共查询到19条相似文献,搜索用时 171 毫秒
1.
2.
支持向量机的混沌序列预测模型及在径流中应用 总被引:2,自引:0,他引:2
混沌和支持向量机理论为研究复杂多变的非线性水文时间序列开辟了新的途径。给出了应用支持向量机回归原理的混沌时间序列非线性的预测建模的思路、特点及关键参数的选取。根据重构相空间理论对月径流过程进行相空间的重构,探讨了支持向量机混沌时间序列非线性预测模型在月径流预测中的应用,在支持向量机建模过程中引入了经向基核函数,简化了非线性问题的求解过程。实例表明,该模型能较好地处理复杂的水文数据序列,且有较好的预测精度。 相似文献
3.
4.
地下水是影响渣土边坡稳定性的关键因素之一,地下水埋深预测对分析渣土边坡稳定性具有重要意义。考虑渣土边坡地下水水位的高度非平稳和非线性特点,提出了一种基于相空间重构的互补集合经验模态分解-随机森林(CEEMD-RF)的地下水埋深预测模型。以广州市某渣土边坡SW2水文观测孔为例,将基于相空间重构的CEEMD-RF模型应用于该渣土边坡的地下水埋深预测,并与相空间重构的RF模型预测结果进行对比分析。结果表明:利用CEEMD-RF模型对地下水埋深预测的拟合优度为0.997,均方根误差为0.03 m,优于相空间重构的RF模型预测结果;基于相空间重构的CEEMD-RF模型预测的地下水埋深序列能很好地反映地下水埋深的尖变点。 相似文献
5.
混沌相空间理论和神经网络用于径流系统的中长期预测,较传统途径可以更多地利用时间序列中包含的丰富信息,更好地揭示水文动力学系统的规律.针对混沌时间序列,结合混沌分析理论和BP神经网络,建立了相空间重构和BP神经网络耦合预测模型.经实例研究初步表明,用神经网络拟合相空间相点演化的非线性关系是可行的;相空间神经网络耦合预测模型在水文中长期预报中的应用是可行的、合理的,有较好的预报精度和应用价值. 相似文献
6.
简述了混沌理论的基本概念,结合水文预测问题介绍了混沌理论应用的基本方法和步骤。在水文预测中,应用混沌分析方法需解决相空间重构、时间序列的混沌性识别和混沌时间序列预测等关键技术。 相似文献
7.
因长期受人类活动、气候变化等多重因素作用,水文时间序列表现出多时间尺度、多频率、动态变化、自记忆性等复杂性特征,增加了水文预报结果的不确定性。本文将经验模态分解模型,核主成分分析模型和支持向量机模型耦合,建立了针对复杂性水文时间序列的预报模型,并采用NASH效率系数、自相关系数、相对误差作为模拟预测精度及参数率定的多目标判断标准。模型应用于黄河花园口水文站径流序列的长期水文预报中,结果表明:模型预报时段长,具有较好的预测准确性和实践应用价值。该模型为多重因素作用的复杂性水文时间序列预报提供了一种方法。 相似文献
8.
9.
参考作物腾发量的混沌性识别及预测 总被引:4,自引:0,他引:4
本文应用饱和关联维数法对海河流域张北站从1966~2005年50年的参考作物腾发量序列进行混沌性识别,结果表明该序列存在一定的混沌特性。同时,运用自相关函数法和饱和关联维数法确定了该序列重构相空间的嵌入维数和延迟时间,并在此基础上进行了相空间的重构。建立了混沌局域法预测模型对相空间的演化进行了计算,实现了参考作物腾发量的预测,并与时间序列自回归(AR)模型和基于气象资料的BP神经网络模型预测结果进行了比较。结果表明,预测效果比BP网络模型稍差,但明显优于AR模型。这为解决缺乏气象资料地区参考作物腾发量预测问题提供了新的思路。 相似文献
10.
将混沌时间序列预测理论应用到大坝变形预测中,根据非线性大坝变形时间序列,运用相空间重构理论,建立了加权一阶局域法、基于最大Lyapunov指数法大坝预测模型,对混沌的大坝变形数据短期预测模型进行了研究,对比分析了各自的特点,并结合实例完成了对大坝变形的预测。计算分析表明,该模型预测误差较小,与传统的自回归模型预测结果相比,基于混沌时间序列的预测方法在大坝变形的短期预测中具有更高的精度。 相似文献
11.
Recently some generalized autoregressive conditional heteroskedasticity (GARCH) models are proposed and applied to various hydrologic variables to capture and remove the ARCH effect, which has been observed frequently in the residuals from linear autoregressive moving average (ARMA) models fitted to hydrologic time series. As a nonlinear phenomenon of variance behavior, the ARCH effect reveals partially nonstationarity and nonlinearity of hydrological processes. This paper deals with the variation of a river basin using the ARMA-GARCH error model, which combines an ARMA model for modelling the mean behavior and a GARCH model for modelling the variance behavior of the residuals from the ARMA model. Based on the heteroscedasticity of hydrological variable series, the time-varying regional variance is proposed to check the variation of a river basin for the first time. As a study case, the method is applied to four deseasonalized daily discharge series from the middle reach of Yangtze River, China. Through the analyses of the conditional variance in different streamflow series, it is concluded that: (1) The ARCH effect exists in all the studied series which means the stream processes is nonstationary in terms of the variance; (2) The variations of time-varying variances are similar for the series from adjacent hydrological stations, and the similarity degree increases from upstream to downstream; (3) The regional variance is time-varying and can be used for further regional research. 相似文献
12.
混合门限自回归模型在长期水文预报中的应用 总被引:5,自引:0,他引:5
在长期水文预报中主要应用两类方法,它们是多元分析法和时间序列分析法,这两种方法各有其局限性;应用混合门限自回归模型模拟水文时间序列可以考虑诸多因素,能避免以上两方法的局限性,并在实践中得到检验. 相似文献
13.
14.
Abstract In non-structural measurement of flood control, hydrologic forecasting plays a very important role. Owing to time-variance, non-linearity, and uncertainty of hydrological processes, realtime forecasting has become an efficient approach. The paper addresses an important practical problem: improving short-term hydrological forecasting based on real-time updating in the operation model. A simple nonlinear model with a variable gain parameter (VGPM) is developed. A separated calibration approach for updating parameters used in the runoff generation process and the response function in the flow routing is proposed. State space equations associated with updating model parameters in a real time scheme were developed. The VGPM approach is verified for three types of representative watersheds. The performances of different updating schemes in rainfall-runoff modeling and real-time forecasting were tested. The results indicate that significant improvement in the efficiency of hydrological modeling can be obtained from the VGPM approach, relative to simple linear models (SLM). For the watersheds with a time-variant characteristic, moreover, significant improvement in the hydrologic forecasting efficiency can be obtained by adaptive schemes. The efficiency of real-time modeling by the self-adaptive Kalman Filtering algorithm was found to be very close to that of the Recursive Least-Square method. 相似文献
15.
16.
Restoration of a constructed stormwater wetland to improve its ecological and hydrological performance. 总被引:1,自引:0,他引:1
Although the vegetation within constructed stormwater wetlands plays an important role in the treatment processes taking place, its density and distribution depends on the wetland bathymetry and the imposed hydrologic regime. This paper describes an ecological and hydrological assessment of a constructed stormwater treatment wetland over a 5 year period. This assessment included the use of a continuous simulation hydrologic model combined with a Digital Elevation Model of the wetland bathymetry, plus a time series of vegetation maps. The combined spatial and temporal analysis indicates that both the frequency and duration of inundation has affected the fate of vegetation throughout the wetland. Restoration strategies have also been investigated to improve the survival of vegetation within the wetland. 相似文献
17.
应用门限自回归(TAR)模型建立了同时受潮汐和径流双重影响的长江下游感潮河段高桥水文站月水位TAR预测模型,建模过程中运用遗传算法来实现模型参数的优化.计算结果显示,门限自回归模型可以拟合感潮河段的非线性特性,拟合及预测精度均满足水文预报规范要求,遗传算法的引入简化了建模过程,提高了模型的预测精度并保证了其预测性能的稳定性.研究结果表明用遗传门限自回归模型预测感潮河段的水位是可行的,该模型在感潮河段其他水文要素的非线性时序预测中也具有广泛的实用价值. 相似文献
18.
水文时间序列突变点分析的启发式分割方法 总被引:3,自引:1,他引:3
水文时间序列突变点的检测是水文统计分析中的一项重要内容,但由于水文时间序列的非线性、非平稳性,导致对检测水文时间序列突变点的技术要求较高。建立了时间序列变点分析的启发式分割数学模型,以此来分析水文时间序列均值的突变。以理想的水文时间序列为仿真基础,结合合水水库的月径流序列,采用启发式分割算法得出去除周期成分后的月径流序列的突变点,以此证明了启发式分割算法检测水文时间序列突变的可行性和优越性。结果表明:4个突变点中,3个突变点集中于水库建成后不久,说明人类活动的剧烈影响为主要驱动力,加剧了对入库月径流量的影响,造成月径流的频繁波动和数次突变点的产生。 相似文献