Evolutionary algorithm for forecastng mean sea level based on meta-heuristic approach
East coast peninsular Malaysia (ECPM) has a sandy shoreline, and is dominated by low-lying regions that are exposed to severe storms, particularly during the Northeast Monsoon, making them vulnerable to erosion. This paper seeks to predict the sea level in ECPM. This study has an important implicati...
Saved in:
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
IAEME Publication
2023
|
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | East coast peninsular Malaysia (ECPM) has a sandy shoreline, and is dominated by low-lying regions that are exposed to severe storms, particularly during the Northeast Monsoon, making them vulnerable to erosion. This paper seeks to predict the sea level in ECPM. This study has an important implication for the population in ECPM since the predicted sea level could be used as an early warning signal to help prevent severe erosion and facilitate early evacuation of affected communities in case of flood inundation. Genetic Programming (GP) algorithm is an example of an evolutionary algorithm (EA) in the field of evolutionally computation (EC) and, more broadly, in Artificial Intelligence. GP is a meta-heuristic search and optimization technique based on natural evolution. The control and optimization parameters in this study are tuned. The findings obtained using the proposed model indicate that GP is able to make a good prediction of monthly mean sea level (MMSL) for a horizon of 10 years ahead for Kerteh, with a testing stage correlation coefficient (C.C) of 0.810 and the 300generation runs. A separate analysis was done for two other regions, Tioman Island and TanjungSedili, to compare the strength and consistency of the model. � 2018 IAEME Publication. |
|---|