Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Smart Dam: Upstream Sensing, Hydro-Blockchain, and Flood Feature Extractions for Dam Inflow Prediction
Heavy rain occurs frequently during extreme weather, and the associated resultant flood damage represents a social problem. The present study aims to redefine the dam watershed as a smart dam and attempts to systematize the technology for flood prediction by integrating upstream sensing, dam inflow prediction, and a hydro-blockchain. In order to detect high water levels, we devise an upstream sensing method to observe the water level at the uppermost stream of a dam watershed, and summarize potential implementation hurdles. We also propose a hydro-blockchain configuration that provides a basis for the fair transaction of water rights. We implement a field study in the Kanto region, Japan, to observe the upstream water level using the devised water level sensor. We analyze the relationship between the measured water levels and the dam inflow and also analyze the response of the time difference. Furthermore, we propose a flood-feature extraction using a 20 year hydro-dataset of rainfall and water levels, and propose dam inflow prediction models using various time series machine learning algorithms. We demonstrate the application of our model results and discuss their usefulness.
Smart Dam: Upstream Sensing, Hydro-Blockchain, and Flood Feature Extractions for Dam Inflow Prediction
Heavy rain occurs frequently during extreme weather, and the associated resultant flood damage represents a social problem. The present study aims to redefine the dam watershed as a smart dam and attempts to systematize the technology for flood prediction by integrating upstream sensing, dam inflow prediction, and a hydro-blockchain. In order to detect high water levels, we devise an upstream sensing method to observe the water level at the uppermost stream of a dam watershed, and summarize potential implementation hurdles. We also propose a hydro-blockchain configuration that provides a basis for the fair transaction of water rights. We implement a field study in the Kanto region, Japan, to observe the upstream water level using the devised water level sensor. We analyze the relationship between the measured water levels and the dam inflow and also analyze the response of the time difference. Furthermore, we propose a flood-feature extraction using a 20 year hydro-dataset of rainfall and water levels, and propose dam inflow prediction models using various time series machine learning algorithms. We demonstrate the application of our model results and discuss their usefulness.
Smart Dam: Upstream Sensing, Hydro-Blockchain, and Flood Feature Extractions for Dam Inflow Prediction
Advs in Intelligent Syst., Computing
Arai, Kohei (Herausgeber:in) / Kapoor, Supriya (Herausgeber:in) / Bhatia, Rahul (Herausgeber:in) / Yasuno, Takato (Autor:in) / Ishii, Akira (Autor:in) / Amakata, Masazumi (Autor:in) / Fujii, Junichiro (Autor:in)
Future of Information and Communication Conference ; 2020 ; San Francisco, CA, CA, USA
25.02.2020
20 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
INFLOW PREDICTION DEVICE, INFLOW PREDICTION METHOD, AND INFLOW PREDICTION PROGRAM
| Europäisches Patentamt | 2017
Inflow Forecasting for Short-Term Hydro Scheduling
| British Library Conference Proceedings | 1997
Flood modelling: parameterisation and inflow uncertainty
| Online Contents | 2014
Investigation of an Ensemble Inflow-Prediction System for Upstream Reservoirs in Sai River, Japan
| DOAJ | 2024
Inflow-height-adjustable pontoon type upstream control check gate
| Europäisches Patentamt | 2015