Abstract:
Water distribution has become a global concern due to its scarcity and inefficient management practices, such as unnecessary water wastage from open taps and neglect of broken pipes. Additionally, the constant supply of water at high pressure to areas with low demand leads to wasteful consumption and inadequate supply to areas in need. To address these challenges, there is a pressing need for an effective water distribution management system. In this study, we propose the Development of an Internet of Things (IoT) based Water Management system utilizing Decision Tree and Deep Neural Network algorithms.
To conduct this research, we developed an efficient IoT water meter to collect consumption data from MI Wushishi Minna, our designated study area. Analyzing the generated data helped us understand consumption patterns. We also simulated a water tank with constant valve resistance on Simulink to ensure a stable supply to the study area. Additionally, based on the consumption behavior of the estate occupants, we conducted another simulation with varied valve resistance in response to demand, resulting in significant water savings of approximately 3000 liters.
To enhance the system’s intelligence, we employed the Deep Neural Decision Tree algorithm for auto selection through classification. Compared to existing approaches, the scheduling achieved by the Decision Tree Algorithm in this research demonstrated a notable improvement with an accuracy of 94.2%.
In conclusion, our IoT-based Water Management system, incorporating Decision Tree and Deep Neural Network algorithms, proves to be a promising solution for addressing water distribution challenges, optimizing consumption, and conserving this invaluable resource.
Developing a Water Management System based on IoT and Employing Decision Tree and Deep Neural Network Algorithms. GET MORE, ACTUARIAL SCIENCE PROJECT TOPICS AND MATERIALS
