Unscented Transformation-based Probabilistic Optimal Power Flow

Renewable energy-based generation causes uncertainties in power system operation and planning due to its stochastic nature. The load uncertainties combined with the increasing penetration of renewable energy-based generation lead to more complicated power system operations. In power system operation, optimal power flow (OPF) is a widely-used tool in Energy Management System (EMS), for scheduling power generation of power plants, to operate the power system with least cost of generation and to ensure the security and reliability of power transmission grids. On the other hand, in order to deal with the stochastic variables (e.g., renewable energy-based generation and load uncertainties), probabilistic optimal power flow (POPF) has been instituted. This thesis introduces a new Unscented Transformation (UT)-based POPF algorithm. UT-based OPF has a key advantage in handling the correlated random variables, and has become an open research area. Integrated wind power and independent or correlated loads are represented using a Gaussian probability distribution function (PDF). The UT is utilized to generate the sigma points that represent the PDF with a limited number of points. The generated sigma points are then used in the deterministic OPF algorithm. The statistical characteristics (i.e. means and variances) of the UT-based POPF solutions are calculated according to the inputs and their corresponding weights. Different UT methods with their corresponding sigma point selection processes are evaluated and compared with Monte Carlo Simulation (MCS) as the solution benchmark. In the thesis, Locational Marginal Price (LMP) in the transmission network is evaluated as the output of the UT-based POPF. The proposed algorithm is successfully verified on the standard IEEE 30- and 118-bus power transmission systems with wind power generation and unspecified loads. These two test cases represent a portion of American Electric Power (AEP) transmission grid