Search results for: accurately

Authors: Nacera Yassa, Abdelmalek Saidoune, Ghania Ouadfel, Hamza Houassine

Abstract:

The rapid advancement in electrical technologies has underscored the increasing importance of multiphase machines across various industrial sectors. These machines offer significant advantages in terms of efficiency, compactness, and reliability compared to their single-phase counterparts. However, early detection and diagnosis of electrical faults remain critical challenges to ensure the durability and safety of these complex systems. This paper presents an advanced analytical model for multiphase machines, with a particular focus on dual stator asynchronous machines. The primary objective is to develop a robust diagnostic tool capable of effectively detecting and locating electrical faults in these machines, including short circuits, winding faults, and voltage imbalances. The proposed methodology relies on an analytical approach combining electrical machine theory, modeling of magnetic and electrical circuits, and advanced signal analysis techniques. By employing detailed analytical equations, the developed model accurately simulates the behavior of multiphase machines in the presence of electrical faults. The effectiveness of the proposed model is demonstrated through a series of case studies and numerical simulations. In particular, special attention is given to analyzing the dynamic behavior of machines under different types of faults, as well as optimizing diagnostic and recovery strategies. The obtained results pave the way for new advancements in the field of multiphase machine diagnostics, with potential applications in various sectors such as automotive, aerospace, and renewable energies. By providing precise and reliable tools for early fault detection, this research contributes to improving the reliability and durability of complex electrical systems while reducing maintenance and operation costs.

Keywords: faults, diagnosis, modelling, multiphase machine

Procedia PDF Downloads 26

Authors: Zahrasadat Hosseini, Jie Yuan

Abstract:

Point-of-care (POC) diagnostic devices based on lab-on-a-chip (LOC) technology have the potential to revolutionize medical diagnostics. However, the development of an ideal microfluidic system based on LOC technology for diagnostics purposes requires overcoming several obstacles, such as improving sensitivity, selectivity, portability, cost-effectiveness, and prototyping methods. While numerous studies have introduced technologies and systems that advance these criteria, existing systems still have limitations. Electrochemical enzyme-linked immunosorbent assay (e-ELISA) in a LOC device offers numerous advantages, including enhanced sensitivity, decreased turnaround time, minimized sample and analyte consumption, reduced cost, disposability, and suitability for miniaturization, integration, and multiplexing. In this study, we present a novel design and fabrication method for a microfluidic diagnostic platform that integrates screen-printed electrochemical carbon/silver chloride electrodes on flexible printed circuit boards with flexible, multilayer, polydimethylsiloxane (PDMS) microfluidic networks to accurately manipulate and pre-immobilize analytes for performing electrochemical enzyme-linked immunosorbent assay (e-ELISA) for multiplexed quantification of blood serum biomarkers. We further demonstrate fast, cost-effective prototyping, as well as accurate and reliable detection performance of this device for quantification of interleukin-6-spiked samples through electrochemical analytics methods. We anticipate that our invention represents a significant step towards the development of user-friendly, portable, medical-grade, POC diagnostic devices.

Keywords: lab-on-a-chip, point-of-care diagnostics, electrochemical ELISA, biomarker quantification, fast prototyping

Procedia PDF Downloads 56

Authors: Zahrasadat Hosseini, Jie Yuan

Abstract:

Point-of-care (POC) diagnostic devices based on lab-on-a-chip (LOC) technology have the potential to revolutionize medical diagnostics. However, the development of an ideal microfluidic system based on LOC technology for diagnostics purposes requires overcoming several obstacles, such as improving sensitivity, selectivity, portability, cost-effectiveness, and prototyping methods. While numerous studies have introduced technologies and systems that advance these criteria, existing systems still have limitations. Electrochemical enzyme-linked immunosorbent assay (e-ELISA) in a LOC device offers numerous advantages, including enhanced sensitivity, decreased turnaround time, minimized sample and analyte consumption, reduced cost, disposability, and suitability for miniaturization, integration, and multiplexing. In this study, we present a novel design and fabrication method for a microfluidic diagnostic platform that integrates screen-printed electrochemical carbon/silver chloride electrodes on flexible printed circuit boards with flexible, multilayer, polydimethylsiloxane (PDMS) microfluidic networks to accurately manipulate and pre-immobilize analytes for performing electrochemical enzyme-linked immunosorbent assay (e-ELISA) for multiplexed quantification of blood serum biomarkers. We further demonstrate fast, cost-effective prototyping, as well as accurate and reliable detection performance of this device for quantification of interleukin-6-spiked samples through electrochemical analytics methods. We anticipate that our invention represents a significant step towards the development of user-friendly, portable, medical-grade POC diagnostic devices.

Keywords: lab-on-a-chip, point-of-care diagnostics, electrochemical ELISA, biomarker quantification, fast prototyping

Procedia PDF Downloads 62

Authors: Chia-Wen Yeh, Ting-Sheng Lin, Chih-Han Chang

Abstract:

Pressure garment has been used to prevent and treat the hypertrophic scars following serious burns since 1970s. The use of pressure garment is believed to hasten the maturation process and decrease the highness of scars. Pressure garment is custom made by reducing circumferential measurement of the patient by 10%~20%, called Reduction Factor. However the exact reducing value used depends on the subjective judgment of the therapist and the feeling of patients throughout the try and error process. The Laplace Law can be applied to calculate the pressure from the dimension of the pressure garment by the circumferential measurements of the patients and the tension profile of the fabrics. The tension profile currently obtained neglects the stress relaxation and hysteresis effect within most elastic fabrics. The purpose of this study was to investigate the influence of the tension attenuation, from stress relaxation and hysteresis effect of the fabrics. Samples of pressure garment were obtained from Sunshine Foundation Organization, a nonprofit organization for burn patients in Taiwan. The wall tension profile of pressure garments were measured on a material testing system. Specimens were extended to 10% of the original length, held for 1 hour for the influence of the stress relaxation effect to take place. Then, specimens were extended to 15% of the original length for 10 seconds, then reduced to 10% to simulate donning movement for the influence of the hysteresis effect to take place. The load history was recorded. The stress relaxation effect is obvious from the load curves. The wall tension is decreased by 8.5%~10% after 60mins of holding. The hysteresis effect is obvious from the load curves. The wall tension is increased slightly, then decreased by 1.5%~2.5% and lower than stress relaxation results after 60mins of holding. The wall tension attenuation of the fabric exists due to stress relaxation and hysteresis effect. The influence of hysteresis is more than stress relaxation. These effect should be considered in order to design and evaluate the pressure of pressure garment more accurately.

Keywords: hypertrophic scars, hysteresis, pressure garment, stress relaxation

Procedia PDF Downloads 490

Authors: Arash Yousefi

Abstract:

Experts suggest that relying exclusively on parameters to convey a situation or establish a condition may not be adequate. Assessing and appraising incidents in a system based on accident parameters, such as accident frequency, lost workdays, or fatalities, may not always be precise and occasionally erroneous. The frequency rate of accidents is a metric that assesses the correlation between the number of accidents causing work-time loss due to injuries and the total working hours of personnel over a year. Traditionally, this has been calculated based on one million working hours, but the American Occupational Safety and Health Organization has updated its standards. The new coefficient of 200/000 working hours is now used to compute the frequency rate of accidents. It's crucial to ensure that the total working hours of employees are equally represented when calculating individual event and incident numbers. The accident severity rate is a metric used to determine the amount of time lost or wasted during a given period, often a year, in relation to the total number of working hours. It measures the percentage of work hours lost or wasted compared to the total number of useful working hours, which provides valuable insight into the number of days lost or wasted due to work-related incidents for each working hour. Calculating the severity of an incident can be difficult if a worker suffers permanent disability or death. To determine lost days, coefficients specified in the "tables of days equivalent to OSHA or ANSI standards" for disabling injuries are used. The accident frequency coefficient denotes the rate at which accidents occur, while the accident severity coefficient specifies the extent of damage and injury caused by these accidents. These coefficients are crucial in accurately assessing the magnitude and impact of accidents.

Keywords: incidents, safety, analysis, frequency, severity, injuries, determine

Procedia PDF Downloads 60

Authors: Winncy Du, Jeremy Nguyen, Harpinder Dhillon, Reinardus Justin Halim, Clayton Haske, Trent Hughes, Marissa Ortiz, Rozy Saini

Abstract:

Stroke often causes death or cerebro-vascular (CV) brain damage. Most patients with CV brain damage lost their motor control on their limbs. This paper focuses on developing a reliable, safe, and non-invasive EEG-based robot-assistant stroke rehabilitation system to help stroke survivors to rapidly restore their motor control functions for their limbs. An electroencephalogram (EEG) recording device (EPOC Headset) and was used to detect a patient’s brain activities. The EEG signals were then processed, classified, and interpreted to the motion intentions, and then converted to a series of robot motion commands. A six-axis articulated robot (AdeptSix 300) was employed to provide the intended motions based on these commends. To ensure the EEG device, the computer, and the robot can communicate to each other, an Arduino microcontroller is used to physically execute the programming codes to a series output pins’ status (HIGH or LOW). Then these “hardware” commends were sent to a 24 V relay to trigger the robot’s motion. A lookup table for various motion intensions and the associated EEG signal patterns were created (through training) and installed in the microcontroller. Thus, the motion intention can be direct determined by comparing the EEG patterns obtaibed from the patient with the look-up table’s EEG patterns; and the corresponding motion commends are sent to the robot to provide the intended motion without going through feature extraction and interpretation each time (a time-consuming process). For safety sake, an extender was designed and attached to the robot’s end effector to ensure the patient is beyond the robot’s workspace. The gripper is also designed to hold the patient’s limb. The test results of this rehabilitation system show that it can accurately interpret the patient’s motion intension and move the patient’s arm to the intended position.

Keywords: brain waves, EEG sensor, motion control, robot-assistant stroke rehabilitation

Procedia PDF Downloads 359

Authors: Anthony Quansah, Tresor Ntaryamira, Shula Mushota

Abstract:

Models for geotechnical centrifuge testing are usually made from re-formed soil, allowing for comparisons with naturally occurring soil deposits. However, there is a fundamental omission in this process because the natural soil is deposited in layers creating a unique structure. Nonlinear dynamics of clay material deposit is an essential part of changing the attributes of ground movements when subjected to solid seismic loading, particularly when diverse intensification conduct of speeding up and relocation are considered. The paper portrays a review of axis shaking table tests and numerical recreations to explore the offshore clay deposits subjected to seismic loadings. These perceptions are accurately reenacted by DEEPSOIL with appropriate soil models and parameters reviewed from noteworthy centrifuge modeling researches. At that point, precise 1-D site reaction investigations are performed on both time and recurrence spaces. The outcomes uncover that for profound delicate clay is subjected to expansive quakes, noteworthy increasing speed lessening may happen close to the highest point of store because of soil nonlinearity and even neighborhood shear disappointment; nonetheless, huge enhancement of removal at low frequencies are normal in any case the forces of base movements, which proposes that for dislodging touchy seaward establishments and structures, such intensified low-recurrence relocation reaction will assume an essential part in seismic outline. This research shows centrifuge as a tool for creating a layered sample important for modelling true soil behaviour (such as permeability) which is not identical in all directions. Currently, there are limited methods for creating layered soil samples.

Keywords: seismic analysis, layered modeling, terotechnology, finite element modeling

Procedia PDF Downloads 131

Authors: V. Manjunath, C. V. Chandrashekara

Abstract:

Ecosystem authority calls for the use of lubricants with less effect on the nature in terms of exhaust emission, while engine user demands more mileage per liter of fuel without any compromise on engine durability. From this viewpoint, engine manufacturers require the optimum combination of materials and lubricant additive package to minimize friction and wear in the engine components like piston, crankshaft and valve train etc. The demands are placed for requirements to operate at higher speeds, loads, temperature and for extended replacement intervals of engine oil. Besides, it is necessary to accurately predict the lubricant life or the replacement interval to prevent lubrication and valve-train components failure. Experimental tribology evaluation of new engine oils requires large amount of time and energy. Hence low cost bench test is necessary for industries and original equipment manufacturing companies (OEM) to study the performance of lubricants. The present work outlines the procedure for the design and development of a valve train wear rig (MCR) to simulate the ASTMD-6891 and to develop new engine test for Indian automobile sector to evaluate lubricants for Indian automobile market. In order to improve the lubrication between cam and follower of internal combustion engine, the influence of materials or oils viscosity and additives on the friction and wear characteristics are examined with test rig by increasing the contact load at two different revolution speed. From the experimentation following results are made obvious. Temperature, Torque, speed and wear plots are used to validate the data obtained from the newly developed multi-cam cam rig (MCR) with follower against a cast iron camshaft. Camshaft lobe wear is measured at seven different locations on cam profile. Tribofilm formed using 5W-30 oil is evaluated and correlated with the standard test results.

Keywords: ASTMD-6891, multi-cam rig (MCR), 5W-30, cam-profile

Procedia PDF Downloads 152

Authors: Abdulaziz Alsadhan, Naveed Khan

Abstract:

In recent years intrusions on computer network are the major security threat. Hence, it is important to impede such intrusions. The hindrance of such intrusions entirely relies on its detection, which is primary concern of any security tool like Intrusion Detection System (IDS). Therefore, it is imperative to accurately detect network attack. Numerous intrusion detection techniques are available but the main issue is their performance. The performance of IDS can be improved by increasing the accurate detection rate and reducing false positive. The existing intrusion detection techniques have the limitation of usage of raw data set for classification. The classifier may get jumble due to redundancy, which results incorrect classification. To minimize this problem, Principle Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Local Binary Pattern (LBP) can be applied to transform raw features into principle features space and select the features based on their sensitivity. Eigen values can be used to determine the sensitivity. To further classify, the selected features greedy search, back elimination, and Particle Swarm Optimization (PSO) can be used to obtain a subset of features with optimal sensitivity and highest discriminatory power. These optimal feature subset used to perform classification. For classification purpose, Support Vector Machine (SVM) and Multilayer Perceptron (MLP) used due to its proven ability in classification. The Knowledge Discovery and Data mining (KDD’99) cup dataset was considered as a benchmark for evaluating security detection mechanisms. The proposed approach can provide an optimal intrusion detection mechanism that outperforms the existing approaches and has the capability to minimize the number of features and maximize the detection rates.

Keywords: Particle Swarm Optimization (PSO), Principle Component Analysis (PCA), Linear Discriminant Analysis (LDA), Local Binary Pattern (LBP), Support Vector Machine (SVM), Multilayer Perceptron (MLP)

Procedia PDF Downloads 343

Authors: Shan-Hong Ying, Ying-Fang Wang

Abstract:

A surrogate approach was deployed for assessing exposures of multiple chemicals at the selected working area of coating processes and applied to assess the exposure concentration of similar exposed groups using the same chemicals but different formula ratios. For the selected area, 6 to 12 portable photoionization detector (PID) were placed uniformly in its workplace to measure its total VOCs concentrations (CT-VOCs) for 6 randomly selected workshifts. Simultaneously, one sampling strain was placed beside one of these portable PIDs, and the collected air sample was analyzed for individual concentration (CVOCi) of 5 VOCs (xylene, butanone, toluene, butyl acetate, and dimethylformamide). Predictive models were established by relating the CT-VOCs to CVOCi of each individual compound via simple regression analysis. The established predictive models were employed to predict each CVOCi based on the measured CT-VOC for each the similar working area using the same portable PID. Results show that predictive models obtained from simple linear regression analyses were found with an R2 = 0.83~0.99 indicating that CT-VOCs were adequate for predicting CVOCi. In order to verify the validity of the exposure prediction model, the sampling analysis of the above chemical substances was further carried out and the correlation between the measured value (Cm) and the predicted value (Cp) was analyzed. It was found that there is a good correction between the predicted value and measured value of each measured chemical substance (R2=0.83~0.98). Therefore, the surrogate approach could be assessed the exposure concentration of similar exposed groups using the same chemicals but different formula ratios. However, it is recommended to establish the prediction model between the chemical substances belonging to each coater and the direct-reading PID, which is more representative of reality exposure situation and more accurately to estimate the long-term exposure concentration of operators.

Keywords: exposure assessment, exposure prediction model, surrogate approach, TVOC

Procedia PDF Downloads 124

Authors: Małgorzata Smolarek

Abstract:

One of the barriers to the development of small and medium-sized enterprises (SME) are difficulties connected with management of human resources. The first part of article defines the specifics of staff management in small and medium enterprises. The practical part presents results of own studies in the area of diagnosis of the state of the human resources management in small and medium-sized enterprises in Poland. It takes into account its impact on the functioning of SME in a variable environment. This part presents findings of empirical studies, which enabled verification of the hypotheses and formulation of conclusions. The findings presented in this paper were obtained during the implementation of the project entitled 'Tendencies and challenges in strategic managing SME in Silesian Voivodeship.' The aim of the studies was to diagnose the state of strategic management and human resources management taking into account its impact on the functioning of small and medium enterprises operating in Silesian Voivodeship in Poland and to indicate improvement areas of the model under diagnosis. One of the specific objectives of the studies was to diagnose the state of the process of strategic management of human resources and to identify fundamental problems. In this area, the main hypothesis was formulated: The enterprises analysed do not have comprehensive strategies for management of human resources. The survey was conducted by questionnaire. Main Research Results: Human resource management in SMEs is characterized by simplicity of procedures, and the lack of sophisticated tools and its specificity depends on the size of the company. The process of human resources management in SME has to be adjusted to the structure of an organisation, result from its objectives, so that an organisation can fully implement its strategic plans and achieve success and competitive advantage on the market. A guarantee of success is an accurately developed policy of human resources management based on earlier analyses of the existing procedures and possessed human resources.

Keywords: human resources management, human resources policy, personnel strategy, small and medium enterprises

Procedia PDF Downloads 220

Authors: Iman Nezamzadeh, Ali Faghih, Behnam Oveisi

Abstract:

In order to reduce earthquake hazards in urban areas, it is necessary to perform comprehensive studies to understand the dynamics of the active faults and identify potentially high risk areas. The fault slip-rates in Late Quaternary sediment are critical indicators of seismic hazard and also provide valuable data to recognize young crustal deformations. To measure slip-rates accurately, is needed to displacement of geomorphic markers and ages of quaternary sediment samples of alluvial deposit that deformed by movements on fault. In this study we produced information about Intermediate term slip rate of Sabzpushan Fault Zone (SPF) within the central part of the Zagros Mountains of Iran using OSL dating technique to make better analysis of seismic hazard and seismic risk reduction for Shiraz city. For this purpose identifiable geomorphic fluvial surfaces help us to provide a reference frame to determine differential or absolute horizontal and vertical deformation. Optically stimulated luminescence (OSL) is an alternative and independent method of determining the burial age of mineral grains in Quaternary sediments. Field observation and satellite imagery show geomorphic markers that deformed horizontally along the Sabzpoushan Fault. Here, drag folds is forming because of evaporites material of Miocen Formation. We estimate 2.8±0.5 mm/yr (mm/y) horizontal slip rate along the Sabzpushan fault zone, where ongoing deformation is involve with drug folding. The Soltan synclinal structure, close to the Sabzpushan fault, shows slight uplift rate due to active core-extrousion.

Keywords: slip rate, active tectonics, OSL, geomorphic markers, Sabzpushan Fault Zone, Zagros, Iran

Procedia PDF Downloads 328

Authors: Chen Wang, Qiuxiao Chen

Abstract:

With the advent of the information age and the era of big data, urban planning research is no longer satisfied with the analysis and application of traditional data. Because of the limitations of traditional urban commercial center system research, big data provides new opportunities for urban research. Therefore, based on the quantitative evaluation method of big data, the commercial center system of the main city of Hangzhou is analyzed and evaluated, and the scale and hierarchical structure characteristics of the urban commercial center system are studied. In order to make up for the shortcomings of the existing POI extraction method, it proposes a POI extraction method based on adaptive adjustment of search window, which can accurately and efficiently extract the POI data of commercial business in the main city of Hangzhou. Through the visualization and nuclear density analysis of the extracted Point of Interest (POI) data, the current situation of the commercial center system in the main city of Hangzhou is evaluated. Then it compares with the commercial center system structure of 'Hangzhou City Master Plan (2001-2020)', analyzes the problems existing in the planned urban commercial center system, and provides corresponding suggestions and optimization strategy for the optimization of the planning of Hangzhou commercial center system. Then get the following conclusions: The status quo of the commercial center system in the main city of Hangzhou presents a first-level main center, a two-level main center, three third-level sub-centers, and multiple community-level business centers. Generally speaking, the construction of the main center in the commercial center system is basically up to standard, and there is still a big gap in the construction of the sub-center and the regional-level commercial center, further construction is needed. Therefore, it proposes an optimized hierarchical functional system, organizes commercial centers in an orderly manner; strengthens the central radiation to drive surrounding areas; implements the construction guidance of the center, effectively promotes the development of group formation and further improves the commercial center system structure of the main city of Hangzhou.

Keywords: business center system, business format, main city of Hangzhou, POI extraction method

Procedia PDF Downloads 117

Authors: Akhila Potluru

Abstract:

Worldwide, more than 250 bodies of water are transboundary, meaning they cross the political boundaries of multiple countries. This creates a system of hydrological, economic, and social interdependence between communities reliant on these water sources. Transboundary water conflicts can occur as a result of this intense interdependence. Many factors contribute to the sparking of transboundary water conflicts, ranging from natural hydrological factors to hydro-political interactions. Previous attempts to predict transboundary water conflicts by analysing changes or trends in the contributing factors have typically failed because patterns in the data are hard to identify. However, there is potential for artificial intelligence and machine learning to fill this gap and identify future ‘hotspots’ up to a year in advance by identifying patterns in data where humans can’t. This research determines the extent to which AI can be used to predict and prevent transboundary water conflicts. This is done via a critical literature review of previous case studies and datasets where AI was deployed to predict water conflict. This research not only delivered a more nuanced understanding of previously undervalued factors that contribute toward transboundary water conflicts (in particular, culture and disinformation) but also by detecting conflict early, governance bodies can engage in processes to de-escalate conflict by providing pre-emptive solutions. Looking forward, this gives rise to significant policy implications and water-sharing agreements, which may be able to prevent water conflicts from developing into wide-scale disasters. Additionally, AI can be used to gain a fuller picture of water-based conflicts in areas where security concerns mean it is not possible to have staff on the ground. Therefore, AI enhances not only the depth of our knowledge about transboundary water conflicts but also the breadth of our knowledge. With demand for water constantly growing, competition between countries over shared water will increasingly lead to water conflict. There has never been a more significant time for us to be able to accurately predict and take precautions to prevent global water conflicts.

Keywords: artificial intelligence, machine learning, transboundary water conflict, water management

Procedia PDF Downloads 80

Authors: Erica Lima

Abstract:

The objective of this paper is to discuss relevant points about teaching translation in Brazilian universities and the possible impacts of blogs and social networks to translator education today. It is intended to analyze the curricula of Brazilian translation courses, contrasting them to information obtained from two social networking groups of great visibility in the area concerning essential characteristics to become a successful profession. Therefore, research has, as its main corpus, a few undergraduate translation programs’ syllabuses, as well as a few postings on social networks groups that specifically share professional opinions regarding the necessity for a translator to obtain a degree in translation to practice the profession. To a certain extent, such comments and their corresponding responses lead to the propagation of discourses which influence the ideas that aspiring translators and recent graduates end up having towards themselves and their undergraduate courses. The postings also show that many professionals do not have a clear position regarding the translator education; while refuting it, they also encourage “free” courses. It is thus observed that cyberspace constitutes, on the one hand, a place of mobilization of people in defense of similar ideas. However, on the other hand, it embodies a place of tension and conflict, in view of the fact that there are many participants and, as in any other situation of interlocution, disagreements may arise. From the postings, aspects related to professionalism were analyzed (including discussions about regulation), as well as questions about the classic dichotomies: theory/practice; art/technique; self-education/academic training. As partial result, the common interest regarding the valorization of the profession could be mentioned, although there is no consensus on the essential characteristics to be a good translator. It was also possible to observe that the set of socially constructed representations in the group reflects characteristics of the world situation of the translation courses (especially in some European countries and in the United States), which, in the first instance, does not accurately reflect the Brazilian idiosyncrasies of the area.

Keywords: cyberspace, teaching translation, translator education, university

Procedia PDF Downloads 363

Authors: Betsy Ng, Seng Chee Tan, Choon Lang Quek, Peter Looker, Jaime Koh

Abstract:

The purpose of the present study was to determine the particular scale rating that stands out for an instrument. The instrument was designed to assess student perceptions of various learning environments, namely face-to-face, online and blended. The original instrument had a 5-point Likert items (1 = strongly disagree and 5 = strongly agree). Alternate versions were modified with a 6-point Likert scale and a bar scale rating. Participants consisted of undergraduates in a local university were involved in the usability testing of the instrument in an electronic setting. They were presented with the 5-point, 6-point and percentage-bar (100-point) scale ratings, in response to their perceptions of learning environments. The 5-point and 6-point Likert scales were presented in the form of radio button controls for each number, while the percentage-bar scale was presented with a sliding selection. Among these responses, 6-point Likert scale emerged to be the best overall. When participants were confronted with the 5-point items, they either chose 3 or 4, suggesting that data loss could occur due to the insensitivity of instrument. The insensitivity of instrument could be due to the discreet options, as evidenced by response interpolation. To avoid the constraint of discreet options, the percentage-bar scale rating was tested, but the participant responses were not well-interpolated. The bar scale might have provided a variety of responses without a constraint of a set of categorical options, but it seemed to reflect a lack of perceived and objective accuracy. The 6-point Likert scale was more likely to reflect a respondent’s perceived and objective accuracy as well as higher sensitivity. This finding supported the conclusion that 6-point Likert items provided a more accurate measure of the participant’s evaluation. The 5-point and bar scale ratings might not be accurately measuring the participants’ responses. This study highlighted the importance of the respondent’s perception of accuracy, respondent’s true evaluation, and the scale’s ease of use. Implications and limitations of this study were also discussed.

Keywords: usability, interpolation, sensitivity, Likert scales, accuracy

Procedia PDF Downloads 390

Authors: Kuo-Chun Chien, Chia-Hui Chang, Ren-Der Sun

Abstract:

Legal Judgment Prediction (LJP) is a subtask for legal AI. Its main purpose is to use the facts of a case to predict the judgment result. In Taiwan's criminal procedure, when prosecutors complete the investigation of the case, they will decide whether to prosecute the suspect and which article of criminal law should be used based on the facts and evidence of the case. In this study, we collected 305,240 indictments from the public inquiry system of the procuratorate of the Ministry of Justice, which included 169 charges and 317 articles from 21 laws. We take the crime facts in the indictments as the main input to jointly learn the prediction model for law source, article, and charge simultaneously based on the pre-trained Bert model. For single article cases where the frequency of the charge and article are greater than 50, the prediction performance of law sources, articles, and charges reach 97.66, 92.22, and 60.52 macro-f1, respectively. To understand the big performance gap between articles and charges, we used a bipartite graph to visualize the relationship between the articles and charges, and found that the reason for the poor prediction performance was actually due to the wording precision. Some charges use the simplest words, while others may include the perpetrator or the result to make the charges more specific. For example, Article 284 of the Criminal Law may be indicted as “negligent injury”, "negligent death”, "business injury", "driving business injury", or "non-driving business injury". As another example, Article 10 of the Drug Hazard Control Regulations can be charged as “Drug Control Regulations” or “Drug Hazard Control Regulations”. In order to solve the above problems and more accurately predict the article and charge, we plan to include the article content or charge names in the input, and use the sentence-pair classification method for question-answer problems in the BERT model to improve the performance. We will also consider a sequence-to-sequence approach to charge prediction.

Keywords: legal judgment prediction, deep learning, natural language processing, BERT, data visualization

Procedia PDF Downloads 99

Authors: Guillaume Richard, Sarra Zaied

Abstract:

Marine waste quantities increase more and more, 5 million tons of plastic waste enter the ocean every year. Their spatiotemporal distribution is never homogeneous and depends mainly on the hydrodynamic characteristics of the environment, as well as the size and location of the waste. As part of optimizing collect of marine plastic wastes, it is important to measure and monitor their evolution over time. In this context, diverse studies have been dedicated to describing waste behavior in order to identify its accumulation in ocean areas. None of the existing tools which track objects at sea had the objective of tracking down a slick of waste. Moreover, the applications related to marine waste are in the minority compared to rescue applications or oil slicks tracking applications. These approaches are able to accurately simulate an object's behavior over time but not during the collection mission of a waste sheet. This paper presents numerical modeling of a boat’s wake impact on the floating marine waste behavior during a collection mission. The aim is to predict the trajectory of a marine waste slick to optimize its collection using meteorological data of ocean currents, wind, and possibly waves. We have made the choice to use Ocean Parcels which is a Python library suitable for trajectoring particles in the ocean. The modeling results showed the important role of advection and diffusion processes in the spatiotemporal distribution of floating plastic litter. The performance of the proposed method was evaluated on real data collected from the Copernicus Marine Environment Monitoring Service (CMEMS). The results of the evaluation in Cape of Good Hope (South Africa) prove that the proposed approach can effectively predict the position and velocity of marine litter during collection, which allowed for optimizing time and more than $90\%$ of the amount of collected waste.

Keywords: marine litter, advection-diffusion equation, sea current, numerical model

Procedia PDF Downloads 63

Authors: Joseph Lillington, Tom Gout, Mike Harrison, Ian Farnan

Abstract:

The vitrification of high-level nuclear waste within borosilicate glass and its incorporation within a multi-barrier repository deep underground is widely accepted as the preferred disposal method. However, for this to happen, any safety case will require validation that the initially localized radionuclides will not be considerably released into the near/far-field. Therefore, accurate mechanistic models are necessary to predict glass dissolution, and these should be robust to a variety of incorporated waste species and leaching test conditions, particularly given substantial variations across international waste-streams. Here, machine learning is used to predict glass material properties (viscosity, density) and glass leaching model parameters from large-scale industrial data. A variety of different machine learning algorithms have been compared to assess performance. Density was predicted solely from composition, whereas viscosity additionally considered temperature. To predict suitable glass leaching model parameters, a large simulated dataset was created by coupling MATLAB and the chemical reactive-transport code HYTEC, considering the state-of-the-art GRAAL model (glass reactivity in allowance of the alteration layer). The trained models were then subsequently applied to the large-scale industrial, experimental data to identify potentially appropriate model parameters. Results indicate that ensemble methods can accurately predict viscosity as a function of temperature and composition across all three industrial datasets. Glass density prediction shows reliable learning performance with predictions primarily being within the experimental uncertainty of the test data. Furthermore, machine learning can predict glass dissolution model parameters behavior, demonstrating potential value in GRAAL model development and in assessing suitable model parameters for large-scale industrial glass dissolution data.

Keywords: machine learning, predictive modelling, pattern recognition, radioactive waste glass

Procedia PDF Downloads 93

Authors: Mrinal Kanti Bhowmik, Debanjana Debnath Jr., Debotosh Bhattacharjee

Abstract:

A facial expression is undeniably the human manners. It is a significant channel for human communication and can be applied to extract emotional features accurately. People in pain often show variations in facial expressions that are readily observable to others. A core of actions is likely to occur or to increase in intensity when people are in pain. To illustrate the changes in the facial appearance, a system known as Facial Action Coding System (FACS) is pioneered by Ekman and Friesen for human observers. According to Prkachin and Solomon, a set of such actions carries the bulk of information about pain. Thus, the Prkachin and Solomon pain intensity (PSPI) metric is defined. So, it is very important to notice that facial expressions, being a behavioral source in communication media, provide an important opening into the issues of non-verbal communication in pain. People express their pain in many ways, and this pain behavior is the basis on which most inferences about pain are drawn in clinical and research settings. Hence, to understand the roles of different pain behaviors, it is essential to study the properties. For the past several years, the studies are concentrated on the properties of one specific form of pain behavior i.e. facial expression. This paper represents a comprehensive study on pain assessment that can model and estimate the intensity of pain that the patient is suffering. It also reviews the historical background of different pain assessment techniques in the context of painful expressions. Different approaches incorporate FACS from psychological views and a pain intensity score using the PSPI metric in pain estimation. This paper investigates in depth analysis of different approaches used in pain estimation and presents different observations found from each technique. It also offers a brief study on different distinguishing features of real and fake pain. Therefore, the necessity of the study lies in the emerging fields of painful face assessment in clinical settings.

Keywords: facial action coding system (FACS), pain, pain behavior, Prkachin and Solomon pain intensity (PSPI)

Procedia PDF Downloads 312

Authors: D. Ramajo, S. Corzo, M. Nigro

Abstract:

A Multi-dimensional computational fluid dynamics (CFD) two-phase model was developed with the aim to simulate the in-core coolant circuit of a pressurized heavy water reactor (PHWR) of a commercial nuclear power plant (NPP). Due to the fact that this PHWR is a Reactor Pressure Vessel type (RPV), three-dimensional (3D) detailed modelling of the large reservoirs of the RPV (the upper and lower plenums and the downcomer) were coupled with an in-house finite volume one-dimensional (1D) code in order to model the 451 coolant channels housing the nuclear fuel. Regarding the 1D code, suitable empirical correlations for taking into account the in-channel distributed (friction losses) and concentrated (spacer grids, inlet and outlet throttles) pressure losses were used. A local power distribution at each one of the coolant channels was also taken into account. The heat transfer between the coolant and the surrounding moderator was accurately calculated using a two-dimensional theoretical model. The implementation of subcooled boiling and condensation models in the 1D code along with the use of functions for representing the thermal and dynamic properties of the coolant and moderator (heavy water) allow to have estimations of the in-core steam generation under nominal flow conditions for a generic fission power distribution. The in-core mass flow distribution results for steady state nominal conditions are in agreement with the expected from design, thus getting a first assessment of the coupled 1/3D model. Results for nominal condition were compared with those obtained with a previous 1/3D single-phase model getting more realistic temperature patterns, also allowing visualize low values of void fraction inside the upper plenum. It must be mentioned that the current results were obtained by imposing prescribed fission power functions from literature. Therefore, results are showed with the aim of point out the potentiality of the developed model.

Keywords: PHWR, CFD, thermo-hydraulic, two-phase flow

Procedia PDF Downloads 445

Authors: Dylan Elliott, Katya Pertsova

Abstract:

Preposition errors are some of the most common errors created by L2 speakers. In addition, improving error correction and detection methods remains an open issue in the realm of Natural Language Processing (NLP). This research investigates whether the bidirectional encoder representations from the transformers model (BERT) have the potential to correct preposition errors accurately enough to be useful in error correction software. This research finds that BERT performs strongly when the scope of its error correction is limited to preposition choice. The researchers used an open-source BERT model and over three hundred thousand edited sentences from Wikipedia, tagged for part of speech, where only a preposition edit had occurred. To test BERT’s ability to detect errors, a technique known as multi-level masking was used to generate suggestions based on sentence context for every prepositional environment in the test data. These suggestions were compared with the original errors in the data and their known corrections to evaluate BERT’s performance. The suggestions were further analyzed to determine if BERT more often agreed with the judgements of the Wikipedia editors. Both the untrained and fined-tuned models were compared. Finetuning led to a greater rate of error-detection which significantly improved recall, but lowered precision due to an increase in false positives or falsely flagged errors. However, in most cases, these false positives were not errors in preposition usage but merely cases where more than one preposition was possible. Furthermore, when BERT correctly identified an error, the model largely agreed with the Wikipedia editors, suggesting that BERT’s ability to detect misused prepositions is better than previously believed. To evaluate to what extent BERT’s false positives were grammatical suggestions, we plan to do a further crowd-sourcing study to test the grammaticality of BERT’s suggested sentence corrections against native speakers’ judgments.

Keywords: BERT, grammatical error correction, preposition error detection, prepositions

Procedia PDF Downloads 120

Authors: Ankit Kamboj, Shikha Talwar, Nilesh Powar

Abstract:

To successfully deliver our products in the market, the employees need to be in a safe environment, especially in an industrial and manufacturing environment. The consequences of delinquency in wearing safety glasses while working in industrial plants could be high risk to employees, hence the need to develop a real-time automatic detection system which detects the persons (violators) not wearing safety glasses. In this study a convolutional neural network (CNN) algorithm called faster region based CNN (Faster RCNN) with rotated bounding box has been used for detecting safety glasses on persons; the algorithm has an advantage of detecting safety glasses with different orientation angles on the persons. The proposed method of rotational bounding boxes with a convolutional neural network first detects a person from the images, and then the method detects whether the person is wearing safety glasses or not. The video data is captured at the entrance of restricted zones of the industrial environment (manufacturing plant), which is further converted into images at 2 frames per second. In the first step, the CNN with pre-trained weights on COCO dataset is used for person detection where the detections are cropped as images. Then the safety goggles are labelled on the cropped images using the image labelling tool called roLabelImg, which is used to annotate the ground truth values of rotated objects more accurately, and the annotations obtained are further modified to depict four coordinates of the rectangular bounding box. Next, the faster RCNN with rotated bounding box is used to detect safety goggles, which is then compared with traditional bounding box faster RCNN in terms of detection accuracy (average precision), which shows the effectiveness of the proposed method for detection of rotatory objects. The deep learning benchmarking is done on a Dell workstation with a 16GB Nvidia GPU.

Keywords: CNN, deep learning, faster RCNN, roLabelImg rotated bounding box, safety goggle detection

Procedia PDF Downloads 112

Authors: Yi-Shiuan Chiu

Abstract:

Deaf signers have been found to have better mental image performance than hearing nonsigners. The goal of this study was to investigate the ability to generate mental images, to maintain them, and to manipulate them in deaf signers of Taiwanese Sign Language (TSL). In the visual image task, participants first memorized digits formed in a cell of 4 × 5 grids. After presenting a cue of Chinese digit character shown on the top of a blank cell, participants had to form a corresponding digit. When showing a probe, which was a grid containing a red circle, participants had to decide as quickly as possible whether the probe would have been covered by the mental image of the digit. The ISI (interstimulus interval) between cue and probe was manipulated. In experiment 1, 24 deaf signers and 24 hearing nonsigners were asked to perform image generation tasks (ISI: 200, 400 ms) and image maintenance tasks (ISI: 800, 2000 ms). The results showed that deaf signers had had an enhanced ability to generate and maintain a mental image. To explore the process of mental image, in experiment 2, 30 deaf signers and 30 hearing nonsigners were asked to do visual searching when maintaining a mental image. Between a digit image cue and a red circle probe, participants were asked to search a visual search task to see if a target triangle apex was directed to the right or left. When there was only one triangle in the searching task, the results showed that both deaf signers and hearing non-signers had similar visual searching performance in which the searching targets in the mental image locations got facilitates. However, deaf signers could maintain better and faster mental image performance than nonsigners. In experiment 3, we increased the number of triangles to 4 to raise the difficulty of the visual search task. The results showed that deaf participants performed more accurately in visual search and image maintenance tasks. The results suggested that people may use eye movements as a mnemonic strategy to maintain the mental image. And deaf signers had enhanced abilities to resist the interference of eye movements in the situation of fewer distractors. In sum, these findings suggested that deaf signers had enhanced mental image processing.

Keywords: deaf signers, image maintain, mental image, visual search

Procedia PDF Downloads 131

Authors: Rohini Hariharan, Yazhini R, Bhamidipati Naga Shrikarti

Abstract:

In the realm of cricket, particularly within the context of the Indian Premier League (IPL), the ability to predict team scores accurately holds significant importance for both cricket enthusiasts and stakeholders alike. This paper presents a comprehensive study on IPL score prediction utilizing various machine learning algorithms, including Support Vector Machines (SVM), XGBoost, Multiple Regression, Linear Regression, K-nearest neighbors (KNN), and Random Forest. Through meticulous data preprocessing, feature engineering, and model selection, we aimed to develop a robust predictive framework capable of forecasting team scores with high precision. Our experimentation involved the analysis of historical IPL match data encompassing diverse match and player statistics. Leveraging this data, we employed state-of-the-art machine learning techniques to train and evaluate the performance of each model. Notably, Multiple Regression emerged as the top-performing algorithm, achieving an impressive accuracy of 77.19% and a precision of 54.05% (within a threshold of +/- 10 runs). This research contributes to the advancement of sports analytics by demonstrating the efficacy of machine learning in predicting IPL team scores. The findings underscore the potential of advanced predictive modeling techniques to provide valuable insights for cricket enthusiasts, team management, and betting agencies. Additionally, this study serves as a benchmark for future research endeavors aimed at enhancing the accuracy and interpretability of IPL score prediction models.

Keywords: indian premier league (IPL), cricket, score prediction, machine learning, support vector machines (SVM), xgboost, multiple regression, linear regression, k-nearest neighbors (KNN), random forest, sports analytics

Procedia PDF Downloads 24

Authors: Gao Youtao, Zhao Tanran, Jin Bingyu, Xu Bo

Abstract:

Global navigation satellite system(GNSS) can deliver navigation information for spacecraft orbiting on low-Earth orbits and medium Earth orbits. However, the GNSS cannot navigate the spacecraft on high-Earth orbit or deep space probes effectively. With the deep space exploration becoming a hot spot of aerospace, the demand for a deep space satellite navigation system is becoming increasingly prominent. Many researchers discussed the feasibility and performance of a satellite navigation system on periodic orbits around the Earth-Moon libration points which can be called Lagrangian point satellite navigation system. Autonomous orbit determination (AOD) is an important performance for the Lagrangian point satellite navigation system. With this ability, the Lagrangian point satellite navigation system can reduce the dependency on ground stations. AOD also can greatly reduce total system cost and assure mission continuity. As the elliptical restricted three-body problem can describe the Earth-Moon system more accurately than the circular restricted three-body problem, we study the autonomous orbit determination of Lagrangian navigation constellation using only crosslink range based on elliptical restricted three body problem. Extended Kalman filter is used in the autonomous orbit determination. In order to compare the autonomous orbit determination results based on elliptical restricted three-body problem to the results of autonomous orbit determination based on circular restricted three-body problem, we give the autonomous orbit determination position errors of a navigation constellation include four satellites based on the circular restricted three-body problem. The simulation result shows that the Lagrangian navigation constellation can achieve long-term precise autonomous orbit determination using only crosslink range. In addition, the type of the libration point orbit will influence the autonomous orbit determination accuracy.

Keywords: extended Kalman filter, autonomous orbit determination, quasi-periodic orbit, navigation constellation

Procedia PDF Downloads 260

Authors: James Ladzekpo

Abstract:

Background: The urgent need to identify new pharmacological targets for diabetes treatment and prevention has been amplified by the disease's extensive impact on individuals and healthcare systems. A deeper insight into the biological underpinnings of diabetes is crucial for the creation of therapeutic strategies aimed at these biological processes. Current predictive models based on genetic variations fall short of accurately forecasting diabetes. Objectives: Our study aims to pinpoint key epigenetic factors that predispose individuals to diabetes. These factors will inform the development of an advanced predictive model that estimates diabetes risk from genetic profiles, utilizing state-of-the-art statistical and data mining methods. Methodology: We have implemented a recursive feature elimination with cross-validation using the support vector machine (SVM) approach for refined feature selection. Building on this, we developed six machine learning models, including logistic regression, k-Nearest Neighbors (k-NN), Naive Bayes, Random Forest, Gradient Boosting, and Multilayer Perceptron Neural Network, to evaluate their performance. Findings: The Gradient Boosting Classifier excelled, achieving a median recall of 92.17% and outstanding metrics such as area under the receiver operating characteristics curve (AUC) with a median of 68%, alongside median accuracy and precision scores of 76%. Through our machine learning analysis, we identified 31 genes significantly associated with diabetes traits, highlighting their potential as biomarkers and targets for diabetes management strategies. Conclusion: Particularly noteworthy were the Gradient Boosting Classifier and Multilayer Perceptron Neural Network, which demonstrated potential in diabetes outcome prediction. We recommend future investigations to incorporate larger cohorts and a wider array of predictive variables to enhance the models' predictive capabilities.

Keywords: diabetes, machine learning, prediction, biomarkers

Procedia PDF Downloads 30

Authors: Mohammad Zaeimzade

Abstract:

The development of communication technologies is clearly and rapidly moving towards reducing the distance between the virtual and real worlds. Of course, living in a two-spatial or two-globalized world or any other interpretation that means mixing real and virtual life is still relevant and debatable. In the present age of communication, where social networks have transformed the message equation and turned the audience out of passivity and turned into a user. Platforms have penetrated widely in various aspects of human life, from culture and education and economy. Among the messengers, Instagram, which is one of the most extensive image-based interactive networks, plays a significant role in the new economic life. It doesn't need much explanation that the era of thinking of every messenger as a non-insulating conductor that is just a neutral load has passed. Every messenger has its own economic, political and of course security background, Instagram is no exception to this rule and of course it leaves its effects in bio-economics as well. Iran, as the 19th largest economy in the world, has not been unaffected by new platforms, including Instagram, and their consequences in the economy. Generally, in the policy-making space, there are two simple and inflexible pessimistic or optimistic views on this issue, and each of the holders of these views usually have their own one-dimensional policy recommendations regarding how to deal with Instagram. Prescriptions that are usually very different and sometimes contradictory. In this article, we show that this confusion of policymakers is the result of not accurately describing the reality of its effect, and the reason for this inaccurate description is the existence of a conflict of interests in the eyes of describers and researchers. In this article, we first take a look at the main indicators of the Iranian economy, estimate the role of the digital economy in Iran's economic growth, then study the conflicting descriptions of the Instagram-based digital economy, the statistics that show the tolerance of economic users of Instagram in Iran. 300 thousand to 9 million have been estimated. Finally, we take a look at the government's actions in this matter, especially in the context of street riots in October and November 2022. And we suggest an intermediate idea.

Keywords: digital economy, instagram, conflict of interest, social networks

Procedia PDF Downloads 54

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 95

Authors: Nia Mair Fry, Matthew Profit, Chenfeng Li

Abstract:

This work focuses on the development and implementation of a fully implicit-implicit, coupled mechanical deformation and porous flow, finite element software tool. The fully implicit software accurately predicts classical fundamental analytical solutions such as the Terzaghi consolidation problem. Furthermore, it can capture other analytical solutions less well known in the literature, such as Gibson’s sedimentation rate problem and Coussy’s problems investigating wellbore stability for poroelastic rocks. The mechanical volume strains are transferred to the porous flow governing equation in an implicit framework. This will overcome some of the many current industrial issues, which use explicit solvers for the mechanical governing equations and only implicit solvers on the porous flow side. This can potentially lead to instability and non-convergence issues in the coupled system, plus giving results with an accountable degree of error. The specification of a fully monolithic implicit-implicit coupled porous media code sees the solution of both seepage-mechanical equations in one matrix system, under a unified time-stepping scheme, which makes the problem definition much easier. When using an explicit solver, additional input such as the damping coefficient and mass scaling factor is required, which are circumvented with a fully implicit solution. Further, improved accuracy is achieved as the solution is not dependent on predictor-corrector methods for the pore fluid pressure solution, but at the potential cost of reduced stability. In testing of this fully monolithic porous media code, there is the comparison of the fully implicit coupled scheme against an existing staggered explicit-implicit coupled scheme solution across a range of geotechnical problems. These cases include 1) Biot coefficient calculation, 2) consolidation theory with Terzaghi analytical solution, 3) sedimentation theory with Gibson analytical solution, and 4) Coussy well-bore poroelastic analytical solutions.

Keywords: coupled, implicit, monolithic, porous media

Procedia PDF Downloads 111