Risk-aware short term hydro-wind-thermal scheduling using a probability interval optimization model

Due to the uncertainty of wind power and complex constraints of available hydro, short term hydro-wind-thermal scheduling (HWTS) is one of the most difficult optimization problems in the operational planning of power systems. This paper presents a risk-aware optimization model, named probability interval optimization (PIO), to reliably evaluate the HWTS from the perspective of risk and profit. In PIO, the uncertain wind power is deemed as a probability interval variable, the risk of wind power is assessed by its probability distribution, and the profit is manifested by the decrease of generation cost between the same system with and without wind power integrated. For solving the PIO model, an evolutionary predator and prey strategy (EPPS) is proposed in this paper. The EPPS focuses on dynamically adjusting the algorithm's exploration and exploitation abilities by introducing an escaping mechanism and a classification mechanism. In addition, a heuristic repair mechanism, instead of penalty function approach, is applied to handle the complex equality and inequality constraints of HWTS. Simulation studies based on three HWTS systems demonstrate that the risk-aware PIO model is well reliable and applicable to solve HWTS considering the uncertain wind power integrated, the EPPS algorithm can obtain superior solutions in comparison with other recently developed algorithms, and the heuristic repair mechanism is efficient for dealing with complex constraints of HWTS.