Interactive simulation-prediction method and interactive simulation-prediction system for hydrologic series of tidal reaches

Abstract

The invention discloses an interactive simulation-prediction method and an interactive simulation-prediction system for the hydrologic series of tidal reaches. According to the method and the system, a prediction process for the hydrologic series of tidal reaches is realized by virtue of comprehensive invoking for an initial ARIMA model configuration module, a real-time data analysis configuration module, a tidal reach water-level prediction module, a predicated result analysis verification module, a predicated result display module and a relearning module. According to the method and the system disclosed by the invention, an emergency plan is equipped for monitoring real-time data, judging abnormal data in real time, and eliminating the abnormal data in time, thus an ARIMA method is capable of correctly exerting a prediction function; the real-time relearning function of a model is further realized, and when verification for the predicated result of the model is not passed, the relearning function can be started to re-acquire the type of the model and new model parameters until the model meets requirements on the predication accuracy.

Description

technical field [0001] The invention relates to a method and system for interactively simulating and predicting a tidal river section hydrological sequence. Background technique [0002] The tidal river section is located at the junction of tidal estuaries and inland rivers, and is affected by both river dynamics (upstream runoff discharge) and ocean dynamics (tidal movement), with dual characteristics, and the state of water flow is complex. It is difficult to grasp the changing laws of elements such as water level and flow. Compared with inland rivers or oceans, its hydrological system is a more complex system influenced by many factors. When forecasting the tide level, the commonly used method of slowly varying and unsteady flow Saint-Venant equations to establish a forecasting model has high data requirements, and requires river topography data that are generally not readily available. Not only is the solution cumbersome, but real-time correction is also difficult. Ther...

AI Technical Summary

Problems solved by technology

[0026] When real-time forecasting is carried out at hydrological stations in tidal reaches, there are two problems in analyzing and predicting purely by a model with fixed parameters: first, the water level in tidal reaches changes with time, so the composition of the data and distribution will continue to change, so the original model may not continue to be applicable; second, when the model monitors abnormal water levels, if such abnormal values ​​are added to the model for fitting, it will not be possible to accurately predict the next time segment water level

Images