Hydrological Changes Driven Flood Disaster Simulation Method

As the consequences of urbanization trends and climate changes, rainstorm-flood disaster processes have become significantly complicated. In the background of ever-graving problems with the rainstorm-flood disaster, the perceived value of rainstorm-flood simulation methods and systems is increasing to stimulate consideration of appropriate flood control and disaster mitigation policies and help the government and public examine future flood events. Under the trend toward more frequent and extreme weather events, traditional simulation methods for evaluating the extent and magnitude of rainstorm-flood hazards have been invalidated, which rely on the assumption of a stationary climate and environment. Traditionally, most simulation paradigms use static or historical data driven the simulation processes. Problems of data delay, analysis delay and decision delay under traditional static data driven passive simulation methods, have become increasingly prominent. This impediment is partially being addressed by the increasingly complex spatial and temporal change characteristics of the rainstorm-flood disaster. Aiming at these increasing critical issues, a novel hydrological changes driven rainstorm-flood simulation method is proposed in this paper.The main contents of this research are organized as follows:(1) Uncertainty analysis of the rainstorm-flood disaster simulation. The existing methods and systems of rainstorm-flood disaster simulations are detailed reviewed in a chronological order following the evolution of methods. Main problems, future trends and the capability of each method, followed by the summary of the key research problems, are analyzed which provides guidelines for related researches.(2) Hydrological changes driven rainstorm-flood simulation mechanism. Deeply analysis of sources of uncertainty of the complex rainstorm-flood disaster spatial and temporal change characteristics is given. Then the framework and connotation of the hydrological changes driven rainstorm-flood simulation mechanism are presented.(3) Universal description model of dynamic hydrological observational data and rainstorm-flood simulation models. After in-depth analysis of spatio-temporal variations involved participants, driving forces and occurrence patterns, which are separately defined as the feature, process and event three domains, as an explicit high-level abstraction over these change-related details, the possible semantic association relationship between the hydrological changes is described, and the spatio-temporal hydrological change-oriented model is presented.(4) Provding semantic awarenss based hydrological changes discovery and intelligence hydrological model parameters calibration. To quantitatively analyze the hydrological spatio-temporal changes and improve the model parameters, three main algrithms are presented. First, semantic awarenss based hydrological changes discovery is given to quantitatively describe and analyze the hydrological spatio-temporal changes; second, change semantics constrained online Kalman filtering is presented to overcome the uncertainty associated with the observational hydrological data; at last, change driven hydrological model parameters calibration is proceeded, in which hydrological spatio-temporal change information capacity coefficient has been taken into consideration to reduce the the uncertainty associated with the hydrological model parameters.Based on the presented method, the research develops prototype using VS208. Case applications indicate the developed framework can be adopted for use by decision-makers for flood control and disaster mitigation.