Comparison of Machine Learning Models Performance on Simulating Reservoir Outflow: A Case Study of Two Reservoirs in Illinois,U.S.A.

Reservoir outflow is an important variable for understanding hydrological processes and water resource management. Natural streamflow variation, in addition to the streamflow regulation provided by dams and reservoirs, can make streamflow difficult to understand and predict. This makes them a challenge to accurately simulate hydrologic processes at a daily scale. In this study, three Machine Learning (ML) algorithms, Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN), were examined and compared to model reservoir outflow. Past, current, and future hydrologic and meteorological data were used as model inputs, and the outflow of next day was used as prediction. Simulation results demonstrated that all three models can reasonably simulate reservoir outflow. For Carlyle Lake, the coefficient of determination and Nash–Sutcliffe efficiency were each close to one for the three models. The coefficient of determination, relative mean bias, and root mean square error indicated that the SVM performed better than the RF and ANN, but the SVM output displayed a larger relative mean bias than that from RF and ANN. For Lake Shelbyville, the ANN model performed better than RF and SVM when considering the coefficient of determination, Nash–Sutcliffe efficiency, relative mean bias, and root mean square error. The study results demonstrate that the three ML algorithms (RF, SVM, and ANN) are all promising tools for simulating reservoir outflow. Both the accuracy and efficacy of the three ML algorithms are considered to support practitioners in planning reservoir management.     

REAL CODED GENETIC ALGORITHM OPTIMIZATION OF LONG TERM RESERVOIR OPERATION1

ABSTRACT: An optimization and simulation model holds promise as an efficient and robust method for long term reservoir operation, an increasingly important facet of managing water resources. Recently, genetic algorithms have been demonstrated to be highly effective optimization methods. According to previous studies, a real coded genetic algorithm (RGA) has many advantages over a binary coded genetic algorithm. Accordingly, this work applies an RGA to obtain the 10‐day (the traditional period of reservoir operation in Taiwan) operating rule curves for the proposed reservoir system. The RGA is combined with an effective and flexible scheme for coding the reservoir rule curves and applied to an important reservoir in Taiwan, considering a water reservoir development scenario to the year 2021. Each rule curve is evaluated using a complex simulation model to determine a performance index for a given flow series. The process of generating and evaluating decision parameters is repeated until no further improvement in performance is obtained. Many experiments were performed to determine the suitable RGA components, including macro evolutionary (ME) selection and blend‐α crossover. Macro evolution (ME) can be applied to prevent the premature problem of the conventional selection scheme of genetic algorithm. The purpose of adjusting a of a crossover scheme is to determine the exploratory or exploitative degree of various subpopulations. The appropriate rule curve searched by an RGA can minimize the water deficit and maintain the high water level of the reservoir. The results also show that the most promising RGA for this problem consists of these revised operators significantly improves the performance of a system. It is also very efficient for optimizing other highly nonlinear systems.     

万峰湖总有机碳空间分布特征

于2012年9月对万峰湖(水库)的总有机碳(TOC)空间分布特征进行研究,结果显示:TOC含量范围为0.41~3.94 mg/L。TOC在纵向分布上的差异不显著,在垂向分布上,总体呈现出上层中层下层的趋势。TOC在垂向上的分布差异主要源于万峰湖属于深水水库,不同深度水层因光照和温度差异所导致生物量的多寡是致使万峰湖出现上层TOC浓度明显高于中层和下层的主要因素。     

官厅水库重金属污染状况调查研究

为确保北京市饮用水源安全,2007年对官厅水库中重金属进行了筛选性调查.采用电感耦合等离子体质谱仪(ICP-MS),对官厅水库枯水期和丰水期水体中的金属元素含量进行了全面的分析.监测结果表明,水库在丰水期砷含量偏高,整体水质状况良好,重金属污染程度轻微.     

丹江口库区及其上游流域水质污染特征及评价

根据丹江口库区及其上游河流5个区域42个断面2012-2013年的水质监测数据，采用主成分分析法确定主要污染因子及权重，对不同流域的水质进行综合评价。第一主成分包括总氮、溶解氧、五日生化需氧量、总磷、氨氮、高锰酸盐指数，第二主成分为氟化物、粪大肠菌群，第三主成分为化学需氧量；其权重分别为5．022，2．256，1．508。评价结果表明，湖北十堰市和丹江口市流域水环境污染相对较重，其次为河南南阳市、陕西商洛市、陕西安康市以及陕西汉中市流域。     

汶川8.0级地震绵阳市游仙区水库震害调查与分析

5.12汶川特大地震给四川省水库造成了严重的震害和巨大的经济损失。依据绵阳市水库地震灾害的现场科学考察资料,对绵阳市游仙区44座高危险情以上的水库土坝的震损现象做了初步总结和分析。发现这些震损水库基本上都建于1950~1980年,未经抗震设计,典型震害现象包括水库坝顶中部及坝轴线两侧纵向裂缝、坝体纵横向裂缝和坝坡局部滑裂等,并有水库泄水构筑物震损、防浪墙倒塌、上游坡护坡下滑、坝坡挡土墙倒塌等。这说明对小型水库土坝进行抗震设防是必要的,今后应深入研究小型水库土坝的震害机制,以指导水库土坝的抗震设计。