Optimum impedance spectroscopy circuit model identification using deep learning algorithms

doi:10.1016/j.jelechem.2022.116854

Optimum impedance spectroscopy circuit model identification using deep learning algorithms

2022-11-01

2

PRE

AI

A

Abdulwadood A. Al–Ali *

DOI：10.1016/j.jelechem.2022.116854

Original

Original request for help

Save

Abstract

En 中文

In this work, a deep learning based optimum impedance spectroscopy model finding algorithm is proposed. The technique uses convolutional neural networks with the long short-term memory to identify the circuit model that is most suitable for fitting measured spectral impedance data. A modified two-stage optimization technique is also proposed to find the optimum circuit model parameters. Experimentally measured battery impedance data at different points on its current-voltage characteristic curve are used to validate the proposed algorithm. The techniques was also applied to experimentally measured cherry tomato bio-impedance data. The procedure can be used in applications requiring high throughput screening and fast diagnoses of large numbers of samples such as in commercial battery testing, and in vegetable/fruit freshness monitors using bio-impedance.

Keywords:

Deep Learning

Electrical Impedance Spectroscopy

Circuit Identification

Convolutional Neural Networks

Recurrent Neural Networks

Optimum impedance spectroscopy circuit model identification using deep learning algorithms

Abstract

Journal

Organization