CERES :: Browsing by Author "Moulitsas, Irene"

Browsing by Author "Moulitsas, Irene"

Now showing 1 - 20 of 50

Open Access
Advancing aviation safety through machine learning and psychophysiological data: a systematic review
(IEEE, 2024-01-03) Alreshidi, Ibrahim; Moulitsas, Irene; Jenkins, Karl W.
In the aviation industry, safety remains vital, often compromised by pilot errors attributed to factors such as workload, fatigue, stress, and emotional disturbances. To address these challenges, recent research has increasingly leveraged psychophysiological data and machine learning techniques, offering the potential to enhance safety by understanding pilot behavior. This systematic literature review rigorously follows a widely accepted methodology, scrutinizing 80 peer-reviewed studies out of 3352 studies from five key electronic databases. The paper focuses on behavioral aspects, data types, preprocessing techniques, machine learning models, and performance metrics used in existing studies. It reveals that the majority of research disproportionately concentrates on workload and fatigue, leaving behavioral aspects like emotional responses and attention dynamics less explored. Machine learning models such as tree-based and support vector machines are most commonly employed, but the utilization of advanced techniques like deep learning remains limited. Traditional preprocessing techniques dominate the landscape, urging the need for advanced methods. Data imbalance and its impact on model performance is identified as a critical, under-researched area. The review uncovers significant methodological gaps, including the unexplored influence of preprocessing on model efficacy, lack of diversification in data collection environments, and limited focus on model explainability. The paper concludes by advocating for targeted future research to address these gaps, thereby promoting both methodological innovation and a more comprehensive understanding of pilot behavior.
Open Access
An enhanced deep autoencoder for flight delay prediction
(Embry-Riddle Aeronautical University, 2024-08-01) Bisandu, Desmond B.; Soviani-Sitoiu, Dan Andrei; Moulitsas, Irene
Accurate and timely flight delay prediction cannot be overemphasized because of the ever-increasing demand for air travel and its importance in deploying intelligent transportation systems. Nonetheless, there has not been a universal solution to the problem, as more intelligent flight decision systems are required for the aviation industry’s future growth. Existing flight delay classification and prediction approaches are mainly shallow traffic models and do not satisfy many applications in the real world. Our motivation to rethink the deep architecture model for predicting flight delays emanates from the problem. In this research, we proposed a technique that modified stacked autoencoder architecture parameters for training the network and understanding the link between space, time and information gained from the flight on-time data. We developed three different types of autoencoders based on the architecture of the modified stacked autoencoder. The models learn the generic flight delay features, and it’s trained greedily in a layer-wise fashion. To the best of our knowledge, this is the first time these performances of vanilla autoencoder, logistic regression autoencoder and Multilayer perceptron for classification were evaluated based on the developed modified stacked autoencoder architecture. Moreover, our experiment demonstrates that the models achieved varying levels of accuracy in the flight delay classifications task. The deep vanilla autoencoder shows superior accuracy, recall and precision performance compared to logistic regression autoencoder and Multilayer perceptron autoencoders at different parameter settings.
Open Access
Analysing the sentiment of air-traveller: a comparative analysis
(IJCTE, 2022-05) Homaid, Mohammed Salih; Bisandu, Desmond B.; Moulitsas, Irene; Jenkins, Karl W.
Airport service quality is considered to be an indicator of passenger satisfaction. However, assessing this by conventional methods requires continuous observation and monitoring. Therefore, during the past few years, the use of machine learning techniques for this purpose has attracted considerable attention for analysing the sentiment of the air traveller. A sentiment analysis system for textual data analytics leverages the natural language processing and machine learning techniques in order to determine whether a piece of writing is positive, negative or neutral. Numerous methods exist for estimating sentiments which include lexical-based methodologies and directed artificial intelligence strategies. Despite the wide use and ubiquity of certain strategies, it remains unclear which is the best strategy for recognising the intensity of the sentiments of a message. It is necessary to compare these techniques in order to understand their advantages, disadvantages and limitations. In this paper, we compared the Valence Aware Dictionary and sentiment Reasoner, a sentiment analysis technique specifically attuned and well known for performing good on social media data, with the conventional machine learning techniques of handling the textual data by converting it into numerical form. We used the review data obtained from the SKYTRAX website for each airport. The machine learning algorithms evaluated in this paper are VADER sentiment and logistic regression. The term frequency-inverse document frequency is used in order to convert the textual review data into the resulting numerical columns. This was formulated as a classification problem, whereby the prediction of the algorithm was compared with the actual recommendation of the passenger in the dataset. The results were analysed according to the accuracy, precision, recall and F1-score. From the analysis of the results, we observed that logistic regression outperformed the VADER sentiment analysis.
Open Access
Bladesense – a novel approach for measuring dynamic helicopter rotor blade deformation
(European Rotorcraft Forum, 2018-12-31) Weber, Simone; Southgate, Dominic; Mullaney, Kevin; James, Stephen; Rutherford, Robert; Sharma, Anuj; Lone, Mudassir; Kissinger, Thomas; Chehura, Edmond; Staines, Stephen; Pekmezci, Huseyin; Fragonara, Luca Zanotti; Petrunin, Ivan; Williams, Dan; Moulitsas, Irene; Cooke, Alastair; Rosales, Waldo; Tatam, Ralph P.; Morrish, Peter; Fairhurst, Mark; Atack, Richard; Bailey, Gordon; Morley, Stuart
Technologies that allow accurate measurement of rotorblade dynamics can impact almost all areas of the rotorcraft sector; ranging from maintenance all the way to blade design. The BladeSense project initiated in 2016 aims to take a step in developing and demonstrating such a capability using novel fibre optic sensors that allow direct shape measurement. In this article the authors summarise key project activities in modelling and simulation, instrumentation development and ground testing. The engineering approach and associated challenges and achievements in each of these disciplines are discussed albeit briefly. This ranges from the use of computational aerodynamics and structural modelling to predict blade dynamics to the development of direct fibre optic shape sensing that allows measurements above 1kHz over numerous positions on the blade. Moreover, the development of the prototype onboard system that overcomes the challenge of transferring data between the rotating main rotor to the fixed fuselage frames is also discussed.
Open Access
Code and data supporting 'A Comprehensive Analysis of Machine Learning and Deep Learning Models for Identifying Pilots' Mental States from Imbalanced Physiological Data'
(Cranfield University, 2023-09-18 16:40) Alreshidi, Ibrahim; Moulitsas, Irene; Jenkins, Karl; Yadav, Satendra
Data: This folder contains: - A dataset called combined_df4, which contains the power spectral density features after employing SMOTE. - A dataset called combined_df5, which contains the power spectral density features after employing SMOTE and cosine similarity. Source code: This folder contains: - A jupyter notebook called AdaBoost.ipynb which was used to generate the results for the AdaBoost algorithm. - A jupyter notebook called CNN.ipynb which was used to generate the results for the CNN algorithm. - A jupyter notebook called CNN+LSTM.ipynb which was used to generate the results for the CNN+LSTMalgorithm. - A jupyter notebook called LSTM.ipynb which was used to generate the results for the LSTMalgorithm. - A jupyter notebook called FNN.ipynb which was used to generate the results for the FNN algorithm. - A jupyter notebook called Random_Forest.ipynb which was used to generate the results for the Random Forest algorithm. - A jupyter notebook called XGBoost.ipynb which was used to generate the results for the XGBoost algorithm.
Open Access
Code and Data: Multimodal Approach for Pilot Mental State Detection Based on EEG
(Cranfield University, 2023-08-23 15:04) Alreshidi, Ibrahim; Moulitsas, Irene; Jenkins, Karl
Data: This folder contains: A dataset called Crews_equalized_dataset_epo.fif which was used to obtain the results presented in the journal paper. It is the preprocessed EEG dataset used to predict four mental states, Channelised Attention, Diverted Attention, Startle/Surprise, and Baseline. A dataset called Example_raw.fif which was used to obtain Figure 6 of the journal paper. Source code: This folder contains a jupyter notebook called python_code.ipynb which implements the proposed EEG preprocessing pipeline and all the algorithms presented and validated in the journal paper. Output: This folder contains: A figure called Confusion Matrices.jpg which shows results from the Random Forest classifier in (A), Extremely Randomized Trees in (B), Gradient Tree Boosting in (C), AdaBoost in (D), and Voting in (E). Figures called Figure 6A.jpg and Figure 6B.jpg which show the EEG signals before applying the preprocessing pipeline, and after applying the preprocessing pipeline, respectively. A text file called ML models evaluation.txt which contains the results produced by all algorithms presented and validated in the journal paper. A figure called The preprocessed EEG signals.jpg which shows the EEG signals, upon completion of our preprocessing pipeline, fed into the machine learning models for training and testing purposes.
Open Access
A comprehensive analysis of machine learning and deep learning models for identifying pilots’ mental states from imbalanced physiological data
(AIAA, 2023-06-08) Alreshidi, Ibrahim; Yadav, Satendra; Moulitsas, Irene; Jenkins, Karl W.
This study focuses on identifying pilots' mental states linked to attention-related human performance-limiting states (AHPLS) using a publicly released, imbalanced physiological dataset. The research integrates electroencephalography (EEG) with non-brain signals, such as electrocardiogram (ECG), galvanic skin response (GSR), and respiration, to create a deep learning architecture that combines one-dimensional Convolutional Neural Network (1D-CNN) and Long Short-Term Memory (LSTM) models. Addressing the data imbalance challenge, the study employs resampling techniques, specifically downsampling with cosine similarity and oversampling using Synthetic Minority Over-sampling Technique (SMOTE), to produce balanced datasets for enhanced model performance. An extensive evaluation of various machine learning and deep learning models, including XGBoost, AdaBoost, Random Forest (RF), Feed-Forward Neural Network (FFNN), standalone 1D-CNN, and standalone LSTM, is conducted to determine their efficacy in detecting pilots' mental states. The results contribute to the development of efficient mental state detection systems, highlighting the XGBoost algorithm and the proposed 1D-CNN+LSTM model as the most promising solutions for improving safety and performance in aviation and other industries where monitoring mental states is essential.
Open Access
CranSLIK v1.0: stochastic prediction of oil spill transport and fate using approximation methods
(2015-07-22T00:00:00Z) Snow, Ben J.; Moulitsas, Irene; Kolios, Athanasios J.; De Dominicis, Michela
Oil spill models are used to forecast the transport and fate of oil after it has been released. CranSLIK is a model that predicts the movement and spread of a surface oil spill at sea via a stochastic approach. The aim of this work is to identify parameters that can further improve the forecasting algorithms and expand the functionality of CranSLIK, while maintaining the run-time efficiency of the method. The results from multiple simulations performed using the operational, validated oil spill model, MEDSLIK-II, were analysed using multiple regression in order to identify improvements which could be incorporated into CranSLIK. This has led to a revised model, namely CranSLIK v2.0, which was validated against MEDSLIK-II forecasts for real oil spill cases. The new version of CranSLIK demonstrated significant forecasting improvements by capturing the oil spill accurately in real validation cases and also proved capable of simulating a broader range of oil spill scenarios.
Open Access
CranSLIK v2.0: improving the stochastic prediction of oil spill transport and fate using approximation methods
(European Geosciences Union, 2015-10-26) Rutherford, R.; Moulitsas, Irene; Snow, Ben J.; Kolios, Athanasios J.; De Dominicis, Michela
Oil spill models are used to forecast the transport and fate of oil after it has been released. CranSLIK is a model that predicts the movement and spread of a surface oil spill at sea via a stochastic approach. The aim of this work is to identify parameters that can further improve the forecasting algorithms and expand the functionality of CranSLIK, while maintaining the run-time efficiency of the method. The results from multiple simulations performed using the operational, validated oil spill model, MEDSLIK-II, were analysed using multiple regression in order to identify improvements which could be incorporated into CranSLIK. This has led to a revised model, namely CranSLIK v2.0, which was validated against MEDSLIK-II forecasts for real oil spill cases. The new version of CranSLIK demonstrated significant forecasting improvements by capturing the oil spill accurately in real validation cases and also proved capable of simulating a broader range of oil spill scenarios.
Open Access
Critical assessment of the lattice Boltzmann method for cavitation modelling based on single bubble dynamics
(Springer, 2024-05-01) Xiong, Xin; Teschner, Tom-Robin; Moulitsas, Irene; Józsa, Tamás István
The lattice Boltzmann Method (LBM) is recognised as a popular technique for simulating cavitation bubble dynamics due to its simplicity. In the validation of LBM results, the Rayleigh-Plesset (R-P) equation is commonly employed. However, most studies to date have neglected the impact of simulation settings on the predictions. This article sets out to quantify the impact of LBM domain size and bubble size, and the initial conditions of the R-P equations on the predicted bubble dynamics. First, LBM results were validated against the classical benchmarks of Laplace’s law and Maxwell’s area construction. LBM results corresponding to these fundamental test cases were found to be in satisfactory agreement with theory and previous simulations. Secondly, a one-to-one comparison was considered between the predictions of the LBM and the R-P equation. The parameters of the two models were matched based on careful considerations. Findings revealed that a good overlap between the predictions is observable only under certain conditions. The warming-up period of the LBM simulations, small domain size, and small bubble radius were identified as key factors responsible for the measured differences. The authors hope that the results will promote good simulation practices for cavitation simulation including both single bubbles and bubble clusters.
Open Access
Data supporting "Analysing the Sentiment of Air-Traveller: A Comparative Analysis"
(Cranfield University, 2022-08-31 12:55) Salih A Homaid, Mohammed; Bala Bisandu, Desmond; Moulitsas, Irene; Jenkins, Karl
Airport service qualityis considered to be an indicator of passenger satisfaction. However, assessingthis by conventional methods requires continuous observation and monitoring.Therefore, during the past few years, the use of machine learning techniquesfor this purpose has attracted considerable attention for analysing thesentiment of the air traveller. A sentiment analysis system for textual dataanalytics leverages the natural language processing and machine learningtechniques in order to determine whether a piece of writing is positive, negativeor neutral. Numerous methods exist for estimating sentiments which includelexical-based methodologies and directed artificial intelligence strategies.Despite the wide use and ubiquity of certain strategies, it remains unclearwhich is the best strategy for recognising the intensity of the sentiments of amessage. It is necessary to compare these techniques in order to understandtheir advantages, disadvantages and limitations. In this paper, we compared theValence Aware Dictionary and sentiment Reasoner, a sentiment analysis techniquespecifically attuned and well known for performing good on social media data,with the conventional machine learning techniques of handling the textual databy converting it into numerical form. We used the review data obtained from theSKYTRAX website for each airport. The machine learning algorithms evaluated inthis paper are VADER sentiment and logistic regression. The termfrequency-inverse document frequency is used in order to convert the textualreview datainto the resulting numerical columns. This was formulated as a classificationproblem, whereby the prediction of the algorithm was compared with the actualrecommendation of the passenger in the dataset. The results were analysedaccording to the accuracy, precision, recall and F1-score. From the analysis ofthe results, we observed that logistic regression outperformed the VADERsentiment analysis.
Open Access
A deep BiLSTM machine learning method for flight delay prediction classification
(Embry-Riddle Aeronautical University, 2023-09-25) Bisandu, Desmond B.; Moulitsas, Irene
This paper proposes a classification approach for flight delays using Bidirectional Long Short-Term Memory (BiLSTM) and Long Short-Term Memory (LSTM) models. Flight delays are a major issue in the airline industry, causing inconvenience to passengers and financial losses to airlines. The BiLSTM and LSTM models, powerful deep learning techniques, have shown promising results in a classification task. In this study, we collected a dataset from the United States (US) Bureau of Transportation Statistics (BTS) of flight on-time performance information and used it to train and test the BiLSTM and LSTM models. We set three criteria for selecting highly important features to train and test the models. The performance evaluation of the models and Confusion matrix shows that BiLSTM outperforms the LSTM model. In evaluating the models using the Mathews Correlation Coefficient (MCC), the BiLSTM model offers a better correlation of 0.99 between the original and predicted classes. Our experiment shows that for predicting flight delays, the BiLSTM model takes advantage of the forward and backward hidden sequences and the deep neural network for performance exploration and exploitation to achieve high accuracy, recall, and F1-Score. Our findings suggest that the BiLSTM model can effectively predict flight delays and provide valuable information for airlines, passengers, and airport managers.
Open Access
A deep feedforward neural network and shallow architectures effectiveness comparison: Flight delays classification perspective
(Association for Computing Machinery (ACM) , 2021-11-22) Bisandu, Desmond B.; Homaid, Mohammed Salih; Moulitsas, Irene; Filippone, Salvatore
Flight delays have negatively impacted the socio-economics state of passengers, airlines and airports, resulting in huge economic losses. Hence, it has become necessary to correctly predict their occurrences in decision-making because it is important for the effective management of the aviation industry. Developing accurate flight delays classification models depends mostly on the air transportation system complexity and the infrastructure available in airports, which may be a region-specific issue. However, no specific prediction or classification model can handle the individual characteristics of all airlines and airports at the same time. Hence, the need to further develop and compare predictive models for the aviation decision system of the future cannot be over-emphasised. In this research, flight on-time data records from the United State Bureau of Transportation Statistics was employed to evaluate the performances of Deep Feedforward Neural Network, Neural Network, and Support Vector Machine models on a binary classification problem. The research revealed that the models achieved different accuracies of flight delay classifications. The Support Vector Machine had the worst average accuracy than Neural Network and Deep Feedforward Neural Network in the initial experiment. The Deep Feedforward Neural Network outperformed Support Vector Machines and Neural Network with the best average percentage accuracies. Going further to investigate the Deep Feedforward Neural Network architecture on different parameters against itself suggest that training a Deep Feedforward Neural Network algorithm, regardless of data training size, the classification accuracy peaks. We examine which number of epochs works best in our flight delay classification settings for the Deep Feedforward Neural Network. Our experiment results demonstrate that having many epochs affects the convergence rate of the model; unlike when hidden layers are increased, it does not ensure better or higher accuracy in a binary classification of flight delays. Finally, we recommended further studies on the applicability of the Deep Feedforward Neural Network in flight delays prediction with specific case studies of either airlines or airports to check the impact on the model’s performance.
Open Access
Development and performance comparison of MPI and Fortran Coarrays within an atmospheric research model
(IEEE, 2018-11-16) Rasmussen, Soren; Gutmann, Ethan D.; Friesen, Brian; Rouson, Damian; Filippone, Salvatore; Moulitsas, Irene
A mini-application of The Intermediate Complexity Research (ICAR) Model offers an opportunity to compare the costs and performance of the Message Passing Interface (MPI) versus coarray Fortran, two methods of communication across processes. The application requires repeated communication of halo regions, which is performed with either MPI or coarrays. The MPI communication is done using non-blocking two-sided communication, while the coarray library is implemented using a one-sided MPI or OpenSHMEM communication backend. We examine the development cost in addition to strong and weak scalability analysis to understand the performance costs.
Open Access
The duality between particle methods and artificial neural networks
(Nature Publishing Group / Nature Research, 2020-10-01) Alexiadis, Alessio; Simmons, M. J. H.; Stamatopoulos, K.; Batchelor, H. K.; Moulitsas, Irene
The algorithm behind particle methods is extremely versatile and used in a variety of applications that range from molecular dynamics to astrophysics. For continuum mechanics applications, the concept of ‘particle’ can be generalized to include discrete portions of solid and liquid matter. This study shows that it is possible to further extend the concept of ‘particle’ to include artificial neurons used in Artificial Intelligence. This produces a new class of computational methods based on ‘particle-neuron duals’ that combines the ability of computational particles to model physical systems and the ability of artificial neurons to learn from data. The method is validated with a multiphysics model of the intestine that autonomously learns how to coordinate its contractions to propel the luminal content forward (peristalsis). Training is achieved with Deep Reinforcement Learning. The particle-neuron duality has the advantage of extending particle methods to systems where the underlying physics is only partially known, but we have observations that allow us to empirically describe the missing features in terms of reward function. During the simulation, the model evolves autonomously adapting its response to the available observations, while remaining consistent with the known physics of the system.
Open Access
Enhancing bank's profitability by applying machine learning techniques on financial data to intelligently predict customer behaviour towards the use of electronic channels.
(2021-09) Alsanousi, Hessa Abdulaziz; Moulitsas, Irene; Filippone, Salvatore
Technology is evolving rapidly, and this represents a huge prospect for investment in business. In the banking industry, new technologies have led to the creation of new electronic communication channels for customers. Therefore, banks need to do consider how to make the most of these channels. This will help them create prediction models to implement better strategies and improve their decision making. In the long run, this will help them to decrease costs and increase revenues. The aim of this study was to determine the appropriate electronic channels for specific customers based on their information. It used a big data prediction model to predict the best online channel for banking customers. The point of the model was to help banks understand their customers’ preferences, thereby increasing customer satisfaction, productivity, and profits. I obtained a substantial amount of financial data from a minimum of 100,000 customers from ten local banks in Kuwait. The data covered a period of ten years. The independent variables I studied were age, gender, number of current accounts, number of savings accounts, number of deposit accounts, income, number of consumer loans, number of instalment loans, credit card limit, outstanding credit card balance, length of relationship with the bank, continent, and nationality. The dependent variables were call centres, websites, and mobile applications. Given the size and type of data, I used machine learning. I used the Statistics and Machine Learning Toolbox, Financial Toolbox, and other functions in MATLAB to run a multinomial logistic regression analysis. I examined four different methods for analyzing the data and chose the one that was most appropriate: multinomial logistic regression. I also considered whether there was any correlation between the independent variables. I discovered one significant correlation between credit card limit and outstanding credit card balance. I explored this finding further by considering the time of execution and the key performance measures. Consequently, I decided to keep both variables in my study. I also studied the relation between the volume of training data and the accuracy of the model to see how sensitive the models were to variations in data size. The results confirmed that my method performed stably across the different sample sizes. I addressed the issue of overfitting by running a sub-sample regression and comparing it to the original model.my results indicated that the model was not overfitted. I differentiated between conventional banks and Islamic banks. This distinction had not been made by previous studies, and my outcomes provided more information about the difference between Islamic banks and conventional banks. I found that there were very few differences between the customers of conventional banks and those of Islamic banks. However, I found that male customers of Islamic banks tended to use internet more often than female customers, who tended to use the mobile applications. I also found that older customers tended to use call centres as their primary way of communicating with their bank. The results showed that clients with more current accounts and savings accounts were more likely to use mobile applications. However, one unexpected finding was that clients with more deposit accounts were more likely to use the internet or call centres rather than mobile applications. The findings also demonstrated that clients with higher incomes were more likely to use mobile applications than other communication platforms. In a couple of banks, customers with more loans tended to use call centres as their primary means of communication. The type of loan had no significant impact on their choices. Also, I found that customers who had been with their bank for longer were more likely to use call centres as their primary communication source. Finally, in a few banks, the results showed that a client’s continent or nationality had no significant impact on their preference for a particular communication channel. These are very important findings that could change how banks operate. They will have a positive influence on decision-making and strategies in the banking industry.
Open Access
Examining the societal impact and legislative requirements of deepfake technology: a comprehensive study
(EJournal Publishing, 2024-03-29) Alanazi, Sami; Asif, Seemal; Moulitsas, Irene
Deepfakes, highly realistic fabricated videos, images, or audios created using artificial intelligence algorithms, have gained widespread attention and raised concerns about their social impact and ethical implications. While initially seen as a source of entertainment and utilized in commercial applications, deepfake technology has increasingly been misused for creating adult content, blackmailing individuals, spreading misinformation, and manipulating memories. The negative consequences of deepfakes extend beyond individuals, impacting society as a whole, particularly during sensitive times like elections, where trust can be undermined. The paper looks at the social implications and legislative considerations for deepfakes content, with the goal of cultivating a thorough understanding of their impact on society. By highlighting the importance of enacting laws and regulations, the paper emphasizes the pressing need to control their widespread dissemination. Deepfakes have broad societal implications, especially during critical events like elections, eroding trust. This survey delves into deepfake complexities, aiming to foster understanding and emphasizing the urgency of regulation. The paper also discusses the positive outcomes of deepfake technology for intellectual property protection, highlighting the FORGE system developed to trick attackers who steal company documents. However, it emphasizes the risks posed by easily accessible websites that facilitate the creation of fake identification documents, increasing the likelihood of identity theft, criminal activities, and fraudulent transactions. Implementing restrictions on the use of deepfake technology involving children is also crucial to prevent harm and manipulation targeting minors.
Open Access
Fortran coarray implementation of semi-Lagrangian convected air particles within an atmospheric model
(MDPI, 2021-05-06) Rasmussen, Soren; Gutmann, Ethan D.; Moulitsas, Irene; Filippone, Salvatore
This work added semi-Lagrangian convected air particles to the Intermediate Complexity Atmospheric Research (ICAR) model. The ICAR model is a simplified atmospheric model using quasi-dynamical downscaling to gain performance over more traditional atmospheric models. The ICAR model uses Fortran coarrays to split the domain amongst images and handle the halo region communication of the image’s boundary regions. The newly implemented convected air particles use trilinear interpolation to compute initial properties from the Eulerian domain and calculate humidity and buoyancy forces as the model runs. This paper investigated the performance cost and scaling attributes of executing unsaturated and saturated air particles versus the original particle-less model. An in-depth analysis was done on the communication patterns and performance of the semi-Lagrangian air particles, as well as the performance cost of a variety of initial conditions such as wind speed and saturation mixing ratios. This study found that given a linear increase in the number of particles communicated, there is an initial decrease in performance, but that it then levels out, indicating that over the runtime of the model, there is an initial cost of particle communication, but that the computational benefits quickly offset it. The study provided insight into the number of processors required to amortize the additional computational cost of the air particles.
Open Access
How to modify LAMMPS: from the prospective of a particle method researcher
(MDPI, 2021-06-13) Albano, Andrea; le Guillou, Eve; Danzé, Antoine; Moulitsas, Irene; Sahputra, Iwan H.; Rahmat, Amin; Duque-Daza, Carlos Alberto; Shang, Xiaocheng; Ng, Khai Ching; Ariane, Mostapha; Alexiadis, Alessio
LAMMPS is a powerful simulator originally developed for molecular dynamics that, today, also accounts for other particle-based algorithms such as DEM, SPH, or Peridynamics. The versatility of this software is further enhanced by the fact that it is open-source and modifiable by users. This property suits particularly well Discrete Multiphysics and hybrid models that combine multiple particle methods in the same simulation. Modifying LAMMPS can be challenging for researchers with little coding experience. The available material explaining how to modify LAMMPS is either too basic or too advanced for the average researcher. In this work, we provide several examples, with increasing level of complexity, suitable for researchers and practitioners in physics and engineering, who are familiar with coding without been experts. For each feature, step by step instructions for implementing them in LAMMPS are shown to allow researchers to easily follow the procedure and compile a new version of the code. The aim is to fill a gap in the literature with particular reference to the scientific community that uses particle methods for (discrete) multiphysics.
Open Access
A hybrid computer vision and machine learning approach for robust vortex core detection in fluid mechanics applications
(American Society of Mechanical Engineers, 2024-06-01) Abolholl, Hazem Ashor Amran; Teschner, Tom-Robin; Moulitsas, Irene
Vortex core detection remains an unsolved problem in the field of experimental and computational fluid dynamics. Available methods such as the Q, delta and swirling-strength criterion are based on a decomposed velocity gradient tensor but detect spurious vortices (false positives and false negatives), making these methods less robust. To overcome this, we propose a new hybrid machine learning approach in which we used a convolutional neural network to detect vortex regions within surface streamline plots and an additional deep neural network to detect vortex cores within identified vortex regions. Furthermore, we propose an automatic labelling approach based on K-means clustering to pre-process our input images. We show results for two classical test cases in fluid mechanics; the Taylor-Green vortex problem and two rotating blades. We show that our hybrid approach is up to 2.6 times faster than a pure deep neural network-based approach and furthermore show that our automatic K-means clustering labelling approach is within 0.45% mean square error of the more labour-intensive, manual labelling approach. At the same time, using a sufficient number of samples, we show that we are able to reduce false positives and negatives entirely and thus show that our hybrid machine learning approach is a viable alternative to currently used vortex detection tools in fluid mechanics applications.

Browsing by Author "Moulitsas, Irene"

Results Per Page

Sort Options