Real-Time Monitoring and Optimization of Recycled Water Irrigation in Coffee Plantations using Bi-Directional RNNs and IoT Sensors

The coffee firms relies heavily on sustainable water management measures due to water scarcity. Present study investigates the utilization of Bi-directional Recurrent Neural Networks (Bi-directional RNNs) combined with Internet of Things (IoT) sensors to monitor and optimize the irrigation of reclaimed water in coffee plants in real-time. The suggested system utilizes a range of IoT sensors to gather up-to-the-minute information on soil moisture, temperature, humidity, pH value, nutrient value, water quality, and potentially weather conditions. Data collected is transmitted through the IoT Lora WAN wireless network, which facilitates long-distance communication for plantations. Bi-directional recurrent neural networks (RNNs) examine the sequential sensor data to detect temporal relationships and determine the best watering schedules and possible irregularities. The proposed model suggests modifications accuracy of 95.66% to irrigation practices based use of water in a more effective manner.