Hybrid predictive based control of precipitation in a water-scarce region: a focus on the application of intelligent learning for green irrigation in agriculture sector
A growing need for irrigation in agriculture results from recent climatic parameter uncertainties brought on by climate change, global warming, and other factors. The present-day tumultuous, unpredictable, ever-changing, and ambiguous nature of the onset, cessation, and duration of adverse weather c...
Saved in:
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
King Saud University
2023
|
Subjects: | |
Online Access: | http://eprints.utm.my/104792/1/AminuYahayaZimit2023_HybridPredictiveBasedControlofPrecipitation.pdf http://eprints.utm.my/104792/ http://dx.doi.org/10.1016/j.jssas.2023.06.001 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | A growing need for irrigation in agriculture results from recent climatic parameter uncertainties brought on by climate change, global warming, and other factors. The present-day tumultuous, unpredictable, ever-changing, and ambiguous nature of the onset, cessation, and duration of adverse weather conditions poses a formidable obstacle for farmers in formulating informed judgments pertaining to agricultural practices. In this study, the metrological simulation was carried out based on different input variables, including wind speed, wind direction, relative humidity, and minimum and maximum temperature, to predict the rainfall in the arid agricultural area of Kano, Nigeria. For this purpose, an adaptive neuro-fuzzy inference system (ANFIS), feed-forward neural network (FFNN), and multi-linear regression (MLR) were utilized. Five evaluation criteria for predictive control, including determination coefficient (R2), Nash–Sutcliffe efficiency (NSE), mean square error (MSE), mean absolute error (MAE), and root mean square error (RMSE), were used to figure out how accurate the models were based on how the features were chosen. The output proved the reliable accuracy of intelligent regression learning. The results depicted that MLR-M1 with R2 = 0.9989, NSE = 0.9872, and RMSE = 0.0016 performs the best at predicting rainfall, even though all three computational models (ANFIS, FFNN, and MLR) produced good results. The predictive models justified reliable tools for the management of water resources, especially in the agricultural sector. |
---|