EPE Association: EPE 2019 - DS1j: Data Analysis, Artificial Intelligence, and Communication Issues

EPE 2019 - DS1j: Data Analysis, Artificial Intelligence, and Communication Issues
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2019 ECCE Europe - Conference > EPE 2019 - Topic 10: Data Analysis, Artificial Intelligence And Communication Issues > EPE 2019 - DS1j: Data Analysis, Artificial Intelligence, and Communication Issues

   [return to parent folder]

   Charging Station Selection through the Analytic Hierarchy Process enabled by OPC-UA for Vehicle-to-Grid Communications
By Nikolaos MILAS [View]
[Download]

Abstract: During the last decade, the technologies related to the electric vehicles capture both the scientific and the industrial interest. In this context, the Smart Grid philosophy with the Vehicle-to-Grid concept poses a major role for the prevalence of the electric vehicles. Despite the presence of extended literature in the topic, the interoperability of different systems through well-adopted standards is still a challenge. Aiming to address this challenge and to support decision-making in a smart grid, this paper proposes a framework for the charging station selection in the case of electric vehicles. The framework employs the Analytic Hierarchy Process (AHP) and the Open Platform Communications-Unified Architecture (OPC-UA) standard. Specifically, the AHP is utilised to solve the multicriteria decision-making problem during the selection of the most appropriate charging station that will fulfil the needs of the electric vehicle driver. In addition, the OPC-UA enables the semantic modelling of the electric vehicle integrating all the interrelationships among the various subsystems and their components. As a result, a seamless information flow can be resulted among systems designed using different engineering disciplines (power electronics, mechanical, microelectronics, etc). The proposed framework is validated in the case study of a laboratory electric vehicle that is designed for urban transportations.

   Detecting Performance Outliers in Fuel Cell Backup Power Systems
By Simon SONDERSKOV [View]
[Download]

Abstract: Fuel cell based backup power systems are finding application in telecommunication applications. Due to the criticality of the application, as well as the effort of keeping operating costs low, appropriate maintenance strategies are of high importance. This paper investigates key performance indicators, derived from numerous fuel cell based backup systems, installed in the field. The methods of principal component analysis and local outlier factor are applied to the KPIs, in order to identify systems that are performing differently from the majority of the systems. These underperforming systems can then be examined closer to identify potential problems.

   Effect of power quality on PEM fuel cells and water electrolyzers: A literature review with Watson Discovery
By Lauri JÄRVINEN [View]
[Download]

Abstract: The growing number of research articles makes the comprehension of a scientific field increasingly difficult. Tools for automated analysis of unstructured data have recently emerged. This paper studies the use of cognitive data analysis tools to perform a literature review on the effects of power quality on electrolytic cells.

   Low-Cost Multi-Channel Data Transmission over a Single Plastic Optic Fibre for Isolated Sensing Applications
By Michael PRITZ [View]
[Download]

Abstract: The growing need for higher efficiency in converter systems and in power electronic systems in general has led to the implementation of more complex and sophisticated control algorithms. Often, the controller unit is separated from the sensors and the data has to be transmitted via an isolated channel. Plastic optical fibers are utilized for this task due to their low cost, high robustness and easy handling. In order to reduce the number of fibers, intensity modulation can be used to transmit multiple signals over a single optical fiber. A recently introduced modulation scheme with an additional intensity level indicating the clock allows a simple clock recovery at the receiver. In this paper, three receiver concepts based on this modulation scheme are presented for medium data rates in the range of 20 - 50 Mbps which is sufficient for most power electronic applications. The low complexity and small footprint of the proposed concepts facilitate the system integration. Furthermore, a more compact system can be designed due to the lower number of necessary fibers and corresponding transmitter and receiver circuits.

   Online Switch Open-Circuit Fault Diagnosis Using Reconfigurable Scheduler for Modular Multilevel Converter with Parallel Connectivity
By Chuang WANG [View]
[Download]

Abstract: The modular multilevel series-parallel converter (MMSPC) was shown to provide advantages in several industrial applications, but the many individual switches may challenge its reliability. This paper proposes an effective fault diagnosis method (FDM) based on temporary hot-swapping of the look-up table of the scheduler and the use of additional interconnection states to detect and locate transistor open-circuit faults. First, the feature table of all possible switch open-circuit faults is collected via temporarily recon_guring the scheduler. The converter can intermittently switch to the fault-diagnosis mode for only a few cycles. Features as simple as the square of the error between the output and reference in the fault-diagnosis mode indicate a switch open-circuit fault. Finally, the output voltage characteristics in the fault diagnosis mode are checked and compared with those in the aforesaid characteristic tables to locate the open switch. Validation results reveal that this proposed fault diagnosis method is efficient, accurate, and fast.

   Tensor Based Algorithm for Automatic Partial Discharges Pattern Classification
By Christian GIANOGLIO [View]
[Download]

Abstract: The development of automated tools capable of monitoring electric motors is important for industrialapplications. The measure of partial discharges is one of the most prominent methodologies for the evaluation of electric motors operating conditions. This work proposes to apply tensor-based classification methods to discriminate between different partial discharges sources. The major advantage of tensor based approaches relies on the fact that they can inherently handle partial discharges pattern using all the geometrical information embedded in the patterns. Secondly, noise filtering can be embedded in classification mechanism, without manual tuning. As a consequence, better performances are obtained in presence of additive sources of noise like 4G, GSM connections and electromagnetic sources that could be detected by conducted and irradiated sensors especially during an online monitoring. To demonstrate the ability of the tensor-based methods to overcome the noise presence, tests involved defects with different levels of addictive noises. Experimental results proved the effectiveness of the proposed approach.

EPE 2019 - DS1j: Data Analysis, Artificial Intelligence, and Communication Issues
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2019 ECCE Europe - Conference > EPE 2019 - Topic 10: Data Analysis, Artificial Intelligence And Communication Issues > EPE 2019 - DS1j: Data Analysis, Artificial Intelligence, and Communication Issues
	[return to parent folder]

	Charging Station Selection through the Analytic Hierarchy Process enabled by OPC-UA for Vehicle-to-Grid Communications By Nikolaos MILAS	[View] [Download]
Abstract: During the last decade, the technologies related to the electric vehicles capture both the scientific and the industrial interest. In this context, the Smart Grid philosophy with the Vehicle-to-Grid concept poses a major role for the prevalence of the electric vehicles. Despite the presence of extended literature in the topic, the interoperability of different systems through well-adopted standards is still a challenge. Aiming to address this challenge and to support decision-making in a smart grid, this paper proposes a framework for the charging station selection in the case of electric vehicles. The framework employs the Analytic Hierarchy Process (AHP) and the Open Platform Communications-Unified Architecture (OPC-UA) standard. Specifically, the AHP is utilised to solve the multicriteria decision-making problem during the selection of the most appropriate charging station that will fulfil the needs of the electric vehicle driver. In addition, the OPC-UA enables the semantic modelling of the electric vehicle integrating all the interrelationships among the various subsystems and their components. As a result, a seamless information flow can be resulted among systems designed using different engineering disciplines (power electronics, mechanical, microelectronics, etc). The proposed framework is validated in the case study of a laboratory electric vehicle that is designed for urban transportations.

	Detecting Performance Outliers in Fuel Cell Backup Power Systems By Simon SONDERSKOV	[View] [Download]
Abstract: Fuel cell based backup power systems are finding application in telecommunication applications. Due to the criticality of the application, as well as the effort of keeping operating costs low, appropriate maintenance strategies are of high importance. This paper investigates key performance indicators, derived from numerous fuel cell based backup systems, installed in the field. The methods of principal component analysis and local outlier factor are applied to the KPIs, in order to identify systems that are performing differently from the majority of the systems. These underperforming systems can then be examined closer to identify potential problems.

	Effect of power quality on PEM fuel cells and water electrolyzers: A literature review with Watson Discovery By Lauri JÄRVINEN	[View] [Download]
Abstract: The growing number of research articles makes the comprehension of a scientific field increasingly difficult. Tools for automated analysis of unstructured data have recently emerged. This paper studies the use of cognitive data analysis tools to perform a literature review on the effects of power quality on electrolytic cells.

	Low-Cost Multi-Channel Data Transmission over a Single Plastic Optic Fibre for Isolated Sensing Applications By Michael PRITZ	[View] [Download]
Abstract: The growing need for higher efficiency in converter systems and in power electronic systems in general has led to the implementation of more complex and sophisticated control algorithms. Often, the controller unit is separated from the sensors and the data has to be transmitted via an isolated channel. Plastic optical fibers are utilized for this task due to their low cost, high robustness and easy handling. In order to reduce the number of fibers, intensity modulation can be used to transmit multiple signals over a single optical fiber. A recently introduced modulation scheme with an additional intensity level indicating the clock allows a simple clock recovery at the receiver. In this paper, three receiver concepts based on this modulation scheme are presented for medium data rates in the range of 20 - 50 Mbps which is sufficient for most power electronic applications. The low complexity and small footprint of the proposed concepts facilitate the system integration. Furthermore, a more compact system can be designed due to the lower number of necessary fibers and corresponding transmitter and receiver circuits.

	Online Switch Open-Circuit Fault Diagnosis Using Reconfigurable Scheduler for Modular Multilevel Converter with Parallel Connectivity By Chuang WANG	[View] [Download]
Abstract: The modular multilevel series-parallel converter (MMSPC) was shown to provide advantages in several industrial applications, but the many individual switches may challenge its reliability. This paper proposes an effective fault diagnosis method (FDM) based on temporary hot-swapping of the look-up table of the scheduler and the use of additional interconnection states to detect and locate transistor open-circuit faults. First, the feature table of all possible switch open-circuit faults is collected via temporarily recon_guring the scheduler. The converter can intermittently switch to the fault-diagnosis mode for only a few cycles. Features as simple as the square of the error between the output and reference in the fault-diagnosis mode indicate a switch open-circuit fault. Finally, the output voltage characteristics in the fault diagnosis mode are checked and compared with those in the aforesaid characteristic tables to locate the open switch. Validation results reveal that this proposed fault diagnosis method is efficient, accurate, and fast.

	Tensor Based Algorithm for Automatic Partial Discharges Pattern Classification By Christian GIANOGLIO	[View] [Download]
Abstract: The development of automated tools capable of monitoring electric motors is important for industrialapplications. The measure of partial discharges is one of the most prominent methodologies for the evaluation of electric motors operating conditions. This work proposes to apply tensor-based classification methods to discriminate between different partial discharges sources. The major advantage of tensor based approaches relies on the fact that they can inherently handle partial discharges pattern using all the geometrical information embedded in the patterns. Secondly, noise filtering can be embedded in classification mechanism, without manual tuning. As a consequence, better performances are obtained in presence of additive sources of noise like 4G, GSM connections and electromagnetic sources that could be detected by conducted and irradiated sensors especially during an online monitoring. To demonstrate the ability of the tensor-based methods to overcome the noise presence, tests involved defects with different levels of addictive noises. Experimental results proved the effectiveness of the proposed approach.