Equipment outfield reliability evaluation method based on factory and repair data

Abstract

The invention discloses an equipment outfield reliability evaluation method based on factory and repair data. The method comprises the following steps: S1, carrying out equipment performance degradation mechanism and degradation characteristic quantity research; s2, cleaning the data, and extracting performance data Y, degradation characteristic quantity X and service time delta T of each piece of equipment; s3, constructing a performance degradation model Y = f (X) according to the performance data Y and the degradation characteristic quantity X; s4, constructing an average performance degradation rate model V = g (X, delta T) according to the degradation characteristic quantity X and the service time delta T; s5, carrying out statistical model parameter estimation and hypothesis testing; s6, determining an equipment performance degradation failure threshold value Yw; s7, solving an extreme value Xw of the equipment degradation characteristic quantity; s8, constructing an equipment reliability evaluation model, and solving the service life of the equipment under the given reliability; and S9, drawing an equipment reliable life graph. The method is high in universality, is not limited by equipment life compliance index distribution any more, and is suitable for various degradation mechanism processes.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment fault and life prediction, and particularly to a method for evaluating the outfield reliability of equipment based on factory and repair data. Background Art

[0002] With the continuous improvement of the complexity and intelligence level of modern equipment systems, reliability evaluation has become a key link in ensuring the safe operation of equipment and optimizing maintenance strategies. The current outfield reliability evaluation of equipment has the following problems:

[0003] 1. The existing technology is based on the equipment life and is only applicable to equipment whose life follows an exponential distribution;

[0004] 2. The existing technology can only obtain the overall life distribution of the equipment and cannot consider the degradation information during the operation of the equipment, resulting in its inability to well reflect the remaining service life distribution of the equipment after working for a period of time;

[0005] 3. The cost required to ensure the accuracy of the existing technology evaluation method is too high;

[0006] 4. The existing technology does not consider the equipment degradation mechanism process and cannot specifically guide the optimization design of the equipment. Summary of the Invention

[0007] In view of the above deficiencies of the existing technology, the technical problem to be solved by this patent application is how to provide a method for evaluating the outfield reliability of equipment based on factory and repair data, which has high universality and is no longer limited to the equipment life following an exponential distribution; can well reflect the remaining service life distribution of the equipment after working for a period of time, and boost the rapid implementation of condition-based maintenance; the verification cost of the model accuracy is controllable; it is applicable to various degradation mechanism processes and can specifically guide the optimization of the equipment design scheme.

[0008] To solve the above technical problems, the present invention adopts the following technical solutions:

[0009] A method for evaluating the outfield reliability of equipment based on factory and repair data, comprising the following steps:

[0010] S1: Conduct research on the equipment performance degradation mechanism and degradation characteristic quantities, and clarify the effective data types to be extracted in the factory and repair data;

[0011] S2: Clean the factory and repair data of multiple equipment obtained in the outfield environment, and extract the performance data Y, degradation characteristic quantity X, and service time ΔT of each equipment, where the factory and repair performance data are respectively denoted as Y0 and Y1, and the factory and repair degradation characteristic quantities are respectively denoted as X0 and X1;

[0012] S3: Construct a performance degradation model Y = f(X) based on the performance data Y and degradation characteristic quantity X extracted in S2;

[0013] S4: Construct an average performance degradation rate model V = g(X, ΔT) based on the degradation characteristic quantity X and service time ΔT extracted in S2;

[0014] S5: Conduct statistical model parameter estimation and hypothesis testing based on the degradation characteristic quantity X0 extracted in S2;

[0015] S6: Determine the performance degradation failure threshold Yw of the equipment;

[0016] S7: Obtain the extreme value Xw of the equipment degradation characteristic quantity according to S3 and S6;

[0017] S8: Construct an equipment reliability evaluation model based on S4, S5, and S7, and calculate the equipment life under a given reliability;

[0018] S9: Draw an equipment reliable life diagram according to S8.

[0019] Among them, step S1 includes the following steps:

[0020] S101: Take the failure samples of the key components of the equipment as the starting point, and study the macroscopic and microscopic morphology characteristics, composition analysis, and physical property inspection of the failure samples with the help of physical and chemical inspection methods;

[0021] S102: Based on the fault tree analysis method, combined with finite element simulation, logical reasoning, and performance test means, analyze the whole process of equipment performance degradation and its corresponding degradation characteristic quantity;

[0022] S103: Clarify the effective data types to be extracted in the factory and repair data.

[0023] Among them, step S2 includes the following steps:

[0024] S201: In the repair data, extract the performance data Y1, degradation quantity characteristic X1, and repair time T1 of each equipment;

[0025] S202: Conduct duplicate checking according to the repair data and clean the duplicate equipment numbers;

[0026] S203: In the factory data, find the equipment information under the corresponding equipment number, and extract the performance data Y0, degradation quantity characteristic X0, and factory time T0 of each equipment;

[0027] S204: According to the extracted factory and repair data of each equipment, calculate the service time ΔT = T1 - T0, performance change quantity ΔY = Y1 - Y0, and degradation characteristic quantity change quantity ΔX = X1 - X0 of each equipment.

[0028] Among them, step S3 includes the following steps:

[0029] S301: Construct a two-column array in which the degradation feature quantity X and the performance data Y correspond one by one;

[0030] S302: Sort any one of the two columns in the two-column array in step S301 in an increasing or decreasing sequence, then generate a scatter plot, and observe the distribution law of the two-column array;

[0031] S303: Use R software to conduct a regression analysis of the performance degradation model, and conduct a significance test on the regression coefficient and the regression equation. Finally, obtain the performance degradation model Y = f(X);

[0032] S304: After completing the calculation of the model, make appropriate corrections to the model according to the background needs of the actual problem.

[0033] Among them, step S4 includes the following steps:

[0034] S401: According to the degradation feature quantity X and the service time ΔT of each equipment extracted in S2, the average performance degradation rate V of each equipment can be obtained according to the formula V = ΔX / ΔT;

[0035] S402: Use the normal distribution model to conduct fitting analysis and parameter estimation on the average degradation rate V of each device calculated in S401;

[0036] S403: Draw a probability plot of the theoretical distribution and the empirical distribution of the data in S402, and observe the fitting effect;

[0037] S404: Judge whether the distribution model obtained in S402 passes the K-S test. If it does not pass, re-correct the distribution form in S402 and repeat steps S402 to S404 until the K-S test is passed to obtain the distribution model of the average degradation rate V of the equipment;

[0038] S405: The distribution model of the average degradation rate V of the equipment is denoted as Among them, μ V and are the mean and variance of the normal distribution that the average degradation rate V of the equipment follows.

[0039] Among them, step S5 includes the following steps:

[0040] S501: Use the normal distribution model to conduct fitting analysis and parameter estimation on the degradation feature quantity X0 extracted according to S2;

[0041] S502: Draw a probability plot of the theoretical distribution and the empirical distribution of the data in S501, and observe the fitting effect;

[0042] S503: Determine whether the distribution model obtained in S501 passes the K-S test. If not, revise the distribution form in S501 and repeat steps S501 to S503 until the K-S test is passed to obtain the distribution model of the degradation feature quantity X0.

[0043] S502: Denote the distribution model of the degradation feature quantity X0 as where, μ X0 and are the mean and variance of the normal distribution that the degradation feature quantity X0 follows.

[0044] Among them, step S6 includes but is not limited to the following steps:

[0045] S601: Determine the equipment performance degradation failure threshold Yw according to the equipment design theory calculation, performance test outline, and acceptance requirements.

[0046] S602: Determine the equipment performance degradation failure threshold Yw according to the general criteria, specifications, and standards.

[0047] S603: Determine the equipment performance degradation failure threshold Yw according to the equipment service health status, similar products, and repair performance retest data.

[0048] Among them, in step S7, to obtain the extreme value Xw of the equipment degradation feature quantity, substitute the equipment performance degradation failure threshold Yw into the performance degradation model Y = f(X), and calculate the result of X when Y = Yw, denoted as the extreme value Xw of the equipment degradation feature quantity.

[0049] Among them, step S8 includes the following steps:

[0050] Construct an equipment reliability evaluation model and calculate the equipment life under a given reliability according to the following formula:

[0051]

[0052] where, T is the equipment life.

[0053] Among them, in step S9, draw the equipment reliable life diagram, and use the R software to complete the drawing of the equipment reliable life diagram by Monte Carlo sampling.

[0054] In summary, the equipment field reliability evaluation method based on factory and repair data has the following beneficial effects:

[0055] 1. High universality, no longer limited to the equipment life following an exponential distribution.

[0056] 2. Can well reflect the remaining service life distribution of the equipment after working for a period of time, and boost the rapid implementation of condition-based maintenance.

[0057] 3. The verification cost of model accuracy is controllable.

[0058] 4. It is applicable to multiple degradation mechanism processes and can specifically guide the optimization of equipment design schemes. Description of the Drawings

[0059] Figure 1 It is a flowchart of a method for evaluating the field reliability of equipment based on factory and repair data according to the present invention.

[0060] Figure 2 It is the life - reliability curve of a certain type of fuel injection pump. Detailed Implementation Manner

[0061] The present invention will be further described in detail below with reference to the drawings. In the description of the present invention, it should be understood that the orientation or positional relationship indicated by orientation words such as "upper, lower" and "top, bottom" is usually based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description. Without contrary description, these orientation words do not indicate and imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and thus cannot be understood as a limitation on the protection scope of the present invention; the orientation words "inner, outer" refer to the inside and outside relative to the contour of each component itself.

[0062] As Figure 1 shown, a method for evaluating the field reliability of equipment based on factory and repair data includes the following steps:

[0063] S1: By studying the performance degradation mechanism and degradation characteristic quantities of a certain type of fuel injection pump, it is clarified that the types of performance data and degradation characteristic quantity data are injection pressure Y (MPa) and plunger couple clearance X (μm) respectively. Taking the failed samples of the fuel injection pump plunger couple as the starting point, physical and chemical inspection and analysis are carried out. Combining the structural design, working principle and environmental load of the fuel injection pump plunger couple, its performance degradation mechanism is finally determined to be fatigue wear.

[0064] S2: The data extracted from the factory and repair data of a certain type of fuel injection pump are shown in Table 1 after cleaning. Since the formats of the factory time and repair time are year / month / day, for the convenience of calculation, Table 1 converts them into the hour (h) format and rounds them according to the formula service time ΔT = repair time T1 - factory time T0.

[0065]

[0066] S3: Construct a two-column array with the degraded characteristic quantity X of the data in Table 1 and the performance data Y in one-to-one correspondence, and sort the two-column array in ascending order of the degraded characteristic quantity X. To fit the equipment performance degradation model with higher quality, filter out the scattered and concentrated data of the degraded characteristic quantity X. The filtered data is shown in Table 2.

[0067]

[0068] Considering that the performance of the fuel injection pump is proportional to the cube of the clearance between the plunger and barrel pair, use the statistical analysis software R to perform a cubic polynomial fit on this set of data. The obtained performance degradation model is Y = A - B*X^3, where Y is the fuel injection pressure (MPa) of the fuel injection pump, X is the clearance between the plunger and barrel pair (um), and A and B are regression coefficients. Both the regression coefficients and the regression equation pass the significance test.

[0069] S4: Calculate the average degradation rate array from the data in Table 1 according to the formula V = (X1 - X0) / ΔT, conduct a normal distribution fit and parameter estimation, observe the fitting effect by plotting a probability plot, and finally obtain the average degradation rate model of this equipment as V ~ N(0.0002206, 0.000074 2 )

[0070] S5: Using the same method steps as in S4, obtain the clearance distribution model of the plunger and barrel pair at the time of factory as X0 ~ N(0.009, 0.0003 2 )

[0071] S6, S7: Determination of the failure threshold: The minimum fuel pressure required by the factory performance test outline is Yw = 90 MPa. According to the performance degradation formula obtained above, the maximum corresponding plunger clearance Xw = 13.9 um can be obtained.

[0072] S8: Calculate the life of a certain type of fuel injection pump at a given reliability according to the formula . The calculation results are shown in Table 3.

[0073] Reliability R(t) Quantile Xp Lifetime T(t) 0.5 0 T_0.5 0.6 0.250 T_0.6 0.7 0.522 T_0.7 0.8 0.841 T_0.8 0.9 1.282 T_0.9 0.95 1.640 T_0.95 0.99 2.326 T_0.99

[0074] S9: The life-reliability curve of a certain type of fuel injection pump drawn through Monte Carlo sampling is shown in Figure 2 as shown.

[0075] The beneficial effects of the present invention are as follows: The embodiment of the present invention proposes a life assessment method based on performance degradation, aiming to extract the performance degradation information of the equipment from the factory and repair data, which can better reflect the remaining service life distribution of the equipment after working for a period of time, and boost the rapid implementation of condition-based maintenance; this technical solution is based on performance degradation and does not require complete monitoring data after the system completely fails. Instead, according to the actual prediction requirements and considering limiting factors such as time and cost, the corresponding performance degradation data obtained by monitoring the system can greatly reduce the verification cost of the model's accuracy; this technical solution has high universality and is applicable to various failure mechanisms, such as wear, erosion, corrosion, cavitation, surface scaling, etc., and can specifically guide the optimization of the equipment design scheme, no longer being limited by the equipment life obeying the exponential distribution.

[0076] Finally, it should be noted that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention also intends to include these changes and modifications.

Claims

1. A method for evaluating equipment reliability in the field based on factory delivery and repair data, characterized in that: The following steps are involved: S1: Conduct research on equipment performance degradation mechanisms and degradation characteristics, and clarify the effective types of data to be extracted from factory and repair data; S2: Clean the factory and repair data of multiple equipment in the field environment, extract the performance data Y, degradation characteristic value X and service time ΔT of each equipment, where the factory and repair performance data are recorded as Y0 and Y1 respectively, and the factory and repair degradation characteristic values ​​are recorded as X0 and X1 respectively; S3: constructing a performance degradation model Y=f(X) based on the performance data Y and degradation feature quantity X extracted in S2; S4: construct an average performance degradation rate model V = g(X, ΔT) based on the degradation feature X extracted in S2 and the service time ΔT; S5: Statistical model parameter estimation and hypothesis testing are performed based on the degradation feature X0 extracted in S2; S6: Determine the equipment performance degradation failure threshold Yw; S7: Obtain the extreme value Xw of the equipment degradation characteristic quantity according to S3 and S6; S8: Construct an equipment reliability assessment model based on S4, S5 and S7 to calculate the equipment life under a given reliability; S9: Draw the equipment reliability life diagram based on S8.

2. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: Step S1 includes the following steps: S101: Taking the failed samples of key equipment components as the starting point, the macroscopic and microscopic morphological characteristics, component analysis, and physical property inspection of the failed samples are studied by physical and chemical testing methods; S102: Based on the fault tree analysis method, combined with finite element simulation, logical reasoning, and performance test methods, analyze the entire process of equipment performance degradation and its corresponding degradation characteristic quantities; S103: Clarify the valid data types to be extracted from the factory and repair data.

3. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: Step S2 includes the following steps: S201: extracting the performance data Y1, degradation feature X1 and repair time T1 of each equipment from the repair data; S202: Check for duplicates based on the repair data and remove duplicate equipment numbers; S203: searching for equipment information under the corresponding equipment number in the factory data, and extracting the performance data Y0, degradation feature X0 and factory time T0 of each equipment; S204: Based on the extracted factory delivery and repair data of each equipment, the service time ΔT=T1-T0, the performance change ΔY=Y1-Y0, and the degradation characteristic change ΔX=X1-X0 of each equipment are calculated.

4. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: Step S3 includes the following steps: S301: constructing two arrays with one-to-one correspondence between degradation feature quantity X and performance data Y; S302: sort any one of the two arrays in S301 in an increasing or decreasing sequence, and then generate a scatter plot to observe the distribution pattern of the two arrays; S303: using R software to carry out regression analysis of the performance degradation model, and performing significance tests on the regression coefficient and the regression equation, and finally obtaining the performance degradation model Y=f(X); S304: After completing the calculation of the model, the model is appropriately modified according to the background requirements of the actual problem.

5. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: Step S4 includes the following steps: S401: According to the degradation characteristic quantity X and the service time ΔT of each equipment extracted in S2, the average performance degradation rate V of each equipment can be obtained according to the formula V=ΔX / ΔT; S402: performing fitting analysis and parameter estimation on the average degradation rate V of each device calculated in S401 using a normal distribution model; S403: Draw a probability graph of the theoretical distribution and empirical distribution of the data in S402 to observe the fitting effect; S404: Determine whether the distribution model obtained in S402 passes the KS test. If not, re-correct the distribution form in S402 and repeat steps S402 to S404 until the KS test is passed to obtain the distribution model of the average degradation rate V of the equipment; S405: The distribution model of the average degradation rate V of the equipment is recorded as Among them, μ V and is the mean and variance of the normal distribution obeyed by the average degradation rate V of the equipment.

6. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: Step S5 includes the following steps: S501: Use a normal distribution model to perform fitting analysis and parameter estimation on the degradation feature quantity X0 extracted according to S2; S502: Draw a probability graph of the theoretical distribution and empirical distribution of the data in S501 to observe the fitting effect; S503: Determine whether the distribution model obtained in S501 passes the KS test. If not, re-correct the distribution form in S501 and repeat steps S501 to S503 until the KS test is passed to obtain the distribution model of the degradation feature quantity X0; S502: The distribution model of the degradation feature quantity X0 is recorded as Among them, μ X0 and is the mean and variance of the normal distribution obeyed by the degenerate feature quantity X0.

7. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: Step S6 includes but is not limited to the following methods: S601: Determine the equipment performance degradation failure threshold Yw based on equipment design theoretical calculations, performance test outline, and acceptance requirements; S602: Determine the equipment performance degradation failure threshold Yw according to general criteria, specifications, and standards; S603: Determine the equipment performance degradation failure threshold Yw based on the equipment service health status, similar products, and repair performance retest data.

8. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1 is characterized in that: In step S7, the extreme value Xw of the equipment degradation characteristic quantity is obtained by substituting the equipment performance degradation failure threshold Yw into the performance degradation model Y=f(X) and calculating the result of X when Y=Yw, which is recorded as the extreme value Xw of the equipment degradation characteristic quantity.

9. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1, characterized in that: Step S8 includes the following steps: Construct an equipment reliability assessment model and calculate the equipment life under a given reliability according to the following formula: Where T is the equipment life.

10. The equipment field reliability assessment method based on factory delivery and repair data according to claim 1, characterized in that: In step S9, a reliable life graph of the equipment is drawn, and the reliable life curve of the equipment is drawn by using R software and Monte Carlo sampling.

Images