US 12,602,526 B2Granted

Battery design optimization

AssigneeAddionics IL Ltd

InventorsXiang Li, Hicham Machrouki, Vladimir Yufit, Farid Tariq

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method for optimising battery design comprising receiving one or more battery performance targets indicating design constraints on one or more battery performance characteristics, applying one or more trained machine learning models to search and determine one or more optimal battery designs. The machine learning model(s) adapted to replace one or more replace complex computing intensive battery design physical/mechanical models for creating a plurality of battery designs complying with the design constraint(s) and compute their performance scores are trained to learn the physical/mechanical model(s), created using experimental data collected for a plurality of batteries, using a plurality of training samples extracted from simulations of the physical/mechanical model(s).

Claims10

Dependent claims

A method of selecting battery designs using machine learning (ML) models, comprising:

receiving at least one design constraint for a battery design defining at least one of a plurality of battery design parameters and at least one performance characteristic, wherein each of the plurality of battery design parameters defines a respective range of values, and wherein the at least one design constraint comprises at least one limitation or precision level of a manufacturing process for the designed battery;
applying at least one trained ML model to compute an estimated performance score for each of a plurality of battery designs, wherein each of the plurality of battery designs is defined by values of a set of battery design parameters selected from a plurality of battery design parameters, and a value of each of the plurality of battery design parameters is within the respective range;
selecting at least one of the plurality of battery designs having an estimated performance score compliant with the at least one design constraint;
adjusting the selection according to a correlation between at least two battery design parameters of the plurality of battery design parameters;
outputting the selected at least one of the plurality of battery designs;
and
manufacturing at least one battery according to the selected at least one of the plurality of battery designs.

4 more elements

Dependencies are inferred from the claim text.

Facts

Application: 18/212,758Filed 22 Jun 2023
Published: US 2024/0426918 A126 Dec 2024
Granted: 14 Apr 2026
Term ends: Not estimatedThe source records don’t support an estimate.

Classification8 codes

G01R 31/3648
Constructional arrangements: comprising digital calculation means, e.g. for performing an algorithm
G06F 30/27
Design optimisation, verification or simulation: using machine learning, e.g. artificial intelligence, neural networks, support vector machines…
G01R 31/367
Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of…
G01R 31/3842
Arrangements for monitoring battery or accumulator variables, e.g. SoC: combining voltage and current measurements

Family

US 2024/0426918 A1Pre-grant publication · published 26 Dec 2024
US 12,602,526 B2This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key