Estimating Evapotranspiration using Machine Learning Techniques

The measurement of evapotranspiration is the most important factor in irrigation scheduling. Evapotranspiration means loss of water from the surface of plant and soil. Evaporation parameters are being used in studying water balances, water resource management, and irrigation system design and for es...

Full description

Saved in:
Bibliographic Details
Published in:International journal of advanced computer science & applications 2017-01, Vol.8 (9)
Main Authors: Adnan, Muhammad, Ahsan, M., -, Abaid-ur-Rehman, Nazir, Maria
Format: Article
Language:English
Subjects:
Accuracy
Agricultural production
Algorithms
Artificial intelligence
Atmospheric models
Climate change
Computer science
Evapotranspiration
Humidity
Hydrology
Irrigation
Irrigation scheduling
Irrigation systems
Machine learning
Mathematical models
Meteorological data
Meteorological parameters
Methods
Microbalances
Neural networks
Precipitation
Principal components analysis
Radiation
Rain
Scheduling
Systems design
Time series
Water resources management
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The measurement of evapotranspiration is the most important factor in irrigation scheduling. Evapotranspiration means loss of water from the surface of plant and soil. Evaporation parameters are being used in studying water balances, water resource management, and irrigation system design and for estimating plant growth and height as well. Evapotranspiration is measured by different methods by using various parameters. Evapotranspiration varies with the climate change and as the climate has a lot of variation geographically, the pre-developed systems have not used all available meteorological data hence not robust models. In this research work, a model is developed to estimate evapotranspiration with more authentic and accurate reduced meteorological parameters using different machine learning techniques. The study reveals to learn and generalize the relationship among different parameters. The dataset with reduced dimension is modeled through time series neural network giving the regression value R=83%.
ISSN:2158-107X
2156-5570
DOI:10.14569/IJACSA.2017.080915