Congestion due to traffic, results in wasted fuel, increase in pollution level, increase in travel time and vehicular queuing. Smart city initiatives are aimed to improve the quality of urban life. Intelligent Transportation System (ITS) provides solution for many smart city projects, as they capture real time data without any fixed infrastructure. The real-time prediction of traffic flow aids in alleviating congestion. Accurate and timely prediction on the future traffic flow helps individual travellers, public transport, and transport planning. Existing systems are designed to predict specific traffic parameters like weekday, weekend, and holidays. This research presents a machine learning based traffic flow forecasting for the city of Bloomington, US not with any precise parameter. The day-wise dataset for the 5 areas is taken from Jan 1, 2017 to Dec 31, 2019. The algorithm used for implementation is Support Vector Regression (SVR) and Long Short-Term Memory (LSTM). LSTM algorithm provides better traffic prediction with least root means square error value.

No abstract

In power grid system power quality improvement plays an important role. Dual Voltage Source Inverter (DVSI) scheme is proposed to improve the aspects of Power and consistency of the micro grid system. Here we use Distributed Energy Resources (DER) for power exchange and unbalanced load compensation and nonlinear load in the system. Load sharing and power injections are done by grid interactive inverters in micro-grid. Based on Instantaneous Symmetrical Component Theory (ISCT), control algorithm is developed. For extraction of positive sequence voltage, dq0 transformation is done. An inverter connected to a 3- phase four- wire distribution combination is employed to test the managing strategy of the system. In this work we mainly focused on the assessment of overall performance of Proportional Integral (PI) controller and Fuzzy Logic Controller (FLC). Thus, the FLC provides increase in reliability, better performance of micro grid, less bandwidth requirements of inverters when compared to PI controller. The proposed system is validated by MATLAB simulation methods using PI controller and Fuzzy Logic Controller.

No abstract