MISSING DAILY RAINFALL PREDICTION USING GREY WOLF OPTIMIZER-BASED NEURAL NETWORK
This research chapter presents the integration of the Grey Wolf Optimizer (GWO) algorithm for training a Feedforward Neural Network (FNN) to address the issue of missing daily rainfall records. A case study was conducted to evaluate the efficacy and reliability of GWO in overcoming the limitations a...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Book Chapter |
| Language: | English |
| Published: |
Cambridge Scholars Publishing
2024
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/46911/4/missing%20daily.pdf http://ir.unimas.my/id/eprint/46911/ https://www.cambridgescholars.com/product/978-1-0364-0804-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research chapter presents the integration of the Grey Wolf Optimizer (GWO) algorithm for training a Feedforward Neural Network (FNN) to address the issue of missing daily rainfall records. A case study was conducted to evaluate the efficacy and reliability of GWO in overcoming the limitations associated with conventional FNN training algorithms, which often get stuck in local optima. The performance of the developed GWOFNN approach was assessed in handling 20% of missing daily rainfall observations at Kuching Third Mile Station. Comparative analyses were conducted against the Levenberg-Marquardt Feedforward Neural Network (LMFNN) and the K-Nearest Neighbour (KNN) algorithm, both of which are recognized for their reliability in addressing missing rainfall data. The results indicate that GWOFNN outperformed KNN and LMFNN in terms of the coefficient of correlation and mean absolute error performance criteria. |
|---|