人工神经网络法预测城市用水量

城市用水量的预测结果,对于城市规划、供水系统的管理及改扩建有着重要的意义,寻求科学合理的预测模型是保障预测结果准确可靠的关键。针对这一问题,利用人工神经网络理论建立了BP（Back—Propagation,反向传播算法）网络预测模型,该模型考虑了反映社会、经济的两个影响因素人口和工业产值对用水量需求的影响,具备系统决策功能。通过实例证明该模型是一种行之有效的用水量预测模型。     

Estimating monthly total nitrogen concentration in streams by using artificial neural network

Artificial Neural Network (ANN) is a flexible and popular tool for predicting the non-linear behavior in the environmental system. Here, the feed-forward ANN model was used to investigate the relationship among the land use, fertilizer, and hydrometerological conditions in 59 river basins over Japan and then applied to estimate the monthly river total nitrogen concentration (TNC). It was shown by the sensitivity analysis, that precipitation, temperature, river discharge, forest area and urban area have high relationships with TNC. The ANN structure having eight inputs and one hidden layer with seven nodes gives the best estimate of TNC. The 1:1 scatter plots of predicted versus measured TNC were closely aligned and provided coefficients of errors of 0.98 and 0.93 for ANNs calibration and validation, respectively. From the results obtained, the ANN model gave satisfactory predictions of stream TNC and appears to be a useful tool for prediction of TNC in Japanese streams. It indicates that the ANN model was able to provide accurate estimates of nitrogen concentration in streams. Its application to such environmental data will encourage further studies on prediction of stream TNC in ungauged rivers and provide a useful tool for water resource and environment managers to obtain a quick preliminary assessment of TNC variations.     

Modeling Flow and Sediment Transport in a River System Using an Artificial Neural Network

A river system is a network of intertwining channels and tributaries, where interacting flow and sediment transport processes are complex and floods may frequently occur. In water resources management of a complex system of rivers, it is important that instream discharges and sediments being carried by streamflow are correctly predicted. In this study, a model for predicting flow and sediment transport in a river system is developed by incorporating flow and sediment mass conservation equations into an artificial neural network (ANN), using actual river network to design the ANN architecture, and expanding hydrological applications of the ANN modeling technique to sediment yield predictions. The ANN river system model is applied to modeling daily discharges and annual sediment discharges in the Jingjiang reach of the Yangtze River and Dongting Lake, China. By the comparison of calculated and observed data, it is demonstrated that the ANN technique is a powerful tool for real-time prediction of flow and sediment transport in a complex network of rivers. A significant advantage of applying the ANN technique to model flow and sediment phenomena is the minimum data requirements for topographical and morphometric information without significant loss of model accuracy. The methodology and results presented show that it is possible to integrate fundamental physical principles into a data-driven modeling technique and to use a natural system for ANN construction. This approach may increase model performance and interpretability while at the same time making the model more understandable to the engineering community.     

Use of artificial neural network black-box modeling for the prediction of wastewater treatment plants performance

A reliable model for any wastewater treatment plant is essential in order to provide a tool for predicting its performance and to form a basis for controlling the operation of the process. This would minimize the operation costs and assess the stability of environmental balance. This process is complex and attains a high degree of nonlinearity due to the presence of bio-organic constituents that are difficult to model using mechanistic approaches. Predicting the plant operational parameters using conventional experimental techniques is also a time consuming step and is an obstacle in the way of efficient control of such processes. In this work, an artificial neural network (ANN) black-box modeling approach was used to acquire the knowledge base of a real wastewater plant and then used as a process model. The study signifies that the ANNs are capable of capturing the plant operation characteristics with a good degree of accuracy. A computer model is developed that incorporates the trained ANN plant model. The developed program is implemented and validated using plant-scale data obtained from a local wastewater treatment plant, namely the Doha West wastewater treatment plant (WWTP). It is used as a valuable performance assessment tool for plant operators and decision makers. The ANN model provided accurate predictions of the effluent stream, in terms of biological oxygen demand (BOD), chemical oxygen demand (COD) and total suspended solids (TSS) when using COD as an input in the crude supply stream. It can be said that the ANN predictions based on three crude supply inputs together, namely BOD, COD and TSS, resulted in better ANN predictions when using only one crude supply input. Graphical user interface representation of the ANN for the Doha West WWTP data is performed and presented.     

Multi-scale analysis of oxygen demand trends in an urbanizing Oregon watershed, USA

Human alteration of the landscape has an extensive influence on the biogeochemical processes that drive oxygen cycling in streams. We estimated trends from the mid-1990s to 2003, using the seasonal Mann-Kendall's test, for percent saturation dissolved oxygen (DO), chemical oxygen demand (COD), total Kjeldahl nitrogen (TKN), and ammonia-nitrogen (NH(3)-N) for 12 sites in the Rock Creek watershed, northwest Oregon, USA. In order to understand the influence of landscape change, scale, and stormwater runoff management on dissolved oxygen trends, we calculated land cover change through aerial photo interpretation at full-basin, local (near sample point) basin, and 100m stream buffer scales, for the years 1994 and 2000. Significant (p < or = 0.05) trends occurred in DO (increasing at five sites), COD (decreasing at seven sites), TKN (decreasing at five sites, increasing at one site), and NH(3)-N (decreasing at one site, increasing at one site). Significant land cover change occurred in agricultural land cover (-8% for the entire basin area) and residential land cover (+10% for the entire basin area) (p < or = 0.05). Correlation results indicated that: (1) forest cover negatively influenced COD at the full basin scale and positively influences NH(3)-N at local scales, (2) residential land cover influenced oxygen demand variables at local scales, (3) agricultural land cover did not influence oxygen demand, (4) local topography negatively influenced TKN and NH(3)-N, and (5) stormwater runoff management infrastructure correlated positively with COD at the local scale. This study indicates that landscape factors influencing DO conditions for the study streams act at multiple scales, suggesting that better knowledge of scale-process interactions can guide watershed managers' decision making in order to maintain improving water quality conditions.     

Analyzing the factors affecting environmental risks of projects using a hybrid approach of DEMATEL‐ANP,artificial neural network: A case study

The increasing growth of the economy in each country necessitates a great amount of investment in infrastructure. The belief that projects involve various uncertainties, such as technical skills, management quality, and the like, indicates that most projects fail to achieve their aims, interests, costs, as well as their timeframes and space requirements. As the environment can pose significant uncertainty to any project, environmental risks should be deeply studied by project management departments. This study intends to analyze as a case the environmental risk management system within a consulting firm. From this analysis, each aspect of a project's environmental risk management is ranked using a fuzzy analytical network process (ANP), a neural network algorithm, and a decision‐making trial and evaluation laboratory (DEMATEL) methodology. From the organizational aspect, budget risk is the most significant. From the technical aspect, the risk of regulations is the most important one. Finally, the risk of project failure from poor communication is another identified main risk in this research. By studying high‐ranking items in this hierarchy, it can be understood that these criteria exist in different aspects; therefore, all aspects of the risk should be taken into account to cover and assess risk. A neural network algorithm for validating and reassessment of ranking is employed. Results of this application showed that, based on Spearman's rank correlation method, two different approaches resulted in similar rankings. Finally, some practical implications for responding to the most highly ranked risks are proposed.     

Landsat ETM+ Images in the Estimation of Seasonal Lake Water Quality in Boreal River Basins

We investigated the use of Landsat ETM+ images in the monitoring of turbidity, colored dissolved organic matter (CDOM), and Secchi disk transparency (Z(SD)) in lakes of two river basins located in southern Finland. The ETM+ images were acquired in May, June, and September 2002 and were corrected for atmospheric disturbance using the simplified method of atmospheric correction (SMAC) model. The in situ measurements consisted of water sampling in the largest lake of the region, routine monitoring results for the whole study area, and Z(SD) observations made by volunteers. The ranges of the water quality variables in the dataset were as follows: turbidity, 0.6-25 FNU; absorption coefficient of CDOM at 400 nm, 1.0-12.2 m(-1); Z(SD), 0.5-5.5 m; and chlorophyll a concentration, 2.4-80 mug L(-1). The estimation accuracies of the image-specific empirical algorithms expressed as relative errors were 23.0% for turbidity, 17.4% for CDOM, and 21.1% for Z(SD). If concurrent in situ measurements had not been used for algorithm training, the average error would have been about 37%. The atmospheric correction improved the estimation accuracy only slightly compared with the use of top-of-atmospheric reflectances. The accuracy of the water quality estimates without concurrent in situ measurements could have been improved if in-image atmospheric parameters had been available. The underwater reflectance simulations of the ETM+ channel wavelengths using water quality typical for Finnish lakes (data from 1113 lakes) indicated that region-specific algorithms may be needed in other parts of the country, particularly in the case of Z(SD). Despite the limitations in the spectral and radiometric resolutions, ETM+ imagery can be an effective aid, particularly in the monitoring and management of small lakes (<1 km(2)), which are often not included in routine monitoring programs.     

Time Series Forecasting of Cyanobacteria Blooms in the Crestuma Reservoir (Douro River, Portugal) Using Artificial Neural Networks

In this work, time series neural networks were used to predict the occurrence of toxic cyanobacterial blooms in Crestuma Reservoir, which is an important potable water supply for the Porto region, located in the north of Portugal. These models can potentially be used to provide water treatment plant operators with an early warning for developing cyanobacteria blooms. Physical, chemical, and biological parameters were collected at Crestuma Reservoir from 1999 to 2002. The data set was then divided into three independent time series, each with a fortnightly periodicity. One training series was used to “teach” the neural networks to predict results. Another series was used to verify the results, and to avoid over-fitting of the data. An additional independently collected data series was then used to test the efficacy of the model for predicting the abundance of cyanobacteria. All of the models tested in this study incorporated a prediction time (look-ahead parameter) equal to the sampling interval (two weeks). Various lag periods, from 2 to 52 weeks, were also investigated. The best model produced in this study provided the following correlations between the target and forecast values in the training, verification, and validation series: 1.000 (P = 0.000), 0.802 (P = 0.000), and 0.773 (P = 0.001), respectively. By applying this model to the three-year data set, we were able to predict fluctuations in cyanobacteria abundance in the Crestuma Reservoir, with a high level of precision. By incorporating a lag-period of eight weeks, we were able to detect secondary fluctuations in cyanobacterial abundance over the annual cycle.     

Modeling the Impacts of Farming Practices on Water Quality in the Little Miami River Basin

Since intensive farming practices are essential to produce enough food for the increasing population, farmers have been using more inorganic fertilizers, pesticides, and herbicides. Agricultural lands are currently one of the major sources of non-point source pollution. However, by changing farming practices in terms of tillage and crop rotation, the levels of contamination can be reduced and the quality of soil and water resources can be improved. Thus, there is a need to investigate the amalgamated hydrologic effects when various tillage and crop rotation practices are operated in tandem. In this study, the Soil Water Assessment Tool (SWAT) was utilized to evaluate the individual and combined impacts of various farming practices on flow, sediment, ammonia, and total phosphorus loads in the Little Miami River basin. The model was calibrated and validated using the 1990–1994 and 1980–1984 data sets, respectively. The simulated results revealed that the SWAT model provided a good simulation performance. For those tested farming scenarios, no-tillage (NT) offered more environmental benefits than moldboard plowing (MP). Flow, sediment, ammonia, and total phosphorus under NT were lower than those under MP. In terms of crop rotation, continuous soybean and corn–soybean rotation were able to reduce sediment, ammonia, and total phosphorus loads. When the combined effects of tillage and crop rotation were examined, it was found that NT with continuous soybean or corn–soybean rotation could greatly restrain the loss of sediments and nutrients to receiving waters. Since corn–soybean rotation provides higher economic revenue, a combination of NT and corn–soybean rotation can be a viable system for successful farming.     

Allocation of supplementary aeration stations in the Chicago waterway system for dissolved oxygen improvement

The Chicago Waterway System (CWS), used mainly for commercial and recreational navigation and for urban drainage, is a 122.8 km branching network of navigable waterways controlled by hydraulic structures. The CWS receives pollutant loads from 3 of the largest wastewater treatment plants in the world, nearly 240 gravity Combined Sewer Overflows (CSO), 3 CSO pumping stations, direct diversions from Lake Michigan, and eleven tributary streams or drainage areas. Even though treatment plant effluent concentrations meet the applicable standards and most reaches of the CWS meet the applicable water quality standards, Dissolved Oxygen (DO) standards are not met in the CWS during some periods. A Use Attainability Analysis was initiated to evaluate what water quality standards can be achieved in the CWS. The UAA team identified several DO improvement alternatives including new supplementary aeration stations. Because of the dynamic nature of the CWS, the DUFLOW model that is capable of simulating hydraulics and water quality processes under unsteady-flow conditions was used to evaluate the effectiveness of new supplementary aeration stations. This paper details the use of the DUFLOW model to size and locate supplementary aeration stations. In order to determine the size and location of supplemental aeration stations, 90% compliance with a 5 mg/l DO standard was used as a planning target. The simulations showed that a total of four new supplementary aeration stations with oxygen supply capacities ranging from 30 to 80 g/s would be sufficient to meet the proposed target DO concentration for the North Branch and South Branch of the Chicago River. There are several aeration technologies, two of which are already being used in the CWS, available and the UAA team determined that the total capital costs of the alternatives range from $35.5 to $89.9 million with annual operations and maintenance costs ranging from $554,000 to $2.14 million. Supplemental aeration stations have been shown to be a potentially effective means to improve DO concentrations in the CWS and will be included in developing an integrated strategy for improving water quality in the CWS.     

A hybrid model based on smooth transition periodic autoregressive and Elman artificial neural network for wind speed forecasting of the Hebei region in China

Wind energy, one of the most promising renewable and clean energy sources, is becoming increasingly significant for sustainable energy development and environmental protection. Given the relationship between wind power and wind speed, precise prediction of wind speed for wind energy estimation and wind power generation is important. For proper and efficient evaluation of wind speed, a smooth transition periodic autoregressive (STPAR) model is developed to predict the six-hourly wind speeds. In addition, the Elman artificial neural network (EANN)-based error correction technique has also been integrated into the new STPAR model to improve model performance. To verify the developed approach, the six-hourly wind speed series during the period of 2000–2009 in the Hebei region of China is used for model construction and model testing. The proposed EANN-STPAR hybrid model has demonstrated its powerful forecasting capacity for wind speed series with complicated characteristics of linearity, seasonality and nonlinearity, which indicates that the proposed hybrid model is notably efficient and practical for wind speed forecasting, especially for the Hebei wind farms of China.     

Estimating the economic value of improvements in river ecology using choice experiments: an application to the water framework directive

The Water Framework Directive is a major regulatory reform of water resources management within the European Union. Integrated catchment management plans must be prepared for all river basins, in order to achieve 'good ecological status' in all EU waters. Ecological status is a broader measure of water quality than the chemical and biological measures that were previously dominant. The Directive calls for a consideration of the economic costs and benefits of improvements to ecological status. In this paper, we use the choice experiment method to estimate the value of improvements in three components of ecological status. Given the high resource cost of valuation studies, benefits transfer methods will be needed in implementing the Directive. We thus also test the ability of choice experiments for benefits transfer across two very similar rivers in the UK.     

Assessing the Effects of Nutrient Management in an Estuary Experiencing Climatic Change: The Neuse River Estuary, North Carolina

Eutrophication is a serious water quality problem in estuaries receiving increasing anthropogenic nutrient loads. Managers undertaking nutrient-reduction strategies aimed at controlling estuarine eutrophication are faced with the challenge that upstream freshwater segments often are phosphorus (P)-limited, whereas more saline downstream segments are nitrogen (N)-limited. Management also must consider climatic (hydrologic) variability, which affects nutrient delivery and processing. The interactive effects of selective nutrient input reductions and climatic perturbations were examined in the Neuse River Estuary (NRE), North Carolina, a shallow estuary with more than a 30-year history of accelerated nutrient loading and water quality decline. The NRE also has experienced a recent increase in Atlantic hurricanes and record flooding, which has affected hydrology and nutrient loadings. The authors examined the water quality consequences of selective nutrient (P but not N) reductions in the 1980s, followed by N reductions in the 1990s and an increase in hurricane frequency since the mid-1990s. Selective P reductions decreased upstream phytoplankton blooms, but increased downstream phytoplankton biomass. Storms modified these trends. In particular, upstream annual N and P concentrations have decreased during the elevated hurricane period. Increased flushing and scouring from storms and flooding appear to have enhanced nutrient retention capabilities of the NRE watershed. From a management perspective, one cannot rely on largely unpredictable changes in storm frequency and intensity to negate anthropogenic nutrient enrichment and eutrophication. To control eutrophication along the hydrologically variable freshwater–marine continuum, N and P reductions should be applied adaptively to reflect point-source–dominated drought and non–point-source–dominated flood conditions.