Seán McLoone | National University of Ireland, Maynooth (original) (raw)
Dr. Sean Mc Loone, a native of Glenties, Co. Donegal, received his third level education at The Queen’s University of Belfast, where he was awarded a Master of Engineering Degree with Distinction in Electrical and Electronic Engineering in 1992 followed by a PhD in Control Engineering in 1996. The subject of his PhD research was ‘Neural Network Training for Modelling and Control’. In 1998 he enrolled part-time for a Postgraduate Certificate in Higher Education Teaching (PGCHET), which he successfully completed in July 2000.
From March 1995 to April 1996 he worked as a research fellow in the School of Electrical Engineering, Queen's University Belfast conducting research on modelling and control of power station plant using neural network techniques. In December 1996 he joined the Intelligent Systems and Control Research Group (ISAC) at Queen's as a postdoctoral research fellow, investigating the application of advanced modelling and multivariable statistical process control techniques to quality control and predictive maintenance of chemical processes, before taking up an appointment as Lecturer in Control Engineering at the University in January 1998. In September 2002 he joined the Department of Electronic Engineering at NUI Maynooth, where he is currently Senior Lecturer and Head of Department.
At a professional level, Dr. McLoone is a Chartered Engineer, a Senior Member of the IEEE and Past Chairman of the UK and Republic of Ireland (UKRI) Section of the IEEE. He is also a Member of the Accreditation Board of Engineers Ireland, Treasurer of the Irish Systems and Control Committee (Ireland’s NMO for IFAC) and a Member of the IFAC Technical Committee on Cognition and Control.
Dr. McLoone's research interests lie in the general area of data based modelling and analysis of dynamical systems. This encompasses techniques ranging from classical system identification, fault diagnosis and statistical process control to modern artificial intelligence inspired adaptive learning algorithms and optimisation techniques. Current research activities include data analysis and signal processing for in-home health monitoring and assisted living technologies, algorithms for unsupervised sparse feature selection and clustering with application to large semiconductor manufacturing and geospatial data sets and virtual metrology and predictive maintenance for manufacturing processes. His research has a strong application focus, with many projects undertaken in collaboration with industry in areas such as process monitoring, control and optimisation, time series prediction and in-line sensor characterisation.
He is a staff member of the Callan Institute, the Dynamics and Control Research Group at NUI Maynooth and an Honorary Senior Research Fellow in the Intelligent Systems and Control Research Group at Queen's University Belfast. He is also a member of Robo Eireann, an inter-disciplinary team of researchers at NUI Maynooth that compete annually in the NAO robot standard platform league of Robocup.
less
Related Authors
National Institute of Technology, Calicut
Graduate Center of the City University of New York
Uploads
Papers by Seán McLoone
IEEE Transactions on Pattern Analysis and Machine Intelligence, Dec 1, 2017
Principal Component Analysis (PCA) is a powerful and widely used tool for dimensionality reductio... more Principal Component Analysis (PCA) is a powerful and widely used tool for dimensionality reduction. However, the principal components generated are linear combinations of all the original variables and this often makes interpreting results and root-cause analysis difficult. Forward Selection Component Analysis (FSCA) is a recent technique that overcomes this difficulty by performing variable selection and dimensionality reduction at the same time. This paper provides, for the first time, a detailed presentation of the FSCA algorithm, and introduces a number of new variants of FSCA that incorporate a refinement step to improve performance. We then show different applications of FSCA and compare the performance of the different variants with PCA and Sparse PCA. The results demonstrate the efficacy of FSCA as a low information loss dimensionality reduction and variable selection technique and the improved performance achievable through the inclusion of a refinement step.
Transactions of the Institute of Measurement and Control, Jul 18, 2013
Moving from combustion engine to Electric Vehicle (EV) based transport is recognized as having a ... more Moving from combustion engine to Electric Vehicle (EV) based transport is recognized as having a major role to play in reducing pollution, combating climate change and improving energy security. However, the introduction of EVs poses major challenges for power system operation. With increasing penetration of EVs uncontrolled coincident charging may overload the grid and substantially increase peak power requirements. Developing smart grid technologies and appropriate charging strategies to support the role out of EVs is therefore a high priority. In this paper we investigate the effectiveness of distributed Additive Increase and Multiplicative Decrease (AIMD) charging algorithms, as proposed in (Stüdli et al., 2012a, 2012b), at mitigating the impact of domestic charging of EVs on low-voltage distribution networks. In particular, a number of enhancements to the basic AIMD implementation are introduced to enable local power system infrastructure and voltage level constraints to be taken into account and to reduce peak power requirements. The enhanced AIMD EV charging strategies are evaluated using power system simulations for a typical low voltage residential feeder network in Ireland. Results show that by using the proposed AIMD based smart charging algorithms 50% EV penetration can be accommodated, compared to only 10% with uncontrolled charging, without exceeding network infrastructure constraints.
The integration of an ever growing proportion of large scale distributed renewable generation has... more The integration of an ever growing proportion of large scale distributed renewable generation has increased the probability of maloperation of the traditional RoCoF and vector shift relays. With reduced inertia due to non-synchronous penetration in a power grid, system wide disturbances have forced the utility industry to design advanced protection schemes to prevent system degradation and avoid cascading outages leading to widespread blackouts. This paper explores a novel adaptive nonlinear approach applied to islanding detection, based on wide area phase angle measurements. This is challenging, since the voltage phase angles from different locations exhibit not only strong nonlinear but also time-varying characteristics. The adaptive nonlinear technique, called moving window kernel principal component analysis is proposed to model the time-varying and nonlinear trends in the voltage phase angle data. The effectiveness of the technique is exemplified using both DigSilent simulated cases and real test cases recorded from the Great Britain and Ireland power systems by the OpenPMU project. Index Terms-Islanding detection, kernel principal component analysis, moving window, phase angle measurements, wide area protection.
We are all familiar with the sound which can be viewed as a wave motion in air or other elastic m... more We are all familiar with the sound which can be viewed as a wave motion in air or other elastic media. In this case, sound is a stimulus. Sound can also be viewed as an excitation of the hearing mechanism that results in the perception of sound. The interaction between the physical properties of sound, and our perception of them, poses delicate and complex issues. It is this complexity in audio and acoustics that creates such interesting problems. Acoustic echo is inevitable whenever a speaker is placed near to a microphone in a
The Vldb Journal, May 28, 2020
Time series indexing plays an important role in querying and pattern mining of big data. This pap... more Time series indexing plays an important role in querying and pattern mining of big data. This paper proposes a novel structure for tightly covering a given set of time series under the dynamic time warping similarity measurement. The structure, referred to as Dynamic Covering with cross-Range Constraints (DCRC), enables more efficient and scalable indexing to be developed than current hypercube based partitioning approaches. In particular, a lower bound of the DTW distance from a given query time series to a DCRC-based cover set is introduced. By virtue of its tightness, which is proven theoretically, the lower bound can be used for pruning when querying on an indexing tree. If the DCR-C based Lower Bound (LB DCRC) of an upper node in an index tree is larger than a given threshold, all child nodes can be pruned yielding a significant reduction in
Machine learning for signal processing ..., Aug 1, 2007
Functional testing of radio frequency integrated circuits is a challenging task and one that is b... more Functional testing of radio frequency integrated circuits is a challenging task and one that is becoming an increasingly expensive aspect of circuit manufacture. Due to the difficulties with bringing high frequency signals off-chip, current automated test equipment (ATE) technologies are approaching the limits of their operating capabilities as circuits are pushed to operate at higher and higher frequencies. This paper explores the possibility of extending the operating range of existing ATEs by using machine learning techniques to infer high frequency circuit performance from more accessible lower frequency and DC measurements. Results from a simulation study conducted on a low noise amplifier (LNA) circuit operating at 2.4 GHz demonstrate that the proposed approach has the potential to substantially increase the operating bandwidth ofATE.
Energy, Dec 1, 2022
Future energy systems rely on integrating renewable energy resources to decarbonise the heating a... more Future energy systems rely on integrating renewable energy resources to decarbonise the heating and cooling sectors and contribute to global net zero targets. Traditional approaches to energy modelling are segregated as focus tends to be on individual objectives such as minimising operational cost. Furthermore, they are limited with respect to computational time, level of precision and scalability. Model complexity is greater for district heating and cooling systems when compared to power systems due to the thermal behaviour and fluid dynamic principles which are present. Prevailing research tends to deliver a detailed analysis of specific elements within the network, but an approach for visualising the whole system is still missing. This study aims to evaluate the current tools and techniques used to model heating and cooling networks and then propose a more up to date hybrid approach that utilises recent technical advancements. A detailed literature review outlines existing modelling methods and assesses the capabilities of available software tools. The results are summarised in a Pugh Matrix using relevant criteria to compare and select the most appropriate methods. The review concludes that energy models must evolve to become interdisciplinary and multi-objective to simulate a smart energy system.
IEEE Transactions on Power Delivery, Dec 1, 2021
Electric Power Systems Research, Apr 1, 2019
This paper investigates the voltage level dependencies of different modern lighting types. Experi... more This paper investigates the voltage level dependencies of different modern lighting types. Experiments are undertaken and the results used to formulate polynomial load models, presented in a ZIP format. These characterize active power, reactive power and harmonic current behaviors; are compatible with established power systems simulation practice; and are generally demonstrated to muster a good fit to the raw experimental data. Despite exceedingly high levels of current distortion being noted for several lamp instances, the attenuating effects of lowering voltage and harmonic diversity, as captured within a pair of diversity factor terms, are shown to help significantly alleviate the associated harmonic pollution impact on low voltage networks.
Polymers
This work investigates real-time monitoring of extrusion-induced degradation in different grades ... more This work investigates real-time monitoring of extrusion-induced degradation in different grades of PLA across a range of process conditions and machine set-ups. Data on machine settings together with in-process sensor data, including temperature, pressure, and near-infrared (NIR) spectra, are used as inputs to predict the molecular weight and mechanical properties of the product. Many soft sensor approaches based on complex spectral data are essentially ‘black-box’ in nature, which can limit industrial acceptability. Hence, the focus here is on identifying an optimal approach to developing interpretable models while achieving high predictive accuracy and robustness across different process settings. The performance of a Recursive Feature Elimination (RFE) approach was compared to more common dimension reduction and regression approaches including Partial Least Squares (PLS), iterative PLS (i-PLS), Principal Component Regression (PCR), ridge regression, Least Absolute Shrinkage and ...
Ninth IEEE International Symposium on Distributed Simulation and Real-Time Applications, 2005
Human-to-human interaction across distributed applications requires that sufficient consistency b... more Human-to-human interaction across distributed applications requires that sufficient consistency be maintained among participants in the face of network characteristics such as latency and limited bandwidth. Techniques and approaches for reducing bandwidth usage can minimise network delays by reducing the network traffic and therefore better exploiting available bandwidth. However, these approaches induce inconsistencies within the level of human perception. Dead reckoning is a well-known technique for reducing the number of update packets transmitted between participating nodes. It employs a distance threshold for deciding when to generate update packets. This paper questions the use of such a distance threshold in the context of absolute consistency and it highlights a major drawback with such a technique. An alternative threshold criterion based on time and distance is examined and it is compared to the distance only threshold. A drawback with this proposed technique is also identified and a hybrid threshold criterion is then proposed. However, the trade-off between spatial and temporal inconsistency remains.
Proceedings of the 19th IFAC World Congress, 2014
Electric Vehicle (EV) technology has developed rapidly in recent years, with the result that incr... more Electric Vehicle (EV) technology has developed rapidly in recent years, with the result that increasing levels of EV penetration are expected on electrical grids in the near future. The increasing electricity demand due to EVs is expected to provide many challenges for grid companies, and it is expected that it will be necessary to reinforce the current electrical grid infrastructure to cater for increasing loads at distribution level. However, by harnessing the power of Vehicle to Grid (V2G) technologies, groups of EVs could be harnessed to provide ancillary services to the grid. Current unbalance occurs at distribution level when currents are unbalanced between each of the phases. In this paper a distributed consensus algorithm is used to coordinate EV charging in order to minimise current unbalance. Simulation results demonstrate that the proposed algorithm is effective in rebalancing phase currents.
SPIE Proceedings, 2005
In this paper a new stable high order Digital Phase Lock Loop (DPLL) design technique is proposed... more In this paper a new stable high order Digital Phase Lock Loop (DPLL) design technique is proposed. This technique uses linear theory to design the DPLL. The stability of the DPLL is guaranteed by placing a restriction on the system gain. This stability boundary is found by transforming the system transfer function to the Z-domain and plotting the root locus of the LPLL for values of gain where all the system poles lie inside the unit circle. The max value of gain where all the poles lie inside the unit circle is the stability boundary. It is shown that the stability boundary of the LPLL is comparable to the stability boundary of the DPLL. Finally where the above Bessel filter system produces slow lock, gear shifting of the DPLL components is considered. This allows the DPLL to start off with a wide loop bandwidth and switch to the narrow bandwidth once the system has locked.
UK-RAS Conference for PhD and Early Career Researchers Proceedings, 2020
Melecon 2010 - 2010 15th IEEE Mediterranean Electrotechnical Conference, 2010
Intrusive temperature sensors such as thermocouples and resistance temperature detectors (RTDs) h... more Intrusive temperature sensors such as thermocouples and resistance temperature detectors (RTDs) have become industry standards for simple and cost-effective temperature measurement. However, many situations require the use of physically robust and therefore low bandwidth temperature sensors. Much work has been published on dual-thermocouple thermometry as a means of obtaining increased sensor bandwidth from relatively robust thermocouples, which are assumed to have firstorder response. This contribution seeks to determine if RTDs, which are known to have approximately first-order response [1], can also be characterised using the dual-thermocouple approach. Experimental results show that the response of an RTD cannot be represented by a first-order model with sufficient accuracy to allow successful application of this method. Furthermore, simulation studies demonstrated that if a sensor exhibits even marginally second-order response, highly inaccurate temperature reconstructions follow. It is concluded that a higher-order model that more accurately reflects RTD response would be required for successful dual-RTD characterisation.
at - Automatisierungstechnik, 2009
The optimization of agricultural and industrial biogas plants with respect to external influences... more The optimization of agricultural and industrial biogas plants with respect to external influences and various process disturbances is essential for efficient plant operation. The fact that most biogas plants are manually operated because of a lack of online-measurements and limited knowledge about the anaerobic digestion process makes it necessary to develop new optimization and control strategies. However, the optimization and control of such plants is a challenging problem due to the underlying highly nonlinear and complex digestion processes. One approach to address this challenge is to exploit the flexibility and power of computational intelligence (CI) methods such as Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO). The use of CI methods in conjunction with a validated plant simulation model, based on the Anaerobic Digestion Model No. 1, allows optimization of the substrate feed mix, a key factor in stable and efficient biogas production. Results show that an imp...
2008 IEEE/SEMI Advanced Semiconductor Manufacturing Conference, 2008
Advanced data mining techniques such as variable selection through matrix factorization have been... more Advanced data mining techniques such as variable selection through matrix factorization have been intensively applied in the last ten years in the area of plasma-etch point detection using optimal emission spectroscopy (OES). OES data sets are enormous, consisting of measurements of over 2000 wavelength recorded at sample rates of 1 − 3 Hertz, and consequently, these techniques are needed in order to generate compact representations of the relevant process characteristics. To date, the main technique employed in this regard has been PCA (Principal Components Analysis), a matrix factorisation technique which generates linear combinations of the original variables that best capture the information in the data (in terms of variance explained). Recently, an alternative matrix factorisation technique, Non-Negative Matrix Factorisation (NMF) [1], has been gaining increasing attention in the fields of image feature extraction and blind source separation due to its tendency to yield sparse representations of data. The aim of this work is to introduce Non-Negative Matrix Factorisation to the semiconductor research community and to provide a comparison with PCA in order to highlight its properties.
Annals of GIS, 2014
The quantity and quality of spatial data are increasing rapidly. This is particularly evident in ... more The quantity and quality of spatial data are increasing rapidly. This is particularly evident in the case of movement data. Devices capable of accurately recording the position of moving entities have become ubiquitous and created an abundance of movement data. Valuable knowledge concerning processes occurring in the physical world can be extracted from these large movement data sets. Geovisual analytics offers powerful techniques to achieve this. This article describes a new geovisual analytics tool specifically designed for movement data. The tool features the classic space-time cube augmented with a novel clustering approach to identify common behaviour. These techniques were used to analyse pedestrian movement in a city environment which revealed the effectiveness of the tool for identifying spatiotemporal patterns.
IET Irish Signals and Systems Conference (ISSC 2012), 2012
Mobile phone networks were traditionally voice and SMS messaging networks which made them easy to... more Mobile phone networks were traditionally voice and SMS messaging networks which made them easy to quantify and model. More recently with the advent of broadband data connections and smart phones, mobile phone networks are being used for a wide range of purposes using ever more bandwidth. This paper utilises data provided by an Irish operator to provide an initial perspective on the current uses of mobile phone networks as a within Ireland. This work serves as a precursor to the development of a quantitative model of existing behaviour which in turn will help predict future requirements.
IET Irish Signals and Systems Conference (ISSC 2012), 2012
This work utilises data from an Irish mobile phone network to provide a preliminary, but novel, a... more This work utilises data from an Irish mobile phone network to provide a preliminary, but novel, analysis of the interurban communication network between twenty five of the largest cities and towns in Ireland. An intuitive technique is applied to a mobile phone operator's call detail records to identify the actual subscriber population of different urban areas with various penetration rates. Weighted communication links are generated between the urban centres based on spatial and temporal metrics of distance, and are examined for different times of the day and for different days of the week. These communication links are compared to the output of a standard gravity model in order to ascertain the latter's ability to accurately represent Ireland's interurban communication network. The results obtained are presented and discussed within.
IEEE Transactions on Pattern Analysis and Machine Intelligence, Dec 1, 2017
Principal Component Analysis (PCA) is a powerful and widely used tool for dimensionality reductio... more Principal Component Analysis (PCA) is a powerful and widely used tool for dimensionality reduction. However, the principal components generated are linear combinations of all the original variables and this often makes interpreting results and root-cause analysis difficult. Forward Selection Component Analysis (FSCA) is a recent technique that overcomes this difficulty by performing variable selection and dimensionality reduction at the same time. This paper provides, for the first time, a detailed presentation of the FSCA algorithm, and introduces a number of new variants of FSCA that incorporate a refinement step to improve performance. We then show different applications of FSCA and compare the performance of the different variants with PCA and Sparse PCA. The results demonstrate the efficacy of FSCA as a low information loss dimensionality reduction and variable selection technique and the improved performance achievable through the inclusion of a refinement step.
Transactions of the Institute of Measurement and Control, Jul 18, 2013
Moving from combustion engine to Electric Vehicle (EV) based transport is recognized as having a ... more Moving from combustion engine to Electric Vehicle (EV) based transport is recognized as having a major role to play in reducing pollution, combating climate change and improving energy security. However, the introduction of EVs poses major challenges for power system operation. With increasing penetration of EVs uncontrolled coincident charging may overload the grid and substantially increase peak power requirements. Developing smart grid technologies and appropriate charging strategies to support the role out of EVs is therefore a high priority. In this paper we investigate the effectiveness of distributed Additive Increase and Multiplicative Decrease (AIMD) charging algorithms, as proposed in (Stüdli et al., 2012a, 2012b), at mitigating the impact of domestic charging of EVs on low-voltage distribution networks. In particular, a number of enhancements to the basic AIMD implementation are introduced to enable local power system infrastructure and voltage level constraints to be taken into account and to reduce peak power requirements. The enhanced AIMD EV charging strategies are evaluated using power system simulations for a typical low voltage residential feeder network in Ireland. Results show that by using the proposed AIMD based smart charging algorithms 50% EV penetration can be accommodated, compared to only 10% with uncontrolled charging, without exceeding network infrastructure constraints.
The integration of an ever growing proportion of large scale distributed renewable generation has... more The integration of an ever growing proportion of large scale distributed renewable generation has increased the probability of maloperation of the traditional RoCoF and vector shift relays. With reduced inertia due to non-synchronous penetration in a power grid, system wide disturbances have forced the utility industry to design advanced protection schemes to prevent system degradation and avoid cascading outages leading to widespread blackouts. This paper explores a novel adaptive nonlinear approach applied to islanding detection, based on wide area phase angle measurements. This is challenging, since the voltage phase angles from different locations exhibit not only strong nonlinear but also time-varying characteristics. The adaptive nonlinear technique, called moving window kernel principal component analysis is proposed to model the time-varying and nonlinear trends in the voltage phase angle data. The effectiveness of the technique is exemplified using both DigSilent simulated cases and real test cases recorded from the Great Britain and Ireland power systems by the OpenPMU project. Index Terms-Islanding detection, kernel principal component analysis, moving window, phase angle measurements, wide area protection.
We are all familiar with the sound which can be viewed as a wave motion in air or other elastic m... more We are all familiar with the sound which can be viewed as a wave motion in air or other elastic media. In this case, sound is a stimulus. Sound can also be viewed as an excitation of the hearing mechanism that results in the perception of sound. The interaction between the physical properties of sound, and our perception of them, poses delicate and complex issues. It is this complexity in audio and acoustics that creates such interesting problems. Acoustic echo is inevitable whenever a speaker is placed near to a microphone in a
The Vldb Journal, May 28, 2020
Time series indexing plays an important role in querying and pattern mining of big data. This pap... more Time series indexing plays an important role in querying and pattern mining of big data. This paper proposes a novel structure for tightly covering a given set of time series under the dynamic time warping similarity measurement. The structure, referred to as Dynamic Covering with cross-Range Constraints (DCRC), enables more efficient and scalable indexing to be developed than current hypercube based partitioning approaches. In particular, a lower bound of the DTW distance from a given query time series to a DCRC-based cover set is introduced. By virtue of its tightness, which is proven theoretically, the lower bound can be used for pruning when querying on an indexing tree. If the DCR-C based Lower Bound (LB DCRC) of an upper node in an index tree is larger than a given threshold, all child nodes can be pruned yielding a significant reduction in
Machine learning for signal processing ..., Aug 1, 2007
Functional testing of radio frequency integrated circuits is a challenging task and one that is b... more Functional testing of radio frequency integrated circuits is a challenging task and one that is becoming an increasingly expensive aspect of circuit manufacture. Due to the difficulties with bringing high frequency signals off-chip, current automated test equipment (ATE) technologies are approaching the limits of their operating capabilities as circuits are pushed to operate at higher and higher frequencies. This paper explores the possibility of extending the operating range of existing ATEs by using machine learning techniques to infer high frequency circuit performance from more accessible lower frequency and DC measurements. Results from a simulation study conducted on a low noise amplifier (LNA) circuit operating at 2.4 GHz demonstrate that the proposed approach has the potential to substantially increase the operating bandwidth ofATE.
Energy, Dec 1, 2022
Future energy systems rely on integrating renewable energy resources to decarbonise the heating a... more Future energy systems rely on integrating renewable energy resources to decarbonise the heating and cooling sectors and contribute to global net zero targets. Traditional approaches to energy modelling are segregated as focus tends to be on individual objectives such as minimising operational cost. Furthermore, they are limited with respect to computational time, level of precision and scalability. Model complexity is greater for district heating and cooling systems when compared to power systems due to the thermal behaviour and fluid dynamic principles which are present. Prevailing research tends to deliver a detailed analysis of specific elements within the network, but an approach for visualising the whole system is still missing. This study aims to evaluate the current tools and techniques used to model heating and cooling networks and then propose a more up to date hybrid approach that utilises recent technical advancements. A detailed literature review outlines existing modelling methods and assesses the capabilities of available software tools. The results are summarised in a Pugh Matrix using relevant criteria to compare and select the most appropriate methods. The review concludes that energy models must evolve to become interdisciplinary and multi-objective to simulate a smart energy system.
IEEE Transactions on Power Delivery, Dec 1, 2021
Electric Power Systems Research, Apr 1, 2019
This paper investigates the voltage level dependencies of different modern lighting types. Experi... more This paper investigates the voltage level dependencies of different modern lighting types. Experiments are undertaken and the results used to formulate polynomial load models, presented in a ZIP format. These characterize active power, reactive power and harmonic current behaviors; are compatible with established power systems simulation practice; and are generally demonstrated to muster a good fit to the raw experimental data. Despite exceedingly high levels of current distortion being noted for several lamp instances, the attenuating effects of lowering voltage and harmonic diversity, as captured within a pair of diversity factor terms, are shown to help significantly alleviate the associated harmonic pollution impact on low voltage networks.
Polymers
This work investigates real-time monitoring of extrusion-induced degradation in different grades ... more This work investigates real-time monitoring of extrusion-induced degradation in different grades of PLA across a range of process conditions and machine set-ups. Data on machine settings together with in-process sensor data, including temperature, pressure, and near-infrared (NIR) spectra, are used as inputs to predict the molecular weight and mechanical properties of the product. Many soft sensor approaches based on complex spectral data are essentially ‘black-box’ in nature, which can limit industrial acceptability. Hence, the focus here is on identifying an optimal approach to developing interpretable models while achieving high predictive accuracy and robustness across different process settings. The performance of a Recursive Feature Elimination (RFE) approach was compared to more common dimension reduction and regression approaches including Partial Least Squares (PLS), iterative PLS (i-PLS), Principal Component Regression (PCR), ridge regression, Least Absolute Shrinkage and ...
Ninth IEEE International Symposium on Distributed Simulation and Real-Time Applications, 2005
Human-to-human interaction across distributed applications requires that sufficient consistency b... more Human-to-human interaction across distributed applications requires that sufficient consistency be maintained among participants in the face of network characteristics such as latency and limited bandwidth. Techniques and approaches for reducing bandwidth usage can minimise network delays by reducing the network traffic and therefore better exploiting available bandwidth. However, these approaches induce inconsistencies within the level of human perception. Dead reckoning is a well-known technique for reducing the number of update packets transmitted between participating nodes. It employs a distance threshold for deciding when to generate update packets. This paper questions the use of such a distance threshold in the context of absolute consistency and it highlights a major drawback with such a technique. An alternative threshold criterion based on time and distance is examined and it is compared to the distance only threshold. A drawback with this proposed technique is also identified and a hybrid threshold criterion is then proposed. However, the trade-off between spatial and temporal inconsistency remains.
Proceedings of the 19th IFAC World Congress, 2014
Electric Vehicle (EV) technology has developed rapidly in recent years, with the result that incr... more Electric Vehicle (EV) technology has developed rapidly in recent years, with the result that increasing levels of EV penetration are expected on electrical grids in the near future. The increasing electricity demand due to EVs is expected to provide many challenges for grid companies, and it is expected that it will be necessary to reinforce the current electrical grid infrastructure to cater for increasing loads at distribution level. However, by harnessing the power of Vehicle to Grid (V2G) technologies, groups of EVs could be harnessed to provide ancillary services to the grid. Current unbalance occurs at distribution level when currents are unbalanced between each of the phases. In this paper a distributed consensus algorithm is used to coordinate EV charging in order to minimise current unbalance. Simulation results demonstrate that the proposed algorithm is effective in rebalancing phase currents.
SPIE Proceedings, 2005
In this paper a new stable high order Digital Phase Lock Loop (DPLL) design technique is proposed... more In this paper a new stable high order Digital Phase Lock Loop (DPLL) design technique is proposed. This technique uses linear theory to design the DPLL. The stability of the DPLL is guaranteed by placing a restriction on the system gain. This stability boundary is found by transforming the system transfer function to the Z-domain and plotting the root locus of the LPLL for values of gain where all the system poles lie inside the unit circle. The max value of gain where all the poles lie inside the unit circle is the stability boundary. It is shown that the stability boundary of the LPLL is comparable to the stability boundary of the DPLL. Finally where the above Bessel filter system produces slow lock, gear shifting of the DPLL components is considered. This allows the DPLL to start off with a wide loop bandwidth and switch to the narrow bandwidth once the system has locked.
UK-RAS Conference for PhD and Early Career Researchers Proceedings, 2020
Melecon 2010 - 2010 15th IEEE Mediterranean Electrotechnical Conference, 2010
Intrusive temperature sensors such as thermocouples and resistance temperature detectors (RTDs) h... more Intrusive temperature sensors such as thermocouples and resistance temperature detectors (RTDs) have become industry standards for simple and cost-effective temperature measurement. However, many situations require the use of physically robust and therefore low bandwidth temperature sensors. Much work has been published on dual-thermocouple thermometry as a means of obtaining increased sensor bandwidth from relatively robust thermocouples, which are assumed to have firstorder response. This contribution seeks to determine if RTDs, which are known to have approximately first-order response [1], can also be characterised using the dual-thermocouple approach. Experimental results show that the response of an RTD cannot be represented by a first-order model with sufficient accuracy to allow successful application of this method. Furthermore, simulation studies demonstrated that if a sensor exhibits even marginally second-order response, highly inaccurate temperature reconstructions follow. It is concluded that a higher-order model that more accurately reflects RTD response would be required for successful dual-RTD characterisation.
at - Automatisierungstechnik, 2009
The optimization of agricultural and industrial biogas plants with respect to external influences... more The optimization of agricultural and industrial biogas plants with respect to external influences and various process disturbances is essential for efficient plant operation. The fact that most biogas plants are manually operated because of a lack of online-measurements and limited knowledge about the anaerobic digestion process makes it necessary to develop new optimization and control strategies. However, the optimization and control of such plants is a challenging problem due to the underlying highly nonlinear and complex digestion processes. One approach to address this challenge is to exploit the flexibility and power of computational intelligence (CI) methods such as Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO). The use of CI methods in conjunction with a validated plant simulation model, based on the Anaerobic Digestion Model No. 1, allows optimization of the substrate feed mix, a key factor in stable and efficient biogas production. Results show that an imp...
2008 IEEE/SEMI Advanced Semiconductor Manufacturing Conference, 2008
Advanced data mining techniques such as variable selection through matrix factorization have been... more Advanced data mining techniques such as variable selection through matrix factorization have been intensively applied in the last ten years in the area of plasma-etch point detection using optimal emission spectroscopy (OES). OES data sets are enormous, consisting of measurements of over 2000 wavelength recorded at sample rates of 1 − 3 Hertz, and consequently, these techniques are needed in order to generate compact representations of the relevant process characteristics. To date, the main technique employed in this regard has been PCA (Principal Components Analysis), a matrix factorisation technique which generates linear combinations of the original variables that best capture the information in the data (in terms of variance explained). Recently, an alternative matrix factorisation technique, Non-Negative Matrix Factorisation (NMF) [1], has been gaining increasing attention in the fields of image feature extraction and blind source separation due to its tendency to yield sparse representations of data. The aim of this work is to introduce Non-Negative Matrix Factorisation to the semiconductor research community and to provide a comparison with PCA in order to highlight its properties.
Annals of GIS, 2014
The quantity and quality of spatial data are increasing rapidly. This is particularly evident in ... more The quantity and quality of spatial data are increasing rapidly. This is particularly evident in the case of movement data. Devices capable of accurately recording the position of moving entities have become ubiquitous and created an abundance of movement data. Valuable knowledge concerning processes occurring in the physical world can be extracted from these large movement data sets. Geovisual analytics offers powerful techniques to achieve this. This article describes a new geovisual analytics tool specifically designed for movement data. The tool features the classic space-time cube augmented with a novel clustering approach to identify common behaviour. These techniques were used to analyse pedestrian movement in a city environment which revealed the effectiveness of the tool for identifying spatiotemporal patterns.
IET Irish Signals and Systems Conference (ISSC 2012), 2012
Mobile phone networks were traditionally voice and SMS messaging networks which made them easy to... more Mobile phone networks were traditionally voice and SMS messaging networks which made them easy to quantify and model. More recently with the advent of broadband data connections and smart phones, mobile phone networks are being used for a wide range of purposes using ever more bandwidth. This paper utilises data provided by an Irish operator to provide an initial perspective on the current uses of mobile phone networks as a within Ireland. This work serves as a precursor to the development of a quantitative model of existing behaviour which in turn will help predict future requirements.
IET Irish Signals and Systems Conference (ISSC 2012), 2012
This work utilises data from an Irish mobile phone network to provide a preliminary, but novel, a... more This work utilises data from an Irish mobile phone network to provide a preliminary, but novel, analysis of the interurban communication network between twenty five of the largest cities and towns in Ireland. An intuitive technique is applied to a mobile phone operator's call detail records to identify the actual subscriber population of different urban areas with various penetration rates. Weighted communication links are generated between the urban centres based on spatial and temporal metrics of distance, and are examined for different times of the day and for different days of the week. These communication links are compared to the output of a standard gravity model in order to ascertain the latter's ability to accurately represent Ireland's interurban communication network. The results obtained are presented and discussed within.