In a market-based power system, the task of the hydropower producers is to make as much profit as possible for the owner and at the same time comply with the restrictions that is given by the license in relation to the operation.
The producer often owns several power stations and reservoirs in the same watercourse and has to make decisions that have consequences for the entire watercourse due to the physical connections. The decision-making problems faced by power producers therefore have complex technical and economic connections, and the use of mathematically based solutions, implemented in computer programs, provides great value.
We will not discuss here why and under what conditions a market-based power system provides the most cost-effective utilization of the power system but note that good forecasts for future prices are very important for making good decisions.
Here are the main decision problems they face:
Deciding when a power station will produce. The producer must decide what to produce for each power station for each hour of the day. The decision for the first hour is made based on an assessment that can look several years into the future depending on the size of the reservoires.
Decisions relating to the upgrading and expansion of existing power plants or investment in new plants.
Determine which power stations are to be rehabilitated/maintained at which times.
How to obtain price forecasts?
The price forecasts used for operational planning purposes can either be based on the prices in the futures/forward market (Nasdaq) or they can be calculated with computer models that simulate the operation of the power system. There are various commercial power system models. SINTEF's "EMPS - multi area power-market simulator" is such a model.
The futures/forward market does not cover the investment horizon and the price forecasts must either be made with power system models or other simpler methodologies. Sensitivity analyzes for different prices are very common.
Operational simulation models
Operation simulation models are a set of optimization and simulation tools that are used to determine operation today and which can often simultaneously calculate optimal operation of the watercourse in the future for different inflows and prices.
The operational simulation models, with minor variations, are used for decision support in both operational planning, revision planning and investment planning, and thus cover all the decision-making problems described above. Input data for the operational simulation models consists of forecasts for future market prices and inflows as well as the physical description of the watercourse.