Application of artificial neural networks for water quality prediction

The term “water quality” is used to describe the condition of water, including its chemical, physical, and biological characteristics. Modeling water quality parameters is a very important aspect in the analysis of any aquatic systems. Prediction of surface water quality is required for proper management of the river basin so that adequate measure can be taken to keep pollution within permissible limits. Accurate prediction of future phenomena is the life blood of optimal water resources management. The artificial neural network is a new technique with a flexible mathematical structure that is capable of identifying complex non-linear relationships between input and output data when compared to other classical modeling techniques. Johor River Basin located in Johor state, Malaysia, which is significantly degrading due to human activities and development along the river. Accordingly, it is very important to implement and adopt a water quality prediction model that can provide a powerful tool to implement better water resource management. Several modeling methods have been applied in this research including: linear regression models (LRM), multilayer perceptron neural networks and radial basis function neural networks (RBF-NN). The results showed that the use of neural networks and more specifically RBF-NN models can describe the behavior of water quality parameters more accurately than linear regression models. In addition, we observed that the RBF finds a solution faster than the MLP and is the most accurate and most reliable tool in terms of processing large amounts of non-linear, non-parametric data.

     

基于多源数据整合的河湖水质智能应用研究

为改善上海市河道水环境,针对现有河湖水质存在的问题,依靠信息化和数字化技术,汇聚各类水质相关数据,开展数据整合与分析,推出河道水质变化三级报警预警,建立相应业务处置机制和流程,探索河湖水质智能化应用,重点研究河湖水质数据监测、变化趋势分析、波动报警预警、整改处置和结果反馈等5个环节闭环管理.通过河湖水质智能应用研究,并在上海市河长制办公室工作平台和城市运行"一网统管"水务专题中进行实际应用.实际应用表明:河湖水质智能应用可有效发挥作用,强化对河湖水质恶化等问题的持续跟踪和有效监督,提高河湖水质治理效率,推进河湖协同治理,有效改善河湖水环境质量.     

A mixing criterion for turbid rivers

A criterion for the formation/destruction of thermal stratification in turbid rivers is developed and tested against an extensive data set in a weir pool on the Murrumbidgee River, Australia. The criterion estimates whether a river section would stratify or not, subject to a prescribed heat flux through the water surface and a prescribed river flow. It can be determined from readily accessible parameters, including solar radiation, turbidity, wind speed and river discharge.It is suggested that this simple mixing criterion be used as a flow management tool in a variety of water quality applications in rivers.     

Selection of genetic algorithm operators for river water quality model calibration

Model calibration is the most important step in the overall model development. Genetic algorithm (GA) was used in this study to optimise model parameters of river water quality models. In general, the efficiency of using GA depends on the proper selection of GA operators, which are the components that make up the overall GA process. A comprehensive investigation on the importance of GA operators on model parameter optimisation was conducted in this study (and presented in this paper) based on the hypothesis that the selection of GA operators depends on the sensitivity of model parameters to model output. A numerical experiment on GA operators was conducted first considering a hypothetical river network water quality model with both insensitive and sensitive model parameters and were later validated using Yarra River Water Quality Model (YRWQM). It was found that a robust GA operator set obtained from the literature was capable of achieving a near-optimum model parameter set for the sensitive model but not for the insensitive model. However, due to insensitivity of water quality model parameters on model output, the ‘not so’ near-optimum parameter set did not contribute a great difference to the overall water quality predictions in the insensitive model. Therefore, based on these numerical experiments, it was concluded that a GA operator set obtained from the literature is adequate for calibration of model parameters of river water quality model using GA.     

Immediate water quality assessment in shrimp culture using fuzzy inference systems

The continuous monitoring of physical, chemical and biological parameters in shrimp culture is an important activity for detecting potential crisis that can be harmful for the organisms. Water quality can be assessed through toxicological tests evaluated directly from water quality parameters involved in the ecosystem; these tests provide an indicator about the water quality. The aim of this study is to develop a fuzzy inference system based on a reasoning process, which involves aquaculture criteria established by official organizations and researchers for assessing water quality by analyzing the main factors that affect a shrimp ecosystem. We propose to organize the water quality parameters in groups according to their importance; these groups are defined as daily, weekly and by request monitoring. Additionally, we introduce an analytic hierarchy process to define priorities for more critical water quality parameters and groups. The proposed system analyzes the most important parameters in shrimp culture, detects potential negative situations and provides a new water quality index (WQI), which describes the general status of the water quality as excellent, good, regular and poor. The Canadian water quality and other well-known hydrological indices are used to compare the water quality parameters of the shrimp water farm. Results show that WQI index has a better performance than other indices giving a more accurate assessment because the proposed fuzzy inference system integrates all environmental behaviors giving as result a complete score. This fuzzy inference system emerges as an appropriated tool for assessing site performance, providing assistance to improve production through contingency actions in polluted ponds.     

Safe river water: A ubiquitous and collaborative water quality monitoring solution

Water quality is vital to human life and economy. However, one sixth of the world’s population suffers from lack of safe drinking and domestic water. Aiming to improve the capability of predicting and responding to river pollution disasters, this project collaborated with local offices of Chinese National Bureau of Water Resource to explore new solutions to coping with the ever-growing threat of river water pollution. We presented a distributed data analysis algorithm, Infinitesimal Dividing and Analysis, to efficiently locate pollution sources with data gathered from a ubiquitous wired/wireless sensor network. We elaborate on a π-calculus based paradigm to enhance collaboration and interaction among individual monitoring stations. Based on these two enabling technologies, we applied our framework to water quality monitoring at two carefully chosen sites in China.     

Embedded processor based automated assessment of quality of the water in an IoT background

For the life of human beings, water is necessary. While nearly 70% of the earth is drained, just 3% are known to be fresh water. Moreover, approximately 2.6% of the cooling water is not accessible to people. They are either trapped in glaciers and polar ice caps, contained in the soil or water, heavily poisoned, or unnecessarily drained below the Earth's surface. So only 0.4% of the drinkable water in the world is shared by the 7 billion inhabitants. Fresh water is therefore a valuable resource to be regulated and properly maintained. Only 80% of liquid fresh water should not be available to the public in many developing countries. By exponentially increasing the population of India, Fresh Water management in the farming, manufacturing and other fields is much more critical because of water requirements. “Physical, biological and chemical” parameters can be analyzed in determining the Fresh Water Quality. The scientists will manually carry out traditional water quality sampling. Nonetheless, this method takes a little bit of time and is cost-effective. Now, the IoT technology is used to track, capture and analyze the data in various fields of research. In this paper, we design a low-cost system to achieve water value in an IOT environment. The system consists of several sensors used for the calculation of chemical and physical water parameters. The machine learning algorithm has also been used to forecast water quality based on its data set, which were learned from a number of water samples. This system was modeled by using the U D00* 86 Ultra and Teensy++2.0 data processors at low budgets.     

Water quality mapping from Landsat digital data

Landsat multispectral scanner (MSS) data, and U-2 colour and colour infrared photographs were combined with in situ data for the assessment of water quality parameters within the San Francisco Bay-Delta. The water quality parameters of interest included turbidity and suspended solids. The U-2 photography and water quality samples were obtained simultaneously and coincidently with Landsat overpass. Regression models were developed between each of the water quality parameter measurements and Landsat digital data for 29 pre-selected sample sites. These regression models were then extended to the entire study area for mapping the water quality parameters of interest. The results included a series of colour-coded maps, each pertaining to one of the water quality parameters, and the statistical summaries. Areas of relatively high biological activity were clearly discernible on digitally enhanced Landsat MSS data.     

城市排污系统对河流水质影响的仿真研究

以城市废水系统的集成化管理为出发点，基于国际水协会河流水质1号模型(RWOMI)，确定了适合仿真与实时控制的简化的动态河流水质模型，在MATLAB仿真环境下建立了一个以河流为核心的集成化城市废水系统的简单仿真环境。以某河流城区段为背景，采用国际水协会发布的污水负荷和组分数据作为原始输入，模拟了排水管网和污水处理厂在不同运行条件下引起的河流水质的变化情况。仿真结果表明排水管网组合下水溢流和污水处理厂排放物的处理程度对河流水质会产生直接影响，为集成化管理提供了依据。     

Wavelet-based multiresolution analysis for data cleaning and its application to water quality management systems

Data cleaning techniques are useful for extracting desirable knowledge or interesting patterns from existing databases in engineering applications. The major problems of conventional techniques (e.g., Fourier Transformation Technique) are that they are (1) more appropriate in linear systems than nonlinear systems, and (2) stringently depend on state space functions. In this study a wavelet-based multiresolution analysis technique (WMAT) is proposed for reducing noises induced by complex uncertainty. The approach is applied to a river water quality simulation system for showing its practicability in data cleaning and parameter estimation. Clean data are prepared through running a Thomas’ river water quality model and polluted data are synthesized by mixing clean data with white Gaussian noises. The results show that WMAT will not distort the clean data, and can effectively reduce the noise in the polluted data. The data denoised by WMAT are furthermore used for estimating the modeling parameters. It is also indicated that the parameters estimated with the denoised data through WMAT are much closer to real values than those (1) with polluted data through WMAT and (2) with data through Fourier analysis technique. It is thus recommended that the prepared data be used for estimating the modeling parameters until being cleaned with WMAT.     

Using Radial Basis Function Neural Networks to identify river water data parameters

The complex conditions of water dynamics create a challenge in selecting an appropriate neuron structure for artificial neural networks to simulate real river parameters. This study proposes an identification model based on Radial Basis Function (RBF) Neural Networks. We applied this identification model to river water quality parameters with different neuron node size scenarios to test network structure characters. Simulation results reveal that the RBF Neural Networks model achieves convergence through neuron iterations and the simulation error is well controlled within a small margin. The adjusting effect is closely related to structure design and the neuron updating strategy.     

Assessment of surface water quality using multivariate statistical techniques: A case study of the Fuji river basin,Japan

Multivariate statistical techniques, such as cluster analysis (CA), principal component analysis (PCA), factor analysis (FA) and discriminant analysis (DA), were applied for the evaluation of temporal/spatial variations and the interpretation of a large complex water quality data set of the Fuji river basin, generated during 8 years (1995–2002) monitoring of 12 parameters at 13 different sites (14 976 observations). Hierarchical cluster analysis grouped 13 sampling sites into three clusters, i.e., relatively less polluted (LP), medium polluted (MP) and highly polluted (HP) sites, based on the similarity of water quality characteristics. Factor analysis/principal component analysis, applied to the data sets of the three different groups obtained from cluster analysis, resulted in five, five and three latent factors explaining 73.18, 77.61 and 65.39% of the total variance in water quality data sets of LP, MP and HP areas, respectively. The varifactors obtained from factor analysis indicate that the parameters responsible for water quality variations are mainly related to discharge and temperature (natural), organic pollution (point source: domestic wastewater) in relatively less polluted areas; organic pollution (point source: domestic wastewater) and nutrients (non-point sources: agriculture and orchard plantations) in medium polluted areas; and organic pollution and nutrients (point sources: domestic wastewater, wastewater treatment plants and industries) in highly polluted areas in the basin. Discriminant analysis gave the best results for both spatial and temporal analysis. It provided an important data reduction as it uses only six parameters (discharge, temperature, dissolved oxygen, biochemical oxygen demand, electrical conductivity and nitrate nitrogen), affording more than 85% correct assignations in temporal analysis, and seven parameters (discharge, temperature, biochemical oxygen demand, pH, electrical conductivity, nitrate nitrogen and ammonical nitrogen), affording more than 81% correct assignations in spatial analysis, of three different sampling sites of the basin. Therefore, DA allowed a reduction in the dimensionality of the large data set, delineating a few indicator parameters responsible for large variations in water quality. Thus, this study illustrates the usefulness of multivariate statistical techniques for analysis and interpretation of complex data sets, and in water quality assessment, identification of pollution sources/factors and understanding temporal/spatial variations in water quality for effective river water quality management.     

OpenMI-based integrated sediment transport modelling of the river Zenne,Belgium

Recent observations show that the river Zenne (Belgium) remains well below the water quality goals stated by the European Union Water Framework Directive. An interuniversity, multidisciplinary research project was therefore launched to evaluate the effects of wastewater management plans on the ecological functioning of the river. To this end, different water quantity and quality processes had to be considered and modelled, e.g., the hydrology in the river basin, hydraulics in the river and sewers, erosion and sediment transport, faecal bacteria transport and decay. This paper considers the development of an Open Modelling Interface (OpenMI) based integrated model for the purpose of simulating the river's sediment dynamics. We used the Soil and Water Assessment Tool (SWAT) to model water and sediment fluxes from rural areas. The Storm Water Management Model (SWMM) was used to simulate the hydraulics of the river, canal, and sewer systems in urban catchments. New model codes for sediment transport and stream water temperature were developed to complement SWMM. The results show that the integrated sediment transport model reproduced the sediment concentrations in the river Zenne with ‘good’ to ‘satisfactory’ accuracy. We may therefore conclude that the OpenMI has been successfully implemented to integrate water quality models into a hydraulic one. While the OpenMI run-time data communication inflicted calculation time overhead, we found that the overhead was not significant with respect to the total run-time of the integrated model.     

宁波市推进河网水资源智能化调度的思路与探索

为改善宁波市河网水环境,基于现有的工程措施,从智能化非工程措施角度,探究平原河网水资源智能化调度的可行方案。根据宁波市中心城区平原河网基本特点及闸泵工程分布情况,分析宁波市近些年在水资源智能化管理调度方面的建设情况及存在问题,以满足河流生态需水量为前提,从监测、模拟、调度方案的形成及调度执行4个层面,提出推进河网水资源智能化调度的思路及实现途径,并对每一个层面的实际可行性进行分析。通过建立宁波市中心城区河网水质水量联合调度模型,结合重要水功能区与河湖断面的水质实时监测数据,对河道水体进行纳污容量分析,并使用水质模型结合调度方案模拟水污染演化进程、影响范围、持续时间,为河流水量调配提供决策支持,为类似区域的水质改善提供借鉴。     

基于无线传感器网络的水质监测系统设计

使用传统的有线水质监测系统进行水环境污染检测时,存在监测点数量多、监测时间长等问题。为此,提出一种基于无线传感器网络的水质监测系统。通过无线传感器节点对被监测水域进行水质参数的数据采集,将采集到的数据经过Zigbee网络进行汇总及处理,并经过GPRS网络及时地远程传送给监管部门,从而实现对河流水质情况的实时、有效的监督和管理。对水质监测系统的软硬件电路设计进行介绍,并实现对系统的软硬件连调。实验结果证明,该系统能够满足组网要求,可较好地应用于水质监测领域。     

Development and application of an integrated ecological modelling framework to analyze the impact of wastewater discharges on the ecological water quality of rivers

Modelling is an effective tool to investigate the ecological state of water resources. In developing countries, the impact of sanitation infrastructures (e.g. wastewater treatment plants) is typically assessed considering the achievement of legal physicochemical quality standards, but ignoring the ecological water quality (EWQ) of the receiving river. In this paper, we developed a generic integrated ecological modelling framework quantifying the impact of wastewater discharges on the EWQ of the Cauca river (Colombia). The framework is flexible enough to be used in conjunction with different approaches/models and integrates a hydraulic and physicochemical water quality model with aquatic ecological models. Two types of ecological models were developed, habitat suitability models for selected macroinvertebrate groups and ecological assessment models based on a macroinvertebrate biotic index. Four pollution control scenarios were tested. It was found that the foreseen investments in sanitation infrastructure will lead to modest improvements of the EWQ, with an increase lower than six units of the ecological index BMWP-Colombia. Advanced investments, such as the collection and treatment of all wastewater produced by the cities of Cali, Yumbo and Palmira and upgrading of the treatment systems should be considered to achieve a good EWQ. The results show that the integration of ecological models in hydraulic and physicochemical water quality models (e.g. MIKE 11) has an added value for decision support in river management and water policy. The integration of models is a key aspect for the success in environmental decision making. The main limitation of this approach is the availability of physicochemical, hydraulic and biological data that are collected simultaneously. Therefore, a change in the river monitoring strategy towards collection of data which include simultaneous measurements of these variables is required.     

Development of an integrated model for assessing the impact of diffuse and point source pollution on coastal waters

Pollution loads from land sources and their impact on the receiving waters can be predicted by using land-use and surface water quality models, respectively. In this paper details are given of the development of an integrated modelling system for managing water quality in coastal basins. The system includes a model linking the pollution input loads to land-use and a model for simulating surface flow and chemical and biological processes. The land-use model consists of two main components: a database and a model base. It uses a GIS system, namely ArcGIS, as the model platform. The database, built in the Geo-database format, includes six data sets containing information on social, economical, bathymetrical and hydrological aspects of a study area. The model base includes a series of transfer functions that link the pollution loads to key sources, including those from: industrial, domestic and agricultural inputs, and fish and animal farming. The water quality model is a numerical model which solves the governing equations representing the physical, chemical and biological processes in coastal waters. The modelling system was set up for Bohai Bay in China, with the land-use model being used to provide nutrient loadings from individual inputs and the water quality model used to predict the impact of these inputs on the receiving waters. Model simulations were also undertaken to predict the water quality conditions for various development and management scenarios.     

Cyclic control of water quality in a tidal river segment

This paper suggests a simple method for maintaining a desired level of dissolved oxygen in a tidal river. The scheme takes advantage of the short time periodic tidal nature of the river in the utilization of the ability of the river to assimilate biological oxygen demand loading. The parameters of the river model are determined on-line using a Powell search to process actual river data. After model parameter identification, the Powell search is again applied to find the parameters of the control effort to be applied to the cyclic control of reaeration effort and BOD discharge into the stream. Identification of the model parameters requires measurements of only BOD and DO. Because the system is designed to keep the model up-dated in the short term sense, it also implicitly accounts for the usual long term influences. The modelling procedure was tested against data obtained from the Penobscot River in Maine, and the control strategies were tested on the river model simulation.     

河道水质监测系统设计

河道通过水闸而形成类似于内湖等的闸控河道,为了充分发挥水闸和河道的运行效益,同时利用闸门调度控制该河道水位水质的变化,实现水环境的优化,设计一种基于物联网的水环境监测系统。系统由水环境监测子系统、闸门调度子系统、系统管理平台和支撑层组成。采取定时轮循检测,并将所得数据导入数据中心。闸门调度系统从数据中心导出数据,通过计算得到调度方案。     

水质监测FPGA嵌入式软件系统设计

设计了一种水质智能检测嵌入式软件系统,该系统以反熔丝数模混合FPGA FUSION系列嵌入的8051IP核为核心,子站能够自动监测水质常规5项参数,将参数无线数传到上位机,在上位机完成数据处理、分析和显示。该系统具有可靠性高、稳定性好与成本低的优点,满足了水产养殖业与河流污染监测的需要。