MONITORING AND FORECASTING SYSTEM FOR HYDROELECTRIC POWER PLANTS USING IoT

Abstract

Electricity is one of the most important blessings that science has given to mankind. It has also become a part of modern life and one cannot think of a world without it. Electricity has many uses in our day to day life. In this report we mainly focus on hydro-electric power plants parameters. A failure of a hydro power machine can lead to a catastrophic incident, like loss of power availability, loss of a machine, or loss of a complete plant. So, in this report we focus mainly on sensing the real time parameters of hydroelectric power plant like power generated, guide vane position, water pressure, and control panel temperature bearing temperature and so on. Using respective transducers. Once the data sensing was done, the conversion of data to numerical digital levels using ADC, SDA, and SPI technologies are processed and store them in AWS RDS MySQL database to enable easy scalability. Building user friendly monitoring system using HTTP protocol and latest IoT web-technologies to facilitate Hydropower plant owners to keep eye on plant parameters and performance from anywhere globally in any device and operating systems. Finally in this report a 3D printed IoT based data transmitter designing and manufacturing using Global System for Mobile Communications (GSM) dual band GPRS method was shown. R-Programming was used in this report to develop forecasting algorithm, in this algorithm Seasonal native model, Exponential smoothing model, and Autoregressive Integrated Moving Average shortly names as ARIMA model are used to predict the range of future values by recording data from SHIRAITO MHPP SAGA, JAPAN for 6 hours. From this predictions prevention measures to mechanical parts can be taken before big damage happen and create big economical loss