Reliability Analysis Of Auto Parts Based On Bayesian Method

With the improving of the environmental protection requirements around the world, the development of electric vehicles is the main development direction of auto industry. And the key technology problem of the electric car is the reliability of the parts especially the reliability of the battery. Therefore, the research of this problem is of great significance for the development of electric vehicles.In this paper, using the Bayes statistics method for the reliability of the battery level for scientific analysis. Reliability analysis and life analysis of the parts need to focus on the failure time and dispersion modeling. Collecting the car battery capacity degradation data and establishing the degradation model.The electric capacity of 60 Ah battery charge and discharge cycle test, and record each cycle of current and voltage test results. Assume that when the actual capacity degradation to 70% of the rated capacity, battery as a failure, the failure of the battery monomer threshold for 42 Ah. The value range from 0.02 to 0.07 of the battery Re and Rct value range for 0.06 to 0.09. To study relationship between car battery impedance and capacity, it can be seen that the battery impedance and capacity of corresponding points is more evenly spread on both sides of regression straight line.According to the car battery failure time data, exponential distribution model is established, the Weibull distribution model and lognormal model. The corresponding model of bayesian test of goodness of fit, and compareing the BIC values.By using Gibbs sampling algorithm solved the Bayes reliability analysis of complicated posterior integral problem. It is concluded that the parameters of the posterior distribution and the reliability index of point estimation and interval estimation. Calculation results as shown in table 4.3, the average car battery failure time of 509.62 cycles after 500 cycles of charging and discharging reliability R(500)=0.82. Using the exponential model analysis car battery working time data reliability is accurate, the effect is good.The innovation of this article is in view of the battery capacity attenuation data types, improved degradation model is established.