Abstract
To solve learning problems with vast number of inputs, this paper proposes a novel learning structure merging a number of small fuzzy neural networks (FNNs) into a hierarchical learning structure called a merged-FNN. In this paper, the merged-FNN is proved to be a universal approximator. This computing approach uses a fusion of FNNs using B-spline membership functions (BMFs) with a reduced-form genetic algorithm (RGA). RGA is employed to tune all free parameters of the merged-FNN, including both the control points of the BMFs and the weights of the small FNNs. The merged-FNN can approximate a continuous nonlinear function to any desired degree of accuracy. For a practical application, a battery state-of-charge (BSOC) estimator, which is a twelve input, one output system, in a lithium-ion battery string is proposed to verify the effectiveness of the merged-FNN. From experimental results, the learning ability of the newly proposed merged-FNN with RGA is superior to that of the traditional neural networks with back-propagation learning.
| Original language | English |
|---|---|
| Pages (from-to) | 697-708 |
| Number of pages | 12 |
| Journal | IEEE Transactions on Energy Conversion |
| Volume | 22 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - 2007 Sep 1 |
Fingerprint
Keywords
- B-spline membership functions (BMFs)
- Battery state-of-charge (BSOC)
- Battery string
- Fuzzy neural networks (FNNs)
- Merged-FNN
- Reduced-form genetic algorithm (RGA)
ASJC Scopus subject areas
- Energy Engineering and Power Technology
- Electrical and Electronic Engineering
Cite this
A merged fuzzy neural network and its applications in battery state-of-charge estimation. / Li, I. Hum; Wang, Wei Yen; Su, Shun Feng; Lee, Yuang Shung.
In: IEEE Transactions on Energy Conversion, Vol. 22, No. 3, 01.09.2007, p. 697-708.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - A merged fuzzy neural network and its applications in battery state-of-charge estimation
AU - Li, I. Hum
AU - Wang, Wei Yen
AU - Su, Shun Feng
AU - Lee, Yuang Shung
PY - 2007/9/1
Y1 - 2007/9/1
N2 - To solve learning problems with vast number of inputs, this paper proposes a novel learning structure merging a number of small fuzzy neural networks (FNNs) into a hierarchical learning structure called a merged-FNN. In this paper, the merged-FNN is proved to be a universal approximator. This computing approach uses a fusion of FNNs using B-spline membership functions (BMFs) with a reduced-form genetic algorithm (RGA). RGA is employed to tune all free parameters of the merged-FNN, including both the control points of the BMFs and the weights of the small FNNs. The merged-FNN can approximate a continuous nonlinear function to any desired degree of accuracy. For a practical application, a battery state-of-charge (BSOC) estimator, which is a twelve input, one output system, in a lithium-ion battery string is proposed to verify the effectiveness of the merged-FNN. From experimental results, the learning ability of the newly proposed merged-FNN with RGA is superior to that of the traditional neural networks with back-propagation learning.
AB - To solve learning problems with vast number of inputs, this paper proposes a novel learning structure merging a number of small fuzzy neural networks (FNNs) into a hierarchical learning structure called a merged-FNN. In this paper, the merged-FNN is proved to be a universal approximator. This computing approach uses a fusion of FNNs using B-spline membership functions (BMFs) with a reduced-form genetic algorithm (RGA). RGA is employed to tune all free parameters of the merged-FNN, including both the control points of the BMFs and the weights of the small FNNs. The merged-FNN can approximate a continuous nonlinear function to any desired degree of accuracy. For a practical application, a battery state-of-charge (BSOC) estimator, which is a twelve input, one output system, in a lithium-ion battery string is proposed to verify the effectiveness of the merged-FNN. From experimental results, the learning ability of the newly proposed merged-FNN with RGA is superior to that of the traditional neural networks with back-propagation learning.
KW - B-spline membership functions (BMFs)
KW - Battery state-of-charge (BSOC)
KW - Battery string
KW - Fuzzy neural networks (FNNs)
KW - Merged-FNN
KW - Reduced-form genetic algorithm (RGA)
UR - http://www.scopus.com/inward/record.url?scp=34548289440&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34548289440&partnerID=8YFLogxK
U2 - 10.1109/TEC.2007.895457
DO - 10.1109/TEC.2007.895457
M3 - Article
AN - SCOPUS:34548289440
VL - 22
SP - 697
EP - 708
JO - IEEE Transactions on Energy Conversion
JF - IEEE Transactions on Energy Conversion
SN - 0885-8969
IS - 3
ER -