A hybrid approach for water utilization in smart cities using machine learning techniques

Abstract

Water is the most important substance on the earth and its conservation is the need of the hour. If there were no water, there would be no life on earth. Apart from drinking it to survive there are abundant, other industrial environmental, economical sustainability persists because of prominent use water. Consumption of water is a major challenge in the modern distribution plants. Very accurate measurements are required in order to safeguard the water if not it leads to major loss not only in terms of money but also the energy. If these calculations are manual then there is a high chance of miss calculation, which has drastic effect. WHO states that more than 840 million people lack even a basic drinking water facility and nearly 160 million people are depending on surface water. By their analysis, they predicted that by the end of 2025 half of the world’s population would be living in after stressed areas. In this paper a hardware equipped hybrid water conservation and monitoring system is discussed which will monitor and forecast the water usage of a particular house. Here we use various sensors for water level and to calculate the rate of flow of water in a tank and data is dispatched and after receiving the values from the sensors, the prediction and forecasting are done using various algorithms like SVM, RFC and on this basis; the monthly or yearly consumption is calculated. Wireless Sensor technology is used to avoid the massive wastage of water, which is caused main due to uncontrolled use of water in huge apartments/offices