International Journal of Digital Signals and Smart Systems (6 papers in press)
Diagnosis of Stator Faults in Permanent Magnet Synchronous Machine Using Finite Element Method
by Manel Fitouri
Abstract: This paper attempts to analyze the characteristics of the inter-turn short circuit fault for the Permanent Magnet Synchronous Motor (PMSM). Based on Finite Element Analysis (FEA), the influence of the stator faults on the behavior of the PM machine is studied. A simple dynamic model for a PM machine with inter-turn winding fault is presented. The precision of the proposed Finite Element (FE) model is verified by a comparison of the simulation results tests. Using the simulated model, a technical method, based on Fast Fourier Transform (FFT) analysis of stator current and electromagnetic torque, is exported to detect the inter turn fault. The technique used and the obtained results show clearly the possibility of extracting signatures to detect and locate faults.
Keywords: Finite element method (FEM); Permanent Magnet Synchronous
Motor (PMSM); winding short-circuit; Fast Fourier Transform (FFT) analysis.
Human Activity Recognition based on Mobile Phone Sensor Data using
Stacking machine learning classifiers
by Mahsa Soufineyestani, Hedieh Sajedi, Vali Tawosi
Abstract: Human activity recognition, which is one of the rapidly growing fields of research, aims to determine which activity is performed by individuals. It has plenty of real-world applications such as health monitoring, abnormal behavior detection, and sports. Therefore, this study focuses on distinguishing and classifying human activities by applying statistical features and using stacking learning methods with the aim of improving the accuracy and precision of the classification. Two main steps are considered at the presented approach. At first, features are extracted from raw sensor data and 26 subsets of the complete feature set are determined and tested to see which subset results in a higher precision classification. Then a feature selection technique based on the Genetic Algorithm is applied to the extracted features to observe if it can improve the results. Comparative results between stacking and single classifiers showed that stacking models have advantages in increasing classification accuracy, especially in the case of those activities that are difficult to distinguish by single classifiers, such as climbing stairs and walking.
Keywords: Human Activity Recognition; Classification; Feature Extraction; Stacking Learning Algorithm.
Home energy management based on Plug-in Electric Vehicle power control in a residential smart grid
by Siwar Khemakhem
Abstract: With the great tendency of Plug in Electric Vehicles (PEVs) development, the integration of this interesting and flexible electric load on the smart grid emerging from energy exchanging between PEV and power grid have drawn a great attention worldwide. In this paper, taking into account smart grid technology, an optimal PEV charge/discharge power management algorithm in residential place is proposed. The purpose of this control strategy is to ensure the energy flow exchanging between PEV and smart home to improve the energy efficiency and to achieve a flattened power load curve. The main contribution of this strategy is to determine the optimal power of the PEV connected at home. The home-to-vehicle operation (H2V) determines the reference power to be absorbed during the off-peak hours and vehicle-to-home operation (V2H) explains the reference power to be injected to home in order to elevate peaks during rush hours. Therefore, this new concept poses important challenges to achieve the smoothness for the power load curve and the stability and the security of smart grid. Simulation results prove the performance of this control approach.
Keywords: PEV; control strategy; vehicle-to-home (V2H); home-to-vehicle (H2V),smoothness.
Using of an Adaptive Fuzzy Speed Controller for Field Oriented Controlled an IPMSM Drives Supplied by a Voltage Source Inverter
by Salma Charmi, Bassem El Badsi, Abderrazak Yangui
Abstract: This paper is aimed to the investigation and performance of the field oriented controlled (FOC) an interior permanent magnet synchronous machine (IPMSM) drives, supplied by a two-level three-phase inverter, using a fuzzy logic toolbox in speed loop system. In recent years, the proposed speed controller has gained more attention for many scientific researchers owing to its several benefits such as more robustness and high precision against reference mechanical speed and/or load torque disturbance effects. Basically, the proposed speed controller consists in adjusting of the proportional and integral gains of the speed controller, in accordance with the speed error and its first time derivative. A simulation-based comparative study between the performance and effectiveness of the developed fuzzy speed controller and the classical one has highlighted the superiority of the adaptive fuzzy control system, in terms of dynamic response, robustness and steady-state error, especially, when there exist motor parameter uncertainties and unexpected load torque changes occur, under forward and reversal motoring.
Keywords: field oriented control strategy; interior permanent magnet synchronous machine; fuzzy logic controller; steady-state error; load torque variations.
Special Issue on: SSD 2018 Knowledge Processing and Information Quality
Petri Net Based Approach for Scheduling and Rescheduling in Multi-Site Manufacturing System
by MAAOUI AMEL
Abstract: In Multi-Site Manufacturing Systems, operations are subject to a great number of unexpected interruptions that may withdraw the original manufacture schedules. In these cases, rescheduling is prevalent to preserve the performance of the whole system. There are many types of disturbances that can upset the schedule, including; machine failures, processing time delays, clients rush orders, quality problems, transportation failures, etc. In this work, we consider only the machine breakdown as disruption that require schedule updating (rescheduling). For that, we develop a Petri Nets (PN) scheduling and rescheduling approach to model and simulate the Multi-Site Manufacturing System while minimizing the makespan. Finally, we give a case study to analyze the efficiency of the proposed model.
Keywords: Keywords: rescheduling; job shop scheduling; Petri nets; Multi-site Manufacturing systems.
Special Issue on: SSD 2018 Knowledge Processing and Information Quality
Fusion of Information and Analytics: A Discussion on Potential Methods to Cope with Uncertainty in Complex Environments (Big Data and IoT)
by Eloi Bosse, Basel Solaiman
Abstract: Information overload and complexity are core problems to both military and civilian complex systems, networks and organizations of today. The advances in networking capabilities have created the conditions of complexity by enabling richer, real-time interactions between and among individuals, objects, systems and organizations. On the other hand, what is considered as problems for system designers is technological opportunities for deciders, for instance, the Internet of Things (IoT) and the Big Data. Fusion of Information and Analytics Technologies (FIAT) are key enablers to bring these benefits to deciders. The design of current and future decision support systems (real-time, online, and near real-time) make use of FIAT to support prognosis, diagnosis, and prescriptive tasks for systems. Hundreds of methods and technologies exist, and several books have been dedicated to either analytics or information fusion so far. However, very few have discussed the methodological aspects and the need of an integrating framework or a computational model for an integration of these techniques coming from multiple disciplines. This paper presents a discussion of potential integrating frameworks as well as a description of elements that will support the development of a computational model to evolve FIAT-based systems capable of meeting the challenges of complex environments such as in Big Data and IoT.
Keywords: Information fusion; analytics; decision support; situation analysis; complex systems: Big Data: IoT.