Estimation of Parameters of Tank Model by Equivalent Gain Method : Application to Watershed of Minafuku Under Groud Dam

DOI IR (HANDLE) JASI Web Site Open Access
  • Tanaka Kohei
    Laboratory of Land-Drainage and Reclamation, Faculty of Agriculture, Kyushu University
  • Shikasho Shiomi
    Laboratory of Land-Drainage and Reclamation, Faculty of Agriculture, Kyushu University

Bibliographic Information

Other Title
  • 等価線形法によるタンクモデルのパラメータ推定 : 皆福地下ダム流域への適用例
  • 等価線形化法によるタンクモデルのパラメータ推定--皆福地下ダム流域への適用例
  • トウカ センケイカホウ ニヨル タンク モデル ノ パラメータ スイテイ カイ

Search this article

Description

This study was carried out to estimate the parameters of water balance model in the experimental ground water dam in Miyako Island. Miyako is a limestone island with an area of 159㎢. Its top layer is of 40m deep of limestone and underneath it is the impoermeable mud stone layer. This experimental ground water dam was constructed in Minafuku in 1977 and is 16.5m high, 500m long and has 1.7㎢ of catchment area. Now that the dam is filled up, it is greatly expected that it may sure to enhance future project of dam of similar nature. The authors have studied the estimation of ground water level in Miyako Island using the system-theory. in the report, water balance model of ground water dam was made and the non-linear smoothing method was used to analyze the inside state variable (storage depth in model, random noise, and measurement noise). As the result, detail analysis of the structure of water circulation process was made clear and possible. However the parameters of this model were approximated and decided by trial method. In this study, system-theoretic method, the equivalent gain in paticular was used to determine the parameters of the model. As the precipitation and un-off which exhibits the precise dynamic character between them. Its analog is found in non-linear filter theory of system theory where, though precise mathematical representation of filter equation is possible, it does not mean precise dynamic characteristic filter representation is also possible. Hence, in engineering, instead of nono-linear fileter approximate filter is made and often used. Likewise, in this study, approximate filter is used to determine the unknown parameters. The analysis show that determination of the unknown parameters could be made with relatively high accuracy with this method.

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top