Hierarchical Bayesian Model for Probabilistic Analysis of Electric Vehicle Battery Degradation

Abstract

This paper proposes a hierarchical Bayesian model for probabilistic estimation of the electric vehicle battery capacity fade. Since the battery aging factors such as temperature, current, and state of charge are not fixed, and they change in different times, locations and by the different users, deterministic models with constant parameters cannot accurately evaluate the battery capacity fade. Therefore, a probabilistic presentation of the capacity fade including uncertainties of the measurements/ observations of the variables can be a proper solution. We have developed a hierarchical Bayesian Network model for the electric vehicle battery capacity fade considering multiple external variables. The mathematical expression of the model is extracted based on Bayes' theorem, the probability distributions for all variables and their dependencies are carefully chosen where the Metropolis-Hastings Markov Chain Monte Carlo sampling method is applied to generate the posterior distributions. The model is trained with a subset of experimental data (85%) to obtain its unseen parameters and tested with other 15% of data to prove its accuracy. Also, three case studies for different drivers, different grid services' frequencies, and different climates are explored to show model's flexibility with different input data. The developed model needs training data for parameter tuning in different conditions. However, after training, it has more than 95% precision in estimating the battery capacity fade percentage. Index Terms-Bayesian Networks, Electric Vehicle Battery, Markov

Switching control strategy for an energy storage system based on multi-level logic judgment

“…The priority weights of each element of each level on the previous level are then obtained by solving the judgment matrix eigenvectors. Finally, the final weights of the indicators at each level to the integrated target are determined by weighted summation (Jafari et al, 2019). According to the comparison of the influence degree of each index, the comparison matrix A is obtained as follows:…”

Baseline Energy Use Modeling and Characterization in Tertiary Buildings Using an Interpretable Bayesian Linear Regression Methodology

“…In recent years, also thanks to the increase in computing power, multilevel models have attracted great interest and have been applied in many different fields of science and technology [25]. In the energy field, they have found application in the calibration of building energy models [26,27], the forecasting of electricity demand [28][29][30], estimation of overhead lines failure rate [31], estimation of photovoltaic potential [32], and the analysis of the degradation of electric vehicle batteries [33]. In the M&V setting, Booth et al [34] used a hierarchical framework to generate energy intensity estimates for various dwelling types.…”

Filter Design and Optimization of Electromechanical Actuation Systems Using Search and Surrogate Algorithms for More-Electric Aircraft Applications

“…(2SA) In this section, we give a novel category strategy for different commonly used algorithms including GA, particle swarm [25], tabu search [16], [26] and artificial neural network (ANN) [18], [19], [23], [24], Bayesian network [27], support vector machine (SVM) [20] etc. Based on this strategy, the relations (x-to-o) between inputs (x 1 , x 2 ...) and outputs (o 1 , o 2 ...) in algorithm categories can be simply described and compared.…”