High Precision Automated Irrigation System using Linear Regression

Aim: The purpose of this research is to use Linear Regression (LR) to design a high-precision automated irrigation system that will maximize water use and increase the accuracy of soil moisture level predictions. The accuracy of the suggested system's performance is contrasted with that of Temporal Fusion Transformation (TFT). Materials and Methods: Group 1 is the automatic irrigation system that uses 26 samples of Linear Regression (LR), The accuracy of the LR-based system ranged from 90.7%, and Group 2 is the system that uses 26 samples of Temporal Fusion Transformation (TFT). With a 95% confidence interval, the threshold is 0.05% and the G Power value is 80%. Result: The accuracy of the high-precision automated irrigation system that employed Linear Regression (LR) was much greater than that of the system that used Temporal Fusion Transformation (TFT). while that of the TFT-based system ranged from 86.2%. The study's results indicate that a high-precision automated irrigation system utilizing linear regression (LR) outperforms one that relies on temporal fusion transformation (TFT). Conclusion: FT predicts soil moisture more accurately (94.1% vs. 81.5%) and with less error (3.2% vs. 7.8%) than LR. Additionally, it exhibits superior memory (91.2%), precision (93.7%), and adaptation (R2: 92.5% vs. 78.3%). However, real-time utilization is limited by its computing cost, necessitating hybrid models for improved scalability.