In this study the performance of a built Dye-Sensitized Solar Cell-Thermoelectric Generator system is investigated using conventional statistical analysis. The two main questions are (i) Whether there is a significant difference between the performance of the Dye-Sensitized Solar Cell and the performance of the Dye-Sensitized Solar Cell-Thermoelectric Generator hybrid system. (ii) Whether there is a significant relationship between the surface temperature and the performance of the hybrid system. If there is any significant relationship, how many temperature-difference it occurs. In order to achieve the above-mentioned aim, in case of the (i) independent samples T-test has been used and in case of (ii) Analysis of variance analysis have been used. According to the results, there is a significant difference between the performance of the Dye-Sensitized Solar Cell and the performance of the Dye-Sensitized Solar Cell-Thermoelectric Generator hybrid system. Because of the violation of the normality and the violation of the homogeneity, Welch-probe was applied and according to the results, there is a significant relationship between the performance of the hybrid system and the surface temperature of the solar cell. Bonferroni Post-Hoc test was applied, and the results showed that there is significant different between the maximum power output values of the hybrid system for each temperature.
This article presents a solution to obtain adequate ECG signals using low-cost devices. We believe that a pilot ECG examination in the home environment can detect certain forms of arrhythmias. If arrhythmias are diagnosed and treated in time, a person can completely avoid serious health problems. Performing an ECG examination at home is hindered by the high cost of professional equipment. That’s why some low-cost devices capable of recording an ECG signal have drawn our attention. The results of our research showed that an affordable ECG can be recorded at home using a 1-lead ECG machine.
Object localization is widely used today in various areas such as transport, and industry, but also in the field of wireless sensor networks (WSN). In the article, the authors focus on the possible use of WSN networks, which are made up of sensors capable of measuring certain environmental parameters (pressure, temperature, CO 2 value, etc.) in connection with their location. The article describes individual methods used in WSN networks and creates an overview of them. At the same time, it describes the proposal for the practical creation of a WSN network using a set of sensors from the company LIBELIUM and suggests the possibilities of data analysis of data obtained from sensors that detect Bluetooth signals and Wi-Fi signals. The aim of the article is to describe the solved problem associated with the creation and testing of a custom system for use in applications related to the protection of the health of persons that can be located in a given monitored environment.
First building comfort, energy and environmental design optimization (BECEDO) research efforts are dated back to the early 70’s 1 , however most in-depth analysis is carried out in non-architectural fields (IT, mathematics and operation technologies). Among building design optimization studies, most attention is focussed on the application of active (mechanical) design variables (HVAC systems), renewable energy harvesting and supply system variables and the combination of active and passive (architectural) design input variables 2 . Regarding the passive design variables, which should be determined for an optimal comfort-energy, as well as environmental impact balance performance, only the numerically easy to beparametrized design variables are taken into consideration, such as opaque and transparent envelope structures and materials, e.g., the thicknesses and thermal properties of the insulation and walls, as well as wall-window ratios (WWR), orientation (ORI), materials, structures (STR) and shading for instance 3 . Though building shape has significant impact on building operation cost 4 , i.e., up to 60-80% energy conservation 56 and up to 80% LCA savings are possible, the investigations dealing with building geometry as a design variable (BGDV) in the BECEDO process is still in its infancy 7 . Estimations 8 predict 60-70% energy consumptions reduction in HVAC and artificial lighting system improvements and up to 20% savings by using intelligent automation systems, however the energy saving potential of optimized space organization and complete building shape design is still missing. Another issue evolves after analysis of the existing BECEDO literature: the stochastic behaviour of the most most frequently applied evolutionary technique (generic algorithms GA) randomly moves in the search space and only near optimum solutions delivers.
Magnetometer-based localization is a challenging task both in indoor and outdoor applications since their reliability depends on the environmental characteristics. The obtained geomagnetic field enables the estimation of both orientation and position if proper calibration and data fusion with inertial measurement units (IMUs), GPS and radio modules are executed. However, the distortions by hard and soft iron effects of both metallic objects and building structures requires additional data processing steps to obtain usable measurement data. This paper addresses the applicability of magnetometers for indoor mobile robot localization purposes. A measurement method is elaborated, which obtains the metallic objects-related disturbance characteristics with the help of both a robotic arm-based setup, which simulates the movement of the mobile robot, and data acquisition steps for the calculation of induced effects of disturbing objects. The experimental setup is applied to investigate four scenarios, namely, measurements with one simpler disturbing object either on the right or left side of the robot, with simpler disturbing objects on both sides, and with one complex object on one side of the robot. To extract the effect of the objects, the measurements collected with no additional object are subtracted from the measurements in the case of each scenario. The obtained measurement results clearly validate that incorporating the measurements of the undisturbed scenario enables the obtainment of the effect of disturbing objects. These results form the basis for the development of intelligent fusion algorithms of magneto-inertial sensors for mobile robot localization tasks.
the article is focused on the parameterization of the selected 2D visual camera system to test the principles of recognition, processing, and aggregation of collected image data on the application example of an automated workplace of the "Pick & Place" type. The proposed approach and the built-in functions demonstrated the design's efficiency and functionality. In addition, experimental tests confirmed the finding of the searched object with the most intensity and its reliable grasping by the collaborative robotic arm.
There has been a ton of study on the use of machine learning for speech processing applications, particularly voice recognition, over the past few decades. However, in recent years, research has concentrated on using deep learning for applications that relate to voice. In several applications, including speech, this new branch of machine learning has outperformed others, making it an extremely appealing topic for research. This paper presents a comprehensive review of the studies on deep learning for speech applications that have been carried out since 2017 when deep learning emerged as a new field of machine learning. This evaluation provides a rigorous statistical analysis that was completed by removing certain data from 184 papers published between 2017 and 2022. The findings presented in this paper shed light on the patterns of research in this field and concentrate attention on fresh research areas.
There is an increased number of Driver Assistance systems on the field, therefore the need of having naturalistic behavior of these functions is increasing. In our work the trajectory planning task is analyzed. A clothoid-based local trajectory planning algorithm is proposed, which relies on node points within the look ahead distance. The node point distances were optimized to yield a global trajectory which is close to the human drivers’ path. Real driving data was used as the optimization reference. As a result of the optimization, we were able to determine a characteristic node point distance set which fits all drivers. We have also shown that three node points within a look ahead distance of 140 m are sufficient to describe the drivers’ trajectory. Later this result will serve as a basis to build a driver model which calculates the lateral coordinates of the node points.
Atherosclerosis is the most frequent cause of cardiovascular disease. Almost all cardiovascular diseases have many common risk factors that are associated with the onset of the disease, its development, complications, or premature death. The aim of this research was to show, how various machine learning and statistical methods can be used for modeling the influence of factors associated with atherosclerosis in a usable and understandable way for medical doctors. The article focuses on examining the entire available dataset, but also groups of patients and individuals. Several approaches have been used. Filtering methods, namely Pearson correlation coefficient, chi-square test, Relief and permutation method, were used to examine the whole dataset and specific groups of patients. Methods for explaining machine learning models, namely LIME and Shapley values, were used to examine the impact of risk factors on individuals. The research also examined the relationships between attributes using Bayesian networks.
Analyzing the tremendous amount of data that the banking industry generates has become a critical step for financial institutions. Firms that fail to leverage the huge stream of data they are able to collect, risk falling behind and missing out on the full benefits that this in-depth and extensive analysis can bring. From better understanding customer behavior and needs to making accurate and profitable decisions, data mining has proven its usefulness and enormous benefits and is now widely recognized as a key component for banks. This study aims to highlight the importance and usefulness of implementing data science algorithms and methods to study customer behavior in the banking sector. The research is based on a case study of a bank in Tunisia where the main focus behind is to assess the risk and creditworthiness of a loan applicant and to eventually determine whether or not these customers are worth keeping.
In-circuit (ICT) and in-board (IBT) testers are commonly used in industrial environments to support electronics manufacturing. Each of these performs a parametric measurement that checks the voltage values at each measured point and compares them to a value in a database. These devices perform measurements based on the method of nodal potentials.The fact that the measured values can be set, parameterized and programmed makes them very convenient to use, making them ideal for performing a fault detection and analysis measurement in support of production. Programming the equipment requires a high level of expertise and knowledge of electronic circuits. To replace these measuring and testing devices, we provide in this article a recommendation, a cost-effective solution.
The lower latency and bandwidth of 5G will cause an increase in the diversity and volume of data, Internet of Things (IoT) devices, and innovation, which can create several challenges. As such, it results in the urge to understand better the potential and risks of 5G networks. This paper’s aim extends beyond the examination of 5G network potential. At first, it gives a categorization and description of the 5G ecosystem elements. Then, in line with the main focus of our work, the potential of cybersecurity assurance in 5G networks is examined. We suggest that organizations embrace security by design and security in process mindset and build trust throughout the ecosystem. Additionally, we shed light on the interdisciplinary scheme for better network equipment security following standard guidelines–NESAS (Network Equipment Security Assurance Scheme). The implementation of 5G networks should ensure security across policy, technology, and standards. Vendors should be positioned to address and lead critical priorities and actions that will guide countries through this transition.
The dynamic development and change of the world also result in the modernization of teacher-student relationships. In the training, educational processes must be planned with innovative strategies that meet the increasingly changing needs of students, which today go beyond the traditional classroom and methodological frameworks. In this article, we present a technological solution to support the renewal efforts in the education world that, on the one hand, takes advantage of the benefits of cloud-based services, and on the other hand, ensures that students can acquire the material of a given course in a unified, stable environment. The isolated environment offered by the virtual space guarantees private work sessions, and multi-project activities can also be handled more easily and uniformly. The analysis of the teaching program presented below provides insight into the practical and appropriate solutions offered by the containerization process. In addition to these, the topic will also cover the widely used and popular Docker application that supports the implementation of the process, as well as the NodeJS environment closely related to the project.
With the rapid development and frequent application of information and communication technology (ICT) in different industrial sectors, the production and manufacturing process relies more on ICT. ICT significantly increases the quality of service and productivity while it comes together with information security risks. As an essential part of a nation and economy, agriculture is provided with many ICT applications. ICT greatly contributes to the weather climate forecast, pest or disease prediction and so on to improve crop yields and food safety. The revolution of the fourth industry accelerates the application of ICT in agriculture and the development of smart agriculture. Smart agriculture is highly based on ICT to improve productivity and reduce waste, loss, and cost, such as big data, sensors, the Internet of things (IoT), artificial intelligence (AI), cloud computing, and so on. All these changes and applications of ICT bring together information security risks. Based on the literature review and content analysis as research methodology, this paper concluded that information security risks are serious and inevitable in agriculture or food security. Therefore the need for information security risks management is significant in agriculture and food security.
This paper explains the basic methods of troubleshooting small industrial and residential scale offgrid systems. The article deals with maintenance and diagnostic questions of industrial equipment, which are powered by an off grid system. Eventually conclusions were drawn about the feasibility and requirements of measuring, monitoring and diagnostics algorithms.
The UVA/Padova Type 1 Diabetes Simulator is a widely applied tool to test control algorithms among diabetes researchers. The academic version is implemented in Matlab; while Matlab is very popular in the academic field, Python is the most popular programming language. We developed an application programming interface (API) for the simulator in Python, and a representational state transfer (REST) API for additional extension route with a Python reference implementation for easier usage. The interface is designed with the specific purpose of testing control algorithms, thus it maps the functionalities that are needed to implement closed-loop control and virtual patient simulation. The developed API is tested for different simulation scenarios, showcasing identical results between the API and the programming interface of the original simulator. Additional information and the source code can be found in https://github.com/NeuroDiab/UVAPadovaAPI.
The theory of capsule networks and the dynamic routing mechanism for capsules was introduced by Geoffrey Hinton and his research team. In this new approach, they tried to solve typical problems of classical convolutional neural networks. For example, that the efficiency of neural networks degrades when a geometric transformation is applied on the input image, or when the data is far away from the training dataset. It became clear early on that capsule networks are state-of-the-art solutions for visual data classification tasks. For other tasks their use is less common and in many cases difficult to apply. For example image segmentation or object detection and localization. The efficiency of the capsule networks theory in the field of pointcloud processing is also an open question. In this work we investigated the pointcloud reconstruction capability of capsule networks. In this approach, three different complexity autoencoder networks was selected. We created a decoder network based on capsules theory, which was fitted to the existing autoencoder networks. The efficiency of the networks was tested using four different datasets. As a result of our work, we show the effectiveness of capsule networks in the field of pointcloud reconstruction compared with the selected autoencoder networks.
Complete database design, implementation, optimization as well as testing is obtained in order to collect and fuse all the morphological, morphometrical, genomic metadata during the whole, automatic, digital pathology workflow without any human interaction. The subsequent purpose of the fused database is to evaluate texture based (external) and biochemical (internal) features of tumor, cancer collectively. Hence, finding correlation between them might make us able to predict and make diagnose at an earlier stage.
Resilience, as a characteristic of a business organization to prepare, adapt, withstand and survive economic distress periods is essential for a continuous operation and also contributes to a balanced macroeconomic state of regions. It is not evident whether being resilient once or several times during the lifetime of a business organization in which aspect can contribute to long-term survival and success. In this study annual revenue information – gathered from more than 26,000 Hungarian companies from processing industry – is utilized in order to construct an empirical resilience metric and to investigate long-term behaviour of companies that had successfully withstood turbulent economic periods. Based on matched-pair analysis our results showed that resilient organizations despite being successful in the short-run statistically tend to lag behind in terms of performance in the long-run.
Dementia is one of the most common diseases among elderly people, which affects patients’ motor skills, memory, and social abilities. Its early detection can be beneficial in controlling the progression of the disease. This paper concerns the use of Alzheimer’s Dementia Recognition through Spontaneous Speech in both 2020 and 2021, as well as external audio data in DementiaBank, to classify Alzheimer’s disease. We suggest a solution using an attention mechanism on the mel-spectrogram audio representation. The perspective of using model pre-training and model fine-tuning to improve classification accuracy is also covered in the research experiments. The created models outperform the 2020 baseline by 22.91% in terms of accuracy and by 11.26% in terms of accuracy score for the 2021 dataset.
