期刊
ADVANCES IN WATER RESOURCES
卷 167, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.advwatres.2022.104274
关键词
Reservoir simulation; Machine learning; Data -driven models; Hidden-Markov-decision tree model
资金
- China Scholarship Council [201906060183]
This study presents a generic data-driven reservoir operation model (GDROM) that can accurately simulate the operation and release of reservoirs. GDROM, using a few input variables and employing hidden Markov-decision tree and classification and regression tree algorithms, exhibits good interpretability and performance, making it applicable to reservoirs in different regions.
This study presents a generic data-driven reservoir operation model (GDROM). The hidden Markov-decision tree (HM-DT) is applied to deriving representative operation modules for a reservoir; a classification and regression tree (CART) algorithm is used to identify the application and transition conditions for the operation modules. These two procedures result in the GDROM that is featured by 1) using a few input variables (inflow, storage, DOY, and PDSI); 2) inheriting merits of decision trees but dramatically reducing model complexity; 3) adopting a consistent and transparent structure (i.e., better interpretability than other machine learning models); and 4) showing a better performance than traditional decision tree models, especially in storage simulation. GDROM is developed for 467 reservoirs with diverse operation purposes in different regions of the Contiguous United States (CONUS), and the testing procedure shows comparable accuracy in release simulation to other ML models; among these reservoirs, 15 are selected for detailed analysis with diverse operational purposes and regulation capacities, from different USGS Water Regions. GDROM presents a ready-to-use reservoir operation model that can be incorporated into a watershed hydrological simulation model.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据