An approximate dynamic programming method for unit-based small hydropower scheduling

Hydropower will become an important power source of China’s grids oriented to carbon neutral. In order fully exploit the potential water resources and achieve low-carbon operation, this paper proposes approximate dynamic programming (ADP) algorithm for unit-based short-term small hydropower scheduling (STSHS) framework considering hydro unit commitment, which can accurately capture physical operational characteristics individual units. Both non-convex non-linearization original STSHS model are retained without any linearization describe production function head effect, especially dependence between net volume in reservoir, thereby avoiding loss actual optimal solution due large error introduced by process. An value problem is formulated via searching table policy iteration process address “curse dimensionally” traditional programming, provides strategy both accuracy computational efficiency. The then tested with a real-world instance plant three identical units demonstrate effectiveness proposed method.