Abstract
Due to the stochastic and correlated attributes of natural inflows, the mid-term generation scheduling problem for cascaded hydro systems is a very challenging issue. This paper proposes a novel stochastic optimization algorithm using Latin hypercube sampling and Cholesky decomposition combined with scenario bundling and sensitivity analysis (LC-SB-SA) to address this problem. To deal with the uncertainty of natural inflows, Latin hypercube sampling is implemented to provide an adequate number of sampling scenarios efficiently, and Cholesky decomposition is introduced to describe the correlated natural inflows among cascaded stations. In addition, to overcome the difficulties in solving the objectives of all the scenarios, scenario bundling and sensitivity analysis algorithms are developed to improve the computational efficiency. Simulation results from both two-station and ten-station systems indicate that the proposed method has the merits in accuracy as well as calculation speed for the mid-term cascaded hydro generation scheduling. The consideration of natural inflow correlation makes the formulated problem more realistic.
Highlights
Mid-term hydro scheduling (MTHS) aims to manage hydropower generation as well as water release with maximum profit while satisfying various system constraints over a yearly horizon [1, 2]
The mean of annual natural inflow is formulated as input and the problem is described by deterministic models [4,5,6]
It is necessary to consider the correlation of random natural inflows in MTHS problems
Summary
Mid-term hydro scheduling (MTHS) aims to manage hydropower generation as well as water release with maximum profit while satisfying various system constraints over a yearly horizon [1, 2]. With the development of sampling and cutting technology, the multi-scenario method [21] is used to solve the MTHS problem In this approach, uncertainty is formulated by random variables with known forecasting information, and many possible scenarios are generated. E.g., fast forward reduction [22], moment based reduction [23], and particle swarm algorithms [24], are employed to select representative discrete scenarios and bundle corresponding probabilities with the purpose of decreasing the computation load Most of these approaches can only obtain the mean value of the optimization objective, and cannot directly obtain other statistical information such as the standard deviation and probability distribution curves.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
