Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{199823,
        author = {RUDRA BHANUSHALI and Asha durfae},
        title = {Smart Irrigation system Using IOT},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {14325-14334},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=199823},
        abstract = {Freshwater mismanagement in irrigated agriculture is not a new problem, but it remains largely unsolved at the field level. Conventional timer-based and threshold-triggered irrigation controllers respond to conditions that have already occurred rather than anticipating what is about to happen—a reactive posture that routinely wastes 30–50% of applied water through over-irrigation and deep percolation. This paper describes a field-tested smart irrigation system built around an ESP32 edge node that fuses soil moisture (volumetric water content, VWC), pH, ambient temperature, humidity, and real-time flow measurement into a predictive control loop. A Random Forest regressor trained on 8,640 labelled samples—collected over a 60-day deployment at SAKEC, Mumbai—forecasts the 60-minute-ahead VWC with a mean absolute error of 1.9% and an R2 of 0.94. Three hand-crafted features (a Drying Factor, a Moisture Velocity, and a pH-adjusted target) proved to be the decisive additions that lifted accuracy from 78% (raw features) to 92.3%. A YF-S201 Hall-effect sensor closes the volumetric loop, cutting water use by 35% against a manual-irrigation baseline (paired t-test: p < 0.001, Cohen’s d =1.24). An L293D H-bridge decouples the inductive pump load from the microcontroller, eliminating the back-EMF failures that are endemic in hobbyist and early-academic implementations. The paper walks through the hardware co-design, the MQTT communication stack, the mathematical basis of the moisture-decay and pH-interaction model, the full firmware state machine, and a head-to-head comparison with fifteen prior works.},
        keywords = {edge computing, ESP32, IoT, Kalman filter, MQTT, precision agriculture, Random Forest, soil pH, volumetric flow, water conservation},
        month = {April},
        }