Multiparameter optimization system with DCNN in precision agriculture for advanced irrigation planning and scheduling...

Area of Science:

• Agricultural Science
• Computer Science
• Environmental Science

Background:

• The agricultural industry is evolving with advancements in sensor technology and precision agriculture.
• Wireless Sensor Networks (WSN) are emerging to simplify horticultural operations management.
• Optimizing irrigation is crucial for sustainable farming, water conservation, and increased crop production.

Purpose of the Study:

• To develop a Deep Convolutional Neural Network (DCNN) model for predicting soil moisture.
• To enable precision agriculture farmers to plan irrigation effectively, reducing water consumption and increasing crop yields.
• To integrate Internet of Things (IoT) applications for optimized water management in agriculture.

Main Methods:

• Utilized a proposed Deep Convolutional Neural Network (DCNN) for soil moisture prediction.
• Employed Apriori and Gated Recurrent Unit (GRU) for data serving and storage in a grid view.
• Integrated various sensor and parameter modeling methodologies for irrigation planning.

Main Results:

• The DCNN model achieved an experimental accuracy rate of 98.5%.
• Mean Squared Error (MSE) was predicted with 99.25% accuracy using DCNN.
• The system demonstrated effective smart irrigation, leading to high crop yields with minimal water usage.

Conclusions:

• The proposed DCNN model significantly enhances precision agriculture through accurate soil moisture prediction.
• Smart irrigation planning based on DCNN predictions optimizes water use and improves crop productivity.
• The integration of IoT and advanced modeling supports sustainable agricultural practices and water stability.