A new research study from Huaiyin Institute of Technology examines how hybrid energy storage can address one of the most persistent bottlenecks in the electric vehicle industry: the difficulty of combining high energy capacity with high power output while protecting battery life. The work focuses on an electric vehicle architecture that pairs lithium-ion batteries with supercapacitors, supported by bidirectional DC/DC converters, optimized fuzzy energy management, and lower-level current and voltage controllers. The study reports that the hybrid configuration can reduce battery current stress, improve state of charge retention, and lower thermal accumulation under the Urban Dynamometer Driving Schedule, or UDDS.
The research arrives as the electric vehicle market continues to expand and as manufacturers search for storage solutions that can satisfy both daily driving range and transient power demands. Lithium-ion batteries remain the dominant energy carrier in electric vehicle applications because of their relatively high energy density, but they face limitations in power density, cycle life, charging time, and heat generation under repeated high-current operation. Supercapacitors offer rapid charge and discharge capability, long cycle life, and strong low-temperature performance, yet their low energy density prevents them from serving as the sole energy source in a production electric vehicle. The hybrid approach seeks to combine the strengths of both devices while mitigating their individual weaknesses.
1. Electric Vehicle Storage Bottleneck Drives Hybrid Architecture Research
The study begins by framing the electric vehicle energy problem in both technical and industrial terms. Modern electric vehicle storage systems must meet multiple requirements at once: sufficient energy for range, sufficient power for acceleration and grade climbing, acceptable cost, long service life, and stable operation across temperature and load conditions. A single lithium-ion battery pack can satisfy many of these requirements, but high-frequency power fluctuations in urban driving can accelerate battery degradation. Repeated high-current discharge and regenerative charging events increase electrochemical and thermal stress, which in turn can reduce capacity retention and shorten pack life.
Supercapacitors are presented as a complementary device for electric vehicle applications. They can deliver and absorb high power within short time windows, making them suitable for peak shaving, transient power buffering, and regenerative braking energy capture. In a hybrid energy storage system, the battery can supply baseline energy while the supercapacitor handles high-frequency and high-power components. This division of labor is intended to reduce the magnitude and frequency of current spikes seen by the battery, smooth battery power, and improve overall electric vehicle energy efficiency.
The research also notes that early work in this field demonstrated the technical value of battery-supercapacitor combinations. Prior studies cited in the work show that hybrid storage can increase effective power density, reduce resistance losses, and lower battery stress when the supercapacitor handles peak power. Other studies have explored cost optimization, dynamic filtering, and rule-based power allocation. The new study builds on this foundation by combining fuzzy logic, a sparrow search algorithm, signal filtering, power ramp limitation, and a lower-level sliding mode control architecture for a more complete electric vehicle hybrid storage design.
2. Modeling the Electric Vehicle Battery, Supercapacitor, and DC/DC Converter
The study first analyzes the operating characteristics of the main components in the electric vehicle hybrid storage system. For the lithium-ion battery, the research compares lead-acid, nickel-metal hydride, and lithium-ion technologies before selecting lithium-ion cells for the electric vehicle application. The battery model uses an equivalent-circuit approach. Among the candidate models, the Rint model is selected because it offers a practical balance between accuracy and parameter identification complexity. The model represents the battery open-circuit voltage, terminal voltage, internal resistance, current, and state of charge through an ampere-hour integration method. This modeling choice supports system-level electric vehicle simulation without imposing excessive computational burden.
For the supercapacitor, the study reviews simplified first-order RC models, classical models, ladder models, and three-branch models. A classical equivalent model is selected because it captures series resistance, parallel leakage resistance, capacitance, terminal voltage, and state of charge while remaining suitable for electric vehicle system simulation. The supercapacitor model accounts for the relationship between voltage, energy state, and state of charge, and it supports the power buffering behavior expected in a hybrid electric vehicle storage system.
The bidirectional DC/DC converter is modeled with attention to efficiency. The study explains that a Buck/Boost converter can operate in both directions, enabling energy transfer between the supercapacitor and the DC bus. Converter efficiency depends on input and output voltage ratio, power level, switching devices, inductors, capacitors, and switching frequency. The research uses a two-dimensional lookup-table efficiency model to represent converter behavior in simulation. This efficiency model is important because the electric vehicle energy management strategy must account for conversion losses when allocating power between the battery and supercapacitor.
| Component | Modeling Approach Used in the Electric Vehicle Study | Key Role in the Hybrid Storage System |
|---|---|---|
| Lithium-ion battery | Rint equivalent-circuit model with open-circuit voltage, internal resistance, and ampere-hour SOC integration | Provides baseline energy for electric vehicle range and steady operation |
| Supercapacitor | Classical equivalent model with series resistance, parallel leakage resistance, and capacitance | Handles transient power, peak load, and regenerative braking energy |
| Bidirectional DC/DC converter | Buck/Boost topology with two-dimensional efficiency lookup table | Matches supercapacitor voltage to the DC bus and manages power flow |
3. Electric Vehicle Hybrid Storage Topology and Operating Modes
The research compares four possible electric vehicle hybrid storage topologies: passive parallel connection, supercapacitor-side semi-active configuration, battery-side semi-active configuration, and fully active configuration. Each topology has different tradeoffs in cost, control flexibility, conversion efficiency, and dynamic response. The passive parallel topology is simple but cannot fully exploit the supercapacitor because its voltage varies directly with the DC bus. The battery-side semi-active topology can buffer battery current, but it places the main energy source behind a converter, which can reduce overall efficiency. The fully active topology offers independent control of both storage units but introduces higher cost and more complex control coordination.
The study selects the supercapacitor-side semi-active topology for the main electric vehicle energy management investigation. In this configuration, the lithium-ion battery is connected directly to the DC bus while the supercapacitor is interfaced through a bidirectional DC/DC converter. This arrangement allows the supercapacitor to handle transient power while the battery provides the main energy supply. The converter controls the supercapacitor current and voltage, enabling flexible power allocation without placing the battery behind an additional conversion stage. The research argues that this topology offers a practical balance among dynamic response, cycle life protection, and system cost for electric vehicle applications.
The study also defines four operating modes for the electric vehicle hybrid storage system. In the first mode, the supercapacitor supplies power alone during high-power events such as startup or rapid acceleration. In the second mode, the battery supplies power alone during steady cruising or low-load conditions. In the third mode, the battery and supercapacitor supply power together during steep grade climbing or aggressive acceleration. In the fourth mode, regenerative braking energy is captured, with the supercapacitor prioritized for rapid energy absorption before the battery is charged. These modes form the basis for the upper-level energy management strategy.
4. Rule-Based and Fuzzy Energy Management for Electric Vehicle Power Allocation
The study first develops a rule-based energy management strategy for the electric vehicle hybrid storage system. The rule-based controller uses thresholds such as required power, supercapacitor state of charge, and battery state of charge to determine whether the battery, the supercapacitor, or both should supply power. A first-order filter is used to smooth the battery power command, allowing the supercapacitor to respond to fast power changes. The rule-based strategy is simple and easy to implement, but its fixed thresholds limit adaptability across different electric vehicle driving conditions.
To overcome the rigidity of rule-based control, the research designs a fuzzy logic energy management strategy. The fuzzy controller uses three inputs: required power, battery state of charge, and supercapacitor state of charge. The output is a battery power allocation factor. Membership functions are constructed for each input and output. Triangular membership functions are used for required power to support fast computation, while Gaussian membership functions are used for state-of-charge variables and the output allocation factor to provide smooth transitions. A rule base is built from engineering experience, and a three-dimensional decision surface represents the relationship among required power, state of charge, and power allocation.
The fuzzy strategy allows the electric vehicle hybrid storage system to adjust power distribution more smoothly than a fixed-threshold controller. When the supercapacitor state of charge is high and the battery state of charge is low, the controller can prioritize the supercapacitor for transient support. When the battery has sufficient energy and the supercapacitor is depleted, the controller can shift more burden to the battery while respecting electric vehicle power limits. The study reports that the fuzzy controller improves system performance compared with the rule-based controller, but it also acknowledges that traditional fuzzy design depends on expert experience and may not achieve a global optimum.
5. Sparrow Search Algorithm Optimization and Signal Conditioning for Electric Vehicle Duty Cycles
The research introduces the sparrow search algorithm, or SSA, to optimize the fuzzy controller. The SSA is a swarm intelligence method that divides the search population into explorers, followers, and warners. Explorers guide global search, followers perform local exploitation, and warners trigger avoidance behavior when risk thresholds are exceeded. The study uses SSA to optimize 36 parameters related to the membership functions of required power, battery state of charge, supercapacitor state of charge, and battery power allocation factor. The optimization objective combines the maximum battery current and the total battery output power per standard cycle distance. Equal weights of 0.5 and 0.5 are assigned to the two objectives.
The SSA configuration uses a population size of 50, a search dimension of 36, a maximum iteration count of 100, and a safety threshold of 0.8. The optimization process evaluates candidate membership-function parameter sets through the electric vehicle hybrid storage simulation model and updates the best solution iteratively. The study reports that SSA optimization improves the fuzzy controller by reducing reliance on manually tuned membership functions and by searching for a better parameter configuration for electric vehicle energy management.
The research also adds two signal-conditioning modules to improve practical behavior. First, a moving average filter is applied to smooth the battery power command. The filter uses a window length of four sampling periods. The study states that a window that is too short cannot suppress high-frequency ripple, while a window that is too long can introduce phase delay and dynamic distortion. The four-sample window is selected to balance ripple suppression and dynamic response in the electric vehicle application. Second, a power ramp limitation module is added after the filter. The power change rate threshold is set to 6.8 kW per second. When the requested battery power changes faster than this threshold, the ramp limiter adjusts the command so that the battery power transitions along a controlled gradient. This reduces high-rate power changes that can accelerate battery aging in an electric vehicle.
| Optimization or Signal-Conditioning Element | Setting Used in the Electric Vehicle Study | Purpose |
|---|---|---|
| Sparrow search algorithm population | 50 individuals | Search a 36-dimensional membership-function parameter space |
| Sparrow search algorithm iterations | 100 | Balance optimization time and convergence |
| Safety threshold | 0.8 | Trigger warning-agent behavior during search |
| Moving average filter window | 4 sampling periods | Suppress high-frequency battery power ripple |
| Battery power ramp limit | 6.8 kW/s | Limit fast power changes and reduce battery stress |
6. ADVISOR Secondary Development for Electric Vehicle Hybrid Storage Simulation
The study uses ADVISOR, a vehicle simulation platform developed on MATLAB/Simulink, to evaluate the electric vehicle hybrid storage system. ADVISOR originally supports conventional vehicles, hybrid vehicles, pure electric vehicles, and fuel-cell vehicles, but its standard configuration includes only one energy storage source. To simulate a hybrid electric vehicle storage system, the research performs secondary development of the platform. The work renames the original energy storage variables, creates a second energy storage data file for the supercapacitor, modifies the energy storage library model, adds the supercapacitor configuration to the vehicle defaults file, updates the graphical menu database, and adapts the block diagram name file so that the new electric vehicle configuration can be loaded correctly.
The vehicle parameters used in the simulation include a mass of 1300 kg, a frontal area of 2.4 square meters, a wheel radius of 0.3015 m, a transmission efficiency of 0.96, an aerodynamic drag coefficient of 0.3, a rolling resistance coefficient of 0.02, and a rotating mass conversion factor of 1.05. The electric vehicle performance targets include a maximum speed of 130 km/h, a 0 to 100 km/h acceleration time of 10 seconds, a maximum gradeability of at least 20 percent, and a driving range target of 210 km. The study selects a 400 V DC bus platform and configures the lithium-ion battery with 125 series cells and 87 parallel strings using 3.2 V, 1500 mAh cells. The supercapacitor pack uses 159 series units of 2.5 V, 1500 F devices, giving a nominal pack voltage of approximately 397 V.
| Electric Vehicle or Storage Parameter | Value Used in the Study |
|---|---|
| Vehicle mass | 1300 kg |
| Frontal area | 2.4 m² |
| Wheel radius | 0.3015 m |
| Transmission efficiency | 0.96 |
| Aerodynamic drag coefficient | 0.3 |
| Rolling resistance coefficient | 0.02 |
| Rotating mass conversion factor | 1.05 |
| Battery cell nominal voltage | 3.2 V |
| Battery cell nominal capacity | 1500 mAh |
| Battery series and parallel configuration | 125 series, 87 parallel |
| Battery pack nominal voltage | 400 V |
| Supercapacitor cell nominal voltage | 2.5 V |
| Supercapacitor cell capacitance | 1500 F |
| Supercapacitor series configuration | 159 series |
| Supercapacitor pack nominal voltage | 397 V |
The study uses the UDDS cycle to evaluate the electric vehicle hybrid storage system. UDDS includes frequent starts, stops, accelerations, and decelerations, making it suitable for urban electric vehicle energy management analysis. The cycle lasts 1369 seconds, covers 11.99 km, reaches a maximum speed of 91.25 km/h, has an average speed of 31.51 km/h, a maximum acceleration of 1.48 m/s², an average acceleration of 0.5 m/s², a maximum deceleration of -1.48 m/s², 259 seconds of idle time, and 17 stops.
| UDDS Parameter | Value |
|---|---|
| Cycle time | 1369 s |
| Distance | 11.99 km |
| Maximum speed | 91.25 km/h |
| Average speed | 31.51 km/h |
| Maximum acceleration | 1.48 m/s² |
| Average acceleration | 0.5 m/s² |
| Maximum deceleration | -1.48 m/s² |
| Idle time | 259 s |
| Number of stops | 17 |
7. Electric Vehicle Simulation Results: Battery Stress, SOC, and Thermal Behavior
The study compares a single-source electric vehicle storage system with the hybrid battery-supercapacitor system under the UDDS cycle. The results show that the hybrid electric vehicle architecture reduces battery current stress and improves state of charge retention. In the single-source case, the battery state of charge falls from its initial value to 0.8212. In the hybrid case, the final state of charge is 0.8297. The study reports an SOC improvement of about 0.85 percentage points for the hybrid electric vehicle configuration. The hybrid system also reduces the peak discharge current from approximately 185.1 A in the single-source case to approximately 156.3 A in the hybrid case. The reported reduction in battery current fluctuation amplitude is about 28.8 A, and the reduction in battery thermal accumulation is about 5.88 °C.
The thermal comparison is particularly important for electric vehicle battery life. The single-source system shows a battery temperature rise of about 12 °C under the UDDS cycle, while the hybrid system shows a temperature rise of about 6.12 °C. The research attributes this improvement to the supercapacitor handling transient power and high-frequency current components, which reduces polarization and heat generation inside the battery. For an electric vehicle, lower thermal stress can support longer battery life, more stable fast-charging behavior, and improved safety margins.
| Metric | Single-Source Electric Vehicle Storage | Hybrid Electric Vehicle Storage |
|---|---|---|
| Final battery state of charge | 0.8212 | 0.8297 |
| Peak discharge current | 185.1 ± 3.2 A | 156.3 ± 2.7 A |
| Battery temperature rise | About 12 °C | About 6.12 °C |
| Reported current fluctuation amplitude reduction | Reference case | About 28.8 A lower |
| Reported SOC improvement | Reference case | About 0.85 percentage points higher |
| Reported thermal accumulation reduction | Reference case | About 5.88 °C lower |
The study then compares four energy management strategies for the hybrid electric vehicle storage system: rule-based control, fuzzy control, SSA-optimized fuzzy control, and SSA composite fuzzy control with moving average filtering and power ramp limitation. The SSA composite fuzzy strategy achieves the highest final battery state of charge and the lowest battery peak current and battery temperature among the four strategies. The rule-based case ends with a state of charge of 0.8297, a peak current of 156.3 A, and a battery temperature of 26.12 °C. The fuzzy case reaches 0.8708, 75.6 A, and 24.98 °C. The SSA-optimized fuzzy case reaches 0.8818, 74.3 A, and 24.81 °C. The SSA composite fuzzy case reaches 0.8938, 74.1 A, and 23.91 °C.
| Control Strategy for the Electric Vehicle Hybrid Storage System | Battery Remaining SOC | Battery Peak Current | Battery Temperature |
|---|---|---|---|
| Rule-based control | 0.8297 | 156.3 A | 26.12 °C |
| Fuzzy control | 0.8708 | 75.6 A | 24.98 °C |
| SSA-optimized fuzzy control | 0.8818 | 74.3 A | 24.81 °C |
| SSA composite fuzzy control | 0.8938 | 74.1 A | 23.91 °C |
The research reports that the SSA composite fuzzy strategy lowers battery current fluctuation amplitude by about 82.2 A, raises SOC by about 6.41 percent, and reduces thermal accumulation by about 2.21 °C in the UDDS test. These results are presented as evidence that optimized fuzzy control with signal conditioning can improve electric vehicle energy management. The study also examines charging and discharging efficiency. Under the rule-based strategy, the electric vehicle battery does not maintain a consistently high discharge efficiency across the cycle. Under the SSA composite fuzzy strategy, the discharge efficiency remains largely within the 0.9 to 1.0 range. This indicates that the optimized strategy can keep the battery in a more efficient operating region while the supercapacitor supports transient power.
8. Lower-Level Control Design for the Electric Vehicle Hybrid Storage System
The upper-level energy management strategy depends on precise power tracking by the lower-level controllers. To address this requirement, the research builds a fifth-order state-space average model for a fully active electric vehicle hybrid storage system. The model includes battery voltage, supercapacitor voltage, battery-side inductor current, supercapacitor-side inductor current, and DC bus voltage. The control inputs are the duty cycles of the two bidirectional DC/DC converters. The study uses this model to design a sliding mode current controller for the battery and supercapacitor, and a PI voltage controller for DC bus stabilization.
The sliding mode controller is designed to track the battery and supercapacitor reference currents. The current error is defined as the difference between the reference current and the actual current. A sliding surface is constructed with proportional and sign-function terms. The control law includes feedforward compensation based on system parameters and feedback correction based on measured states. A Lyapunov function is used to prove asymptotic stability. The study states that the sliding mode controller provides fast current tracking and strong robustness against disturbances, which is important for electric vehicle power management under rapidly changing load conditions.
The PI voltage controller regulates the DC bus voltage. The voltage error is defined as the difference between the reference bus voltage and the measured bus voltage. The PI control law combines proportional and integral actions to compensate for line impedance changes and power fluctuations. A Lyapunov function is again used to prove stability. The supercapacitor reference current is generated from the PI voltage controller and the battery reference current from the energy management strategy. This hierarchical control structure allows the electric vehicle hybrid storage system to coordinate power allocation and bus voltage regulation.
| Controller or Simulation Parameter | Value Used in the Study |
|---|---|
| DC bus reference voltage | 220 V |
| Battery initial state of charge | 0.9 |
| Battery initial voltage | 180 V |
| Supercapacitor initial voltage | 150 V |
| Minimum output power threshold | 2 kW |
| Supercapacitor charging power upper limit | 100 W |
| Switching frequency | 10 kHz |
| PI proportional gain | 20 |
| PI integral gain | 3 |
| Battery-side inductance | 255 µH |
| Battery internal resistance | 0.255 Ω |
| Battery-side filter capacitance | 69 µF |
| Supercapacitor-side inductance | 188 µH |
| Supercapacitor internal resistance | 0.161 Ω |
| Supercapacitor-side filter capacitance | 69 µF |
| Load-side capacitance | 14.85 µF |
| IGBT on-resistance | 21 mΩ |
The Simulink validation applies a load profile with a step change and a superimposed sinusoidal ripple to represent electric vehicle load variation. The simulation results show that the battery current and supercapacitor current track their references with fast dynamic response. The DC bus voltage follows the reference with a maximum deviation of about 0.25 V and no overshoot. The study reports that the current ripple amplitude meets the expected requirement and that the controllers recover quickly after disturbance. These results support the claim that the sliding mode current controller and PI voltage controller can provide the power-following capability needed by the upper-level electric vehicle energy management strategy.
9. Industry Implications for Electric Vehicle Energy Storage
The research carries several implications for electric vehicle storage design. First, the hybrid battery-supercapacitor architecture offers a practical way to reduce battery current stress without sacrificing energy capacity. By assigning high-frequency and peak power to the supercapacitor, the battery can operate with smoother current and lower thermal load. This can support longer battery life in electric vehicle applications where stop-and-go traffic and frequent acceleration are common.
Second, the study shows that energy management strategy matters as much as hardware configuration. The rule-based strategy is simple but leaves performance on the table. Fuzzy control improves adaptability, and SSA optimization improves the fuzzy controller further. The addition of moving average filtering and power ramp limitation provides an extra layer of protection for the battery. For electric vehicle manufacturers, this suggests that software optimization can deliver measurable benefits in battery stress, state of charge retention, and thermal behavior.
Third, the lower-level control design demonstrates that hybrid electric vehicle storage requires coordinated current and voltage control. The battery and supercapacitor must follow power allocation commands while the DC bus remains stable. Sliding mode control provides fast current tracking, while PI voltage control maintains bus stability. This hierarchical structure is relevant to electric vehicle powertrain design because it connects high-level energy management with low-level power electronics control.
Fourth, the use of ADVISOR secondary development shows how established electric vehicle simulation tools can be adapted for hybrid storage research. By adding a second energy storage source and modifying the data files and model libraries, the research creates a simulation environment for electric vehicle hybrid storage analysis. This approach can be useful for further studies on electric vehicle energy management, component sizing, and driving-cycle evaluation.
10. Limitations and Future Research Directions for Electric Vehicle Hybrid Storage
The study acknowledges several limitations. The UDDS cycle represents a standard urban driving pattern, but real electric vehicle operation includes variations in temperature, road grade, traffic density, and driver behavior. These factors can affect battery aging, supercapacitor performance, and converter efficiency. Future work could use multi-source vehicle data to build driving-cycle databases that include temperature, road conditions, and driver habits, thereby improving the representativeness of electric vehicle energy management validation.
The fuzzy rule base is built from engineering experience, and the membership functions are optimized by SSA. However, the rule base may still have limited generalization across widely different electric vehicle use cases. Future research could compare multi-objective optimization methods, combine Markov decision processes with adaptive parameter adjustment, or apply deep reinforcement learning for online strategy updating. Such methods could make electric vehicle energy management more adaptive to unknown driving conditions.
The study also notes that a full life-cycle economic analysis of the hybrid electric vehicle storage system has not yet been completed. The additional cost of supercapacitor modules, auxiliary power electronics, and dual-storage maintenance should be evaluated against the benefits of extended battery life and improved energy efficiency. Future work could build a multi-dimensional economic model that includes initial purchase cost, maintenance cycles, and replacement frequency. This would help electric vehicle manufacturers and fleet operators make better decisions about hybrid storage adoption.
Overall, the research presents a comprehensive electric vehicle hybrid storage framework that spans component modeling, topology selection, fuzzy energy management, SSA optimization, signal conditioning, ADVISOR simulation, and lower-level controller design. The reported results indicate that hybrid storage can reduce battery current stress, improve state of charge retention, and lower thermal accumulation in electric vehicle applications. The work also shows that optimized control and careful signal conditioning can enhance the benefits of hybrid storage. As the electric vehicle industry continues to seek higher efficiency, longer battery life, and safer operation, hybrid energy storage remains a promising direction for both research and product development.