Weekly planning of hydropower in systems with large volumes of varying power generation

Sammanfattning: Hydropower is the world’s largest source of renewable electricity generation. Hydropower plants with reservoirs provide flexibility to the power systems. Efficient planning techniques improve the flexibility of the power systems and reduce carbon emissions, which is needed in power systems experiencing a rapid change in balance between power production and consumption. This is due to increasing amount of renewable energy sources, such as wind and solar power. Hydropower plants have low operating costs and are used as base power. This thesis focuses on weekly planning of hydropower in systems with large volumes and varying power generation and a literature review and a maintenance scheduling method are presented.The topic of hydropower planning is well investigated and various research questions have been studied under many years in different countries. Some of the works are summarized and discussed in literature reviews, which are presented in this thesis. First, some reviews are presented, which covers several aspects of hydropower planning. Literature reviews for long term, mid term and short term planning, respectively, are described.Maintenance scheduling in power systems consists of preventive and corrective maintenance. Preventive maintenance is performed at predetermined intervals according to a prescribed criteria. This type of maintenance is important for power producers to avoid loss in electricity production and loss in income. The maintenance scheduling for hydropower plants prevent these phenomena since spill in the reservoirs and wear on the turbines can be avoided. Usually, the maintenance in hydropower plants is performed on the turbines or at the reservoir intake. A deterministic and a stochastic method to solve a mid term maintenance scheduling problem formulated as a Mixed Integer Linear Programming using dynamic programming is presented. The deterministic method works well in terms of computational time and accuracy. The stochastic method compared to the deterministic method yields a slightly better result at the cost of a need for larger computational resources.