PI: Priyanka Sharma Co-PI: Swati Jain

25.8 Lacs INR (36539.25 USD)

Thermonuclear fusion is one of the alternative sources of energy. Fusionreactorsusea device called a tokamak. Classification of favourable and non-favourable dischargesin a tokamak is very important from a plasma operation point of view. Non-favorabledischarges are mainly disruptive in nature, which causes certain losses of confinementthat take place abruptly and affect the integrity of Tokamak. During disruptions, theplasma energy gets transferred to the surrounding structures of the vacuumvessel,which causes massive heat and serious damage. The proposed work aims toclassifysuch plasma discharges in tokamak among the other favourable discharges andmakesome suitable classifiers. The deep learning algorithms like Convolutional NeuralNetworks and RNN for time series analysis can be implemented as the most viableand responsive tool to classify the disruption.