determined based on the embodiments of the present disclosure is basically consistent with the sample oil return flow rate at the corresponding time, indicating that the target remaining life prediction model determined in the embodiments of the present disclosure has a high prediction accuracy for degradation data; as shown in Table 2, Table 2 shows the target model parameters associated with the target remaining life prediction model of each sample piston pump determined at different observation times.

[0110] Table 2

[0111]

[0112] Meanwhile, the prediction error of the oil return flow rate of each sample piston pump determined by the solution provided in the embodiments of the present disclosure is compared with the prediction error of the oil return flow rate determined by the method in the related art. The results are as Figure 5 shown, Figure 5 shows the prediction error graphs of the remaining life prediction models constructed by the related art and the present application at different time points. The prediction error of the oil return flow rate determined based on the solution provided in the present disclosure is smaller; and, the comparison of the average prediction errors of the remaining life prediction models constructed by the related art and the present application at different time points is as Figure 6 shown. Obviously, the average error determined based on the method provided in the present disclosure is smaller, that is, the method provided in the present disclosure achieves a higher prediction accuracy of the remaining life value.

[0113] Next, based on the solution provided in the embodiments of the present disclosure, the degradation parameter threshold D is set to 5.1 L / min, and the respective multiple target model parameter intervals of a sample piston pump at multiple observation times (t are 160, 170, 180, 190, 200, 210) are determined. And based on updating each target model parameter interval to the life prediction model to be solved, the multiple RUL intervals determined at multiple observation times, and the BPA corresponding to each RUL interval, as Figure 7As shown Figure 7 shows a schematic diagram of multiple RUL intervals determined for a sample piston pump at multiple observation times, and the BPA corresponding to each RUL interval; wherein, after converting the multiple target model parameter intervals at each observation time into target model parameters described by random variables based on the homogenization formula, the probability density function of the target model parameters is determined, and the relationship between the RUL determined based on the target model parameters described by random variables is as Figure 8 shown. As can be seen from Figure 8 it, compared with the observed degradation, the average errors of the predicted RUL are 6.21%, 7.24%, and 6.53%.

[0114] Furthermore, in order to infer the maintenance decision, it is assumed that the cost of repairing each failure of the device is c r = 500. The cost of preventive maintenance of the system is c m = 100, and the cost of component unavailability per unit time of the device is c q = 10. The component storage cost per unit time of the aero-engine is c c = 0.1. The delivery time of the component order is t ar -t r = 20; then the relationship between the expected maintenance cost per unit time and the ordering time and maintenance time is as Figure 9 shown, where sample m-n is the nth sample oil return flow rate and the corresponding time point in the mth sample oil return pump; obviously, the expected maintenance cost per unit time is affected by the ordering time and maintenance time; specifically, the expected maintenance cost per unit time initially decreases and then increases with the change of the ordering and maintenance time, that is, appropriate ordering time and maintenance time can effectively reduce the maintenance cost. Therefore, it is necessary to optimize the ordering time and maintenance time to reduce the usage cost.

[0115] Among them, under different preventive maintenance intervals and spare part reservation times, the expected unit time maintenance cost of the piston pump determined based on the solution provided by this embodiment of the present disclosure is as Figure 10 shown; obviously, the expected maintenance cost per unit time is significantly affected by the ordering and maintenance time, initially decreasing and then gradually increasing; the optimal maintenance time is closely matched with the failure occurrence time, thus effectively avoiding the cost waste caused by over-maintenance.

[0116] Furthermore, comparing the maintenance strategy determined based on the embodiment of the present disclosure and the maintenance strategy determined in the related art, the results are as Figure 11 shown Figure 11 shows a schematic diagram of the savings rate of the sample piston pump provided by the embodiment of the present disclosure at different observation times. Obviously, the maintenance strategy determined through the embodiment of the present disclosure has a lower cost. Among them, the calculation method of the savings rate is as

[0117]

[0118] In Formula 19, S is the savings rate, cost1 is the expected maintenance cost per unit time determined based on the solution provided by the embodiments of the present disclosure, and cost2 is the expected maintenance cost per unit time determined in the related art.

[0119] An exemplary embodiment of the present disclosure provides a device maintenance strategy determination device, which may be a server or a chip applied to a server. Figure 12 The schematic block diagram of the functional modules of the device maintenance strategy determination device according to the exemplary embodiment of the present disclosure is shown. As Figure 12 shown, the device maintenance strategy determination device 1200 includes:

[0120] A determination module 1201, configured to determine multiple target model parameter intervals of a to-be-solved life prediction model constructed based on a degradation model algorithm based on a sample degradation data set of a target device, and determine a basic probability assignment value corresponding to each of the target model parameter intervals;

[0121] A conversion module 1202, configured to convert the multiple target model parameter intervals described by evidence variables into target model parameters described by random variables based on the multiple target model parameter intervals and the basic probability assignment values corresponding to each of the target model parameter intervals;

[0122] An update module 1203, configured to update the to-be-solved life prediction model through the target model parameters to obtain a target remaining life prediction model associated with the target device;

[0123] An optimization module 1204, configured to optimize the expected cost model per unit time by combining the remaining life value of the target device predicted based on the target remaining life prediction model to obtain a maintenance strategy of the target device at a target time.

[0124] Optionally, the degradation model algorithm is the Wiener algorithm, and the determination module 1201 is configured to:

[0125] Construct the to-be-solved remaining life prediction model based on the Wiener algorithm, and determine the probability density function of the remaining life;

[0126] Determine the likelihood function of the to-be-solved model parameters in the to-be-solved remaining life prediction model based on the sample degradation data set and the probability density function, where the to-be-solved model parameters are the drift coefficient and the diffusion coefficient;

[0127] Solve the likelihood function to obtain multiple target model parameter intervals of the remaining life prediction model to be solved.

[0128] Optionally, the device further includes a processing module 1205, configured to:

[0129] Update the basic probability assignment value corresponding to each target model parameter interval based on the Bayesian algorithm to obtain the updated basic probability assignment value corresponding to each target model parameter interval.

[0130] Optionally, the determining module 1201 is configured to:

[0131] Determine the multiple target model parameter intervals as evidence variables based on the evidence theory, and construct a credibility assignment function of the evidence variables, where the credibility assignment function is:

[0132] m[λ i |T,X]=Me[λ i |T,X];

[0133] where m is the credibility assignment function, λ i is the i-th target parameter interval, X is the degradation parameter value in the sample degradation dataset, and T is the time point corresponding to the degradation parameter value in the sample degradation dataset;

[0134] Solve the credibility assignment function to obtain the basic probability assignment value corresponding to each target model parameter interval.

[0135] Optionally, the conversion module 1202 is configured to:

[0136] Input the multiple target model parameter intervals and the basic probability assignment value corresponding to each target model parameter interval into the evidence variable homogenization formula to obtain the probability density function of the target model parameter described by a random variable, where the homogenization formula is:

[0137]

[0138] where f δ (δ′) is the probability density function of the model parameter δ′ described by a random variable obtained by homogenizing the model parameter interval δ described by the evidence variable, ζ i (·) is the indicator function, m(δ i ) is the basic probability assignment value of the i-th target parameter interval, is the maximum value of the i-th target parameter interval, δ i is the minimum value of the i-th target parameter interval, and K(δ) is the number of focal elements of the model parameter interval δ described by the evidence variable;

[0139] Solve the probability density function of the target model parameters to obtain the target model parameters described by random variables.

[0140] Optionally, the device further includes a modeling module 1206, configured to:

[0141] Construct a maintenance cost model of the device regarding the cost of each failure replacement and repair, the cost of each preventive maintenance, and the maintenance cost of the target remaining life value, where the target remaining life value is the remaining life value of the device at the maintenance time;

[0142] Construct a spare part cost model of the device regarding the spare part storage cost and spare part shortage cost per unit time, and the spare part shortage time and storage time;

[0143] Construct an expected total operating life model regarding the current operating time, the expected time of spare part shortage, and the storage time, and the remaining life value;

[0144] Based on the maintenance cost model, the spare part cost model of the device, and the expected total operating life model, construct the expected cost model per unit time.

[0145] Optionally, the optimization module 1204 is configured to:

[0146] Based on the genetic optimization algorithm, with the goal of minimizing the expected cost per unit time, optimize the expected cost model per unit time to obtain the target ordering time and target maintenance time of the target device;

[0147] Determine the target ordering time and target maintenance time of the target device as the maintenance strategy of the target device at the target time.

[0148] An exemplary embodiment of the present disclosure further provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, it is used to cause the electronic device to execute the method according to the embodiment of the present disclosure.

[0149] An exemplary embodiment of the present disclosure further provides a non-transitory computer-readable storage medium storing a computer program, where when the computer program is executed by a processor of a computer, it is used to cause the computer to execute the method according to the embodiment of the present disclosure.

[0150] As Figure 13As shown, an exemplary embodiment of the present disclosure further provides a computer program product 1300, including a computer program 1301, wherein the computer program, when executed by a processor of a computer, is configured to cause the computer to execute the method according to the embodiments of the present disclosure.

[0151] Referring Figure 14 , a block diagram of an electronic device 1400 that can be used as a terminal device of the present disclosure will now be described. It is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0152] As Figure 14 shown, the electronic device 1400 includes a computing unit 1401, which can execute various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1402 or a computer program loaded from a storage unit 1408 into a random access memory (RAM) 1403. In the RAM 1403, various programs and data required for the operation of the electronic device 1400 can also be stored. The computing unit 1401, the ROM 1402, and the RAM 1403 are connected to each other through a bus 1404. An input / output (I / O) interface 1405 is also connected to the bus 1404.

[0153] Multiple components in the electronic device 1400 are connected to the I / O interface 1405, including: an input unit 1406, an output unit 1407, a storage unit 1408, and a communication unit 1409. The input unit 1406 can be any type of device capable of inputting information into the electronic device 1400. The input unit 1406 can receive input digital or character information and generate key signal inputs related to user settings and / or function controls of the electronic device. The output unit 1407 can be any type of device capable of presenting information and can include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 1408 can include, but is not limited to, magnetic disks and optical disks. The communication unit 1409 allows the electronic device 1400 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks and can include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0154] The computing unit 1401 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1401 executes the various methods and processes described above. For example, in some embodiments, the methods of the exemplary embodiments of the present disclosure can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 1408. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 1400 via the ROM 1402 and / or the communication unit 1409. In some embodiments, the computing unit 1401 can be configured to execute the methods of the exemplary embodiments of the present disclosure by any other suitable means (e.g., by means of firmware).

[0155] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, executed as an independent software package partially on the machine and partially on a remote machine, or executed entirely on a remote machine or server.

[0156] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0157] As used in this disclosure, the terms “machine-readable medium” and “computer-readable medium” refer to any computer program product, apparatus, and / or device (e.g., a disk, optical disk, memory, programmable logic device (PLD)) that can be used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal that can be used to provide machine instructions and / or data to a programmable processor.

[0158] To provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic, speech, or tactile input).

[0159] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of the communication network include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0160] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present disclosure are executed in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a terminal, a user device, or other programmable devices. The computer program or instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program or instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless manner. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center integrating one or more available media. The available medium can be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; it can also be an optical medium, such as a digital video disc (DVD); or it can be a semiconductor medium, such as a solid state drive (SSD).

[0161] Although the present disclosure has been described in conjunction with specific features and their embodiments, it is obvious that various modifications and combinations can be made without departing from the spirit and scope of the present disclosure. Accordingly, this specification and the drawings are merely exemplary illustrations of the present disclosure defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present disclosure. Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalent technologies, the present disclosure is also intended to include these changes and modifications.

Claims

1. A method for determining a device maintenance strategy, characterized in that Including: Based on the sample degradation dataset of the target device, determining multiple target model parameter intervals of the to-be-solved life prediction model constructed by the degradation model algorithm, and determining the basic probability assignment value corresponding to each of the target model parameter intervals; Based on the multiple target model parameter intervals and the basic probability assignment value corresponding to each of the target model parameter intervals, converting the multiple target model parameter intervals described by the evidence variable into the target model parameters described by the random variable; Updating the to-be-solved life prediction model with the target model parameters to obtain the target remaining life prediction model associated with the target device; Combining the remaining life value of the target device at the target time predicted based on the target remaining life prediction model to optimize the expected cost model per unit time to obtain the maintenance strategy of the target device at the target time.

2. The method for determining an equipment maintenance strategy according to claim 1, wherein, The degradation model algorithm is the Wiener algorithm, and the determining of the multiple target model parameter intervals of the to-be-solved remaining life prediction model constructed based on the degradation model algorithm based on the sample degradation dataset of the target device includes: Constructing the to-be-solved remaining life prediction model based on the Wiener algorithm and determining the probability density function of the remaining life; Based on the sample degradation dataset and the probability density function, determining the likelihood function of the to-be-solved model parameters in the to-be-solved remaining life prediction model, where the to-be-solved model parameters are the drift coefficient and the diffusion coefficient; Solving the likelihood function to obtain the multiple target model parameter intervals of the to-be-solved remaining life prediction model.

3. The method for determining an equipment maintenance strategy according to claim 1, wherein, Before converting the multiple target model parameter intervals described by the evidence variable into the target model parameters described by the random variable, the method further includes: Updating the basic probability assignment value corresponding to each of the target model parameter intervals based on the Bayesian algorithm to obtain the updated basic probability assignment value corresponding to each of the target model parameter intervals.

4. The method for determining an equipment maintenance strategy according to claim 1, wherein The determining of the basic probability assignment value corresponding to each of the target model parameter intervals includes: Based on the evidence theory, determining the multiple target model parameter intervals as evidence variables and constructing the credibility assignment function of the evidence variables, and the credibility assignment function is: m[λ i |T,X] = Me[λ i |T,X]; where m is the credibility assignment function, λ i is the i-th target parameter interval, X is the degradation parameter value in the sample degradation dataset, and T is the time point corresponding to the degradation parameter value in the sample degradation dataset; Solving the credibility assignment function to obtain the basic probability assignment value corresponding to each of the target model parameter intervals.

5. The method for determining an equipment maintenance strategy according to claim 1, wherein The converting of the multiple target model parameter intervals described by the evidence variable into the target model parameters described by the random variable based on the multiple target model parameter intervals and the basic probability assignment value corresponding to each of the target model parameter intervals includes: Inputting the multiple target model parameter intervals and the basic probability assignment value corresponding to each of the target model parameter intervals into the evidence variable homogenization formula to obtain the probability density function of the target model parameters described by the random variable, and the homogenization formula is: where f δ (δ′) is the probability density function of the model parameter δ′ described by a random variable, obtained by homogenizing the model parameter interval δ described by the evidence variable, ζ i (·) is the indicator function, m(δ i ) is the basic probability assignment value of the i-th target parameter interval, is the maximum value of the i-th target parameter interval, δ i is the minimum value of the i-th target parameter interval, and K(δ) is the number of focal elements of the model parameter interval δ described by the evidence variable; Solving the probability density function of the target model parameters to obtain the target model parameters described by the random variable.

6. The method for determining an equipment maintenance strategy according to claim 1, wherein The method further includes: Construct a maintenance cost model for the equipment, which includes the cost of replacement and repair for each failure, the cost of each preventive maintenance, and the maintenance cost of the target remaining life value, where the target remaining life value is the remaining life value of the equipment at the maintenance time; Construct an equipment spare part cost model, which includes the spare part storage cost and the spare part shortage cost per unit time, and the equipment spare part cost model considering the spare part shortage time and the storage time; Construct an expected total operating life model, which is based on the current operating time, the expected time of spare part shortage and the storage time, and the expected total operating life considering the remaining life value; Based on the maintenance cost model, the equipment spare part cost model and the expected total operating life model, construct the expected cost model per unit time.

7. The method for determining an equipment maintenance strategy according to any one of claims 1 to 6, characterized in that, Optimize the expected cost model per unit time to obtain the maintenance strategy of the target equipment at the target time, including: Based on the genetic optimization algorithm, with the goal of minimizing the expected cost per unit time, optimize the expected cost model per unit time to obtain the target ordering time and the target maintenance time of the target equipment; Determine the target ordering time and the target maintenance time of the target equipment as the maintenance strategy of the target equipment at the target time.

8. An apparatus for determining a device maintenance strategy, characterized in that Including: A determination module, configured to determine multiple target model parameter intervals of a life prediction model to be solved constructed based on a degradation model algorithm based on a sample degradation data set of a target equipment, and determine a basic probability assignment value corresponding to each target model parameter interval; A conversion module, configured to convert the multiple target model parameter intervals described by evidence variables into target model parameters described by random variables based on the multiple target model parameter intervals and the basic probability assignment values corresponding to each target model parameter interval; An update module, configured to update the life prediction model to be solved through the target model parameters to obtain a target remaining life prediction model associated with the target equipment; An optimization module, configured to optimize the expected cost model per unit time by combining the remaining life value of the target equipment predicted based on the target remaining life prediction model, to obtain the maintenance strategy of the target equipment at the target time.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method according to any one of claims 1 to 7.

Images