KR20210123729A - Acceleration Life Prediction Method of Inverter Stack in Railway Electric Train - Google Patents

Abstract

The present invention relates to an inverter stack acceleration lifespan prediction method for a railway vehicle for which is derived by multiplying an acceleration lifespan time by an acceleration factor in order to know a lifespan under a current operating condition. Provided is a method for which the acceleration factor is applied to the Arrhenius model, allows the acceleration lifespan test time to be derived through a lifespan ratio model between the lifespan time and B10, and allows the lifespan of the inverter stack for the railway vehicle to be predicted through a relationship between a temperature and the acceleration test time.

Description

Acceleration Life Prediction Method of Inverter Stack in Railway Electric Train

The present invention relates to a method for predicting the accelerated life of an inverter stack for a railway vehicle. In order to know the lifespan under the current operating conditions, it is derived by multiplying the accelerated life time by the acceleration factor. As shown in Fig. 1, the present invention applies the Arrhenius model to the acceleration coefficient and derives the accelerated life test time through the life ratio model between _{the life time and B 10.}

Inverter stack, one of the electronic components for railway vehicles, is a product with complicated electrical and electronic components, so high reliability, durability, and maintainability are required as the basics, but failures still occur. , chemical reaction, dust, vibration, etc. are factors. Recently, electronic devices have been reduced in size and weight, and as a result, the environment for factors such as temperature has become harsher than before. Therefore, it is necessary to develop a life prediction method considering cost and time to analyze and verify the failure mode, failure distribution, and lifespan of a product for the purpose of robust design of the product.

Although the inverter has past failure data, it lacks the ability to systematically analyze and interpret it, so data analysis methods and reliability index discovery are required for product reliability through statistical techniques. Therefore, the present invention relates to a method for predicting and interpreting the life of an inverter stack for a railway vehicle.

By analyzing the failure histories of trains of domestic urban railroad operating organizations for the past three years, it was analyzed that temperature was the most important cause of the failure.

As shown in FIG. 2 , failure histories related to the inverter module for the last 3 years were analyzed for electric vehicles of domestic operating organizations. As shown in FIG. 2 , on average for the last three years from 2015 to 2017, 2 to 3 failures occurred per month, and 4 to 5 failures occurred in May to August. It is judged that the occurrence of more failures in summer shows that there is a correlation between temperature and failure.

The inverter stack during operation is under thermal stress, and it is a known result that the inverter stack is damaged by thermal degradation and its lifespan is reached due to a decrease in dielectric strength. The known Arrhenius model is used. The Arrhenius equation is derived from the kinetics of a chemical reaction expressed as a function of temperature, and approximately expresses the relationship between the lifespan of an insulator and temperature as follows.

where k is the Boltzmann gas constant, E is the activation energy, and T is the absolute temperature.

It is very important to evaluate the lifespan of electrical equipment in a short period of time. In the reliability test, a test that simulates the stress under actual operating conditions is performed, but it usually takes a very long time to reach the service life. Therefore, it is to provide a method of predicting the lifespan of an inverter stack for railroad vehicles through the relationship between temperature and accelerated test time.

The purpose of this study is to predict the lifespan of the inverter stack for railway vehicles through the temperature acceleration coefficient, acceleration test time, and B _{10 life ratio model to which the Arrhenius model is applied.}

It relates to a method for predicting the accelerated lifespan of an inverter stack for railroad vehicles, and in order to know the lifespan under current operating conditions, it is derived by multiplying the accelerated life time by the acceleration factor. It is analyzed as the domestic city railway train failure caused by a temperature most important element of the drive stack of the acceleration factor is applied to the fringing nuiwooseu model to derive the accelerated life test time over the life time and the B ₁₀ life of the non-liver model. In the case of no failure, the lifespan according to the reliability level can be predicted, and even if one or more failures occur, the effect of predicting the lifespan is obtained.

1 is a flowchart of an inverter stack accelerated life prediction method for a railway vehicle;
2 is a failure history of a railway vehicle inverter stack;
3 is a train running speed and power diagram
4 is a train operation speed and power diagram between one station
5 is a prediction diagram of a railway vehicle inverter stack life;

To predict the accelerated life, the actual operation pattern of Busan Line 2 was analyzed and the accelerated life test time was predicted. As shown in FIG. 2 , by analyzing the driving pattern of the vehicle, the driving time of the inverter module when moving between stations was analyzed. In the case of Busan Line 2, it was analyzed to be about 89 seconds.

As shown in FIG. 3 , it was confirmed that the inverter module operates during powering and regeneration in one section. It can be seen that the inverter module continuously operates when there is no other running section during operation.

As a result of the analysis of the train operation, the IGBT operation time of the Busan Line 2 train is 89 seconds on average, and when the daily train operation time is 19 hours from 5:00 to 1:00 a.m., it operates 12 round trips, which is 44,856 seconds, which is 12.46 hours per month, 4,548 hours per year. appeared to do

The number of prototypes should be three depending on the research project conditions, and the test time for them can be calculated. Considering the shape parameter and the reliability level in the Weibull distribution, _{the lifespan ratio of the B 10} lifespan and the reliability level was calculated as shown in Table 1. The formula for the ratio between test time and B ₁₀ life is as follows:

division Reliability level 90% Reliability level 80% Reliability level 60% B10 1.488 1.385 1.237 B20 1.280 1.192 1.065 B40 1.085 1.010 0.902

Table 2 shows the life ratio at a confidence level of 90% according to the shape parameter.

Reliability level 90% shape parameter B=3 B=5 B=10 B10 life ratio 1.939 1.488 1.220 B20 life ratio 1.510 1.280 1.131 B40 life ratio 1.145 1.085 1.042

In other words, if there is no failure by testing 1,488 times the lifetime of _{B 10} to be guaranteed with three prototypes, the _{lifetime of B 10} can be guaranteed with a 90% confidence level.

○ Accelerated life test time to guarantee 90% of reliability, B _{10 life}

The number of prototypes is 3, considering the shape parameter β=5.0

· 10-year lifespan guarantee (reliability level 90%, B ₁₀ lifespan)

- Calculation of switching operation time based on 10 years: Set and calculate as 4,548 hours/year

_{- B 10} lifespan for 10 years warranty:

4,548 hours x 10 years x 1,488 = 67,654 hours/10 years

Accelerated life test is required because 67,654 hours of test are required to guarantee the lifespan of the product for 10 years.

○ Derivation of accelerated test time by temperature stress

Accelerated life test:

Life-stress relationship: Since it is a temperature stress condition, the test time is derived using the Arrhenius model. The acceleration factor AF is calculated as follows.

Using the Arrhenius model, the test time is derived as follows.

where Lu = life at actual operating environment temperature

La = Life at accelerated temperature

E = activation energy (unit: eV), average activation energy 0.94 eV Consideration

k = gas constant (8.6173 X 10 ^-5 eV/K )

Tu = Operating environment temperature

Ta = accelerated temperature

division operating temperature
(℃) test temperature
(℃) acceleration factor
(AF) acceleration factor
(AF) Test period
(Work) One 40 40 One 67,654 2,819 2 40 65 13.2 5,138 214 3 40 70 21.1 3,210 134 4 40 75 33.3 2,033 85 5 40 80 51.9 1,304 54

As shown in Table 3, the accelerated test time can be expected to be 54 days during the 80℃ test. Assuming that there is no failure for a total of 54 days (1,304 hours) as a result of the accelerated life test, the relationship between the failure rate λ, the acceleration factor (AF), and the accelerated test time (Ta) is as follows when the confidence level is 90%. . λ = x ² _(1-CL) (2r+2)/2x(1/(nx AF x Ta)) and life expectancy MTTF = 1/λ = 2 xnx AF x Ta /x ² _(1-CL) (2r +2) becomes If the life expectancy is predicted for the test results assuming no failure, there is no failure for 1,304 hours by performing an accelerated test with 3 units at 80°C, and the acceleration factor is 51.9. x 3 x 51.9 x 1,304 / x ² _(0.1) (2) = 2 x 202,962 / 4.61 = 88,145 hours (about 10.06 years). When the reliability level is 90%, 80%, 70%, and 60%, the life expectancy is predicted as shown in FIG. 5 . Therefore, in the case of 90% with the highest level of confidence, it is possible to predict about 10 years.

If one or more failures occur, life prediction divides the normal sample from the failed sample and substitutes the sum of the failure times (ΣTr) in the MTTF equation to MTTF = 2 x (ΣTr + (nr) x Tz)*AF/x ² _(1-CL) (2r+2) It is predicted by the equation.

110: Inverter stack switching time analysis for railway vehicles
120: Acceleration coefficient calculation
200 : Bx life ratio model applied
201: Reliability level and shape parameter applied life ratio calculation
300 10-Year Guarantee Test Time Calculation
400: Accelerated life time calculation
500: Calculate the current lifespan

Claims (2)

It relates to a method for predicting the accelerated lifespan of an inverter stack for railroad vehicles, and in order to know the lifespan under current operating conditions, it is derived by multiplying the accelerated life time by the acceleration factor. Acceleration coefficient is a method of deriving the accelerated life test time by applying the Arrhenius model and using the _{life ratio model between the life time and B 10}
The method of claim 1, wherein the lifespan according to the reliability level can be predicted in the case of no failure by using the test method, and the lifespan can be predicted even when one or more failures occur.

Images