Breakthrough battery algorithm could double lithium lifecycles, Flux Power says

Electric trucks already cost more and have far less range than their diesel counterparts, so for fleets to do more than dabble in electrification, they need to know how long these new investments will last and what their resale value will be. These are critical variables in the total cost of ownership equation, and right now, there isn’t enough real road data or experience to make any firm conclusions.

A recent Geotab analysis estimated that in 2024, light-duty EVs degraded annually at a 1.8% clip, a solid improvement over the 2.3% rate in a 2019 study. However, DC fast-charging and hot climates accelerated the process, Geotab found. That means a trucking company would have to sacrifice uptime for longer life by charging at slower Level-2 stations. And if you’re operating in the South during the summer, you’d want to run more at night to avoid the heat.

Those aren’t ideal—or even realistic—concessions for fleets to make, but maybe they won’t have to. A manufacturer and developer of battery storage solutions called Flux Power says it has a “groundbreaking” battery management algorithm that can conceivably double certain batteries’ theoretical cycle life.

The newly patented Intelligent Cycle Life Maximization Algorithm controls the state of charge (SOC) operations based on that asset’s specific behavior, training each battery pack to find the true capacity it needs, according to the company. It can work in both mobile and stationary battery storage applications, and could be applied to other third-party battery management solutions, such as for electric APUs.

“This algorithm uses real-time machine learning to tune battery cycling behavior based on actual usage patterns in the field, significantly extending system life and efficiency,” Flux Power CTO Paulus Geantil explained.

It’s all about finding a specific battery’s sweet spot. In an application that consistently drains the battery from 100% to 50%, for example, Geantil said the InCycLiMax algorithm would pick up on this and adjust its operating range to be from 75% to 25%. He said this shift minimizes degradation and allows the battery to be more efficient.

“In its simplest form, this technology will allow a battery to ‘learn’ how it is being used—how much of its capacity is actually needed between charges—and then allow the battery to center its operation around 50%,” he detailed to Fleet Maintenance.

While it is best suited for lithium-type batteries, the CTO said the algorithm could be effective for “on just about any chemistry,” because its main function is to “narrow the SOC window” to fight the primary battery degradation accelerants, which include:

         High State of Charge: When charging near or over 100%, electrolyte oxidation occurs and creates a deposit over the cathode area. At the anode electrode, electrolyte decomposition may create lithium, or a surface film that affects performance.

      Deep discharge: Dropping down to near 0% SOC can increase cathode structural damage and solid-electrolyte interphase (SEI) growth

·       Large Depth of Discharge: This can increase mechanical wear because of the more extreme volume changes.

Flux has found that a 75%-25% range is more thermodynamically stable than 100%-50%, even when the depth of discharge (DoD) is the same, and mitigates these accelerants. Geantil noted this also avoids full-depth cycling and extends the theoretical cycle life by 50–100% (or more in some chemistries).

The solution works well with an opportunity charging strategy, where a user “fills up” the battery capacity to around 80% versus topping off, which takes longer as the state of charge reduces to prevent overloading the capacity.

According to Geantil, InCycLiMax “will ease the strain on opportunity charging as it will only take the energy needed to keep up with the operation”… and “will not arbitrarily and simply charge off to 100% when it is not necessary to complete the work.”

In the case of an eAPU specifically, Geantil said the algorithm calculates average actual usage to center at 50%. If the average run time is eight hours and total capacity allows for a 12-hour runtime, InCycLiMax would figure 66% of capacity is needed.

“The pack would learn to cycle itself from 83% down to 17% SOC,” Geantil said. “This would effectively increase the lifetime of this asset from about 2,500 cycles to approximately 3500-4000 cycles (until the battery reaches 80% of its original capacity).” 

The development of the patented algorithm is part of Flux Power’s strategy to transition from a battery manufacturer to more of an energy solutions provider.

“Our customers want more than hardware — they want insight, control, and adaptability,” said Krishna Vanka, CEO of Flux Power. “We are building the future of energy through artificial intelligence, where each battery becomes part of a connected network of self-optimizing assets. This patent is a cornerstone of that vision, enabling us to deliver software-driven solutions, which generate additional customer value that goes far beyond traditional energy storage.”