BuiVuDuyTruong 20202543

Sizing of Battery/Supercapacitor Hybrid Energy Storage System for Electric
Vehicles
This paper focuses on optimizes a hybrid energy storage system (HESS) for electric vehicles,
comprising a battery and supercapacitor. Using the NSGA-II algorithm, the study aims to maximize travel
distance and battery lifetime. Optimization criteria include battery state of charge, current characteristics,
supercapacitor voltage, and regenerative energy. Results are evaluated using different driving cycles,
providing optimal HESS configurations for enhanced efficiency and extended battery life in electric
vehicles.
I.Introduction
This paper focuses on optimizing the sizing of a hybrid energy storage system (HESS) for electric
vehicles using the NSGA-II algorithm. The optimization objectives include minimizing the
supercapacitor voltage difference, maximizing travel distance, and minimizing battery degradation. The
UDDS driving cycle is used for optimization and simulation, and the optimized results are compared with
non-Pareto configurations and validated with different driving cycles. The paper covers the system
configuration, optimization problem formulation, application of the NSGA-II algorithm, optimization
results and validation, and conclusion with an examination of power losses in system components.
II. System Configuration
This study is based on the 2012 Mitsubishi iMiEV electric vehicle, which has a 49kW IPM motor and a
three-phase inverter. The original 16kWh battery system with 330V is replaced by a battery/SC HESS
configuration. The HESS connects to the DC bus via bidirectional DC/DC converters. The study includes
modeling and control design, considering power losses in converters, motor, and transmission. The HESS
performance depends on the optimal configuration of cell/module count and parallel branches.
III. Problem Formulation:
Optimize configurations to increase travel distance and reduce battery degradation.
A.Objective functions
The objectives of this study are to minimize the degraded SoC of the battery cell (SoC) after the test in
order to prolong the travel distance of the EV. The final voltage of the SC pack should be as close to the
initial value as possible, and the difference between the initial value (usc_0) and the final value (usc_f
inal) should be minimized. The standard deviation of battery cell current (cell) is the fourth criteria that
effects to both the travel distance and battery lifetime, and the optimal solutions must reduce RMScell.
Unexpected battery cell charging energy (Ebat_gen) is the last criteria which should be minimized, with
zero being the best case.
B. Constraints
The most important details in this text are the constraints on the nominal voltage of the battery and SC,
the volume of the whole energy storage system, the final SoC of the battery, and the charge/discharge
capability of the battery and SC. The nominal voltage of the battery and SC must not exceed 330V, while
the lower limit of voltage (Vbat_min, Vsc_min) depends on the power of the DC/DC converter. The
volume of the whole energy system must be determined according to the cell/module nominal voltage
(Vcell_nom, Vmodul_nom). The SoC of the battery must be greater than 20% after the test, and the
charge/discharge capability of the battery and SC must be expressed by the maximum current (Icell_max,
Imodul_max) and the minimum current (Icell_min, Imodul_min). The optimal configurations must
minimize (1)-(5) simultaneously and satisfy the constraints in (6), (8), (10) and (11).
IV. SIZING FOR HESS USING NSGA-II
The paper utilizes the NSGA-II algorithm to find optimal configurations in a hybrid energy storage
system. The algorithm creates an initial population and generates child offspring through tournament
selection, crossover, and mutation. The new population is formed by selecting the best solutions from
both the parent population and the offspring based on rank and crowding distance. The result is a set of
optimal configurations known as the Pareto front. The study uses Panasonic NCR18650GA battery cells
and Maxwell BMOD0058 E016 B02 supercapacitor modules, evaluated under UDDS driving cycle
conditions.
V. RESULT
A. Pareto front
The Pareto front is a curved surface with relationship between SoC, Usc and Ebat_gen. The range of
SoC is from 5.726% to 45.998%, the range of Usc is from 9.248V to 25.702V, the range of Ebatgen is
from 0.068As to 14.266As, the cell is from 0.911A to 7.255A, and the RMScell is from 1.114A to
8.899A. The optimization trend of both standard deviation and root-mean-square value of battery cell is
similar to the degraded SoC. When the number of battery cells increases, the total energy (kWh) of the
battery pack increases accordingly. The unexpected charging energy of the battery cell depends on both
the number of battery cells and SC modules. Based on the particular purpose, the manufacturer can
choose a suitable configuration.
B. Validation scenarios
This paper selects three optimal configurations to compare with other non-Pareto configurations, which
are 74-7-16-3, 77-21-11-4, and 58-33-8-2. The optimal configurations are validated with other
performance conditions of WLTC class 2 and Artemis road driving cycle.
C. Numerical Results
Table III compares the optimal configurations with non-Pareto configurations, which have better or
equal objective function values. Table IV presents validation results of the optimal configurations, which
still perform best under different driving conditions. The difference in SC voltage is 10.131V and the
minimum unexpected charging energy is 29.37As.
IV. CONCLUSION
This research designs the HESS configuration with the NSGA-II technique. The algorithm approaches
the Pareto front, containing 40 optimum variations, after 150 generations. Three suitable configurations
are identified that represent the best values for the battery's state of charge, SC pack's ultimate voltage,
and regenerative energy. This comparison indicates that the Pareto front method is the only possible
arrangement, perfect for optimizing goals. Future studies will establish the ideal arrangement that
simultaneously decreases battery degradation and the performance cost of HESS.

BuiVuDuyTruong 20202543

Uploaded by

BuiVuDuyTruong 20202543

Uploaded by

Sizing of Battery/Supercapacitor Hybrid Energy Storage System for Electric

You might also like