文昌13-1-A9井酸化后含水上升分析及预测

文昌13-1-A9井酸化后综合含水达到54%,综合地质资料和油藏生产动态分析认为,导致文昌13-1-A9井酸化后含水上升加快的主要因素有:油水边界上升加快;酸化后渗透率提高,使油层与水体连通能力增强,加剧了边水的突进速度;酸化后采液指数提高,压降增大,导致底水锥进;油井位于油水过渡带内。在分析含水上升原因的基础上利用logistic模型对含水进行了预测,预测结果认为到2008年含水将达到80%以上。     

马尔可夫链在产品可靠性预测中的应用

事物未来的状态仅仅受事物现状的影响，而与过去的状态无关，也就是具有马尔可夫性。用马尔可夫链的理论与方法，对具有马尔可夫性产品的可靠性进行预测，既为产品的可靠性设计和产品售后服务的经济性分析提供了参考，也为马尔可夫模型的实际应用拓广了范围。通过实例说明在数学软件的帮助下这种预测方法的简单可行性。     

对我国石油可采资源量的预测研究

回顾了世界上现行的可采资源量预测方法,总结分析了在我国石油勘探开发过程中大型油气田的发现规律及储采比变化规律,探讨了累积发现可采储量、累积产量与可采资源量三者之间的内在关系及储采比的变化对它们的影响.在此基础上,利用累积发现的可采储量与累积产量建立了预测可采资源量的有效方法———全程历史拟合法和储采比递减阶段曲线拟合法.利用上述两种方法预测出中国石油的经济可采资源量为107.8×10⁸t,次经济的可采储量约为(27～36)×10⁸t,还对部分油区的可采资源量进行了预测.预测结果表明,两种方法对比使用,能获得较好的预测效果.     

改进BP网络在航材需求预测中的应用

针对航材备件需求预测问题,在对影响航材备件需求量的多个因素进行分析研究的基础上,运用改进BP神经网络算法进行预测的仿真实验。实验结果表明,改进BP神经网络能够对积累的历史数据进行充分的应用,并且有较高的预测准确性。     

Improvement of rainwater infiltration and storage capacity by an enhanced seepage well: From laboratory investigation to HYDRUS-2D numerical analysis

Seepage well is an emerging Low Impact Development (LID) technology that can effectively control the storm runoff. However, its rainwater infiltration rate and storage capacity still require further enhancement. By setting a horizontal infiltration structure at the bottom of conventional rainwater seepage well (CSW), an enhanced seepage well (ESW) was proposed in this study, and its infiltration performances compared with the permeable pavement (PP) and the CSW were systemically investigated using static infiltration experiment and HYDRUS-2D simulation. The results showed that the infiltration efficiency of ESW was significantly higher than that of PP and CSW, and the process of water infiltrated through soil mainly controlled the macroscopic infiltration rate. The Nash-Sutcliff Efficient (NSE) index was used to evaluate the accuracy and reliability of the HYDRUS-2D model, and the results of NSE values greater than 0.75 (varied between 0.75 and 0.91) confirmed the applicability of HYDRUS-2D to describe correctly the hydraulic behavior of the ESW system. Simulation infiltration tests showed that the ESW performed a higher average infiltration rate and fewer total runoff volume than the CSW, indicating the effectively enhancement of the infiltration and water retention capacity of ESW, especially under heavy rainfall intensities. Additionally, the ESW system exhibited an excellent runoff-control and rainwater retention capacity in an actual rainfall scenario.     

Stock Price Forecasting: An Echo State Network Approach

Forecasting stock prices using deep learning models suffers from problems such as low accuracy, slow convergence, and complex network structures. This study developed an echo state network (ESN) model to mitigate such problems. We compared our ESN with a long short-term memory (LSTM) network by forecasting the stock data of Kweichow Moutai, a leading enterprise in China’s liquor industry. By analyzing data for 120, 240, and 300 days, we generated forecast data for the next 40, 80, and 100 days, respectively, using both ESN and LSTM. In terms of accuracy, ESN had the unique advantage of capturing nonlinear data. Mean absolute error (MAE) was used to present the accuracy results. The MAEs of the data forecast by ESN were 0.024, 0.024, and 0.025, which were, respectively, 0.065, 0.007, and 0.009 less than those of LSTM. In terms of convergence, ESN has a reservoir state-space structure, which makes it perform faster than other models. Root-mean-square error (RMSE) was used to present the convergence time. In our experiment, the RMSEs of ESN were 0.22, 0.27, and 0.26, which were, respectively, 0.08, 0.01, and 0.12 less than those of LSTM. In terms of network structure, ESN consists only of input, reservoir, and output spaces, making it a much simpler model than the others. The proposed ESN was found to be an effective model that, compared to others, converges faster, forecasts more accurately, and builds time-series analyses more easily.     

Two new characteristic parameters for runoff computation

A method capable of estimating the hydrograph from a prescribed storm for a practical mild slope upstream catchment is proposed. This method makes use of two new characteristic parameters, andS, in conjunction with the kinematic wave equation to compute lateral inflows of the main stream of the catchment. The depth profile of overland flow at any instant within the catchment and hydrograph at any location can be easily found. Lag times for individual lateral inflows are then considered and are linearly combined to obtain the hydrograph at the outlet of the catchment or depth profile of the main stream at any instant. The validity of the excess rainfall-surface runoff linear relationship in this study has also been verified with Tatsunokuchiyama catchment, and it shows good results for this computed runoff.     

人工神经网络技术在电工产品可靠性实验中的应用(理论部分)

本文简要介绍了人工神经网络，BP神经网络的结构、训练和学习规则以及人工神经网络非线性时间序列预测的应用概况。     

降水和植被变化对龙川江径流量的影响

基于龙川江大峡谷进出口水文控制站楚雄站和小黄瓜园站的长系列水文资料和流域森林盖度变化资料,结合小波分析方法,对降水、植被与径流量变化间相互关系以及径流量各时间尺度准周期变化的本质和形成机制进行了研究。结果表明:研究区降水量序列存在2 a、4 a和14.5 a左右的主周期,径流量序列具有2 a、4 a、6 a和22 a左右的主周期;其中径流量低频振荡所反映的是总径流中来自于壤中水径流和存在于裂隙中的地下水径流的准周期变化,而高频振荡反映的是总径流中来自于地面径流的准周期变化;径流量的年内变化和年际高频振荡主要是由降水量变化所引起的,而径流量序列6 a和22 a的主周期是由土壤和裂隙对降水量序列4 a和14.5 a的主周期滞后放大作用所形成的,径流量序列4 a的主周期是由干热河谷特殊自然地理环境所形成的;径流量与降水量间相互关系在1969年左右发生了突变