Presenter
Dr. Boran Zhu, China Institute of Water Resources and Hydropower Research
Co-author(s)
Prof. Jun Liu, College of Hydrology and Water Resources, Hohai University
Prof. Junqiang Lin, China Institute of Water Resources and Hydropower Research
Prof. Yi Liu, China Institute of Water Resources and Hydropower Research
Dr. Di Zhang, China Institute of Water Resources and Hydropower Research
Dr. Yufeng Ren, Hubei Key Laboratory of Intelligent Yangtze and Hydroelectric Science
Prof. Qidong Peng, China Institute of Water Resources and Hydropower Research
Dr. Huaijie He, College of Hydrology and Water Resources, Hohai University
Miss. Qiong Feng, College of Hydrology and Water Resources, Hohai University
Sub-theme
6. Innovation for Water Governance and Management
Topic
6-8. Creative technology and tools for water resources management
Body
Cascade reservoirs are complex engineering systems. The operation of these reservoirs is not only affected by external effects such as natural flow conditions but is also related to the state of the reservoir project, the operating rules of the reservoirs and the operating mode of the discharge facilities. The operating conditions of the system under different working conditions are very different due to the influence of human dispatching operation. The realization of refined simulations that can consider various natural and engineering factors is of great significance to improve the reservoir management level. As a method to study the dynamic behavior of complex systems, system dynamics (SD) simulation models can describe the basic logic law of the internal operation of the system, and have the potential to realize the refined simulation of cascade reservoirs. However, when the SD model is controlled by deterministic rules, the various functional objectives of reservoirs cannot be weighed and emergent scenarios cannot be addressed, which limits the practical application of the model. In this paper, a mechanical-artificial intelligence (AI) coupling modeling paradigm is proposed, in which the AI algorithm is used to extract adaptive operating rules, and then the cascade reservoirs system is divided into three parts: system structure, system state and system operation. The SD method is used to refined simulate the operations of cascade reservoirs, and the mechanism model and the AI algorithm are coupled to build the cascade reservoir adaptive refined (CRAR) simulation model. The example application shows that the proposed model can be used to effectively reflect the dynamic change in system operations. Compared with other models, it is easy to understand, intelligent and extensible, and can realize the refined simulation of system operations and adaptive scheduling decisions under various conditions. From this study, a new idea and a reference for the simulation of cascade reservoir operations are provided.