| 摘要: |
| 【目的】 应用最大熵原理研究水文频率分布参数估计和分布线型优选，为陕北地区水利工程的规划设计提供科学依据。【方法】 基于最大熵原理，以陕北地区主要水文站年径流系列为研究对象，采用14种水文频率分布线型进行参数估计，并选取最小信息熵准则和AIC信息准则进行线型优选。【结果】 在陕北地区7个水文站中，采用最小信息熵准则进行线型优选时，有5个站的最优分布线型为P-Ⅲ分布；采用AIC信息准则时，有3个站的最优分布线型为对数正态分布，表明P-Ⅲ分布和对数正态分布对陕北地区多数年径流资料拟合较好。【结论】 最大熵原理方法简便、合理可行，可用于水文频率参数估计。陕北地区年径流系列的理论分布线型为P Ⅲ分布或对数正态分布。 |
| 关键词: 最大熵原理 水文频率 线型选择 参数估计 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金项目（50579065，50879070）;西北农林科技大学优秀博士论文基金项目 |
|
| Application of maximum entropy principle in the estimation of hydrological frequency parameter |
|
|
| Abstract: |
| 【Objective】 Study on the parameter estimation and distribution curve optimization of hydrological frequency distribution based on the maximum entropy principle can provide scientific basis for the planning and designing of water resources projects.【Method】 The method based on maximum entropy principle is applied to estimate parameters of annual runoff sequence of main stations in northern Shaanxi for 14 kinds of hydrological frequency distribution curve.The criteria of minimum information entropy and AIC information criteria are selected to optimize the hydrological frequency distribution curve.【Result】 It proves that 5 stations'optimal distribution is P-Ⅲ distribution when using the criteria of minimum information entropy to optimize curves and 3 stations’optimal distribution is lognormal distribution when using the criteria of AIC information for 7 stations in northern Shaanxi,that is,P-Ⅲ distribution and lognormal distribution can better fit the majority of annual runoff data in northern Shaanxi.【Conclusion】 The method based on maximum entropy principle is simple,reasonable and feasible,and provides a new way of estimating hydrological frequency distribution parameters.It is suggested to use P-Ⅲ distribution or lognormal distribution as the theoretical distribution curve of annual runoff in northern Shaanxi. |
| Key words: maximum entropy principle hydrological frequency option of curve type parameter estimation |
