Search results for: atmospheric pollutant prediction
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 3098

Search results for: atmospheric pollutant prediction

3008 Effect of Substrate Type on Pollutant Removal and Greenhouse Gases Emissions in Constructed Wetlands with Ornamental Plants

Authors: Maria E. Hernnadez, Elizabeth Ramos, Claudia Ortiz

Abstract:

Pollutant removal (N-NH4, COD, S-SO4, N-NO3 and P-PO4) and greenhouse gases (methane and nitrous oxide) emissions were investigated in constructed wetlands CW mesocosms with four types of substrate (gravel (G) zeolite (Z), Gravel+Plastic (GP) and zeolite+plastic), all planted with the ornamental plant lily (Lilium sp). Significantly higher N-NH4 removal was found in the CW-Z (97%) and CW-ZP (85%) compared with CW-G (61%) and CW-GP (17%), also significantly lower emissions of nitrous oxide were found in CW-Z (2.2 µgm-2min-1) and CW-ZP (2.5 µgm-2min-1) compared with CW-G(7.4 µgm-2min-1 ) and CW-GP (6.30 µgm-2min-1).

Keywords: methane, nitrous oxide, lily, zeolite

Procedia PDF Downloads 379
3007 Screening of Ionic Liquids for Hydrogen Sulfide Removal Using COSMO-RS

Authors: Zulaika Mohd Khasiran

Abstract:

The capability of ionic liquids in various applications makes them attracted by many researchers. They have potential to be developed as “green” solvents for gas separation, especially H2S gas. In this work, it is attempted to predict the solubility of hydrogen sulfide (H2S) in ILs by COSMO-RS method. Since H2S is a toxic pollutant, it is difficult to work on it in the laboratory, therefore an appropriate model will be necessary in prior work. The COSMO-RS method is implemented to predict the Henry’s law constants and activity coefficient of H2S in 140 ILs with various combinations of cations and anions. It is found by the screening that more H2S can be absorbed in ILs with [Cl] and [Ac] anion. The solubility of H2S in ILs with different alkyl chain at the cations not much affected and with different type of cations are slightly influence H2S capture capacities. Even though the cations do not affect much in solubility of H2S, we still need to consider the effectiveness of cation in different way. The prediction results only show their physical absorption ability, but the absorption of H2S need to be consider chemically to get high capacity of absorption of H2S.

Keywords: H2S, hydrogen sulfide, ionic liquids, COSMO-RS

Procedia PDF Downloads 115
3006 Cross Project Software Fault Prediction at Design Phase

Authors: Pradeep Singh, Shrish Verma

Abstract:

Software fault prediction models are created by using the source code, processed metrics from the same or previous version of code and related fault data. Some company do not store and keep track of all artifacts which are required for software fault prediction. To construct fault prediction model for such company, the training data from the other projects can be one potential solution. The earlier we predict the fault the less cost it requires to correct. The training data consists of metrics data and related fault data at function/module level. This paper investigates fault predictions at early stage using the cross-project data focusing on the design metrics. In this study, empirical analysis is carried out to validate design metrics for cross project fault prediction. The machine learning techniques used for evaluation is Naïve Bayes. The design phase metrics of other projects can be used as initial guideline for the projects where no previous fault data is available. We analyze seven data sets from NASA Metrics Data Program which offer design as well as code metrics. Overall, the results of cross project is comparable to the within company data learning.

Keywords: software metrics, fault prediction, cross project, within project.

Procedia PDF Downloads 317
3005 A Deep Learning-Based Pedestrian Trajectory Prediction Algorithm

Authors: Haozhe Xiang

Abstract:

With the rise of the Internet of Things era, intelligent products are gradually integrating into people's lives. Pedestrian trajectory prediction has become a key issue, which is crucial for the motion path planning of intelligent agents such as autonomous vehicles, robots, and drones. In the current technological context, deep learning technology is becoming increasingly sophisticated and gradually replacing traditional models. The pedestrian trajectory prediction algorithm combining neural networks and attention mechanisms has significantly improved prediction accuracy. Based on in-depth research on deep learning and pedestrian trajectory prediction algorithms, this article focuses on physical environment modeling and learning of historical trajectory time dependence. At the same time, social interaction between pedestrians and scene interaction between pedestrians and the environment were handled. An improved pedestrian trajectory prediction algorithm is proposed by analyzing the existing model architecture. With the help of these improvements, acceptable predicted trajectories were successfully obtained. Experiments on public datasets have demonstrated the algorithm's effectiveness and achieved acceptable results.

Keywords: deep learning, graph convolutional network, attention mechanism, LSTM

Procedia PDF Downloads 42
3004 Development of Prediction Models of Day-Ahead Hourly Building Electricity Consumption and Peak Power Demand Using the Machine Learning Method

Authors: Dalin Si, Azizan Aziz, Bertrand Lasternas

Abstract:

To encourage building owners to purchase electricity at the wholesale market and reduce building peak demand, this study aims to develop models that predict day-ahead hourly electricity consumption and demand using artificial neural network (ANN) and support vector machine (SVM). All prediction models are built in Python, with tool Scikit-learn and Pybrain. The input data for both consumption and demand prediction are time stamp, outdoor dry bulb temperature, relative humidity, air handling unit (AHU), supply air temperature and solar radiation. Solar radiation, which is unavailable a day-ahead, is predicted at first, and then this estimation is used as an input to predict consumption and demand. Models to predict consumption and demand are trained in both SVM and ANN, and depend on cooling or heating, weekdays or weekends. The results show that ANN is the better option for both consumption and demand prediction. It can achieve 15.50% to 20.03% coefficient of variance of root mean square error (CVRMSE) for consumption prediction and 22.89% to 32.42% CVRMSE for demand prediction, respectively. To conclude, the presented models have potential to help building owners to purchase electricity at the wholesale market, but they are not robust when used in demand response control.

Keywords: building energy prediction, data mining, demand response, electricity market

Procedia PDF Downloads 295
3003 Prediction of CO2 Concentration in the Korea Train Express (KTX) Cabins

Authors: Yong-Il Lee, Do-Yeon Hwang, Won-Seog Jeong, Duckshin Park

Abstract:

Recently, because of the high-speed trains forced ventilation, it is important to control the ventilation. The ventilation is for controlling various contaminants, temperature, and humidity. The high-speed train route is straight to a destination having a high speed. And there are many mountainous areas in Korea. So, tunnel rate is higher then other country. KTX HVAC block off the outdoor air, when entering tunnel. So the high tunnel rate is an effect of ventilation in the KTX cabin. It is important to reduction rate in CO2 concentration prediction. To meet the air quality of the public transport vehicles recommend standards, the KTX cabin of CO2 concentration should be managed. In this study, the concentration change was predicted by CO2 prediction simulation in route to be opened.

Keywords: CO2 prediction, KTX, ventilation, infrastructure and transportation engineering

Procedia PDF Downloads 518
3002 Determining the City Development Based on the Modeling of the Pollutant Emission from Power Plant by Using AERMOD Software

Authors: Abbasi Fakhrossadat, Moharreri Mohammadamir, Shadmanmahani Mohammadjavad

Abstract:

The development of cities can be influenced by various factors, including air pollution. In this study, the focus is on the city of Mashhad, which has four large power plants operating. The emission of pollutants from these power plants can have a significant impact on the quality of life and health of the city's residents. Therefore, modeling and analyzing the emission pattern of pollutants can provide useful information for urban decision-makers and help in estimating the urban development model. The aim of this research is to determine the direction of city development based on the modeling of pollutant emissions (NOX, CO, and PM10) from power plants in Mashhad. By using the AERMOD software, the release of these pollutants will be modeled and analyzed.

Keywords: emission of air pollution, thermal power plant, urban development, AERMOD

Procedia PDF Downloads 52
3001 Statistical Analysis with Prediction Models of User Satisfaction in Software Project Factors

Authors: Katawut Kaewbanjong

Abstract:

We analyzed a volume of data and found significant user satisfaction in software project factors. A statistical significance analysis (logistic regression) and collinearity analysis determined the significance factors from a group of 71 pre-defined factors from 191 software projects in ISBSG Release 12. The eight prediction models used for testing the prediction potential of these factors were Neural network, k-NN, Naïve Bayes, Random forest, Decision tree, Gradient boosted tree, linear regression and logistic regression prediction model. Fifteen pre-defined factors were truly significant in predicting user satisfaction, and they provided 82.71% prediction accuracy when used with a neural network prediction model. These factors were client-server, personnel changes, total defects delivered, project inactive time, industry sector, application type, development type, how methodology was acquired, development techniques, decision making process, intended market, size estimate approach, size estimate method, cost recording method, and effort estimate method. These findings may benefit software development managers considerably.

Keywords: prediction model, statistical analysis, software project, user satisfaction factor

Procedia PDF Downloads 95
3000 Calibration Model of %Titratable Acidity (Citric Acid) for Intact Tomato by Transmittance SW-NIR Spectroscopy

Authors: K. Petcharaporn, S. Kumchoo

Abstract:

The acidity (citric acid) is one of the chemical contents that can refer to the internal quality and the maturity index of tomato. The titratable acidity (%TA) can be predicted by a non-destructive method prediction by using the transmittance short wavelength (SW-NIR). Spectroscopy in the wavelength range between 665-955 nm. The set of 167 tomato samples divided into groups of 117 tomatoes sample for training set and 50 tomatoes sample for test set were used to establish the calibration model to predict and measure %TA by partial least squares regression (PLSR) technique. The spectra were pretreated with MSC pretreatment and it gave the optimal result for calibration model as (R = 0.92, RMSEC = 0.03%) and this model obtained high accuracy result to use for %TA prediction in test set as (R = 0.81, RMSEP = 0.05%). From the result of prediction in test set shown that the transmittance SW-NIR spectroscopy technique can be used for a non-destructive method for %TA prediction of tomatoes.

Keywords: tomato, quality, prediction, transmittance, titratable acidity, citric acid

Procedia PDF Downloads 249
2999 Ground Surface Temperature History Prediction Using Long-Short Term Memory Neural Network Architecture

Authors: Venkat S. Somayajula

Abstract:

Ground surface temperature history prediction model plays a vital role in determining standards for international nuclear waste management. International standards for borehole based nuclear waste disposal require paleoclimate cycle predictions on scale of a million forward years for the place of waste disposal. This research focuses on developing a paleoclimate cycle prediction model using Bayesian long-short term memory (LSTM) neural architecture operated on accumulated borehole temperature history data. Bayesian models have been previously used for paleoclimate cycle prediction based on Monte-Carlo weight method, but due to limitations pertaining model coupling with certain other prediction networks, Bayesian models in past couldn’t accommodate prediction cycle’s over 1000 years. LSTM has provided frontier to couple developed models with other prediction networks with ease. Paleoclimate cycle developed using this process will be trained on existing borehole data and then will be coupled to surface temperature history prediction networks which give endpoints for backpropagation of LSTM network and optimize the cycle of prediction for larger prediction time scales. Trained LSTM will be tested on past data for validation and then propagated for forward prediction of temperatures at borehole locations. This research will be beneficial for study pertaining to nuclear waste management, anthropological cycle predictions and geophysical features

Keywords: Bayesian long-short term memory neural network, borehole temperature, ground surface temperature history, paleoclimate cycle

Procedia PDF Downloads 108
2998 Hybrid Fuzzy Weighted K-Nearest Neighbor to Predict Hospital Readmission for Diabetic Patients

Authors: Soha A. Bahanshal, Byung G. Kim

Abstract:

Identification of patients at high risk for hospital readmission is of crucial importance for quality health care and cost reduction. Predicting hospital readmissions among diabetic patients has been of great interest to many researchers and health decision makers. We build a prediction model to predict hospital readmission for diabetic patients within 30 days of discharge. The core of the prediction model is a modified k Nearest Neighbor called Hybrid Fuzzy Weighted k Nearest Neighbor algorithm. The prediction is performed on a patient dataset which consists of more than 70,000 patients with 50 attributes. We applied data preprocessing using different techniques in order to handle data imbalance and to fuzzify the data to suit the prediction algorithm. The model so far achieved classification accuracy of 80% compared to other models that only use k Nearest Neighbor.

Keywords: machine learning, prediction, classification, hybrid fuzzy weighted k-nearest neighbor, diabetic hospital readmission

Procedia PDF Downloads 165
2997 Examination of Corrosion Durability Related to Installed Environments of Steel Bridges

Authors: Jin-Hee Ahn, Seok-Hyeon Jeon, Young-Bin Lee, Min-Gyun Ha, Yu-Chan Hong

Abstract:

Corrosion durability of steel bridges can be generally affected by atmospheric environments of bridge installation, since corrosion problem is related to environmental factors such as humidity, temperature, airborne salt, chemical components as SO₂, chlorides, etc. Thus, atmospheric environment condition should be measured to estimate corrosion condition of steel bridges as well as measurement of actual corrosion damage of structural members of steel bridge. Even in the same atmospheric environment, the corrosion environment may be different depending on the installation direction of structural members. In this study, therefore, atmospheric corrosion monitoring was conducted using atmospheric corrosion monitoring sensor, hygrometer, thermometer and airborne salt collection device to examine the corrosion durability of steel bridges. As a target steel bridge for corrosion durability monitoring, a cable-stayed bridge with truss steel members was selected. This cable-stayed bridge was located on the coast to connect the islands with the islands. Especially, atmospheric corrosion monitoring was carried out depending on structural direction of a cable-stayed bridge with truss type girders since it consists of structural members with various directions. For atmospheric corrosion monitoring, daily average electricity (corrosion current) was measured at each monitoring members to evaluate corrosion environments and corrosion level depending on structural members with various direction which have different corrosion environment in the same installed area. To compare corrosion durability connected with monitoring data depending on corrosion monitoring members, monitoring steel plate was additionally installed in same monitoring members. Monitoring steel plates of carbon steel was fabricated with dimension of 60mm width and 3mm thickness. And its surface was cleaned for removing rust on the surface by blasting, and its weight was measured before its installation on each structural members. After a 3 month exposure period on real atmospheric corrosion environment at bridge, surface condition of atmospheric corrosion monitoring sensors and monitoring steel plates were observed for corrosion damage. When severe deterioration of atmospheric corrosion monitoring sensors or corrosion damage of monitoring steel plates were found, they were replaced or collected. From 3month exposure tests in the actual steel bridge with various structural member with various direction, the rust on the surface of monitoring steel plate was found, and the difference in the corrosion rate was found depending on the direction of structural member from their visual inspection. And daily average electricity (corrosion current) was changed depending on the direction of structural member. However, it is difficult to identify the relative differences in corrosion durability of steel structural members using short-term monitoring results. After long exposure tests in this corrosion environments, it can be clearly evaluated the difference in corrosion durability depending on installed conditions of steel bridges. Acknowledgements: This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2017R1D1A1B03028755).

Keywords: corrosion, atmospheric environments, steel bridge, monitoring

Procedia PDF Downloads 335
2996 Using High Performance Computing for Online Flood Monitoring and Prediction

Authors: Stepan Kuchar, Martin Golasowski, Radim Vavrik, Michal Podhoranyi, Boris Sir, Jan Martinovic

Abstract:

The main goal of this article is to describe the online flood monitoring and prediction system Floreon+ primarily developed for the Moravian-Silesian region in the Czech Republic and the basic process it uses for running automatic rainfall-runoff and hydrodynamic simulations along with their calibration and uncertainty modeling. It takes a long time to execute such process sequentially, which is not acceptable in the online scenario, so the use of high-performance computing environment is proposed for all parts of the process to shorten their duration. Finally, a case study on the Ostravice river catchment is presented that shows actual durations and their gain from the parallel implementation.

Keywords: flood prediction process, high performance computing, online flood prediction system, parallelization

Procedia PDF Downloads 473
2995 Prediction of PM₂.₅ Concentration in Ulaanbaatar with Deep Learning Models

Authors: Suriya

Abstract:

Rapid socio-economic development and urbanization have led to an increasingly serious air pollution problem in Ulaanbaatar (UB), the capital of Mongolia. PM₂.₅ pollution has become the most pressing aspect of UB air pollution. Therefore, monitoring and predicting PM₂.₅ concentration in UB is of great significance for the health of the local people and environmental management. As of yet, very few studies have used models to predict PM₂.₅ concentrations in UB. Using data from 0:00 on June 1, 2018, to 23:00 on April 30, 2020, we proposed two deep learning models based on Bayesian-optimized LSTM (Bayes-LSTM) and CNN-LSTM. We utilized hourly observed data, including Himawari8 (H8) aerosol optical depth (AOD), meteorology, and PM₂.₅ concentration, as input for the prediction of PM₂.₅ concentrations. The correlation strengths between meteorology, AOD, and PM₂.₅ were analyzed using the gray correlation analysis method; the comparison of the performance improvement of the model by using the AOD input value was tested, and the performance of these models was evaluated using mean absolute error (MAE) and root mean square error (RMSE). The prediction accuracies of Bayes-LSTM and CNN-LSTM deep learning models were both improved when AOD was included as an input parameter. Improvement of the prediction accuracy of the CNN-LSTM model was particularly enhanced in the non-heating season; in the heating season, the prediction accuracy of the Bayes-LSTM model slightly improved, while the prediction accuracy of the CNN-LSTM model slightly decreased. We propose two novel deep learning models for PM₂.₅ concentration prediction in UB, Bayes-LSTM, and CNN-LSTM deep learning models. Pioneering the use of AOD data from H8 and demonstrating the inclusion of AOD input data improves the performance of our two proposed deep learning models.

Keywords: deep learning, AOD, PM2.5, prediction, Ulaanbaatar

Procedia PDF Downloads 25
2994 Direct Synthesis of Composite Materials Type MCM-41/ZSM-5 by Hydrothermal at Atmospheric Pressure in Sealed Pyrex Tubes

Authors: Zoubida Lounis, Naouel Boumesla, Abd El Kader Bengueddach

Abstract:

The main objective of this study is to synthesize a composite materials by direct synthesis at atmospheric pression having the MFI structure and MCM-41 by using double structuring. In the first part of this work we are interested in the study of the synthesis parameters, in addition to temperature, the crystallization time and pH. The second part of this work is to vary the ratio of the concentrations of both structuring C9 [C9H19(CH3)3NBr] and C16 [C16H33(CH3)3NBr] and determining the area of formation of the two materials (microporous and mesoporous at same time), for this reason we performed a battery of experiments ranging from 0 to 100% for both structural. To enhance the economic purposes of this study, the experiments were carried out by using very cheap and simple process, the pyrex tubes were used instead of the reactors, and the synthesis were done at atmospheric pressure and moderate temperature. The final products (composite materials) were obtained at high and pure quality.

Keywords: composite materials, syntheisis, catalysts, mesoporous materials, microporous materials

Procedia PDF Downloads 362
2993 Life Prediction of Condenser Tubes Applying Fuzzy Logic and Neural Network Algorithms

Authors: A. Majidian

Abstract:

The life prediction of thermal power plant components is necessary to prevent the unexpected outages, optimize maintenance tasks in periodic overhauls and plan inspection tasks with their schedules. One of the main critical components in a power plant is condenser because its failure can affect many other components which are positioned in downstream of condenser. This paper deals with factors affecting life of condenser. Failure rates dependency vs. these factors has been investigated using Artificial Neural Network (ANN) and fuzzy logic algorithms. These algorithms have shown their capabilities as dynamic tools to evaluate life prediction of power plant equipments.

Keywords: life prediction, condenser tube, neural network, fuzzy logic

Procedia PDF Downloads 328
2992 Slope Effect in Emission Evaluation to Assess Real Pollutant Factors

Authors: G. Meccariello, L. Della Ragione

Abstract:

The exposure to outdoor air pollution causes lung cancer and increases the risk of bladder cancer. Because air pollution in urban areas is mainly caused by transportation, it is necessary to evaluate pollutant exhaust emissions from vehicles during their real-world use. Nevertheless their evaluation and reduction is a key problem, especially in the cities, that account for more than 50% of world population. A particular attention was given to the slope variability along the streets during each journey performed by the instrumented vehicle. In this paper we dealt with the problem of describing a quantitatively approach for the reconstruction of GPS coordinates and altitude, in the context of correlation study between driving cycles / emission / geographical location, during an experimental campaign realized with some instrumented cars. Finally the slope analysis can be correlated to the emission and consumption values in a specific road position, and it could be evaluated its influence on their behaviour.

Keywords: air pollution, driving cycles, GPS signal, slope, emission factor, fuel consumption

Procedia PDF Downloads 369
2991 Effect of UV Radiation to Change the Properties of the Composite PA+GF

Authors: Lenka Markovičová, Viera Zatkalíková, Tomasz Garbacz

Abstract:

The development of composite materials and the related design and manufacturing technologies is one of the most important advances in the history of materials. Composites are multifunctional materials having unprecedented mechanical and physical properties that can be tailored to meet the requirements of a particular application. Some composites also exhibit great resistance to high-temperature corrosion, oxidation, and wear. Polymers are widely used indoors and outdoors, therefore they are exposed to a chemical environment which may include atmospheric oxygen, acidic fumes, acidic rain, moisture heat and thermal shock, ultra-violet light, high energy radiation, etc. Different polymers are affected differently by these factors even though the amorphous polymers are more sensitive. Ageing is also important and it is defined as the process of deterioration of engineering materials resulting from the combined effects of atmospheric radiation, heat, oxygen, water, micro-organisms and other atmospheric factors.

Keywords: composites with glass fibers, mechanical properties, polyamides, UV degradation

Procedia PDF Downloads 256
2990 Wind Speed Prediction Using Passive Aggregation Artificial Intelligence Model

Authors: Tarek Aboueldahab, Amin Mohamed Nassar

Abstract:

Wind energy is a fluctuating energy source unlike conventional power plants, thus, it is necessary to accurately predict short term wind speed to integrate wind energy in the electricity supply structure. To do so, we present a hybrid artificial intelligence model of short term wind speed prediction based on passive aggregation of the particle swarm optimization and neural networks. As a result, improvement of the prediction accuracy is obviously obtained compared to the standard artificial intelligence method.

Keywords: artificial intelligence, neural networks, particle swarm optimization, passive aggregation, wind speed prediction

Procedia PDF Downloads 426
2989 An Artificial Intelligence Framework to Forecast Air Quality

Authors: Richard Ren

Abstract:

Air pollution is a serious danger to international well-being and economies - it will kill an estimated 7 million people every year, costing world economies $2.6 trillion by 2060 due to sick days, healthcare costs, and reduced productivity. In the United States alone, 60,000 premature deaths are caused by poor air quality. For this reason, there is a crucial need to develop effective methods to forecast air quality, which can mitigate air pollution’s detrimental public health effects and associated costs by helping people plan ahead and avoid exposure. The goal of this study is to propose an artificial intelligence framework for predicting future air quality based on timing variables (i.e. season, weekday/weekend), future weather forecasts, as well as past pollutant and air quality measurements. The proposed framework utilizes multiple machine learning algorithms (logistic regression, random forest, neural network) with different specifications and averages the results of the three top-performing models to eliminate inaccuracies, weaknesses, and biases from any one individual model. Over time, the proposed framework uses new data to self-adjust model parameters and increase prediction accuracy. To demonstrate its applicability, a prototype of this framework was created to forecast air quality in Los Angeles, California using datasets from the RP4 weather data repository and EPA pollutant measurement data. The results showed good agreement between the framework’s predictions and real-life observations, with an overall 92% model accuracy. The combined model is able to predict more accurately than any of the individual models, and it is able to reliably forecast season-based variations in air quality levels. Top air quality predictor variables were identified through the measurement of mean decrease in accuracy. This study proposed and demonstrated the efficacy of a comprehensive air quality prediction framework leveraging multiple machine learning algorithms to overcome individual algorithm shortcomings. Future enhancements should focus on expanding and testing a greater variety of modeling techniques within the proposed framework, testing the framework in different locations, and developing a platform to automatically publish future predictions in the form of a web or mobile application. Accurate predictions from this artificial intelligence framework can in turn be used to save and improve lives by allowing individuals to protect their health and allowing governments to implement effective pollution control measures.Air pollution is a serious danger to international wellbeing and economies - it will kill an estimated 7 million people every year, costing world economies $2.6 trillion by 2060 due to sick days, healthcare costs, and reduced productivity. In the United States alone, 60,000 premature deaths are caused by poor air quality. For this reason, there is a crucial need to develop effective methods to forecast air quality, which can mitigate air pollution’s detrimental public health effects and associated costs by helping people plan ahead and avoid exposure. The goal of this study is to propose an artificial intelligence framework for predicting future air quality based on timing variables (i.e. season, weekday/weekend), future weather forecasts, as well as past pollutant and air quality measurements. The proposed framework utilizes multiple machine learning algorithms (logistic regression, random forest, neural network) with different specifications and averages the results of the three top-performing models to eliminate inaccuracies, weaknesses, and biases from any one individual model. Over time, the proposed framework uses new data to self-adjust model parameters and increase prediction accuracy. To demonstrate its applicability, a prototype of this framework was created to forecast air quality in Los Angeles, California using datasets from the RP4 weather data repository and EPA pollutant measurement data. The results showed good agreement between the framework’s predictions and real-life observations, with an overall 92% model accuracy. The combined model is able to predict more accurately than any of the individual models, and it is able to reliably forecast season-based variations in air quality levels. Top air quality predictor variables were identified through the measurement of mean decrease in accuracy. This study proposed and demonstrated the efficacy of a comprehensive air quality prediction framework leveraging multiple machine learning algorithms to overcome individual algorithm shortcomings. Future enhancements should focus on expanding and testing a greater variety of modeling techniques within the proposed framework, testing the framework in different locations, and developing a platform to automatically publish future predictions in the form of a web or mobile application. Accurate predictions from this artificial intelligence framework can in turn be used to save and improve lives by allowing individuals to protect their health and allowing governments to implement effective pollution control measures.Air pollution is a serious danger to international wellbeing and economies - it will kill an estimated 7 million people every year, costing world economies $2.6 trillion by 2060 due to sick days, healthcare costs, and reduced productivity. In the United States alone, 60,000 premature deaths are caused by poor air quality. For this reason, there is a crucial need to develop effective methods to forecast air quality, which can mitigate air pollution’s detrimental public health effects and associated costs by helping people plan ahead and avoid exposure. The goal of this study is to propose an artificial intelligence framework for predicting future air quality based on timing variables (i.e. season, weekday/weekend), future weather forecasts, as well as past pollutant and air quality measurements. The proposed framework utilizes multiple machine learning algorithms (logistic regression, random forest, neural network) with different specifications and averages the results of the three top-performing models to eliminate inaccuracies, weaknesses, and biases from any one individual model. Over time, the proposed framework uses new data to self-adjust model parameters and increase prediction accuracy. To demonstrate its applicability, a prototype of this framework was created to forecast air quality in Los Angeles, California using datasets from the RP4 weather data repository and EPA pollutant measurement data. The results showed good agreement between the framework’s predictions and real-life observations, with an overall 92% model accuracy. The combined model is able to predict more accurately than any of the individual models, and it is able to reliably forecast season-based variations in air quality levels. Top air quality predictor variables were identified through the measurement of mean decrease in accuracy. This study proposed and demonstrated the efficacy of a comprehensive air quality prediction framework leveraging multiple machine learning algorithms to overcome individual algorithm shortcomings. Future enhancements should focus on expanding and testing a greater variety of modeling techniques within the proposed framework, testing the framework in different locations, and developing a platform to automatically publish future predictions in the form of a web or mobile application. Accurate predictions from this artificial intelligence framework can in turn be used to save and improve lives by allowing individuals to protect their health and allowing governments to implement effective pollution control measures.

Keywords: air quality prediction, air pollution, artificial intelligence, machine learning algorithms

Procedia PDF Downloads 96
2988 SNR Classification Using Multiple CNNs

Authors: Thinh Ngo, Paul Rad, Brian Kelley

Abstract:

Noise estimation is essential in today wireless systems for power control, adaptive modulation, interference suppression and quality of service. Deep learning (DL) has already been applied in the physical layer for modulation and signal classifications. Unacceptably low accuracy of less than 50% is found to undermine traditional application of DL classification for SNR prediction. In this paper, we use divide-and-conquer algorithm and classifier fusion method to simplify SNR classification and therefore enhances DL learning and prediction. Specifically, multiple CNNs are used for classification rather than a single CNN. Each CNN performs a binary classification of a single SNR with two labels: less than, greater than or equal. Together, multiple CNNs are combined to effectively classify over a range of SNR values from −20 ≤ SNR ≤ 32 dB.We use pre-trained CNNs to predict SNR over a wide range of joint channel parameters including multiple Doppler shifts (0, 60, 120 Hz), power-delay profiles, and signal-modulation types (QPSK,16QAM,64-QAM). The approach achieves individual SNR prediction accuracy of 92%, composite accuracy of 70% and prediction convergence one order of magnitude faster than that of traditional estimation.

Keywords: classification, CNN, deep learning, prediction, SNR

Procedia PDF Downloads 114
2987 Evaluation of Spatial Distribution Prediction for Site-Scale Soil Contaminants Based on Partition Interpolation

Authors: Pengwei Qiao, Sucai Yang, Wenxia Wei

Abstract:

Soil pollution has become an important issue in China. Accurate spatial distribution prediction of pollutants with interpolation methods is the basis for soil remediation in the site. However, a relatively strong variability of pollutants would decrease the prediction accuracy. Theoretically, partition interpolation can result in accurate prediction results. In order to verify the applicability of partition interpolation for a site, benzo (b) fluoranthene (BbF) in four soil layers was adopted as the research object in this paper. IDW (inverse distance weighting)-, RBF (radial basis function)-and OK (ordinary kriging)-based partition interpolation accuracies were evaluated, and their influential factors were analyzed; then, the uncertainty and applicability of partition interpolation were determined. Three conclusions were drawn. (1) The prediction error of partitioned interpolation decreased by 70% compared to unpartitioned interpolation. (2) Partition interpolation reduced the impact of high CV (coefficient of variation) and high concentration value on the prediction accuracy. (3) The prediction accuracy of IDW-based partition interpolation was higher than that of RBF- and OK-based partition interpolation, and it was suitable for the identification of highly polluted areas at a contaminated site. These results provide a useful method to obtain relatively accurate spatial distribution information of pollutants and to identify highly polluted areas, which is important for soil pollution remediation in the site.

Keywords: accuracy, applicability, partition interpolation, site, soil pollution, uncertainty

Procedia PDF Downloads 122
2986 Removal of Gaseous Pollutant from the Flue Gas in a Submerged Self-Priming Venturi Scrubber

Authors: Manisha Bal, B. C. Meikap

Abstract:

Hydrogen chloride is the most common acid gas emitted by the industries. HCl gas is listed as Title III hazardous air pollutant. It causes severe threat to the human health as well as environment. So, removal of HCl from flue gases is very imperative. In the present study, submerged self-priming venturi scrubber is chosen to remove the HCl gas with water as a scrubbing liquid. Venturi scrubber is the most popular device for the removal of gaseous pollutants. Main mechanism behind the venturi scrubber is the polluted gas stream enters at converging section which accelerated to maximum velocity at throat section. A very interesting thing in case of submerged condition, venturi scrubber is submerged inside the liquid tank and liquid is entered at throat section because of suction created due to large pressure drop generated at the throat section. Maximized throat gas velocity atomizes the entered liquid into number of tiny droplets. Gaseous pollutant HCl is absorbed from gas to liquid droplets inside the venturi scrubber due to interaction between the gas and water. Experiments were conducted at different throat gas velocity, water level and inlet concentration of HCl to enhance the HCl removal efficiency. The effect of throat gas velocity, inlet concentration of HCl, and water level on removal efficiency of venturi scrubber has been evaluated. Present system yielded very high removal efficiency for the scrubbing of HCl gas which is more than 90%. It is also concluded that the removal efficiency of HCl increases with increasing throat gas velocity, inlet HCl concentration, and water level height.

Keywords: air pollution, HCl scrubbing, mass transfer, self-priming venturi scrubber

Procedia PDF Downloads 120
2985 Uplink Throughput Prediction in Cellular Mobile Networks

Authors: Engin Eyceyurt, Josko Zec

Abstract:

The current and future cellular mobile communication networks generate enormous amounts of data. Networks have become extremely complex with extensive space of parameters, features and counters. These networks are unmanageable with legacy methods and an enhanced design and optimization approach is necessary that is increasingly reliant on machine learning. This paper proposes that machine learning as a viable approach for uplink throughput prediction. LTE radio metric, such as Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), and Signal to Noise Ratio (SNR) are used to train models to estimate expected uplink throughput. The prediction accuracy with high determination coefficient of 91.2% is obtained from measurements collected with a simple smartphone application.

Keywords: drive test, LTE, machine learning, uplink throughput prediction

Procedia PDF Downloads 135
2984 Study on the Model Predicting Post-Construction Settlement of Soft Ground

Authors: Pingshan Chen, Zhiliang Dong

Abstract:

In order to estimate the post-construction settlement more objectively, the power-polynomial model is proposed, which can reflect the trend of settlement development based on the observed settlement data. It was demonstrated by an actual case history of an embankment, and during the prediction. Compared with the other three prediction models, the power-polynomial model can estimate the post-construction settlement more accurately with more simple calculation.

Keywords: prediction, model, post-construction settlement, soft ground

Procedia PDF Downloads 409
2983 An Auxiliary Technique for Coronary Heart Disease Prediction by Analyzing Electrocardiogram Based on ResNet and Bi-Long Short-Term Memory

Authors: Yang Zhang, Jian He

Abstract:

Heart disease is one of the leading causes of death in the world, and coronary heart disease (CHD) is one of the major heart diseases. Electrocardiogram (ECG) is widely used in the detection of heart diseases, but the traditional manual method for CHD prediction by analyzing ECG requires lots of professional knowledge for doctors. This paper introduces sliding window and continuous wavelet transform (CWT) to transform ECG signals into images, and then ResNet and Bi-LSTM are introduced to build the ECG feature extraction network (namely ECGNet). At last, an auxiliary system for coronary heart disease prediction was developed based on modified ResNet18 and Bi-LSTM, and the public ECG dataset of CHD from MIMIC-3 was used to train and test the system. The experimental results show that the accuracy of the method is 83%, and the F1-score is 83%. Compared with the available methods for CHD prediction based on ECG, such as kNN, decision tree, VGGNet, etc., this method not only improves the prediction accuracy but also could avoid the degradation phenomenon of the deep learning network.

Keywords: Bi-LSTM, CHD, ECG, ResNet, sliding window

Procedia PDF Downloads 65
2982 4-Chlorophenol Degradation in Water Using TIO₂-X%ZnS Synthesized by One-Step Sol-Gel Method

Authors: M. E. Velásquez Torres, F. Tzompantzi, J. C. Castillo-Rodríguez, A. G. Romero Villegas, S. Mendéz-Salazar, C. E. Santolalla-Vargas, J. Cardoso-Martínez

Abstract:

Photocatalytic degradation, as an advanced oxidation technology, is a promising method in organic pollutant degradation. In this sense, chlorophenols should be removed from the water because they are highly toxic. The TiO₂ - X% ZnS photocatalysts, where X represents the molar percentage of ZnS (3%, 5%, 10%, and 15%), were synthesized using the one-step sol-gel method to use them as photocatalysts to degrade 4-chlorophenol. The photocatalysts were synthesized by a one-step sol-gel method. They were refluxed for 36 hours, dried at 80°C, and calcined at 400°C. They were labeled TiO₂ - X%ZnS, where X represents the molar percentage of ZnS (3%, 5%, 10%, and 15%). The band gap was calculated using a Cary 100 UV-Visible Spectrometer with an integrating sphere accessory. Ban gap value of each photocatalyst was: 2.7 eV of TiO₂, 2.8 eV of TiO₂ - 3%ZnS and TiO₂ - 5%ZnS, 2.9 eV of TiO₂ - 10%ZnS and 2.6 eV of TiO2 - 15%ZnS. In a batch type reactor, under the irradiation of a mercury lamp (λ = 254 nm, Pen-Ray), degradations of 55 ppm 4-chlorophenol were obtained at 360 minutes with the synthesized photocatalysts: 60% (3% ZnS), 66% (5% ZnS), 74% (10% ZnS) and 58% (15% ZnS). In this sense, the best material as a photocatalyst was TiO₂ -10%ZnS with a degradation percentage of 74%.

Keywords: 4-chlorophenol, photocatalysis, water pollutant, sol-gel

Procedia PDF Downloads 101
2981 Understanding Health-Related Properties of Grapes by Pharmacokinetic Modelling of Intestinal Absorption

Authors: Sophie N. Selby-Pham, Yudie Wang, Louise Bennett

Abstract:

Consumption of grapes promotes health and reduces the risk of chronic diseases due to the action of grape phytochemicals in regulation of Oxidative Stress and Inflammation (OSI). The bioefficacy of phytochemicals depends on their absorption in the human body. The time required for phytochemicals to achieve maximal plasma concentration (Tₘₐₓ) after oral intake reflects the time window of maximal bioefficacy of phytochemicals, with Tₘₐₓ dependent on physicochemical properties of phytochemicals. This research collated physicochemical properties of grape phytochemicals from white and red grapes to predict their Tₘₐₓ using pharmacokinetic modelling. The predicted values of Tₘₐₓ were then compared to the measured Tₘₐₓ collected from clinical studies to determine the accuracy of prediction. In both liquid and solid intake forms, white grapes exhibit a shorter Tₘₐₓ range (0.5-2.5 h) versus red grapes (1.5-5h). The prediction accuracy of Tₘₐₓ for grape phytochemicals was 33.3% total error of prediction compared to the mean, indicating high prediction accuracy. Pharmacokinetic modelling allows prediction of Tₘₐₓ without costly clinical trials, informing dosing frequency for sustained presence of phytochemicals in the body to optimize the health benefits of phytochemicals.

Keywords: absorption kinetics, phytochemical, phytochemical absorption prediction model, Vitis vinifera

Procedia PDF Downloads 127
2980 Artificial Neural Network in FIRST Robotics Team-Based Prediction System

Authors: Cedric Leong, Parth Desai, Parth Patel

Abstract:

The purpose of this project was to develop a neural network based on qualitative team data to predict alliance scores to determine winners of matches in the FIRST Robotics Competition (FRC). The game for the competition changes every year with different objectives and game objects, however the idea was to create a prediction system which can be reused year by year using some of the statistics that are constant through different games, making our system adaptable to future games as well. Aerial Assist is the FRC game for 2014, and is played in alliances of 3 teams going against one another, namely the Red and Blue alliances. This application takes any 6 teams paired into 2 alliances of 3 teams and generates the prediction for the final score between them.

Keywords: artifical neural network, prediction system, qualitative team data, FIRST Robotics Competition (FRC)

Procedia PDF Downloads 489
2979 Apparent Ageing Mechanism of Polyurethane Coating in Typical Atmospheric Environment

Authors: Jin Gao, Jin Zhang, Xiaogang Li

Abstract:

Outdoor exposure experiments were conducted in three extreme environments, namely the Chinese plateau mountain environment (Lhasa), the cold–temperate environment (Mohe), and the marine atmospheric environment (Wanning), to track a new long-life environment-friendly polyurethane coating. The relationship between apparent properties, namely gloss and microstructural changes, was analyzed, and the influence of typical climatic environment on the aging mechanism of polyurethane coatings was discussed. Results show that the UV radiation in the Lhasa area causes photoaging degradation, micropores are formed on the coating surface, and the powdering phenomenon is obvious. Photodegradation occurs in the Wanning area, and a hydrolysis reaction is observed. The hydrolysis reaction catalyzes the photoaging, the coating surface becomes yellow, and the powdering becomes serious. Photoaging is also present in the Mohe area, but it is mainly due to temperature changes that in turn change the internal stress of the coating. Microcracks and bumps form on the coating surface.

Keywords: aging, atmospheric environment, outdoor exposure, polyurethane coating

Procedia PDF Downloads 101