Hybrid Power Management Breakthrough Enhances EV Battery Life
A groundbreaking energy management strategy combining wavelet transform and fuzzy logic control has demonstrated significant improvements in the performance and longevity of lithium-ion batteries in electric vehicles (EVs). Developed by researchers at Liaoning University of Technology, the new control approach leverages the complementary strengths of lithium batteries and supercapacitors to optimize power distribution, reduce peak current stress, and extend battery life—all while maintaining robust vehicle dynamics.
The study, led by Jiangkun Shi, Caiying Shen, and Xin Chen, introduces a real-time control framework designed specifically for hybrid power systems that integrate lithium-ion batteries with supercapapacitors. This configuration addresses one of the most persistent challenges in EV engineering: the degradation of battery performance due to frequent high-frequency power demands during acceleration, regenerative braking, and rapid load changes. Traditional battery-only systems are particularly vulnerable to these transient loads, which accelerate chemical aging and reduce overall service life.
The research team’s innovation lies in its dual-layer control architecture. At its core is a real-time wavelet transform algorithm that decomposes the vehicle’s total power demand into high-frequency and low-frequency components. High-frequency transients—such as those generated during sudden acceleration or deceleration—are routed to the supercapacitor, which excels at absorbing and delivering short bursts of energy with minimal efficiency loss. Meanwhile, the lithium battery handles the steady, low-frequency power requirements, operating within a more stable and efficient range.
This signal separation is achieved through a multi-level Haar wavelet decomposition process, which was optimized for both accuracy and computational speed. The system employs a sliding window mechanism to ensure real-time responsiveness, with a window length of 32 data points determined through extensive testing under the New European Driving Cycle (NEDC) conditions. Five decomposition levels were selected to maximize the filtering effectiveness, ensuring that the most damaging high-frequency current spikes are effectively diverted away from the battery.
However, wavelet-based filtering alone cannot account for the state of charge (SOC) of the energy storage units—a critical factor in maintaining system reliability and longevity. To address this limitation, the researchers integrated a fuzzy logic controller that dynamically adjusts the power split based on real-time SOC readings from both the battery and the supercapacitor. This hybrid control strategy ensures that the supercapacitor operates within a safe SOC range (above 50%), preventing over-discharge while maximizing its ability to buffer transient loads.
The fuzzy controller uses a Mamdani-type inference system with carefully designed membership functions for input variables including vehicle power demand, battery SOC, and supercapacitor SOC. Separate rule sets govern driving and braking modes, allowing the system to prioritize supercapacitor charging during regenerative braking events. This not only improves energy recovery efficiency but also shields the lithium battery from high-current charging pulses, which are known to contribute to lithium plating and capacity fade.
To validate the effectiveness of their approach, the team built a comprehensive vehicle model using MATLAB/Simulink and ADVISOR software, incorporating detailed representations of the battery, supercapacitor, bidirectional DC/DC converter, and drivetrain components. The model was calibrated using real-world vehicle parameters, including curb weight, aerodynamic drag, rolling resistance, and motor power output. Simulations were conducted under the NEDC driving cycle, a standardized test procedure widely used in automotive research.
The simulation results revealed remarkable improvements in battery performance. Compared to a conventional single-lithium-battery system, the hybrid configuration under the wavelet-fuzzy control strategy reduced the peak discharge current by 20.68%. This reduction translates directly into lower thermal stress and reduced degradation rates within the battery cells. Using an established battery life model that accounts for temperature, discharge rate, and cycling depth, the researchers calculated that the hybrid system extended battery lifespan by 16.74% over the same driving cycle.
Further analysis showed that during high-power events—such as hard acceleration at the 1,120-second mark in the NEDC cycle—the supercapacitor absorbed the majority of the transient load, delivering over 92 A while the battery supplied only 20 A. This “peak-shaving” effect was consistent throughout the test, confirming the controller’s ability to maintain smooth battery current profiles even under aggressive driving conditions.
In addition to enhancing battery longevity, the strategy also improved overall energy efficiency. By offloading high-power transients to the supercapacitor—which has superior charge/discharge efficiency at high rates—the system minimized resistive losses in the battery pack. The simulations indicated that the supercapacitor handled power demands up to 46 kW, while the battery operated within a more conservative 0–29 kW range, primarily during steady-state cruising.
To verify the simulation findings, the research team constructed a scaled-down experimental platform. The test bench included a 24-cell lithium iron phosphate (LFP) battery pack, a series-connected pair of 48 V/165 F supercapacitor modules, a 20 kW bidirectional DC/DC converter, and a programmable load emulator. Control algorithms developed in Simulink were deployed to a RapidECU real-time controller, enabling precise execution of the wavelet-fuzzy logic strategy under real-world electrical conditions.
The experimental results closely mirrored the simulation outcomes. Under NEDC-like driving conditions, the hybrid system reduced peak battery current by 20.85%, from 34.05 A in the single-battery configuration to 26.95 A in the composite system. Battery lifetime loss, measured through cumulative degradation metrics, decreased from 8.96% to 7.54%, representing a 15.86% improvement in longevity. These results confirm the robustness and practical applicability of the control strategy across different scales and operating environments.
One of the most notable achievements of the study was the successful maintenance of the supercapacitor’s SOC within a safe operating window. Throughout the test cycle, the supercapacitor’s voltage—which correlates directly with SOC—remained above 53 V (approximately 58% of full charge), even after supplying high currents during acceleration phases. During regenerative braking events, the controller prioritized supercapacitor charging, causing the voltage to rebound to 64 V by the end of the cycle. This dynamic balancing act demonstrates the intelligence of the fuzzy control layer in managing energy flow based on real-time system conditions.
The implications of this research extend beyond laboratory performance metrics. By significantly reducing battery degradation, the proposed strategy could lower the total cost of ownership for electric vehicles. Battery replacement is one of the most expensive maintenance items for EVs, and extending pack life by nearly 17% could translate into thousands of dollars in savings over a vehicle’s lifetime. Moreover, longer-lasting batteries contribute to improved sustainability by reducing the frequency of resource-intensive manufacturing and recycling processes.
From a system design perspective, the semi-active topology adopted in this study—where the supercapacitor is connected in parallel with the battery via a bidirectional DC/DC converter—offers a practical balance between performance and complexity. Unlike passive parallel configurations, which rely solely on impedance matching, the active control allows for precise power allocation. And unlike fully active systems with complex multi-converter architectures, this setup remains cost-effective and scalable for mass production.
The integration of wavelet transform and fuzzy logic represents a sophisticated yet pragmatic solution to a longstanding challenge in EV power management. Wavelet analysis provides the mathematical rigor needed for accurate signal decomposition, while fuzzy logic brings adaptability and resilience to uncertain or nonlinear system behaviors. Together, they form a synergistic control framework that outperforms traditional rule-based or optimization-only approaches.
While the current study focused on the NEDC cycle, the underlying principles are applicable to a wide range of driving conditions, including urban stop-and-go traffic, highway cruising, and mixed driving patterns. Future work could explore the strategy’s performance under more aggressive test cycles such as the WLTP or real-world driving data. Additionally, the control framework could be adapted for use in plug-in hybrid electric vehicles (PHEVs) or fuel cell electric vehicles (FCEVs), where similar power buffering challenges exist.
Another promising direction is the incorporation of machine learning techniques to further refine the fuzzy rule base or optimize wavelet parameters in real time. Adaptive control systems that learn from driver behavior and route characteristics could enhance efficiency even further, tailoring power distribution to individual usage patterns.
Despite its success, the study acknowledges certain limitations. The absence of direct battery aging measurements in the experimental phase means that lifetime improvements were inferred rather than directly observed. Future iterations of the test platform could include advanced diagnostics such as impedance spectroscopy or capacity fade tracking to provide more granular insights into long-term degradation mechanisms.
Moreover, the research did not compare the wavelet-fuzzy strategy against other advanced control methods such as model predictive control (MPC) or reinforcement learning. Such comparisons would help establish the relative advantages and trade-offs of different approaches, guiding future development in the field.
Nonetheless, the work stands as a significant contribution to the evolving landscape of electric vehicle energy management. It demonstrates that intelligent control strategies can unlock substantial performance gains without requiring fundamental changes to battery chemistry or vehicle architecture. As automakers continue to push the boundaries of range, efficiency, and durability, hybrid power systems with smart control will likely play an increasingly central role.
The findings also underscore the importance of interdisciplinary research in advancing sustainable transportation. By combining signal processing theory (wavelet transforms), control engineering (fuzzy logic), and electrochemical modeling (battery life prediction), the team was able to develop a holistic solution that addresses multiple aspects of the problem simultaneously.
For the automotive industry, this research offers a clear path toward more durable and efficient electric drivetrains. As battery costs remain a key barrier to EV adoption, strategies that extend component life and improve energy utilization will be crucial in making electric mobility more accessible and sustainable.
In conclusion, the wavelet transform-fuzzy control strategy developed by Shi, Shen, and Chen at Liaoning University of Technology represents a major step forward in hybrid power system management. By effectively isolating high-frequency power demands and intelligently managing energy flow between battery and supercapacitor, the system achieves significant reductions in battery stress and notable gains in longevity. Validated through both simulation and physical experimentation, the approach proves that advanced control algorithms can deliver tangible benefits in real-world automotive applications.
Jiangkun Shi, Caiying Shen, Xin Chen, Liaoning University of Technology, Chinese Journal of Automotive Engineering, DOI: 10.3969/j.issn.2095‒1469.2024.01.02