公开(公告)号：US20200090285A1

公开(公告)日：2020-03-19

申请号：US16609177

申请日：2018-03-16

Applicant: DALIAN UNIVERSITY OF TECHNOLOGY

Inventor： Jianjian SHEN ,  Chuntian CHENG ,  Nannan ZHANG ,  Congtong ZHANG

IPC: G06Q50/06 ,  G05B13/04 ,  G05B13/02 ,  G06N5/00 ,  G06Q10/06

Abstract: The invention relates to the field of hydropower scheduling, and relates to a method for short-term generation scheduling of cascade hydropower plants coupling cluster analysis and decision tree. Linear regression technique is adopted to determine a matching relationship of energy production between upstream and downstream hydropower plants. With the relationship, typical generation curves of each hydropower plant can be clustered from lots of historical data. These typical generation curves and impact factors of generation scheduling such as planned daily electricity, reservoir level, and power grid characteristics, are together trained to obtain decision-making library for generation scheduling of cascaded hydropower plants. Thus, the decision tree method can be used to rapidly determine an operational scheme for generation scheduling of cascaded hydropower plants. Finally, a correction strategy for operation constraints is introduced to ensure feasibility of the operation scheme. The invention can quickly obtain generation schedules of cascade hydropower plants.