Development of water system models with step-response tests
Lummikko, Ristomatti (2014)
Lummikko, Ristomatti
2014
Ympäristö- ja energiatekniikka
Teknis-luonnontieteellinen tiedekunta - Faculty of Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2014-04-05
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201703221198
https://urn.fi/URN:NBN:fi:tty-201703221198
Tiivistelmä
More than 50 percent of the electricity in the Nordics is produced with hydropower. Hydropower production is flexible and it is capable of responding to the fluctuating electricity demand. Hydropower production is dependent on current hydrological situation and the supply of hydropower is a significant price driver in deregulated Nordic electricity wholesale markets. The variation in electricity price requires the producer to utilize price-dependent production planning, which is important for the producer to succeed, but also to respond to the variations in electricity demand. In addition, hydropower is useful in managing water levels in reservoirs which helps to mitigate flooding.
Optimal hydropower planning requires precise price forecasts and knowledge of the available amount water, but also detailed knowledge of the hydro system limitations is essential. Hydropower production is planned with optimization models which utilize mathematical methods to form the optimal power production combination. All the water systems are unique and the specific characteristics of the water system must be included to the models.
This thesis is focused on a single river system, and especially to a specific part of it. The river system is located in Finland. Experimental knowledge shows that the river section is hard to model with existing data. Thus, step-response tests are planned and implemented in the river system. More precise stream flow routing and reservoir storing capacity are modelled with both historical data and data acquired from step-response tests. The end result of this thesis is a forecasting tool, which strives to model the stream flow routing and water level behavior.
The function of the forecasting model created in this study is to simplify the operation and short-term planning of the river system. The forecasting tool based on step-response tests results is compared to other alternative or prior forecasting tool results. The forecasting tool predicts the water level behavior more precisely than antecedent models. The errors in water levels in production planning are decreased and the modelled water levels are closer to realized when created forecasting tool is used.
Optimal hydropower planning requires precise price forecasts and knowledge of the available amount water, but also detailed knowledge of the hydro system limitations is essential. Hydropower production is planned with optimization models which utilize mathematical methods to form the optimal power production combination. All the water systems are unique and the specific characteristics of the water system must be included to the models.
This thesis is focused on a single river system, and especially to a specific part of it. The river system is located in Finland. Experimental knowledge shows that the river section is hard to model with existing data. Thus, step-response tests are planned and implemented in the river system. More precise stream flow routing and reservoir storing capacity are modelled with both historical data and data acquired from step-response tests. The end result of this thesis is a forecasting tool, which strives to model the stream flow routing and water level behavior.
The function of the forecasting model created in this study is to simplify the operation and short-term planning of the river system. The forecasting tool based on step-response tests results is compared to other alternative or prior forecasting tool results. The forecasting tool predicts the water level behavior more precisely than antecedent models. The errors in water levels in production planning are decreased and the modelled water levels are closer to realized when created forecasting tool is used.