Technical University of Cluj-Napoca

Policy-gradient-based actor-critic algorithms are amongst the most popular algorithms in the reinforcement learning framework. Their advantage of being able to search for optimal policies using low-variance gradient estimates has made them useful in several real-life applications, such as robotics, power control, and finance. Although general surveys on reinforcement learning techniques already exist, no survey is specifically dedicated to actor-critic algorithms in particular. This paper, therefore, describes the state of the art of actor-critic algorithms, with a focus on methods that can work in an online setting and use function approximation in order to deal with continuous state and action spaces. After starting with a discussion on the concepts of reinforcement learning and the origins of actor-critic algorithms, this paper describes the workings of the natural gradient, which has made its way into many actor-critic algorithms over the past few years. A review of several standard and natural actor-critic algorithms is given, and the paper concludes with an overview of application areas and a discussion on open issues.

In this paper, we investigate the use of decentralized blockchain mechanisms for delivering transparent, secure, reliable, and timely energy flexibility, under the form of adaptation of energy demand profiles of Distributed Energy Prosumers, to all the stakeholders involved in the flexibility markets (Distribution System Operators primarily, retailers, aggregators, etc.). In our approach, a blockchain based distributed ledger stores in a tamper proof manner the energy prosumption information collected from Internet of Things smart metering devices, while self-enforcing smart contracts programmatically define the expected energy flexibility at the level of each prosumer, the associated rewards or penalties, and the rules for balancing the energy demand with the energy production at grid level. Consensus based validation will be used for demand response programs validation and to activate the appropriate financial settlement for the flexibility providers. The approach was validated using a prototype implemented in an Ethereum platform using energy consumption and production traces of several buildings from literature data sets. The results show that our blockchain based distributed demand side management can be used for matching energy demand and production at smart grid level, the demand response signal being followed with high accuracy, while the amount of energy flexibility needed for convergence is reduced.

In the last decade, research on the synthesis and characterization of nanosized ferrites has highly increased and a wide range of new applications for these materials have been identified. The ability to tailor the structure, chemical, optical, magnetic, and electrical properties of ferrites by selecting the synthesis parameters further enhanced their widespread use. The paper reviews the synthesis methods and applications of MFe2O4 (M = Co, Cu, Mn, Ni, Zn) nanoparticles, with emphasis on the advantages and disadvantages of each synthesis route and main applications. Along with the conventional methods like sol-gel, thermal decomposition, combustion, co-precipitation, hydrothermal, and solid-state synthesis, several unconventional methods, like sonochemical, microwave assisted combustion, spray pyrolysis, spray drying, laser pyrolysis, microemulsion, reverse micelle, and biosynthesis, are also presented. MFe2O4 (M = Co, Cu, Mn, Ni, Zn) nanosized ferrites present good magnetic (high coercivity, high anisotropy, high Curie temperature, moderate saturation magnetization), electrical (high electrical resistance, low eddy current losses), mechanical (significant mechanical hardness), and chemical (chemical stability, rich redox chemistry) properties that make them suitable for potential applications in the field of magnetic and dielectric materials, photoluminescence, catalysis, photocatalysis, water decontamination, pigments, corrosion protection, sensors, antimicrobial agents, and biomedicine.

The rapid growth of the Internet of Things (IoT) has accelerated strong interests in the development of low-power wireless sensors. Today, wireless sensors are integrated within IoT systems to gather information in a reliable and practical manner to monitor processes and control activities in areas such as transportation, energy, civil infrastructure, smart buildings, environment monitoring, healthcare, defense, manufacturing, and production. The long-term and self-sustainable operation of these IoT devices must be considered early on when they are designed and implemented. Traditionally, wireless sensors have often been powered by batteries, which, despite allowing low overall system costs, can negatively impact the lifespan and the performance of the entire network they are used in. Energy Harvesting (EH) technology is a promising environment-friendly solution that extends the lifetime of these sensors, and, in some cases completely replaces the use of battery power. In addition, energy harvesting offers economic and practical advantages through the optimal use of energy, and the provisioning of lower network maintenance costs. We review recent advances in energy harvesting techniques for IoT. We demonstrate two energy harvesting techniques using case studies. Finally, we discuss some future research challenges that must be addressed to enable the large-scale deployment of energy harvesting solutions for IoT environments.

Background: Over the last years Deep Learning has shown to yield remarkable results when compared to traditional computer vision algorithms, in a large variety of computer vision applications. The deeplearning models outperformed in both accuracy and processing time. Thus, once a deeplearning models won the Image Net Large Scale Visual Recognition Contest, it proved that this area of research is of great potential. Furthermore, these increases in recognition performance resulted in more applied research and thus, more applications where deeplearning is useful: one of which is defect detection (or visual defect detection). In the last few years, deeplearning models achieved higher and higher accuracy on the complex testing datasets used for benchmarking. This surge in accuracy and usage is also supported (besides swarms of researchers pouring into the race), by incremental breakthroughs in computing hardware: such as more powerful GPUs(Graphical processing units), CPUs(central processing units) and better computing procedures (libraries and frameworks). Aim of the review: To offer a structured and analytical overview(stating both advantages and disadvantages) of the existing popular object detection models that can be re-purposed for defect detection: such as Region based CNNs(Convolutional neural networks), YOLO(You only look once), SSD(single shot detectors) and cascaded architectures. A further brief summary on model compression and acceleration techniques that enabled the portability of deeplearning detection models is included. Key Scientific Concepts of Review: It is of great use for future developments in the manufacturing industry that many of the popular, above mentioned models are easy to re-purpose for defect detection and, thus could really contribute to the overall increase in productivity of this sector. Moreover, in the experiment performed the YOLOv4 model was trained and re-purposed for industrial cable detection in several hours. The computing needs could be fulfilled by a general purpose computer or by a high-performance desktop setup, depending on the specificity of the application. Hence, the barrier of computing shall be somewhat easier to climb for all types of businesses.

Abstract This paper discusses the unsteady flow and heat transfer of a Casson fluid over a moving flat plate with a parallel free stream. The analytic solutions of the system of nonlinear partial differential equations valid for all times in the whole spatial domain are constructed in the series form by a homotopic approach. The influences of the governing parameters on the velocity, temperature, skin friction coefficient, and local Nusselt number are thoroughly investigated. It is revealed that an increase in the dimensionless time decreases the velocity and enhances the temperature. The surface shear stress and surface heat transfer are enhanced by increasing the Casson fluid parameter (β) and Eckert number ( Ec ), respectively. © 2011 Wiley Periodicals, Inc. Heat Trans Asian Res; Published online in Wiley Online Library ( wileyonlinelibrary.com/journal/htj ). DOI 10.1002/htj.20358

The recent changes in climate have increased the importance of environmental monitoring, making it a topical and highly active research area. This field is based on remote sensing and on wireless sensor networks for gathering data about the environment. Recent advancements, such as the vision of the Internet of Things (IoT), the cloud computing model, and cyber-physical systems, provide support for the transmission and management of huge amounts of data regarding the trends observed in environmental parameters. In this context, the current work presents three different IoT-based wireless sensors for environmental and ambient monitoring: one employing User Datagram Protocol (UDP)-based Wi-Fi communication, one communicating through Wi-Fi and Hypertext Transfer Protocol (HTTP), and a third one using Bluetooth Smart. All of the presented systems provide the possibility of recording data at remote locations and of visualizing them from every device with an Internet connection, enabling the monitoring of geographically large areas. The development details of these systems are described, along with the major differences and similarities between them. The feasibility of the three developed systems for implementing monitoring applications, taking into account their energy autonomy, ease of use, solution complexity, and Internet connectivity facility, was analyzed, and revealed that they make good candidates for IoT-based solutions.

The flywheel energy-storage systems (FESSs) are suitable for improving the quality of the electric power delivered by the wind generators and for helping these generators to contribute to the ancillary services. Supervisors must be used for controlling the power flow from a variable-speed wind generator (VSWG) to the power grid or to an isolated load. This paper investigates the control method and the energetic performances of a low-speed FESS with a classical squirrel-cage induction machine in the view of its association to a VSWG. A test bench is developed, and experimental results are presented and discussed

BACKGROUND AND AIMS: The current study aims to evaluate the performance of a new elastographic method (ARFI) in noninvasive fibrosis assessment and to compare it to another validated technology (transient elastography, TE). METHOD: 112 consecutive chronic hepatitis C patients (histologically proven according to the Metavir scoring system: 12.5% F0, 26.6% F1, 16.1% F2, 7.1% F3, 37.5% F4) were prospectively included in this study. They were examined on the same day, using both ARFI (with shear wave velocity--SWV-quantification) and TE (with liver stiffness quantification). RESULTS: SWV is correlated only with fibrosis (r=0.717, p less than 0.0001) and necroinflammatory activity (r=0.328, p=0.014), but not with steatosis (r=0.122, p=0.321). There is a significant increase of SWV in parallel with the increase in the fibrosis stage: 1.079+/-0.150 (F0-F1), 1.504+/-0.895 (F2), 1.520+/-0.575 (F3), 2.552+/-0.782 (F4), p<0.0001, but there is a certain degree of overlap between the consecutive stages F1-F2 (p=0.072), F2-F3 (p=0.965). SWV cut-off values (m/s) that were predictive for each fibrosis stage were: 1.19 (F more or equal to 1), 1.34 (F>or=2), 1.61 (F more or equal to 3) and 2.00 (F4). AUROC for ARFI vs TE were: 0.709 vs 0.902, p=0.006 (F>or=1), 0.851 vs 0.941, p=0.022 (F>or=2), 0.869 vs 0.926, p=0.153 (F>or=3) and 0.911 vs 0.945, p=0.331 (F4). CONCLUSIONS: ARFI allows SWV quantification, in strong correlation with the fibrosis stage. Steatosis does not influence SWV. The maximal performance of the method consists of the prediction in severe fibrosis and cirrhosis. The diagnostic accuracy is strongly comparable to TE only for the prediction of severe fibrosis and cirrhosis, whereas for earlier stages, TE performs better.

Due to the Heisenberg-Gabor uncertainty principle, finite oscillation transients are difficult to localize simultaneously in both time and frequency. Classical estimators, like the short-time Fourier transform or the continuous-wavelet transform optimize either temporal or frequency resolution, or find a suboptimal tradeoff. Here, we introduce a spectral estimator enabling time-frequency super-resolution, called superlet, that uses sets of wavelets with increasingly constrained bandwidth. These are combined geometrically in order to maintain the good temporal resolution of single wavelets and gain frequency resolution in upper bands. The normalization of wavelets in the set facilitates exploration of data with scale-free, fractal nature, containing oscillation packets that are self-similar across frequencies. Superlets perform well on synthetic data and brain signals recorded in humans and rodents, resolving high frequency bursts with excellent precision. Importantly, they can reveal fast transient oscillation events in single trials that may be hidden in the averaged time-frequency spectrum by other methods.

An analysis is carried out to study the steady two-dimensional stagnation-point flow of a micropolar fluid over a shrinking sheet in its own plane. The shrinking velocity and the ambient fluid velocity are assumed to vary linearly with the distance from the stagnation point. The features of the flow characteristics are analyzed and discussed. Different from a stretching sheet, it is found that the solutions for a shrinking sheet are nonunique.

An analysis is made for the steady mixed convection boundary layer flow near the two-dimensional stagnation-point flow of an incompressible viscous fluid over a stretching vertical sheet in its own plane. The stretching velocity and the surface temperature are assumed to vary linearly with the distance from the stagnation-point. Two equal and opposite forces are impulsively applied along the x-axis so that the wall is stretched, keeping the origin fixed in a viscous fluid of constant ambient temperature. The transformed ordinary differential equations are solved numerically for some values of the parameters involved using a very efficient numerical scheme known as the Keller-box method. The features of the flow and heat transfer characteristics are analyzed and discussed in detail. Both cases of assisting and opposing flows are considered. It is observed that, for assisting flow, both the skin friction coefficient and the local Nusselt number increase as the buoyancy parameter increases, while only the local Nusselt number increases but the skin friction coefficient decreases as the Prandtl number increases. For opposing flow, both the skin friction coefficient and the local Nusselt number decrease as the buoyancy parameter increases, but both increase as Pr increases. Comparison with known results is excellent.

Rolling bearing faults are generally slowly progressive; therefore, the development of an effective diagnostic technique could be worth detecting such faults in their incipient phase and preventing complete failure of the motor. The methods proposed in the literature for this purpose are mainly based on measuring and analyzing vibration and current. Here, a novel technique based on the stray flux measurement in different positions around the electrical machine is proposed. The main advantages of this method are due to the simplicity and the flexibility of the custom flux probe with its amplification and filtering stage. The flux probe can be easily positioned on the machines and adapted to a wide range of power levels. This paper also reports an extensive survey on the stray-flux-based fault detection methods for induction motors, prior to introducing a novel sensor/diagnostic scheme.

Environmental education and education for the environment today play an important role toward sustainability. Environmental education provided by higher education institutions has an important impact on training and preparing the future generation for a green society. The purpose of this study is to examine the relationship among perception, attitude, and environmental behavior of the university students enrolled in different specialization fields (engineering electrical, mechanical, and economic). A total of 358 students participated in this survey conducted at the North Center University of Baia Mare. To collect data to measure students’ environmental education, perception, students’ attitudes, and behavior a Likert scale was used. In this study, it was revealed that students receiving academic education are involved in activities regarding environmental protection (volunteer, warning, participation, recycling of materials) using the new product and “greener” alternative energy. As a result of the t-test performed, it was put forward that there was no difference in their level of perception regarding the importance of environmental education. As a result of the correlation analysis, a positive relation was identified between the perception, attitude, and behavior variables.

Recent advances in automatic machine learning (aML) allow solving problems without any human intervention. However, sometimes a human-in-the-loop can be beneficial in solving computationally hard problems. In this paper we provide new experimental insights on how we can improve computational intelligence by complementing it with human intelligence in an interactive machine learning approach (iML). For this purpose, we used the Ant Colony Optimization (ACO) framework, because this fosters multi-agent approaches with human agents in the loop. We propose unification between the human intelligence and interaction skills and the computational power of an artificial system. The ACO framework is used on a case study solving the Traveling Salesman Problem, because of its many practical implications, e.g. in the medical domain. We used ACO due to the fact that it is one of the best algorithms used in many applied intelligence problems. For the evaluation we used gamification, i.e. we implemented a snake-like game called Traveling Snakesman with the MAX–MIN Ant System (MMAS) in the background. We extended the MMAS–Algorithm in a way, that the human can directly interact and influence the ants. This is done by “traveling” with the snake across the graph. Each time the human travels over an ant, the current pheromone value of the edge is multiplied by 5. This manipulation has an impact on the ant’s behavior (the probability that this edge is taken by the ant increases). The results show that the humans performing one tour through the graphs have a significant impact on the shortest path found by the MMAS. Consequently, our experiment demonstrates that in our case human intelligence can positively influence machine intelligence. To the best of our knowledge this is the first study of this kind.

The aim of this study is to gain a deeper understanding of the micromorphology characteristics of thin titanium nitride (TiN) films sputtered on glass substrates by using ion beam sputtering (IBS). For that purpose, TiN samples were deposited onto glass substrates from gas mixtures with different contents of molecular nitrogen and argon atoms. Atomic force microscopy (AFM) was used to characterize the surface microtexture of obtained thin films at high magnification. The detailed analysis of AFM images using Minkowski functionals and fractal analysis reveals a significant effect of the preparation conditions on the surface features with non-monotonic dependences. Presented results suggest that non-classical spatial characteristics of the variability of the surface topography, including fractal dimension, corner frequency, roughness, and feature shape and size, can be tuned having control over the relative flow rates of the gas mixture during film deposition.

The current worldwide energy directives are oriented toward reducing energy consumption and lowering greenhouse gas emissions. The exponential increase in the production of electrified vehicles in the last decade are an important part of meeting global goals on the climate change. However, while no greenhouse gas emissions directly come from the operations of the electrical vehicles, the electrical vehicle production process results in much higher energy consumption and greenhouse gas emissions than in the case of a classical internal combustion vehicle; thus, to reduce the environment impact of electrified vehicles, they should be used for as long as possible. Using only batteries for electric vehicles can lead to a shorter battery life for certain applications, such as in the case of those with many stops and starts but not only in these cases. To increase the lifespan of the batteries, couplings between the batteries and the supercapacitors for the new electrical vehicles in the form of the hybrid energy storage systems seems to be the most appropriate way. For this, there are four different types of converters, including rectifiers, inverters, AC-AC converters, and DC-DC converters. For a hybrid energy storage system to operate consistently, effectively, and safely, an appropriate realistic controller technique must be used; at the moment, a few techniques are being used on the market.

This article explores the potential of Artificial Intelligence (AI) chatbots for creating positive change by supporting customers in the digital realm. Our study, which focuses on the customer and his/her declarative psychological responses to an interaction with a virtual assistant, will fill a gap in the digital marketing research, where little attention has been paid to the impact of Error and Gender, as well as the extent to which Social Presence and Perceived Competence mediate the relationships between Anthropomorphic design cues and Trust. We provide consistent evidence of the significant negative effect of erroneous conversational interfaces on several constructs considered in our conceptual model, such as: perceived competence, trust, as well as positive consumer responses. We also provide support to previous research findings and confirm that people employ a biased thinking across gender and this categorization also influences their acceptance of chatbots taking social roles. The results of an empirical study demonstrated that highly anthropomorphized female chatbots that engage in social behaviors are significantly shaping positive consumer responses, even in the error condition. Moreover, female virtual assistants are much more commonly forgiven when committing errors compared to male chatbots.

This paper presents the analytical, numerical, and experimental results obtained with a double excited synchronous machine (DESM) prototype designed and constructed for an electric vehicle traction system. To obtain a wide speed range, the flux-weakening technique is implemented. Analytical design, 2-D finite element method (FEM) analysis, thermal analysis, and prototype construction of the DESM are discussed, and the performances are assessed with experimental data.

Unmanned aerial vehicles (UAVs) have gained significant attention in recent years. Low-cost platforms using inexpensive sensor payloads have been shown to provide satisfactory flight and navigation capabilities. In this report, we survey vision and control methods that can be applied to low-cost UAVs, and we list some popular inexpensive platforms and application fields where they are useful. We also highlight the sensor suites used where this information is available. We overview, among others, feature detection and tracking, optical flow and visual servoing, low-level stabilization and high-level planning methods. We then list popular low-cost UAVs, selecting mainly quadrotors. We discuss applications, restricting our focus to the field of infrastructure inspection. Finally, as an example, we formulate two use-cases for railway inspection, a less explored application field, and illustrate the usage of the vision and control techniques reviewed by selecting appropriate ones to tackle these use-cases. To select vision methods, we run a thorough set of experimental evaluations.