Research on life distribution model of electrical protection cover for energy meter in charging pile based on accelerated test technology

The electric protection cover for the energy meter in the charging pile is an important part to protect the power line terminal and signal line terminal from being damaged by pollution. However, due to the complex and diverse environment in which the charging pile is located, it is easy to cause damage such as damage to electrical protection cover in high temperatures, high humidity, and other harsh environments, resulting in pollution and corrosion of electrical terminals, affecting normal electricity metering. To solve this problem, this paper studies the important factors that affect the life distribution of the electric protective cover used for energy meters in charging piles in the process of use. The accelerated stress model and method are constructed by the comprehensive application of the accelerated test theory, and the accelerated life test is completed. Through experiments, the failure time distribution of typical electrical protection covers for energy meters in a natural service environment is found, the model is hypothesized, tested, and estimated, and the key parameters of life distribution are determined. This paper also provides a good technical idea for the research of aging failure data analysis of electric plastic materials.


Introduction
As a tool for people's daily travel, cars have gradually become popular in the world.With the improvement in the quality of life, the total number of cars in the country has exceeded 250 million.However, traditional fuel vehicles will bring about problems such as intensified greenhouse effect and energy shortage.Electric vehicles have been rapidly promoted and applied at home and abroad due to their advantages of environmental protection, energy conservation, freedom from pollution, and low noise.At present, the electric vehicle charging pile is under rapid construction.The electric energy metering device of the charging pile belongs to the transaction settlement instrument, and its accuracy is related to the fairness of the electric energy trade settlement.As one of the non-electrical components of energy meters, the electrical protection cover is mainly used for the electrical protection of the power line terminal and signal line terminal of smart meters to reduce the influence of water vapor, corrosive atmosphere, and dust in the air on wiring terminals, contact points and insulated areas between wires.The installation state of the electrical protection cover on the energy meter is shown in Figure 1.The electrical protection cover is generally made of PC materials by pressing molding.In theory and engineering practice, there is the possibility of naturally aging them.From the application of energy meters, the most worrying thing is cracking and even falling off, resulting in the risk of fire and casualties.Energy meters are exposed to uncontrolled climatic conditions under the protection of non-sealed metal table boxes for a long time.Such long-term exposure is generally longer than 10 years, and the natural aging risk under such conditions is more prominent.To study the life distribution characteristics of electrical protection covers for energy meters in charging piles in the time range of engineering concern, it is necessary to carry out technical exploration.
One way to obtain the life distribution characteristics of electrical protection covers is to conduct statistical analysis on the cracking and damage data of electrical protection covers for batches of energy meters in use [1].However, this method requires long-term field data as the basis, but the results of onsite fault data collection cannot support the analysis requirements well.For the time being, the real failure data of the electrical protection covers used in energy meters cannot be obtained.The second method is based on the theoretical life data of the material of the electrical protection cover used in energy meters.The reliability prediction method is used to calculate the life data of the electrical protection cover [2].However, as there is no theoretical basic data with good reference value at present, the reliability of the results obtained by this method is insufficient.The third way is to use the test method for research [3].Although the data obtained by the test method is not as real as the actual field data, it is more reliable than the results obtained by the prediction method.In this way, on the one hand, the meter manufacturers can reasonably anticipate the electrical protection covers they use based on the life distribution data.On the other hand, it also enables the State Grid company to better grasp the rhythm of replacing the electrical protection cover with batches of energy meters.

Basic test scheme
The study on the life distribution of the electrical protection cover for energy meters is not an identification test [4] [5], so the reliability index cannot be defined in advance.The use state of the energy meter with the electric protection cover is listed, the working condition is analyzed, and then the accelerated aging test is considered.

Analysis of the working condition of the smart meter
1) The energy meter with the electrical protection cover is used continuously in the field, that is, it is in the working state 24 hours a day, not intermittently.
2) The typical stress conditions and working voltage of the smart meter with the electrical protection cover: ~220 V; frequency: 50Hz; the load is 2000W.

Routine test scheme
In this study, the object of study is the electrical protection cover used for energy meters in charging piles.However, in order to simulate the actual use state as much as possible, a fully assembled energy meter is used as the carrier during the test, and the meter is kept in a working state according to the normal electrical stress condition.Only the non-accelerated conventional scheme is described in this section.
1) The test object is the electrical protection cover for the 2019 new energy meter.The actual implementation uses the energy meter as the carrier.The test samples are 80 energy meters, and the sample numbers are 1#~80#.
2) The electrical protection cover is a normal material, and its size, color, and strength are all in line with the requirements before the test; the meter samples are produced in normal batches, and the function and appearance are normal after the manufacturer's inspection.
3) In the life test mentioned in this paper, the sample of the energy meter is in the working state of the above typical conditions.
4) During the test, a full-function test of all meter samples shall be carried out every three days, focusing on the appearance, buckle function, and size of the electrical protection cover.
5) To obtain a sufficient number of failure numbers to describe the life distribution model, the test truncation method is a fixed number truncation, and at least 50% of the electrical protection cover samples can be guaranteed to fail before the test is finished [4].

Fault definition
If deformation is found in the inspection of the electrical protection cover during or after the test, especially the failure of the buckle function, it is regarded as a fault.

Accelerated test scheme
As the purpose of the test is to explore the life distribution characteristics of the product, referring to GB/T 34986-2017 [6], it is advisable to adopt the stress acceleration method of Class B, that is, to apply additional stress to accelerate the cumulative effect of time based on actual working conditions.
The material characteristics of the electrical protection cover are analyzed, and the main factors affecting the service life of the electrical protection cover used for energy meters involve the power consumption and heating temperature rise of the energy meter itself and the external environmental conditions of the application site [7].Based on this, Class B acceleration can be carried out by applying external environmental stress such as temperature, humidity, and electrical stress to the meter to heat up the electrical protection cover [6] [7] [8].However, the reliability of the acceleration model under various stress acceleration conditions is low.Therefore, only the conventional stress is considered in the influence of the stress and humidity stress that the electrical protection cover of the energy meter heats up.
Referring to MIL-HDBK-338B [9] and GB/T 34986-2017, in this study, only high temperature is used as the accelerating stress, and at the same time, samples of energy meters are allowed to work under normal electrical stress levels, without additional high humidity conditions.
It is needed to inquire high-temperature limit of the electrical protection cover materials used by various manufacturers, refer to GB/T 34986-2017 high-stress value selection principle, and take comprehensive consideration into the test risk.The high temperature is determined as 85℃.
The temperature acceleration stress condition is adopted in the test, which can be referred to as the Arrhenius acceleration model.
The temperature acceleration factor is calculated by the Arrhenius model: Where, AF T --temperature acceleration factor, k --Boltzmann constant (k=8.617×10-5eV/K), a E -activation energy (eV), normal T --the absolute temperature at room temperature, stress T --the absolute temperature at high temperature under test conditions [9].
In this test, the typical value of 25℃ is taken as a reference to the conventional temperature in the product operating condition.
Considering the material type of the electrical protection cover for energy meters, the common failure mechanism and the recommended range of activation energy are given in Table 8.7.1 of MIL-HDBK-338B.In this test scheme, the value of activation energy E is defined as a relatively conservative 0.6eV, which is lower than medium.According to formula (1), AF T is 50.13.

Analysis of Results
In this test, 80 energy meter samples from different manufacturers were connected in series on a singlephase AC line, and a 2000W electric heater was shared as the load, that is, make them in the normal use of the electrical stress level.80 samples were inspected once every three days, and the way was taken out of the test chamber in groups, so as to minimize the influence of the intermediate inspection action on the temperature of the test chamber.The high-temperature test condition was 85℃, and the operation method of the test was carried out according to "GB-T2423.2-2008Test B: High-Temperature Test Method" [10].The fixed number truncation test method was adopted, and the intermediate inspection time was deducted.The actual longest working time in the high-temperature chamber is t=3951 hours.At this time, 42 samples of the electrical protection cover failed, which exceeded 50%, and the test was stopped.
All the faults of the electrical protection cover in the test are buckle fractures, and the fault mode is the same.The test data is valuable for life analysis.
Since the sample failure was found during the intermediate detection every 3 days, that is, the inspection cycle was 72 hours, the time data recorded in the experiment needed to be corrected based on the median rank.Taking the electrical protection cover of the smart meter as the object, the acceleration factor of this acceleration test and the failure time data of each sample are combined.The data are preliminarily sorted out, as shown in Table 1  Based on the above data, it is assumed that the life distribution follows the Weibull model [11]: Where ( ) F t is the failure distribution function; m and  are two parameters to be solved The above equation is converted into a linear equation: The aforementioned data is converted to take time ln(t) as the independent variable and 1 1 as the dependent variable, so that fitting can be carried out in the Weibull linear coordinate system.[12] The aforementioned data is transformed correspondingly and fitted in the Weibull linear coordinate system, and the results are shown in the figure below.

Figure 2. Weibull linear fitting
It can be seen from the above fitting results that: 1) The goodness of fit is 0.9883, indicating that the buckle fracture failure of this batch of electrical protective cover samples follows the distribution of Weibull; 2) The slope of the linear model reaches 5.1727, which is the shape parameter m value of the Weibull distribution, indicating that the batch property of the electrical protection cover enters the period of loss and failure, following the normal distribution.
3) If the intercept is 63.323, the Eigen life η of Weibull distribution is 207268 hours, equivalent to 23.6 years.The average life expectancy is 190,680 hours, equivalent to about 21.8 years.
4) The typical reliable life of a project can be calculated, i.e., 134151 hours with a reliability of 0.9, equivalent to about 15.3 years.
According to the above research, the following aspects can be summarized and analyzed: 1) The typical failure mode of the electrical protection cover for an energy meter in charging piles is buckle fracture.
2) The failure of the buckle of the electrical protection cover can be accelerated by high-temperature stress.
3) In this study, 85℃ was used as the high temperature to accelerate, and buckle fracture failure occurred.The failure mode is consistent with the most common failure mode of the electrical protection cover under the conventional use conditions of energy meters.It shows that the choice of acceleration stress is appropriate.However, it is impossible to infer whether the life distribution parameters are completely consistent because there is no analysis and comparison of the buckle fracture data of the electrical protection cover under normal working conditions.
4) The activation energy of high-temperature acceleration stress is 0.6eV, which is the value of engineering experience.The multi-stress group test was not carried out in this study.From the data analysis results, it is more reasonable.The acceleration model parameters or acceleration factors obtained in any way cannot be regarded as having high reliability.Ln(t) 5) According to the analysis results of the data obtained from the acceleration test, the failure mode of the buckle of the electrical protection cover of the energy meter begins to appear after 124,000 hours, that is, about 14 years later, and then the failure probability increases with the typical Weibull distribution dissipation failure data characteristics, which will appear about 50% in about 20 years.

Conclusion
This paper studies the important factors that affect the life distribution of the electrical protection cover used for energy meters in charging piles and makes an accelerated life test scheme by comprehensive use of accelerated test theory, acceleration model, and algorithm.After implementation, it can explore the failure time distribution of the current typical electrical protection cover used for energy meters under a natural service environment.Then the hypothesis, test, and parameter estimation of the model are carried out, and the key parameters of life distribution are obtained with statistical significance.This paper also provides a good technical idea for the research of aging failure data analysis of electric plastic materials.

Figure 1 .
Figure 1.Installation state of the electrical protection cover

Table 1 .
below.Data collation result