Artificial Intelligence Technique in Hydrological Forecasts Supporting for Water Resources Management of a Large River Basin in Vietnam

Hydrological forecasting plays an important role in water resource management, supporting socio-economic development and managing water-related risks in river basins. There are many flow forecasting techniques that have been developed several centuries ago, ranging from physical models, physics-based models, conceptual models, and data-driven models. Recently, Artificial Intelligence (AI) has become an advanced technique applied as an effective data-driven model in hydrological forecasting. The main advantage of these models is that they give results with compatible accuracy, and require short computation time, thus increasing forecasting time and reducing human and financial effort. This study evaluates the applicability of machine learning and deep learning in Hanoi water level forecasting where it is controlled for flood management and water supply in the Red River Delta, Vietnam. Accordingly, SANN (machine learning algorithm) and LSTM (deep learning algorithm) were tested and compared with a Physics-Based Model (PBM) for the Red River Delta. The results show that SANN and LSTM give high accuracy. The R-squared coefficient is greater than 0.8, the mean squared error (MSE) is less than 20 cm, the correlation coefficient of the forecast hydrology is greater than 0.9 and the level of assurance of the forecast plan ranges from 80% to 90% in both cases. In addition, the calculation time is much reduced compared to the requirement of PBM, which is its limitation in hydrological forecasting for large river basins such as the Red River in Vietnam. Therefore, SANN and LSTM are expected to help increase lead time, thereby supporting water resource management for sustainable development and management of water-related risks in the Red River Delta.

Water quality forecast through application of BP neural network at Yuqiao reservoir

This paper deals with the study of a water quality forecast model through application of BP neural network technique and GUI (Graphical User Interfaces) function of MATLAB at Yuqiao reservoir in Tianjin. To overcome the shortcomings of traditional BP algorithm as being slow to converge and easy to reach extreme minimum value,the model adopts LM (Leven-berg-Marquardt) algorithm to achieve a higher speed and a lower error rate. When factors affecting the study object are identified,the reservoir's 2005 measured values are used as sample data to test the model. The number of neurons and the type of transfer functions in the hidden layer of the neural network are changed from time to time to achieve the best forecast results. Through simulation testing the model shows high efficiency in forecasting the water quality of the reservoir.

Gray comprehensive assessment and optimal selection of water consumption forecasting model

A comprehensive assessing method based on the principle of the gray system theory and gray relational grade analysis was put forward to optimize water consumption forecasting models. The method provides a better accuracy for the assessment and the optimal selection of the water consumption forecasting models. The results show that the forecasting model built on this comprehensive assessing method presents better self-adaptability and accuracy in forecasting.

System Dynamics Approach to Urban Water Demand Forecasting—A Case Study of Tianjin

A system dynamics approach to urban water demand forecasting was developed based on the analysis of urban water resources system, which was characterized by multi-feedback and nonlinear interactions among sys-tem elements. As an example, Tianjin water resources system dynamic model was set up to forecast water resources demand of the planning years. The practical verification showed that the relative error was lower than 10%. Fur-thermore, through the comparison and analysis of the simulation results under different development modes pre-sented in this paper, the forecasting results of the water resources demand of Tianjin was achieved based on sustain-able utilization strategy of water resources.

The Water-Bearing Numerical Model and Its Operational Forecasting Experiments PartII: The Operational Forecasting Experiments

おhe water-bearing numerical model is undergone all round examinations during the operational forecasting experiments from 1994 to 1996. A lot of difficult problems arising from the model′s water-bearing are successfully resolved in these experiments through developing and using a series of technical measures. The operational forecasting running of the water-bearing numerical model is realized stably and reliably, and satisfactory forecasts are obtained.

The Application of a Grey Markov Model to Forecasting Annual Maximum Water Levels at Hydrological Stations

Compared with traditional real-time forecasting,this paper proposes a Grey Markov Model(GMM) to forecast the maximum water levels at hydrological stations in the estuary area.The GMM combines the Grey System and Markov theory into a higher precision model.The GMM takes advantage of the Grey System to predict the trend values and uses the Markov theory to forecast fluctuation values,and thus gives forecast results involving two aspects of information.The procedure for forecasting annul maximum water levels with the GMM contains five main steps:1) establish the GM(1,1) model based on the data series;2) estimate the trend values;3) establish a Markov Model based on relative error series;4) modify the relative errors caused in step 2,and then obtain the relative errors of the second order estimation;5) compare the results with measured data and estimate the accuracy.The historical water level records(from 1960 to 1992) at Yuqiao Hydrological Station in the estuary area of the Haihe River near Tianjin,China are utilized to calibrate and verify the proposed model according to the above steps.Every 25 years' data are regarded as a hydro-sequence.Eight groups of simulated results show reasonable agreement between the predicted values and the measured data.The GMM is also applied to the 10 other hydrological stations in the same estuary.The forecast results for all of the hydrological stations are good or acceptable.The feasibility and effectiveness of this new forecasting model have been proved in this paper.

Water inflow forecasting for tunnel considering nonlinear variation of permeability coefficient

To assess the water inflow which is more suitable to the actual conditions of tunnel,an empirical correlation about the permeability coefficient changing with depth is introduced.Supposing that the surrounding rock is heterogeneous isotropy,the formula for calculating water inflow of tunnel with the nonlinear variation of permeability coefficient is deduced.By the contrast analysis with the existing formulas,the presented method has the similar value to them;moreover,the presented method has more simple form and easy to use.Due to parameter analysis,the water inflow decreases after considering the nonlinear variation of permeability coefficient.When the attenuation coefficient a>0,the water inflow increases first till reaches the maximum at a certain depth,then decreases and is close to 0 finally if deep enough.Thus,it is better to keep away from the certain depth where it is with the maximum water inflow for safe operation and economical construction,and reduce the water damage.Based on the analysis,the radius of tunnel has less impact on the amount of water inflow,and the water inflow just increases by 6.7% when the radius of tunnel increases by 1 m.

Considering the Influence of Multi-weather-factors to Forecasting the Water Requirement of Well Irrigation Rice Based on ANN Model

The author considered the influences of several weather factors, such as air temperature, sunlight, saturation deficiency, wind speed and so on to forecasting the water requirement of well irrigation rice based on Artificial Neutron Network. Through dealing with the time series of water requirement and its influence factors, the author applied the multi-dimension data correlation analysis to ensure the net structure. Thus, the ANN model to forecast the water requirement of well irrigation rice has been built. By means of the ANN model, uncertainty relation between water requirement and many influence factors among the interior and exterior can be discovered. The results of ANN model is good, and can provide some references for establishing the water saving irrigation system.

The Water-Bearing Numerical Model and Its Operational Forecasting Experiments Part I: The Water-Bearing Numerical Model

In first paper of articles, the physical and calculating schemes of the water-bearing numerical model are described. The model is developed by bearing all species of hydrometeors in a conventional numerical model in which the dynamic framework of hydrostatic equilibrium is taken. The main contributions are: the mixing ratios of all species of hydrometeors are added as the prognostic variables of model, the prognostic equations of these hydrometeors are introduced, the cloud physical framework is specially designed, some technical measures are used to resolve a series of physical, mathematical and computational problems arising from water-bearing; and so on. The various problems (in such aspects as the designs of physical and calculating schemes and the composition of computational programme) which are exposed in feasibility test, in sensibility test, and especially in operational forecasting experiments are successfully resolved using a lot of technical measures having been developed from researches and tests. Finally, the operational forecasting running of the water-bearing numerical model and its forecasting system is realized stably and reliably, and the fine forecasts are obtained. All of these mentioned above will be described in second paper.

Synthetic Reconstruction of Water Demand Time Series for Real Time Demand Forecasting

The forecasting of the demand applied to water supply systems has been an important tool to realize time control. The use of the time series to do the forecasting of the demand is the main way that has been used by researchers. By this way, the need of a complete time demand series increases. This work presents two ways to reconstruct the water demand time series synthetically, using the Average Reconstruction Method and Fourier Method. Both the methods were considered interesting to do the synthetic reconstruction and able to complete the time series, but the Fourier Method showed better results and a better fitness to approximation of the water consumption pattern.

Forecasting of water yield of deep-buried iron mine in Yanzhou, Shandong

This paper compares analytical and numerical methods by taking the forecasting of water yield of deep-buried iron mine in Yanzhou, Shandong as an example. Regarding the analytical method, the equation of infinite and bilateral water inflow boundary is used to forecast the water yield, and in the case of numerical simulation, we employed the GMS software to establish a model and further to forecast the water yield. On the one hand, through applying the analytical method, the maximum water yield of mine 1 500 m deep below the surface was calculated to be 13 645.17 m3/d; on the other hand, through adopting the numerical method, we obtained the predicted result of 3 816.16 m3/d. Meanwhile, by using the boundary generalization in the above-mentioned two methods, and through a comparative analysis of the actual hydro-geological conditions in this deep-buried mine, which also concerns the advantages and disadvantages of the two methods respectively, this paper draws the conclusion that the analytical method is only applicable in ideal conditions, but numerical method is eligible to be used in complex hydro-geological conditions. Therefore, it is more applicable to employ the numerical method to forecast water yield of deep-buried iron mine in Yanzhou, Shandong.

Nerual Network Expert System and Their Application to Forecasting Water Invasion of Colliery

In this paper, we propose a formal definition, general structure and work principle of the Neural Network Expert System (NNES) based on joint-type knowledge representation, and show a practical application example using NNES for forecasting the water invasion of coal mine.

Determination of the Real Dotation and Forecast of the Potable Water System in the Huancavelica City, Peru

The research entitled,“Determination of the Real Dotation and Forecast of the Potable Water System in the Huancavelica city-Peru”,has as objective to determine and analyze the actual endowment and prognosis of the drinking water system in the city of Huancavelica.The city of Huancavelica has a population of approximately 50,000.00 inhabitants and 10,680.00 users of drinking water and sewerage according to the source Companies Providing This Service,the same one that is located in the western chain and central saw range of Perúand belongs to a cold climate.The sample was extracted through Companies Providing This Service EMAPA Huancavelica SA(Anonymous Society),which corresponds to 1,296 data from the consumption of water in cubic meters of each dwelling located in the various categories of each sector.The results obtained from the research indicate that the category that consumed more water during the period 2004-2018 was the domestic category,the same one that had an average of 1,152,123 cubic meters,this is due to the population growth of each year.On the other side,the category that consumed the least water was the social category,the average consumption of which was 11,903.47 cubic meters.In addition,the greatest variability in water consumption is in the state category,the total variation being equal to 21.6%.In the Huancavelica city,water consumption has a growing trend and to predict the volume of drinking water consumption by 2030,water consumption was transformed by first differences,using the ARIMA(Autoregressive Integrated Moving Average)model for total consumption model(0-1-1)(2-0-0)-12 and for the domestic consumption the ARIMA model(0-1-2)(0-0-2)-12.On the other hand,water consumption was analyzed on a monthly basis in which there is greater variability for the months June-6 and December-12.It was taken into account for the calculation of the endowment the domestic consumption,having values ranging from 6.45 L/inhab/day until 1,460 L/inhab/day,which reflects a considerable variation to consider.The average value in the spring station is 109.44 L/inhab/day,in the summer station it is 116.93 L/inhab/day,in the autumn station it is 117.17 L/inhab/day and in the winter station it is 108.48 L/inhab/day.Also,the average minimum endowment equal to 7.66 L/inhab/day was obtained,the average maximum endowment is of 569.03 L/inhab/day and the actual endowment 113.01 L/inhab/day.Finally,it is concluded that the estimated average actual endowment for the Huancavelica city,that is to say,113.01 L/inhab/day,is much less than that indicated in the Norma OS.100 of the Basic Considerations of Sanitary Infrastructure Design of the RNE(Reglamento Nacional de Edificaciones),the same one that mentions the endowment equal to 180 L/inhab/day for cold climates,which means a reduction of costs in the dimensioning of hydraulic structures and sanitation and drinking water works.

Research on Forecasting Water Requirement of Well Irrigation Rice by Time Series Analysis Method

The paper builds up the forecasting model of air temperature according to the data (1994～1998) of Fu Jin area.At the same time,the writer inquires into the relation of water requirement of well irrigation rice (ET) and average air temperature (T).Furthermore,the rice irrigation water requirement (ET) of Fu Jin area has been forecast in 1999.Thus,we can apply the model in irrigation management.

皖南农村地区需水量特征分析及预测模型研究

需水量的准确预测对于提高农村地区水资源利用率、优化农饮水厂日常调度有着重要作用,然而现有预测算法面临农村部分地区可用历史数据少、质量差等问题,同时部分预测算法的计算负荷较高,无法有效适应农饮水厂运维环境。选取2021-2022年安徽省南部不同地区3处农饮水厂日供水量数据作为研究对象。经过特征分析发现,皖南农村地区日需水量存在季节性变化趋势明显而短期内波动较大的特点。针对该数据特点及实际计算需求,选取了使用网格搜索法优化后的ARIMA模型进行预测,并与SVR、LSTM、RF等模型进行对比分析。研究结果表明,使用网格搜索法的ARIMA模型可以更好地适应农饮水厂运维环境,利用其对于周期性、季节性数据较强的学习能力,对农饮水厂供水量变化趋势和规律进行准确预测,具有一定的普适性,且计算负荷明显优于其他方法。模型对比分析显示,ARIMA模型的预测精度最高,在3处农饮水厂测试集数据上的平均绝对百分比误差(MAPE)分别为1.827、1.454、2.714。

城市内涝预警预报系统研发及应用

城市内涝问题具有成因多元、影响面广的特点,是基础性社会问题。分析了城市内涝的成因,提出通过信息化手段提升城市内涝灾害应对能力。构建基于内涝实时物联感知、数据汇聚、数学模型预报、大数据分析、水利数字孪生的城市内涝预警预报系统,并在宁波市进行了应用。通过搭建“天基”监测预警、“落地雨”监测预警、内涝预报预警、内涝成因分析、内涝影响评估等功能模块,实现了城区内涝的精准模拟、风险图的在线绘制以及内涝损失的高效评估,有力支撑了城市暴雨内涝灾害预报预警,提升了城市内涝风险管控能力。

智慧灌溉大数据管理平台设计与应用

以粗放型为主的漫灌方式浪费水资源,探索高效节水农业灌溉意义重大。本研究研制了集数据采集、数据管理、数据可视化、分析决策和远程管理为一体的智慧灌溉大数据管理平台,旨在以数字化管理与智能化控制方式提升水资源利用效率。平台利用嵌入式、物联网、互联网、3S(GIS、GPS、RS)等技术,采取“1+1+N”的设计模式,构建作物需水预报模型与灌溉决策模型,基于B/S构架和Java语言,设计了1个灌溉数据中心、1个灌溉数据管理系统、4个应用系统,打造了智慧灌溉大数据管理平台。平台在河北、河南、山东、江苏等地区设有多个示范基地,汇聚了8个科研小组、24个试验基地的有关多种作物的生长、灌溉、气象、土壤等数据,平均采集数据18829条/天,帮助管理人员摸清家底;集成了多个团队的软件系统和62套物联网设备,及时、定量地呈现了农作物生长状况及环境状态,实现了农作区动态监测,助力生产决策;生成农作物需水预报和灌溉决策方案,完成了远程灌溉目标,并且经过实地试验,验证了自动灌溉的有效性,将灌溉水有效利用系数最高提升了31%以上;形成了大田产业、温室中心、数字科研和区域灌溉等不同专题可视化布局,满足多种农业场景应用需求。为农业生产和跨区域管理提供了便捷工具,为农业灌溉数字化系统搭建和应用提供了参考。