Short-term Hydro Scheduling: A Comparison of Linear with Non-linear Network Mathematical Programming

J.P.S. Catalão, S.J.P.S. Mariano, V.M.F. Mendes, and L.A.F.M. Ferreira (Portugal)


Hydro Scheduling, ShortTerm, Cascaded Reservoirs, Variable Head, Applied Optimization


This paper is on the problem of short-term hydro scheduling and the algorithmic aspects of its solution. Resources based on hydroelectric power producers with small storage capacities are known as run-of-the-river. Typically, run-of-the-river resources are considered to operate under stationary conditions at the maximum level of the reservoirs, corresponding by design to the optimum operating point. However, these stationary conditions are often not met, due to the head change effect. Because of the reservoirs low volume, the head may change and the operating efficiency becomes sensitive to the head. This non-linear effect coupled with the cascaded hydro configuration tends to give to this problem complexity and huge dimension. We propose a non-linear network mathematical programming approach to solve this problem, and we compare it with a linear network mathematical programming approach. Using numerical simulation results, we conclude that linear network mathematical programming is not suitable, thus non-linear network mathematical programming is more suitable.

Important Links:

Go Back