扬压力测压管水位预测的稳健回归方法

本文结合献县枢纽的测压管实际观测资料，采用稳健回归方法建立了闸基扬压力测压管水位与闸前水位之间的线性关系，并与传统的最小二乘线性回归进行了对比分析。结果表明，对于测压管水位与闸前水位关系而言，稳健回归方法精度高、抵抗数据污染的能力强，优于传统的最小二乘回归。     

偏最小二乘回归在大坝位移资料分析中的应用

介绍了偏最小二乘回归的基本原理、建模基本思路和交叉有效性判别方法，编制了偏最小二乘回归程序；指出偏最小二乘回归可以很好地应用于大坝位移监测中。通过算例分析表明：建立的模型有很好的拟合和预报功能。     

偏最小二乘回归在大坝位移资料分析中的应用

介绍了偏最小二乘回归的基本原理、建模基础思路和交叉有效性判别方法，编制了偏最小二乘回归程序；指出偏最小二乘回归可以很好地应用于大坝位移监测中。通过算例分析表明：建立的模型有很好的拟合和预报功能。     

大坝安全监控模型因子相关性及不确定性研究

针对最小二乘法难以克服因子多重共线性对回归模型精度影响的不足，本文对大坝安全监控模型因子间的相关性及其不确定性进行了研究。引进偏最小二乘法，对大坝安全监测变量及其影响因子进行偏最小二乘回归分析，将建模预测分析方法与非模型式的数据内涵分析有机结合，可同时实现回归建模、数据结构简化以及因子相关的不确定性分析，所建立的大坝安全监控模型，其精度可通过交叉有效性检验来控制。工程应用实例和模型对比分析研究表明，偏最小二乘回归模型能有效克服各类因子变量间的多重共线性对模型拟合精度及其预测能力的影响，因而比目前常用的最小二乘回归模型更具广泛适用性。     

五强溪水电站大坝变形预测模型研究

基于逐步回归法、偏最小二乘回归法和长短期记忆(LSTM)循环神经网络，构建了五强溪水电站大坝变形预测模型。采用拉伊特准则确定可靠的监测数据，基于可靠的监测数据，构建考虑水压、温度、时效因素的混凝土重力坝变形预测逐步回归和偏最小二乘回归模型，根据五强溪大坝坝顶J23测点2006年～2020年的监测资料获得该测点的沉陷曲线逐步回归和偏最小二乘回归预测模型。根据数值试验，选定的LSTM模型包括2个LSTM层，激活函数采用整流线性单元函数，输入序列长度为20。训练集数据取2006年～2017年的监测值，2018年～2020年的监测数据作为测试集数据。采用随机搜索对LSTM循环神经网络的超参数进行优化。比较3种模型结果可知：3种模型在沉降曲线的预测效果均较好；偏最小二乘回归法能合理地解释各分量；训练数据足够时，LSTM循环神经网络的预测精度非常高；采用偏最小二乘法回归模型或LSTM模型预测J23测点变形更为妥当。     

基于PLSR的静态灰色模型在大坝安全监控中的应用

本文将偏最小二乘回归运用到静态灰色模型中，消除了因子间多重共线性的影响，建立了一种新型的大坝安全监控模型，并对黑河水库的应变资料进行了分析。研究分析表明，基于偏最小二乘回归的静态灰色模型优于传统的静态灰色模型，这为大坝安全监控提供了一种新的思路和研究方法。     

递阶偏最小二乘回归在大坝安全监测中的应用

偏最小二乘回归能有效地消除因子间的多重相关性,但从其算法特点和实际应用来看,也存在不足.例如,在算法方面,偏最小二乘提取的主成分不一定能同时保证方差和相关程度最大;在应用方面,含有较多自变量的偏最小二乘回归模型的可解释性不高.递阶偏最小二乘回归是偏最小二乘回归后续研究的成果之一,在一定程度上克服了上述不足.算例表明,递阶偏最小二乘回归模型较其他回归模型的可解释性强,较为合理.     

将偏最小二乘回归模型应用于非点源负荷预测

介绍了偏最小二乘回归的基本原理,建模思路和方法,将偏最小二乘回归模型应用于泾河流域非点源污染年负荷量预算,对计算结果的代表性和有效性进行了分析,并将其与最小二乘的多元回归模型预测结果进行了对比.实例计算分析结果表明,偏最小二乘回归分析对于反映因变量与多个相关性自变量之间的关系有较高的精度.     

偏最小二乘法回归的建模思路及其在水库大坝监测中的应用分析

常规最小二乘法回归的不足之处是:难以有效识别和消除自变量因子间的多重相关性影响;然而偏最小二乘法回归模型却能够有效消除因子相关性对模型回归系数估计和回归分析效果的影响。文章介绍了偏最小二乘法回归基本原理和建模思路,并结合水库大坝监测实例分析了偏最小二乘法回归,实例表明,偏最小二乘法分离效果更好,反演结果精度更高,能满足对大坝安全监控的要求,在水利工程安全监测及有关数据的统计分析方面具有广阔的应用前景。     

基于新偏最小二乘回归法的系列水文资料分析

在实际问题中,经常会碰到海量数据或者样本点较少,自变量较多的数据。对此可以利用递阶偏最小二乘回归来建立线性模型。但是一个直接的问题是如何对自变量进行分组。由此提出了基于聚类分析的递阶偏最小二乘回归方法,在对解释变量分组时引入聚类分析。通过对长江宜昌段水沙观测数据作实证分析后发现,基于聚类分析的递阶偏最小二乘回归方法是有效可行的,而且用该方法建立的回归模型比一般的偏最小二乘回归模型拟合能力更强。     

An effective method to predict oil recovery in high water cut stage

The water flooding characteristic curve method based on the traditional regression equation between the oil and water phase permeability ratio and the water saturation is inappropriate to predict the oil recovery in the high water cut stage. Hence, a new water flooding characteristic curve equation adapted to the high water cut stage is proposed to predict the oil recovery. The water drive phase permeability experiments show that the curve of the oil and water phase permeability ratio vs. the water saturation, in the semi-logarithmic coordinates, has a significantly lower bend after entering the high water cut stage, so the water flooding characteristic curve method based on the traditional regression equation between the oil and water phase permeability ratio and the water saturation is inappropriate to predict the oil recovery in the high water cut stage; therefore, a new water flooding characteristic curve equation based on a better relationship between ln( k_(ro)/k_(rw)) and S w is urgently desirable to be established to effectively and reliably predict the oil recovery of a water drive reservoir adapted to a high water cut stage. In this paper, by carrying out the water drive phase permeability experiments, a new mathematical model between the oil and water phase permeability ratio and the water saturation is established, with the regression analysis method and an integration of the established model, the water flooding characteristic curve equation adapted to a high water cut stage is obtained. Using the new water flooding characteristic curve to predict the oil recovery of the GD3-block of the SL oilfield and the J09-block of the DG oilfield in China, results with high predicted accuracy are obtained.     

基于 Embedding-GRU 的水库水位预测模型

为解决因水库数据采集设备能力有限、水文数据不全导致预测水库水位时预测精度较低的问题，以四岭水 库每小时水位监测数据为例，提出基于嵌入式-门控循环单元（Embedding-gated?recurrent?unit，Embedding-GRU）的 水库水位预测模型，即利用 Embedding 方法将单维降雨量数据升维至多维数据，扩大降雨的气候特征，结合 GRU 算法进行水库水位预测。将该模型与传统深度学习算法长短期记忆（long?short-term?memory，LSTM）、门控循环单 元（gated?recurrent?unit，GRU）、双向门控循环单元（bidirectional?recurrent?neural?network，BiGRU）这 3 种模型对比， 结果显示：Embedding-GRU 模型的预测效果均优于其他传统模型，平均绝对误差 EMA和均方根误差 ERMS分别平均 下降 19.6% 和 7.7%，并且在预测次日水库水位的应用场景中决定系数 R2能够达到 0.989?37。结果表明：该模型耦 合多种算法，扩大单变量的气候特征，具有较高预测精度和泛化能力。相较传统模型，基于 Embedding-GRU 的水 库水位预测模型能够对缺少温度、气压、风速、蒸发量等监测数据的水库进行可靠度较高的预测，适用水库范围 更广，为水库日常运维、除险加固提供参考。     

基于偏最小二乘回归和最小二乘支持向量机的大坝渗流监控模型

利用偏最小二乘回归法对影响大坝渗流的诸多因素进行分析,提取对因变量影响强的成分,克服了变量间的多重相关性问题,降低了最小二乘支持向量机的输入维数,从而可以较好的解决非线性问题,建立了基于PLS-LSSVM的大坝渗流监控模型。实例分析表明,PLS-LSSVM模型的拟合与预测精度均优于独立使用PLS或LSSVM建模的精度;PLS-LSSVM模型的学习训练效率比LSSVM模型有较大的优势,更适合于大规模的数据建模。     

某土石坝帷幕灌浆渗流监测成果分析

为研究土石坝帷幕灌浆的渗控效果,了解加固后渗压水位的变化规律,针对某黏土心墙坝,采用逐步回归分析和有限元分析方法,对3年来的监测成果进行了分析。逐步回归分析表明:影响坝体渗流场的主要因素是库水位、降雨及时效;往背水方向,库水位的影响程度逐渐减弱,而降雨及时效的影响程度逐渐增强。有限元计算表明:全封闭式帷幕灌浆使大坝渗漏量减小了37.6%,而悬挂式帷幕灌浆只减小了10.7%,全封闭式帷幕灌浆的渗控效果更为显著。     

Forecasting Hourly Intermittent Rainfall by Deep Belief Networks with Simple Exponential Smoothing

Accurate rainfall forecasting is essential in planning and managing water resource systems efficiently. However, intermittent rainfall patterns increase the difficulty of accurately forecasting rainfall values. Deep learning techniques have recently been popular and powerful in forecasting. Thus, this study employed deep belief networks with a simple exponential smoothing procedure (DBNSES) to forecast hourly intermittent rainfall values in Taiwan. Weather factors were used as independent variables to forecast rainfall volume. The simple exponential smoothing data preprocessing procedure was used to deal with the intermittent data patterns. The other three forecasting models, namely the least squares support vector regression (LSSVR), the generalized regression neural network (GRNN), and the backpropagation neural network (BPNN), were employed to forecast rainfall using the same data sets. In addition, genetic algorithms were utilized to determine the parameters of four forecasting models. The empirical results indicate that the developed DBNSES models are superior to the other forecasting models in terms of forecasting accuracy. In addition, the DBNSES can obtain smaller values of RMSE than those in the previous studies. Therefore, the DBNSES model is a suitable and effective way of forecasting rainfall with intermittent data patterns.

     

基于 1D CNN-LSTM 模型的红旗水库水位序列模拟

为了提升水库水位模拟的精度,通过 1D CNN-LSTM 模型与五种常用的机器学习模型对安徽省红旗水库历史水位数据和降雨量数据实现未来 7 天的水位模拟并进行对比验证。 CNN 和 LSTM 能够表现出比较好的模拟性能,结合两种模型的优势能够更加显著的提升模型的模拟效果;1D CNN-LSTM 具有较高鲁棒性,对于未来 3 天以内水位模拟都有较好的预测效果和精度,虽然 3 天以后的模拟效果有明显下降,但对未来第 7 天的模拟 NSE 和 KGE 依然能够达到 0.8 以上,在不发生极端天气的情况下,模型对于水位趋势的模拟依然具有相当的参考价值。1D CNN-LSTM 模型对于红旗水库的水位模拟优于其他五种传统的机器学习模型,并具有相当高的精度。     

基于SVM的河道洪峰水位校正预报方法

针对平原河网地区河道洪峰水位预报中经验模型可靠性不足的问题,提出一种基于SVM的河道洪峰水位校正预报方法。采用谱系聚类法对历史洪水过程数据按降雨特性分类,选择与预报降雨过程最接近的历史数据训练预报模型;采用PCA对输入数据降维以提取有效特征;基于支持向量回归机建立河道洪峰水位预报模型;采用滚动模式对洪峰水位预报,每小时根据最新水位以及降水信息预报未来洪峰水位,不断提高预报精度。通过对京杭运河拱宸桥站的洪峰水位实例预测验证了该研究方法的有效性。     

基于GAMLSS-GLO及随机森林算法的土石坝渗流监测模型

根据土石坝渗流原型观测,厘清水位、降水等因素的影响,模拟大坝渗流的真实状态,合理评估其渗流监测结果,是土石坝安全监控亟待解决的关键问题。基于此,根据数理统计原理,采用随机森林算法构建无降水条件下渗流量与上下游水位的回归模型;考虑渗流量受前期累积降水的综合影响,引入广义可加模型（GAMLSS-GLO）,模拟降水影响下土石坝渗流监测值的波动区间,并将其与渗流-水位回归模型叠加,预测土石坝渗流监测的可靠区间;最后,将该方法应用于糯扎渡心墙堆石坝的渗流监测。结果表明：所提模型方法对渗流的水位、降水响应表现出良好的适用性,显著提高了渗流模拟预测质量。同时求解了渗流量置信区间,有利于土石坝的运行工况判断及安全监控。     

降雨条件下岩溶槽谷泉水的水文地球化学特征

在示踪试验判断岩溶含水介质形态的基础上，通过观测降雨条件下姜家泉泉水的电导率、pH值、水温、泉口实时水位、降雨量以及降雨的pH值、钙离子浓度、电导率和计算出的泉水方解石饱和指数及CO2分压，分析了降雨条件下岩溶槽谷泉水文地球化学行为的动态变化。研究结果表明在以管道为主的岩溶含水介质槽谷中，泉水水位随降雨暴涨暴落，其水温、电导率、pH值、方解石饱和指数和CO2分压对降雨响应迅速，能敏锐地感应和反馈环境的变化。酸雨将可能引起泉水pH值的降低和电导率的升高，因而应重视酸雨对岩溶泉水文地球化学行为和岩溶含水介质演化过程的影响。降雨期间雨水将地表土壤带入地下河，引起泉水浊度的升高，表明此时泉水可能受到微生物的污染。     

Short-Term Water Demand Forecast Modelling at IIT Kanpur Using Artificial Neural Networks

The efficient operation and management of an existing water supply system require short-term water demand forecasts as inputs. Conventionally, regression and time series analysis have been employed in modelling short-term water demand forecasts. The relatively new technique of artificial neural networks has been proposed as an efficient tool for modelling and forecasting in recent years. The primary objective of this study is to investigate the relatively new technique of artificial neural networks for use in forecasting short-term water demand at the Indian Institute of Technology, Kanpur. Other techniques investigated in this study include regression and time series analysis for comparison purposes. The secondary objective of this study is to investigate the validity of the following two hypotheses: 1) the short-term water demand process at the Indian Institute of Technology, Kanpur campus is a dynamic process mainly driven by the maximum air temperature and interrupted by rainfall occurrences, and 2) occurrence of rainfall is a more significant variable than the rainfall amount itself in modelling the short-term water demand forecasts. The data employed in this study consist of weekly water demand at the Indian Institute of Technology, Kanpur campus, and total weekly rainfall and weekly average maximum air temperature from the City of Kanpur, India. Six different artificial neural network models, five regression models, and two time series models have been developed and compared. The artificial neural network models consistently outperformed the regression and time series models developed in this study. An average absolute error in forecasting of 2.41% was achieved from the best artificial neural network model, which also showed the best correlation between the modelled and targeted water demands. It has been found that the water demand at the Indian Institute of Technology, Kanpur campus is better correlated with the rainfall occurrence rather than the amount of rainfall itself.