Development method of preventive maintenance strategy for electronic equipment considering performance degradation prediction error

Abstract

The present invention discloses a method for developing a preventive maintenance strategy for electronic equipment that takes into account performance degradation prediction errors. The method comprises the following steps: step S1, conducting a performance degradation prediction verification test for the electronic equipment; step S2, determining the performance degradation prediction error interval; step S3, constructing a maintenance cost model; step S4, calculating the maintenance cost under different prediction errors; and step S5, determining the optimal maintenance strategy. The present invention can obtain an effective maintenance strategy applicable to all performance degradation prediction errors and reduce maintenance costs. In a three-phase inverter application case, the maintenance strategy determined by the method of the present invention, compared with the maintenance strategy that does not take into account the performance degradation prediction error, achieves the lowest maintenance cost while avoiding failures.

Description

Technical Field

[0001] The present invention belongs to the field of electronic equipment maintenance, and relates to a method for developing a preventive maintenance strategy for electronic equipment, and in particular to a method for developing a preventive maintenance strategy for electronic equipment that takes into account performance degradation prediction errors. Background Art

[0002] After long-term operation, electronic equipment can experience performance degradation, reducing operational stability and increasing the risk of failure. Predicting performance degradation can be used to develop maintenance strategies to ensure long-term reliable operation. However, due to discrepancies between predicted and actual degradation, maintenance strategies that fail to account for these discrepancies can lead to increased maintenance costs or even prevent long-term stable operation due to equipment failure. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for developing a preventive maintenance strategy for electronic equipment that takes into account performance degradation prediction errors, aiming to solve the inaccurate maintenance strategy for electronic equipment caused by performance degradation prediction errors, thereby reducing maintenance costs and the risk of long-term stable operation of equipment.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A method for developing a preventive maintenance strategy for electronic equipment considering performance degradation prediction errors includes the following steps:

[0006] Step S1: conducting a performance degradation prediction verification test of an electronic device;

[0007] Step S2: determining the expected error range of performance degradation;

[0008] Step S3: constructing a maintenance cost model;

[0009] Step S4: Calculate the maintenance cost under different estimated errors;

[0010] Step S5: Determine the optimal maintenance strategy.

[0011] Compared with the prior art, the present invention has the following advantages:

[0012] 1. The present invention can avoid catastrophic failures of electronic equipment caused by neglecting performance degradation prediction errors in developing maintenance strategies, thereby achieving high reliability and long life operation of electronic equipment.

[0013] 2. By quantifying and analyzing the estimated error and incorporating it into the maintenance decision-making process, this invention can significantly improve the practicality and robustness of maintenance strategies. For a certain range of estimated errors, this invention can determine the maintenance plan with the best overall cost.

[0014] 3. The present invention can obtain an effective maintenance strategy applicable to all performance degradation prediction errors and reduce maintenance costs.

[0015] 4. In the three-phase inverter application case, the maintenance strategy determined by the method of the present invention achieves the lowest maintenance cost while avoiding failures compared with the maintenance strategy that does not consider the performance degradation prediction error. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the process for developing a preventive maintenance strategy for electronic equipment that considers performance degradation prediction errors.

[0017] Figure 2 This is a graph showing the cost results of each maintenance strategy calculated when the estimated error is 0.5 in the embodiment.

[0018] Figure 3 Graph showing the union results of effective maintenance strategies for different predicted errors in the embodiment.

[0019] Figure 4 This is a cost result diagram of each maintenance strategy without considering the estimated error in the embodiment. DETAILED DESCRIPTION

[0020] The technical solution of the present invention is further described below with reference to the accompanying drawings, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention that does not depart from the spirit and scope of the technical solution of the present invention should be included in the scope of protection of the present invention.

[0021] The present invention provides a method for developing a preventive maintenance strategy for electronic equipment taking into account the predicted error of performance degradation, such as Figure 1 As shown, the method includes the following steps:

[0022] Step S1: Conducting electronic equipment performance degradation prediction verification test:

[0023] The total duration of the expected verification test of electronic equipment performance degradation is t a , a total of n test samples, each sample parameter P is measured at different times t i (t), where i is the i-th sample, 1≤i≤n, n≥5, 0≤t≤t a .

[0024] Step S2: Determine the expected error range of performance degradation:

[0025] Step S21: Calculate the degradation increment D of each sample at time t e,i (t):

[0026] D e,i (t)=|P i (t)-P i,0

[0027] Among them, P i,0 is the initial value of the parameter before the start of the i-th sample test.

[0028] Step S22: Use the Bootstrap method to calculate the degradation increment D under the significance level α at time t of the experiment. eL (t) is the lower bound, the degenerate increment D eU (t) is the upper bound of the degenerate increment mean interval [D eL (t),D eU (t)].

[0029] Step S23: Calculate the expected performance degradation error e interval:

[0030]

[0031] Among them, D p (t) is the expected mean incremental degradation value of electronic equipment parameters at time t.

[0032] Step S3: Build a maintenance cost model:

[0033] Step S31: Determine the cost C for each inspection i .

[0034] Step S32: Determine the cost C for each maintenance m .

[0035] Step S33: Set maintenance strategy: The electronic equipment is regularly inspected at the inspection interval Δt. When the increment of equipment parameter degradation is greater than or equal to the maintenance threshold D m Maintenance is carried out during the entire operating time of the electronic equipment w To ensure that the parameter degradation increment does not exceed the failure threshold D f .

[0036] Step S34: Establish a maintenance cost model for the total cost:

[0037] C=N i C i +N m C m

[0038] Among them, N i The total running time t w Number of checks within, N m The number of maintenance times.

[0039] Step S4: Calculate the maintenance cost under different estimated errors:

[0040] Step S41: Setting the maintenance threshold value range D m =[DmL ,D mU ], and take A values evenly within this range.

[0041] Step S42: Set the inspection interval Δt to a value range of Δt=[Δt L ,Δt U ], and take B values evenly within this range.

[0042] Step S42: uniformly select C values within the interval of the performance degradation expected error e determined in step S23.

[0043] Step S43: Maintain the threshold D m , the inspection interval Δt, and the performance degradation prediction error e are combined to obtain A×B×C situations. For each situation, steps S44 to S410 are performed.

[0044] Step S44: Setting the failure threshold D f , Maintenance times N m =0, parameters used for counting m=1, h=0.

[0045] Step S45: Calculate the total number of inspections:

[0046]

[0047] Step S46: Calculate the degradation increment of the electronic device during the mth inspection as (1+e)D p [(mh)Δt].

[0048] Step S47: Determine whether the degradation increment of the electronic device exceeds the failure threshold D. f If it exceeds, the calculation stops, and the maintenance threshold D m , the inspection interval Δt, and the estimated error e value combination are invalid. Otherwise, it is a valid combination, and the process goes to step S48.

[0049] Step S48: Determine whether the degradation increment of the electronic device exceeds the maintenance threshold D. m If exceeded, perform maintenance. m =N m +1, h=m. Otherwise, no maintenance is performed.

[0050] Step S49: Update the count m=m+1.

[0051] Step S410: Determine whether m is greater than the total number of inspections N. i If it is greater, the cost is calculated according to the cost model of step S34. Otherwise, steps S46 to S410 are repeated.

[0052] Step S5: Determine the optimal maintenance strategy:

[0053] Step S51: Determine the effective maintenance threshold D under C different performance degradation prediction errors e m , check the union of interval Δt combinations.

[0054] Step S52: Calculate the same maintenance threshold D under C different performance degradation prediction errors e m , the sum of costs calculated at the inspection interval Δt.

[0055] Step S53: Maintenance threshold D with minimum total cost m , the inspection interval Δt is the optimal maintenance strategy.

[0056] Example:

[0057] This embodiment develops a preventive maintenance strategy for electronic equipment taking into account performance degradation prediction errors, taking a three-phase inverter as the object.

[0058] S1. Conduct expected verification tests on electronic equipment performance degradation.

[0059] The three-phase inverter performance degradation prediction verification test is carried out at 80℃ for a total time of t a = 4000 hours, with n = 10 test samples in total, and the effective value of the phase A current of each sample is measured at different times t at intervals of about 166 hours.

[0060] Step S2: Determine the expected error range of performance degradation.

[0061] Step S21: Calculate the degradation increment D of each sample at time t e,i (t).

[0062] Step S22: Use the Bootstrap method to calculate the degradation increment D under the significance level α at time t of the experiment. eL (t) is the lower bound, the degenerate increment D eU (t) is the upper bound of the degenerate increment mean interval [D eL (t),D eU Taking 4000 hours as an example, the degradation increment mean interval obtained by the Bootstrap method at the significance level α = 0.05 is [0.0995, 0.1383].

[0063] Step S23: Calculate the range of the expected error e. Taking 4000 hours as an example, the expected mean value is 0.0635. Based on the mean range of the actual degradation increments in the test, the error range can be calculated as [0.565, 1.175]. Calculate the expected error range at different times, taking the minimum lower bound and the maximum upper bound to obtain the range that encompasses the expected error at all test times. Here, [0.5, 1.2] is used.

[0064] Step S3: Construct a maintenance cost model.

[0065] Step S31: Determine the cost C for each inspection i =20.

[0066] Step S32: Determine the cost C for each maintenance m =38.

[0067] Step S33: Set the maintenance strategy as follows: the electronic equipment is regularly inspected at an inspection interval Δt. When the increment of equipment parameter degradation is greater than or equal to the maintenance threshold D m Maintenance is carried out during the entire operating time of the electronic equipment w = Within 10 years, ensure that the parameter degradation increment does not exceed the failure threshold D f =0.3.

[0068] Step S34: Establish a maintenance cost model as the total cost.

[0069] Step S4: Calculate the maintenance cost under different estimated errors.

[0070] Step S41: Setting the maintenance threshold value range D m = [0.11, 0.3], and A = 20 values are uniformly taken within this range (with an interval of 0.01).

[0071] Step S42: Set the inspection interval Δt value range Δt=[1,700], and evenly take B=700 values within the range (with 1 day as the interval).

[0072] Step S42, in step S23, the estimated error e=[0.5, 1.2] is determined to be uniformly taken as C=8 values (with an interval of 0.1).

[0073] Step S43: Maintain the threshold D m , the inspection interval Δt, and the expected error e are combined to obtain 20×700×8 situations. For each situation, steps S44 to S410 are performed.

[0074] Step S44: Setting the failure threshold D f =0.3, maintenance times N m =0, parameters used for counting m=1, h=0.

[0075] Step S45: Calculate the total number of inspections. Taking Δt = 2 days as an example, the total number of inspections is rounded down to 182 times by rounding down 365 ÷ 2.

[0076] Step S46: Calculate the degradation increment of the electronic device during the mth inspection as (1+e)D p [(mh)Δt].

[0077] Step S47: Determine whether the degradation increment of the electronic device exceeds the failure threshold D. f If it exceeds, the calculation stops, and the maintenance threshold D m , the check interval Δt, and the estimated error e value combination are invalid. Otherwise, it is a valid combination and proceed to the next step.

[0078] Step S48: Determine whether the degradation increment of the electronic device exceeds the maintenance threshold D. m If exceeded, perform maintenance. m =N m +1, h=m. Otherwise, no maintenance is performed.

[0079] Step S49: Update the count m=m+1.

[0080] Step S410: Determine whether m is greater than the total number of inspections N. i If it is greater, the cost is calculated according to the cost model of step S34. Otherwise, steps S46 to S410 are repeated.

[0081] The cost result obtained with e=0.5 is as follows Figure 2 As shown, the white area is the invalid maintenance measurement, which will cause the degradation increment during operation to exceed the failure threshold D f =0.3.

[0082] Step S5: Determine the optimal maintenance strategy.

[0083] Step S51: Determine the effective maintenance threshold D under C different estimated errors e m , check the union of interval Δt combinations.

[0084] Step S52: Calculate the same maintenance threshold D under different estimated errors e m , the sum of costs calculated at the inspection interval Δt.

[0085] Step S53: Maintenance threshold D with minimum total cost m , the inspection interval Δt is the optimal maintenance strategy.

[0086] like Figure 3 As shown, taking the union of the results obtained with e = 0.5 and e = 1.2 as an example, the minimum cost is Δt = 513, D m =[0.11,0.13]. That is, according to the inspection interval Δt=513, the maintenance threshold D m =[0.11,0.13] for maintenance can get the minimum maintenance cost for any case where the expected error belongs to [0.5,1.2].

[0087] Without considering the estimated error, that is, e = 0, the cost calculation result is as follows Figure 4 As shown. The minimum cost is D m =0.11, Δt=[609,716]. This result is not included in the cost result union considering the expected error, which means that maintenance with this strategy will result in the failure to meet the 10-year working requirement. During operation, the degradation increment is greater than 0.3, and the three inverters fail.

[0088] It can be seen that the maintenance strategy that considers the performance degradation prediction error avoids electronic equipment failures, improves equipment operation stability and reduces maintenance costs.

Claims

1. A method for developing a preventive maintenance strategy for electronic equipment considering performance degradation prediction errors, characterized in that The method comprises the following steps: Step S1: Conducting a performance degradation prediction verification test for electronic equipment; Step S2: Determine the expected error range of performance degradation. The specific steps are as follows: Step S21: Calculation test t The degradation increment of each sample at time D e,i ( t ): in, P i, 0 For the i The initial parameter value before the start of the sample test, P i ( t ) is the time t Next i Sample parameters; Step S22: Calculate the test using the Bootstrap method t Momentary significance level α Next, with degenerate increments D eL ( t ) is the lower bound, degenerate increment D eU ( t ) is the upper bound of the degenerate increment mean interval [ D eL ( t ), D eU ( t )]; Step S23: Calculate the expected performance degradation error e Range: in, D p ( t )for t The expected mean value of the incremental degradation of electronic equipment parameters at the time instant; Step S3: Build a maintenance cost model. The specific steps are as follows: Step S31: Determine the cost of each inspection C i ; Step S32: Determine the cost of each maintenance C m ; Step S33: Set maintenance strategy: electronic equipment according to the inspection interval Δ t Perform regular inspections and check if the equipment parameter degradation increment is greater than or equal to the maintenance threshold. D m Maintenance is performed during the entire operating time of the electronic equipment t w To ensure that the parameter degradation increment does not exceed the failure threshold D f ; Step S34: Establish a maintenance cost model for the total cost: in, N i For the entire running time t w The number of checks within N m The number of maintenance times; Step S4: Calculate the maintenance cost under different estimated errors. The specific steps are as follows: Step S41: Setting the maintenance threshold value range D m = [ D mL , D mU ], take uniform A values; Step S42: Set the inspection interval Δ t Value range Δ t =[Δ t L , Δ t U ], take uniform B values; Step S42: the performance degradation prediction error determined in step S23 e Uniformly take the C values; Step S43: Maintain the threshold D m , Inspection interval Δ t , performance degradation prediction error e The values are combined to get A × B × C For each case, steps S44 to S410 are performed; Step S44: Setting the failure threshold D f , maintenance times N m =0, parameter used for counting m =1, h =0; Step S45: Calculate the total number of inspections: Step S46, calculate the m The increment of electronic equipment degradation at the time of the second inspection is (1+ e ) D p [( m - h )Δ t ]; Step S47: Determine whether the degradation increment of the electronic device exceeds the failure threshold. D f If it exceeds, the calculation stops. This maintenance threshold D m , Inspection interval Δ t , expected error e The value combination is invalid, otherwise, it is a valid combination and proceeds to step S48; Step S48: Determine whether the degradation increment of the electronic device exceeds the maintenance threshold. D m , if exceeded, maintenance is performed. N m = N m +1, count h = m , otherwise, no maintenance is performed; Step S49: Update count m = m +1; Step S410: Determine m Is it greater than the total number of checks? N i If it is greater than, calculate the cost according to the cost model of step S34, otherwise, repeat steps S46 to S410; Step S5: Determine the optimal maintenance strategy.

2. The method for developing a preventive maintenance strategy for electronic equipment considering performance degradation prediction errors according to claim 1, characterized in that In step S1, the total duration of the electronic equipment performance degradation verification test is expected to be t a ,common n experimental samples at different times t Measure the parameters of each sample P i ( t ), in, i For the i samples, 1 ≤ i ≤ n , n ≥ 5, 0≤ t ≤ t a .

3. The method for developing a preventive maintenance strategy for electronic equipment considering performance degradation prediction errors according to claim 1, characterized in that The specific steps of step S5 are as follows: Step S51: Determine C Different performance degradation prediction errors e Lower effective maintenance threshold D m , Inspection interval Δ t union of combinations; Step S52: Calculate C Different performance degradation prediction errors e Same maintenance threshold D m , Inspection interval Δ t The calculated sum of costs; Step S53: Maintenance threshold with minimum total cost D m , Inspection interval Δ t This is the optimal maintenance strategy.

Images