Our adaptive charging algorithms sit at the intersection of battery chemistry, data and software

Charging is generally the most important factor in accelerating battery degradation. Adaptive algorithms utilize charging to diagnose the internal properties of every battery in real time. The data provide continuous diagnostics of the battery's aging. Our algorithms then optimize the rate and degree of charging to reduce battery wear and maximize battery life.  The adaptive charging algorithms repeat this process every time the battery is plugged into the charger.

Field data, chemical models and learning algorithms provide accurate predictions of battery health

Though statistically infrequent, batteries can explode or cause deadly fires.  Our predictive algorithms anticipate such hazardous failures before they occur.  These algorithms rely on detailed chemical models of the battery and machine-learning from a vast amount of field data to detect the presence of latent defects or excessive internal degradation. Defects may originate during the manufacturing of the battery, or during the assembly process of the battery pack or the smartphone. Defects may include the presence of lithium metal plating inside the battery,  external mechanical damage or even the use of counterfeit batteries.  

Our innovation

Real-time diagnostics of the battery’s electrochemistry and intrinsic material properties using Electrochemical Impedance Spectroscopy (EIS) on the device itself. Diagnostic data collected over the life of the battery provide a unique insight into aging and other degradation mechanisms.

Chemical battery models use real-time diagnostic data as input to determine optimal charging. These models are constantly learning from data accumulated in the field. They are core to making accurate predictions of battery health and safety, and determining the optimal rate and degree of charging in order to maximize battery life and lifespan.

Closed-loop feedback acts in response to the chemical battery models and controls the charging process in real time. It changes the rate and degree of charging for each individual battery to maximize battery life and lifespan. Closed-loop controls adapt to the uniqueness of each battery and address inherent variability in battery manufacturing. 

Predictive health and safety algorithms combine field data with chemical battery models to anticipate excessive degradation or failure. When they detect a safety fault, they shut down the battery before an explosion or fire occurs. The algorithms provide accurate present and future chemical state-of-health of the battery.

Proven. Consistent. Safe.

1 1 1 1 1 0 0 0 2 2 0 0
10
lithium suppliers tested
35
patents granted
140
cell families tested
2000000
lab test cycles
100000000
Qnovo enabled smartphones
0
safety failures

Qnovo Patents

    See Qnovo's full list of patents.