scholarly journals Applicability Evaluation of the Hydrological Image and Convolution Neural Network for Prediction of the Biochemical Oxygen Demand and Total Phosphorus Loads in Agricultural Areas

Agriculture ◽  
2020 ◽  
Vol 10 (11) ◽  
pp. 529
Author(s):  
Chul Min Song ◽  
Jin Soo Kim
Keyword(s):  
Neural Network ◽  
Total Phosphorus ◽  
Biochemical Oxygen Demand ◽  
Input Data ◽  
Pearson Correlation ◽  
Oxygen Demand ◽  
Convolution Neural Network ◽  
Stable Model ◽  
Model Learning ◽  
Load Model

This study employed a convolution neural network (CNN) model, hitherto used only for solving classification problems, with two-dimensional input data to predict the pollution loads and evaluate the CNN model’s applicability. A CNN model generally requires two-dimension input data, such as photographs in previous studies. However, this study’s CNN model necessitates the numerical images that reflect hydrological phenomena due to the nature of the study. A hydrological image was used as the input data for the CNN model in this study to address this issue. The last layer of the CNN model was also transformed into a linear function to derive the continuous variable. As a result, the Pearson correlation coefficient, which represents the relationship between the measured and predicted values, demonstrated a Biochemical Oxygen Demand (BOD) load model of 0.94 and a Total Phosphorus (TP) load model of 0.87. Nash–Sutcliffe efficiency was used to evaluate the model performance; the BOD load model was 0.83, while the TP load model was 0.79, respectively, indicating good performance. These results demonstrate that the hydrological images led to stable model learning and generalization, and the proposed CNN model is suitable for predicting the pollution load, with potential future applications in various fields.

scholarly journals Hydrological Image Building Using Curve Number and Prediction and Evaluation of Runoff through Convolution Neural Network

Water ◽  
2020 ◽  
Vol 12 (8) ◽  
pp. 2292 ◽  
Author(s):  
Chul Min Song
Keyword(s):  
Neural Network ◽  
Input Data ◽  
Pearson Correlation ◽  
Curve Number ◽  
Convolution Neural Network ◽  
Classification Problems ◽  
Image Building ◽  
Predicted Values ◽  
The Relationship ◽  
Runoff Model

This study developed a runoff model using a convolution neural network (CNN), which had previously only been used for classification problems, to get away from artificial neural networks (ANNs) that have been extensively used for the development of runoff models, and to secure diversity and demonstrate the suitability of the model. For this model’s input data, photographs typically used in the CNN model could not be used; due to the nature of the study, hydrological images reflecting effects such as watershed conditions and rainfall were required, which posed further difficulties. To address this, the method of a generating hydrological image using the curve number (CN) published by the Soil Conservation Service (SCS) was suggested in this study, and the hydrological images using CN were found to be sufficient as input data for the CNN model. Furthermore, this study was able to present a new application for the CN, which had been used only for estimating runoff. The model was trained and generalized stably overall, and R2, which indicates the relationship between the actual and predicted values, was relatively high at 0.82. The Pearson correlation coefficient, Nash–Sutcliffe efficiency (NSE), and root mean square error (RMSE), were 0.87, 0.60, and 16.20 m3/s, respectively, demonstrating a good overall model prediction performance.

Reed beds: constructed wetlands for municipal wastewater treatment plant sludge dewatering

2001 ◽  
Vol 44 (11-12) ◽  
pp. 393-398 ◽  
Author(s):  
J.S. Begg ◽  
R.L. Lavigne ◽  
P.L.M. Veneman
Keyword(s):  
Wastewater Treatment ◽  
Total Phosphorus ◽  
Biochemical Oxygen Demand ◽  
Suspended Solids ◽  
Oxygen Demand ◽  
Total Suspended Solids ◽  
Sludge Dewatering ◽  
Sludge Treatment ◽  
Reed Beds ◽  
Reed Bed

Reed beds are an alternative technology wastewater treatment system that mimic the biogeochemical processes inherent in natural wetlands. The purpose of this project was to determine the effectiveness of a reed bed sludge treatment system (RBSTS) in southern New England after a six-year period of operation by examining the concentrations of selected metals in the reed bed sludge biomass and by determining the fate of solids and selected nutrients. Parameters assessed in both the reed bed influent and effluent: total suspended solids, biochemical oxygen demand, nitrate-nitrogen and total phosphorus. In addition, the following metals were studied in the reed bed influent, effluent and Phragmites plant tissue and the sludge core biomass: boron, cadmium, chromium, copper, iron, lead, manganese, molybdenum, nickel, and zinc. The removal efficiencies for sludge dewatering, total suspended solids and biochemical oxygen demand were all over 90%. Nitrate and total phosphorus removal rates were 90% and 80% respectively. Overall metals removal efficient was 87%. Copper was the only metal in the sludge biomass that exceeded the standards set by the Massachusetts Department of Environmental Protection for land disposal of sludge. The highest metal concentrations, for the most part, tended to be in the lower tier of the sludge profile. The exception was boron, which was more concentrated in the middle tier of the sludge profile. The data and results presented in this paper support the notion that reed bed sludge treatment systems and the use of reed beds provide an efficient and cost effective alternative for municipal sludge treatment.

Quality Issues Regarding the Watercourses from Middle Basin of Jiu River

2017 ◽  
Vol 68 (1) ◽  
pp. 72-76
Author(s):  
Daniela Cirtina ◽  
Camelia Capatina
Keyword(s):  
Dissolved Oxygen ◽  
River Basin ◽  
Total Phosphorus ◽  
Surface Waters ◽  
Biochemical Oxygen Demand ◽  
Oxygen Demand ◽  
Chloride Content ◽  
Quality Issues

The study aims to characterize the quality of surface waters in the middle of the river basin Jiu by monitoring physicochemical indicators of their quality, in 2013-2015. In this regard, the pH, dissolved oxygen (DO), biochemical oxygen demand (BOD5) of nitrate (NO3-), nitrite (NO2-), ammonium (NH4+), total phosphorus (Ptotal), chlorides and sulphates from water have been determined. Water of Jiu River and its tributaries of Gorj county have been monitored on representative sections for the evolution of their quality. It was found that the water from natural reservoirs monitored shows an evolution in the limits permitted by the regulations in force except biochemical oxygen demand and nitrites indicators for river Jiu and Tismana and nitrate and chloride content for Gilort River.

scholarly journals Patent Automatic Classification Based on Symmetric Hierarchical Convolution Neural Network

Symmetry ◽  
2020 ◽  
Vol 12 (2) ◽  
pp. 186 ◽  
Author(s):  
Huiming Zhu ◽  
Chunhui He ◽  
Yang Fang ◽  
Bin Ge ◽  
Meng Xing ◽  
...  
Keyword(s):  
Neural Network ◽  
Input Data ◽  
Large Scale ◽  
Patent Application ◽  
Input Sequence ◽  
Automatic Classification ◽  
Convolution Neural Network ◽  
Coarse Grained ◽  
Short Text ◽  
Patent Classification

With the rapid growth of patent applications, it has become an urgent problem to automatically classify the accepted patent application documents accurately and quickly. Most previous patent automatic classification studies are based on feature engineering and traditional machine learning methods like SVM, and some even rely on the knowledge of domain experts, hence they suffer from low accuracy problem and have poor generalization ability. In this paper, we propose a patent automatic classification method via the symmetric hierarchical convolution neural network (CNN) named PAC-HCNN. We use the title and abstract of the patent as the input data, and then apply the word embedding technique to segment and vectorize the input data. Then we design a symmetric hierarchical CNN framework to classify the patents based on the word embeddings, which is much more efficient than traditional RNN models dealing with texts, meanwhile keeping the history and future information of the input sequence. We also add gated linear units (GLUs) and residual connection to help realize the deep CNN. Additionally, we equip our model with a self attention mechanism to address the long-term dependency problem. Experiments are performed on large-scale datasets for Chinese short text patent classification. Experimental results prove our proposed model’s effectiveness, and it performs better than other state-of-the-art models significantly and consistently on both fine-grained and coarse-grained classification.

scholarly journals Nutrient Level of a Young Tropical Hydroelectric Dam Reservoir in Sarawak, Malaysia

2018 ◽  
Vol 8 (1) ◽  
pp. 14-22 ◽  
Author(s):  
Teck Yee Ling ◽  
Norliza Gerunsin ◽  
Chen Lin Soo ◽  
Nyanti Lee ◽  
Siong Fong Sim ◽  
...  
Keyword(s):  
Total Phosphorus ◽  
Biochemical Oxygen Demand ◽  
Nitrate Nitrogen ◽  
Oxygen Demand ◽  
Nutrient Level ◽  
Dam Construction ◽  
Hydroelectric Reservoir ◽  
Nitrite Nitrogen ◽  
Hydroelectric Dam ◽  
Kjeldahl Nitrogen

Monitoring the nutrient level of a reservoir is crucial as excess nutrients can lead to hypoxia and fish kills in the reservoir. Hence, this study was carried out to examine the nutrient level of the Bakun reservoir, which is a newly built hydroelectric reservoir in Sarawak, Malaysia. Water samples were taken at five stations in the reservoir at three different depths (surface layer, 10 m, and 20 m) in November 2013. The present study demonstrated that Bakun reservoir contained low nitrite-nitrogen (≈ 0.005 mg/L) and nitrate-nitrogen (≈ 0.005 mg/L) concentrations but high five-day biochemical oxygen demand (≈ 4.73 mg/L) and organic Kjeldahl nitrogen (≈ 0.16 mg/L) concentrations indicating that organic pollution occurred in the reservoir. On the other hand, a mean total phosphorus concentration of 98.3 μg/L in the Bakun reservoir complied with the 200 μg/L standard value of Class II according to National Water Quality Standards in Malaysia. The nutrient level in the Bakun reservoir differed according to sampling stations and depths. Samplings stations located at Murum River downstream of the Murum dam construction site showed peak value of turbidity (182 FNU) and organic Kjeldahl nitrogen (0.45 mg/L) particularly at deeper water column. Batang Balui and Linau River were observed with higher five-day biochemical oxygen demand (> 6 mg/L) compared to other stations. Station near to the Bakun hydroelectric dam contained relatively high nitrite-nitrogen and total phosphorus concentrations but low nitrate-nitrogen and organic Kjeldahl nitrogen concentrations. Anthropogenic activities such as floating house and Murum dam construction have influenced the nutrients level in the reservoir. Keywords : Bakun hydroelectric reservoir, dam construction, nitrogen, phosphorus, turbidity

scholarly journals The Use of Permeable Interlocking Concrete Pavement to Filter Stormwater for Non-Potable Uses in Buildings

Water ◽  
2020 ◽  
Vol 12 (7) ◽  
pp. 2045
Author(s):  
Enedir Ghisi ◽  
Thiago Belotto ◽  
Liseane Thives
Keyword(s):  
Total Phosphorus ◽  
Biochemical Oxygen Demand ◽  
Potable Water ◽  
Oxygen Demand ◽  
Model System ◽  
Subsequent Treatment ◽  
Concrete Pavement ◽  
Model Systems ◽  
Fecal Coliforms ◽  
Permeable Interlocking Concrete Pavement

A reduction in potable water demand in buildings could be made by using non-potable water for certain uses, such as flushing toilets. This represents a sustainable strategy that results in potable water savings while also using an underutilised resource. This work assesses the use of permeable interlocking concrete pavement to filter stormwater that could be used for non-potable purposes in buildings. Two pavement model systems were tested. One of the model systems presents a filter course layer with coarse sand and the other model system has no filter course layer. In order to evaluate the filtering capacity, the model systems were exposed to rain events. The amount of water infiltrated through the layers was measured to represent the potential quantity available for use. Stormwater runoff samples were collected from a parking lot paved with impermeable interlocked blocks and then, these were tested in both model systems. Water samples were subjected to quality tests according to the parameters recommended by the Brazilian National Water Agency. The model system with no filter course showed filtering capacity higher (88.1%) than the one with a filter course layer (78.8%). The model system with a filter course layer was able to reduce fecal coliforms (54.7%), total suspended solids (62.5%), biochemical oxygen demand (78.8%), and total phosphorus concentrations (55.6%). Biochemical oxygen demand (42.4%) and total phosphorus concentrations (44.4%) increased in the model system with no filter course layer. In conclusion, one can state that the filter course layer used in permeable interlocking concrete pavement can contribute to decreasing pollutants and can improve stormwater quality. The use of permeable interlocking concrete pavement showed to be a potential alternative for filtering stormwater prior to subsequent treatment for non-potable uses in buildings.

Ship Type Recognition via a Coarse-to-Fine Cascaded Convolution Neural Network

2020 ◽  
Vol 73 (4) ◽  
pp. 813-832 ◽  
Author(s):  
Xinqiang Chen ◽  
Yongsheng Yang ◽  
Shengzheng Wang ◽  
Huafeng Wu ◽  
Jinjun Tang ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Input Data ◽  
Recognition Performance ◽  
Convolution Neural Network ◽  
Bulk Carrier ◽  
Merchant Ship ◽  
General Cargo ◽  
Oil Tanker ◽  
Coarse To Fine

Most previous research has handled the task of ship type recognition by exploring hand-craft ship features, which may fail to distinguish ships with similar visual appearances. This situation motivates us to propose a novel deep learning based ship type recognition framework which we have named coarse-to-fine cascaded convolution neural network (CFCCNN). First, the proposed CFCCNN framework formats the input training ship images and data, and provides trainable input data for the hidden layers of the CFCCNN. Second, the coarse and fine steps are run in a nesting manner to explore discriminative features for different ship types. More specifically, the coarse step is trained in a similar manner to the traditional convolution neural network, while the fine step introduces regularisation mechanisms to extract more intrinsic ship features, and fine tunes parameter settings to obtain better recognition performance. Finally, we evaluate the performance of the CFCCNN model for recognising the most common types of merchant ship (oil tanker, container, LNG tanker, chemical carrier, general cargo, bulk carrier, etc.). The experimental results show that the proposed framework obtains better recognition performance than the conventional methods of ship type recognition.

Sign in / Sign up

Export Citation Format

Share Document