End of discharge prediction;
SMART BATTERY MANAGEMENT
The proposed system is composed of two neural networks. - the first model is a simple model of a discharge curve; - the second one is responsible for selecting the best internal parameters of the first net wrt contextual information. This second is also able to adapt itself on-line to the behavior of the actual battery currently used. |
||
The proposed Model: two neural nets |
This figure display one particular weight of the first net, which is also the output of the second net. |
||
One neural weight for different training conditions (battery, cycle) |