摘要
探讨了水文系统和输入存在不确定性的灰色系统分析问题。基于输出观测的灰区间信息，提出了一类显式的灰参数模型识别方法。通过灰参数分析，获得了反映系统不确定度、过程模拟合格率与精确性之间的量化关系，通过对黄河流域龙山峡水库月径流的预报分析与检验，说明了这类方法的意义与实用性。
This paper is concerned with a hydrological grey system analysis involving the uncer-tainty of both the system model and input-output.Agrey parameter model that takes intoaccount general uncertainty has been developed.Apractical penalty function method is usedto estimate optimal whitening parameters and constrained interval radius of grey parameters,which cohsists in an extension of the classical least squares solution to arbitrarily definedweights with a penalty function and adjusted to ensure satisfaction of the constraints. Theparameters evaluated from historical hydrological and meteorological data for the Long-yangvalley catchrnent of the Yellow River in China are tested by grey system analysis and verifica-tion of monthly average runoff forecasting form l959 to l985.It has been found that thereexists evident uncertainty in both input-output observation and system modelling。The greyinterval operation employing the grey target concept can not only increase the rate of quali-fied operation,but it also provides more system information about modelling and forecast-ing.Also,for the application of hydrology,the key problem is how to provide uncertaintyinformation and give the relationship between the modelling accuracy and feasibility.Thegrey system proposed in this paper can answer some of these questions by the identificationof grey parameters and a set of quantitative indices,that is,the rate of qualified operationPm,accuracy PD and system uncertaintv DGP.
出处
《华中理工大学学报》
CSCD
北大核心
1994年第1期55-60,共6页
Journal of Huazhong University of Science and Technology
基金
国家自然科学基金青年基金
关键词
灰色系统
水文预报
参数识别
grey system
hydrological forecasting
parameter identification
