Back to results
US 2024/0003979 A1Published application
Neural network-based monitoring of components and subsystems for battery-enabled devices
OwnerJohn Olsoninventor
Life of this patent
Estimating the term…
Abstract
A neural network model is in communication with a energy storage device sensor that monitors, evaluates, and determines an operating condition and/or life expectancy of one or more electrical or mechanical subsystems of an energy storage device, such as a battery, and an associated device. A battery sensor includes a microcontroller that processes information from the neural network model and evaluates the battery subsystem operation to determine component status and generate an output indicative of one of component state of health or subsystem operation.
Claims20
An energy storage device sensor comprising:
- a shunt defining a resistance area;
- a printed circuit board in electrical communication with the resistance area through two spaced-apart electrical contacts that establish a resistance datum, the printed circuit board having a powered data connection and further including a microcontroller having a memory and a processor, the microcontroller including a data acquisition input to receive raw data from the shunt resistance area;
- and
- a static neural network model resident in the memory and configured to calculate and output at least one of a energy storage device state of charge or a energy storage state of health.
1 more element
A battery sensor comprising:
- a shunt defining a resistance area;
- a printed circuit board in electrical communication with the resistance area through two spaced-apart electrical contacts that establish a resistance datum, the printed circuit board having a powered data connection and further including a microcontroller having a memory and a processor, the microcontroller including a data input to receive raw data from the shunt resistance area and a data output;
- and
- a neural network model resident in the memory and configured with training data directed to a battery chemistry and including test-derived life cycle data, the neural network model configured to calculate and output at least one of a battery state of charge or a battery state of health.
1 more element
Dependencies are inferred from the claim text.