Runoff Analysis for a Small Watershed of Tono Area Japan by Back Propagation Artificial Neural Network with Seasonal Data

The rainfall runoff (R-R) process was studied for two small sub-basins having different sizes in a mountainous catchment of Tono area Japan. The runoff and other meterological data have been collected in this catchment for the last 14 years. The major objective of this study was to construct numerical models for these sub-basins to predict runoff after 1/2 and 1 h. The effects of season and the size of the catchment on R-R process were also investigated. The hydrogeological conditions of the catchment were studied prior to the analyses. The data obtained for summer (rainy) and winter (dry) seasons were treated separately in order to study the seasonal effects on the model development. The back propagation artificial neural network technique (BPANN) and the multivariate autoregressive and moving average models (ARMA) were adopted for the analysis. It was found that for very small catchments the seasonal effects are dominant and therefore separate models should be developed for each season to obtain better forecasting estimates. It was also found that the predictions by BPANN models were better than multivariate ARMA models for intense rains having complex R-R relationships in summer. On the other hand, both the modelling techniques yielded almost similar results for smaller rains in winter. It was also found clearly that the accuracy of prediction decreased with the increase of the time period for prediction.     

Evaluation of Artificial Neural Network Techniques for Municipal Water Consumption Modeling

Various Artificial Neural Network techniques such as Generalized Regression Neural Networks (GRNN), Feed Forward Neural Networks (FFNN) and Radial Basis Neural Networks (RBNN) have been evaluated based on their performance in forecasting monthly water consumptions from several socio-economic and climatic factors, which affect water use. The data set including total 108 data records is divided into two subsets, training and testing. The models consisting of the combination of the independent variables are constructed and the best fit input structure is investigated. The performance of ANN models in training and testing stages are compared with the observed water consumption values to identify the best fit forecasting model. For this purpose, some performance criteria such as Normalized Root Mean Square Error (NRMSE), efficiency (E) and correlation coefficient (CORR) are calculated for all models. The best fit models are also trained and tested by Multiple Linear Regression (MLR). The results indicated that GRNN outperforms all other methods in modeling monthly water consumptions.     

人工神经网络在我国流域水沙分析预测中的应用研究进展

从人工神经网络独特的特点和类型入手,详细地介绍了人工神经网络模型在我国流域水沙分析预测研究中的应用情况．并对其在研究中存在的问题和以后的研究方向作了探讨。     

人工神经网络在水电机组故障诊断中的应用研究

水电设备状态检修的关键是状态监测和故障诊断。人工神经网络具有分布并行，容错性和记忆功能等特点，用人工神经网络方法进行故障诊断具有明显优势，通过对发电机设备故障诊断的具体应用，证明此方法是有效可行的。     

人工神经网络在水利工程方案综合评价中的应用

方案优选是目前水利工程中普遍面临的问题,但由于方案中指标的多样性和不确定性,综合评价结果略有偏差。引入人工神经网络模型对水利工程方案进行优选。结果证明,此模型能自适应确定权重,具有评价结果合理、客观、精度高等特点,能够有效地用于水利工程方案的综合评价。     

长距离输水系统的神经网络模型研究

常规特征线算法在仿真模拟长距离有压输水系统流量调节过渡过程时，存在计算时间较长等不足，故无法很好地满足实时控制和优化调度的需要．本文利用人工神经网络(ANN)所具有的高度非线性全局作用、良好的自适应性及联想记忆功能等优点，将人工神经网络引入到过渡过程的分析中，建立合适的神经网络模型，并通过该模型得出一系列计算成果．该成果与传统算法所得的结果相比较表明，人工神经网络在过渡流的分析中是可行的，并具有明显的优势．     

深基坑支护结构位移的神经网络预测

针对深基坑系统的复杂性和变形非线性,将人工神经网络技术引入其中。在分析讨论了人工神经网络中应用最为广泛的BP网络基本原理的基础上,建立了基坑变形预测的神经网络模型,并应用Bayesian方法对实例加以论证。研究表明,神经网络是解决基坑变形预测的有效方法之一。     

灯泡贯流式机组轴承温度的人工神经网络建模探讨

本文在现场监测数据的基础上应用人工神经网络方法,建立了灯泡贯流式机组轴承温度的数学模型,并与实际的监测数据进行了对比分析。对比分析表明,所建立的人工神经网络模型,具有较高的计算精度,能够在已知机组工作参数的情况下,快速求出机组轴承的温度,该模型可以用于机组的计算机辅助运行系统开发中。     

灯泡贯流式机组轴承温度的人工神经网络建模探讨

本文在现场监测数据的基础上应用人工神经网络方法,建立了灯泡贯流式机组轴承温度的数学模型,并与实际的监测数据进行了对比分析。对比分析表明,所建立的人工神经网络模型,具有较高的计算精度,能够在已知机组工作参数的情况下,快速求出机组轴承的温度,该模型可以用于机组的计算机辅助运行系统开发中。     

Finite Element Method and GIS Based Distributed Model for Soil Erosion and Sediment Yield in a Watershed

The objective of this study is to develop a soil erosion and sediment yield model based on the kinematic wave approximation using the finite element method, remote sensing and geographical information system (GIS) for calculating the soil erosion and sediment yield in a watershed. Detachment of soil particles by overland flow occurs when the shear stress at the surface overcomes the gravitational forces and cohesive forces on the particles. Deposition occurs when the sediment load is greater than the transport capacity. Beasley et al.’s (Trans ASAE 23:938–944, 1980) transport equations for laminar and turbulent flow conditions are used to calculate the transport capacity. The model is capable of handling distributed information about land use, slope, soil and Manning’s roughness. The model is applied to the Catsop watershed in the Netherlands and the Harsul watershed in India. Remotely sensed data has been used to extract land use/land cover map of the Harsul watershed, and other thematic maps are generated using the GIS. The simulated results for both calibration and validation events are compared with the observed data for the watersheds and found to be reasonable. Statistical evaluation of model performance has been carried out. Further, a sensitivity analysis has also been carried out to study the effect of variation in model parameter values on computed volume of sediment, peak sediment and the time to peak sediment. Sensitivity analysis has also been carried out for grid size variation and time step variation of the Catsop watershed. The proposed model is useful in predicting the hydrographs and sedigraphs in the agricultural watersheds.     

基于人工神经网络的钢筋混凝土受弯构件正截面抗裂预测

钢筋混凝土构件抗烈度预测受多种条件因素的影响,现有的方法由试验实测数据建立的数学模型误差较大,因此有必要寻求一种精度较高的方法进行抗烈度预测。通过实测试验数据,训练形成一个三层BP网络,其中78组数据作为学习样本,另外9组数据则作为测试样本,建立了人工神经网络预测钢筋混凝土正截面抗裂性能的方法,还对其他模型抗裂性能的计算值与实测值进行了比较。该方法预测值与试验值吻合良好。结果表明,提出的人工神经网络预测钢筋混凝土正截面抗裂度预测方法具有对直接参与训练的数据仿真效果好,整体预测精度高,与理论分析得出的结论基本一致,可用于受弯构件抗烈度预测。