Shida Jiang, Shengyu Tao, Vincent Molina, Junzhe Shi, Scott Moura
7 min
Battery aging remains a key bottleneck for EV adoption. This paper develops a P2D (pseudo-2D) electrochemical-thermal-degradation model that consistently parameterizes coupled calendar (storage-induced) and cycle (usage-induced) aging across 81 experimental conditions. It predicts capacity fade, SOH, and remaining useful life (RUL) with mechanistic insights into SEI growth, lithium plating, and particle cracking.
Updated synthesis with cell-level control: Complements 'Model-Agnostic Energy Throughput Control' by providing the detailed degradation physics that cell-level inverters can exploit. While the throughput controller is model-agnostic (avoiding such complexity), P2D models validate its SOH-balancing assumptions for LMO/LFP chemistries.
Calendar aging: Voltage/temperature-driven SEI growth during storage. Cycle aging: Charge/discharge-induced mechanical stress, plating.
The framework unifies them via shared parameters, fitting 81 conditions (various T, SOC, C-rates) with high fidelity.
SOH = present max capacity / nominal capacity. The model outputs spatially-resolved degradation, enabling per-cell SOH tracking—crucial for the cell-level inverter paper's throughput routing to healthier cells.
Integration insight: Cell-level control could use P2D-derived SOH estimates to optimize Ah-throughput, achieving the reported 7-38% lifetime gains while respecting chemistry-specific limits (e.g., LFP's slower degradation).
A conventional electric vehicle (EV) powertrain relies on a centralized high-voltage DC-AC inverter, thereby limiting cell-level control and potentially reducing overall driving range and battery lifetime. This paper studies an H-bridge-based cell-level inverter topology that performs power conversion at the cell level, enabling independent control of individual cells and expanding the design space for battery management. Leveraging these additional degrees of freedom, we propose a model-agnostic energy-throughput control strategy that extends EV range while improving battery-pack lifetime. Because usable energy (and thus driving range) and lifetime are governed by the cells with the lowest state-of-charge (SOC) and state-of-health (SOH), respectively, the proposed controller preferentially routes energy throughput to healthier cells. Specifically, during charging, it permits cell SOCs to diverge to promote SOH equalization; during discharging, it rebalances SOC to maximize usable capacity under per-cell constraints. The proposed SOC-SOH-aware control strategy is evaluated on two aging models representing lithium manganese oxide and lithium iron phosphate chemistries, using a Tesla Model 3 charge-discharge profile across 14 different parameter settings. Simulations show a 7-38% improvement in lifetime relative to a conventional SOC-only balancing baseline. More broadly, the results suggest a software-defined pathway to extend EV pack life through routine charging, with minimal reliance on specific degradation models or discharge profiles.
Alex: Huh, so it can create uneven work deliberately. But what about keeping the battery balanced overall for the drive afterward?
Sam: During charging, it allows the charge levels to spread out temporarily, so healthier cells fill up more and even out wear over time. Then, while driving, it rebalances those levels to pull the most usable energy from what's left, respecting each cell's limits. No need for guesses about aging or routes—it optimizes on the spot.
Alex: Wait, and for really worn-out cells—does it just shut them off somehow?
Sam: Yes, once a cell hits end-of-life, around 70 percent state-of-health, the inverters bypass it entirely during use—no interruption to the pack. Healthier cells carry more load, so the pack acts healthier as a group.
Alex: Right... so it's like the pack keeps going by ignoring the dead weight, without losing drive range.
Alex: But how does the software actually figure out those target charges for each cell during an overnight plug-in—without getting bogged down in endless calculations?
Sam: The system simplifies the problem by planning just the final amount of energy each cell should gain by the end of charging, rather than deciding every tiny step—like sketching the finish line for a race instead of mapping every footstep. It breaks the charging into a few stages: first a steady push of current until voltages near limits, then slower topping off. They turn it into a puzzle where the goal is to add up to a set total energy, but penalize putting too much into weak cells or rushing too fast, since fast charging wears batteries more.
Alex: Okay, so it's like assigning portions at a meal to bigger appetites first, within dinner time. What stops it from just overloading the strong ones?
Sam: Exactly—stronger cells get more, but rules enforce balance. One rule orders cells by health so healthier ones end fuller. Another checks the plan fits the switching setup and guarantees the spread-out charges can be evenly drained later for full driving range.
Alex: Huh, so the penalties make it skip weak cells almost entirely once they're too far gone?
Sam: Yes—for cells below usable health, around 70 percent, the penalty skyrockets, effectively bypassing them during charge, just like in driving. Healthier cells handle the load, slowing pack-wide aging.
Alex: So those endpoint plans with penalties—did they test how it plays out in real driving patterns and over years of use?
Sam: Yes, the paper runs computer tests mimicking real trips and 255 charging sessions from a Tesla Model 3 over 2.5 years—mixing slow overnight plugs and faster DC stops—until the pack wears out. It models two battery types: lithium iron phosphate, which ages steadily, and lithium manganese oxide, which wears quicker under heavy use. Wear comes from time-based storage decay and cycle stress from charging speed, depth, and temperature variations across cells.
Alex: Okay, so they bake in real-world messiness like uneven temps and cell differences. Versus just matching charge levels every time?
Sam: The proposed way shows packs reaching end-of-life notably later than standard charge-balancing alone—like a clear step ahead in simulated years of use. Healthier cells share more load, slowing the weakest from dragging everyone down. For both battery types, it holds full range while extending life.
Alex: Those aging sims sound solid for everyday use... but how does it hold up if things vary, like different driving habits or noisier health readings?
Sam: The paper tests robustness with sensitivity analysis under varied scenarios—like more fast charging or higher temperatures. Results show lifetime improvements of 7 to 38 percent compared to SOC-only balancing, even with added noise in health estimates.
Alex: Okay, so pretty consistent gains... what cuts the benefits most?
Sam: Two scenarios noticeably trim the edge: bumping fast charging to half the sessions, since the approach skips SOH balancing there to prioritize speed; and doubling calendar aging.
Alex: Right, so it's robust but not magic—tuned best for typical cycle-heavy wear, and range stays full either way. Makes routine overnight charging a practical way to ease pack aging without hardware overhauls.
Sam: Precisely. The evidence suggests this SOC-SOH-aware control offers meaningful lifetime extension through software alone, robust to real-world noise and no need for aging forecasts—a notable step for EV packs.
Alex: That's a grounded look at managing battery life smarter. Thanks, Sam—clear insights as always. Thanks for listening to ResearchPod.