Available transfer capability (ATC) between the interfaces needs to be planned well in advance for a secure power transaction in a deregulated power system.Independent system operator (ISO) is responsible for determining ATC at least 24 hours before and made available for the decision makers. This work proposes an effective method to find probabilistic day ahead dynamic ATC (PDA-DATC) in a solar integrated deregulated power system. Solar photo voltaic (PV) output is stochastic which brings uncertainties in PDA-DATC calculation. Hence, it is important to predict solar PV output accurately. In this work, the solar PV output is predicted using neural network. The time-series historical data such as hourly global direct radiation, global diffusion radiation, air temperature, and sun height are taken as the input variable. The input feature selection methods such as mutual information and random forest (RF) techniques are applied to choose the input features. The impact of solar PV on PDA-DATC is analyzed. Also, the penetration of solar PV on a single location and scattered locations are studied and the results are compared for different loading conditions such as high peak, moderate, and low peak periods. The dynamic voltage stability limit namely the Hopf bifurcation limit is used as a limiting criterion for DATC computation. A new swarm intelligent-based optimization algorithm namely dragon fly algorithm (DFA) is used as an optimization algorithm for PDA-DATC

No abstract