Embedded Files
(Volume: 4, Issue: 4
Image courtesy: www.freepik.com
Solid-State Sensors for pH sensing
pH sensing, which indicates the measuring of potential of Hydrogen in a fluid to determine its acidity or alkalinity levels, is an extremely essential task of various applications, such as water quality assessment, biological monitoring, waste water treatment, diagnostic procedures, chemical process control and various other environmental, industrial or agricultural applications. A general pH sensing setup commonly employs a reference electrode and a working electrode that optionally include a conductive base and/or a sensing membrane dipped in the electrolyte test solution. It is this working electrode that interacts with the H+ ions in the test solution, causing a difference in potential between the reference electrode and itself. This potential difference is then measured and correlated to a pH value that the fluid truly owns. Field-Effect Transistors (FETs), like Metal-Oxide-Semiconductor FET (MOSFETs), Ion-Sensitive FET (ISFET) and Extended- Gate FETs (EG-FETs) are a few solid-state sensors, which provide the pH-value in proportion to their gate voltage or drain current, which changes with the working electrode potential. There are ample researches on pH sensors in the past with two main intentions and they are: (i) Finding a sensing membrane, which is capable of effectively absorbing or reacting with H+ ions without complete dissolution and (ii) Finding a solid-state device that rapidly responds with low-drift characteristics to convert the surface potential into pH levels. The research by Uvanesh Kasiviswanathan, Lucky Agarwal, Chandan Kumar, Ajay Kumar Dwivedi and Shweta Tripathi too has similar intentions. Actually, these researchers were the first to develop an extended gate with SnS2 (Tin disulfide) thin film for pH sensing. As per their research in Materials Letters, vol.391, Elsevier, the extended gate was connected to a commercial MOSFET IC, the gate voltage of which was controlled by the extended gate potential. In fact, the latter commercial IC only provided the gate voltage or drain current changes for the pH measurement. Secluding the commercial IC from the test solution with an extended gate, the researchers have actually ensured longer reusability of pH sensors across different testing solutions with pH ranging between 1 and 11. Despite the fact that the researchers have achieved a pH sensitivity of ∼41.89mV/pH with SnS2-based extended gate, similar materials with high-stability, temperature-insensitivity and high-H+ reactive power with less oxidizing effect are yet to be uncovered for pH sensing, mandating further research in future.
Machine Learning to Aid Earlier CAD/CVD Detection
Fatty plaque deposits on the coronary artery, which supplies oxygen-rich blood to the heart muscle, results in Coronary Artery Diseases (CADs). Though CAD can never be fatal always, it sets the onset of numerous life-threatening Cardiovascular Diseases (CVDs). In fact, the World Health Organization emphasizes that “It is important to detect cardiovascular disease as early as possible so that management with counselling and medicines can begin”. So, if early detection is needed, Machine Learning (ML) approaches might seem to be a better choice. The prime reason is that the ML approaches, when trained with massive health records or imaging files of patients, only enhance earlier disease detection, when compared to the time-consuming, manual interpretations by an expert. There are an umpteen number of ongoing researches, which tend to apply machine learning approaches for diagnosing cardiovascular diseases. However, not all the ML systems provide an accurate diagnosis, as their training might have been confined to a specific demographic or population-based conditions, varying precisions of the data-acquiring equipment and many other biological or non-biological factors. This is clearly-depicted by N Kaushik, Sumukh G, Mridul G, Ankush Ramesh and Venu Gopal BT in their study comparing several classifiers, like Support Vector Machines, Artificial Neural Networks, Random Forests, Naïve Bayes classifier, k-Nearest Neighbor classifiers and Decision Trees. “A comprehensive approach, considering both population-level dynamics and biological factors, is indispensable to mitigate the escalating CVD epidemic among diverse populations”, the researchers state while presenting their article in the 5th IEEE International Conference for Emerging Technology (INCET). The same goal can be set by the future ML or Artificial Intelligence (AI)-based healthcare researchers as well to reduce or eradicate the globally-dreadful disease with earlier disease detection systems.
Image courtesy: www.vecteezy.com
Empowering Flexibility in Manufacturing
The consumer demands might alter at any time that the manufacturing systems require to be flexible enough to satisfy their needs. This flexibility can be achieved, only if the manufacturing systems are automated and controlled in any of its operational phases, such as inventory control, planning, scheduling, material handling, load balancing, decision making or task rescheduling. The name given to such a system is known as Flexible Manufacturing System (FMS), which specifically aims at improved production in less cost and/or lead time. Implementing FMS reaps several benefits as: (i) Optimal usage of resources and tools; (ii) Proper job scheduling with less machine idle-times; (iii) Parallel and effective material handling or load balancing with reduced system wear and tear;(iv) Suitable decision making on job production or rescheduling and (v) Decreased lead time. But, how does the FMS achieve these advantages? It is by incorporating mathematical and simulated models, optimization techniques, fuzzy controls, Petri nets, artificial intelligence approaches or their hybridizations. Kanchan D. Ganvir and Rakesh Kumar Jain from Himalyan University, Itanagar Chimpu, Arunachal Pradesh, India have reported several modern approaches for successful FMS realization in the Journal of Advanced Manufacturing System, vol. 21(4), World Scientific Publishing. The researchers have pointed out that majority of the past researches on FMS were focussed on scheduling. Hence, FMS issues pertaining to material handling and machine loading are required to be further explored in the forthcoming years. Additionally, the researchers have also stated that the novel approaches to solve the FMS issues should not blindly aim to reduce cost or manufacturing time. Instead, the carbon emissions and the employee risks should also be addressed to achieve eco-friendly and amicable flexible manufacturing. As flexible manufacturing is a rapidly expanding field, the industrial engineers and their associated computer or business personnel have wider opportunities to integrate digital twins, artificial intelligence, additive manufacturing and other industry 4.0 technologies, in view of adapting and augmenting production based on the customer’s changing needs. Image courtesy: www.freepik.com
User Preference for Network Selection from Heterogeneous Wireless Networks
Seamless network connection and incessant communication is of paramount importance in this world of smart devices and Heterogeneous Wireless Networks (HWNs). Generally, the HWNs allow the users to access the mobile networks, Wireless Metropolitan Area Networks (WMANs) or Wireless Local Area Networks (WLANs) in an interchangeable manner to achieve uninterrupted communication. However, it is not a single user, who is accessing only one of the networks in HWNs. So, an optimal choice on the network becomes necessary for the user to effectively transfer his/her data with limited handovers. Network selection has become a key research area with the gradual cellular network transformation from 2G to the New Radio (NR). Notably, the user preferences to switch between networks have a crucial role in achieving an adaptive, smarter and contented communication system. S. Dinesh Krishnan, A. Daniel, S. Ayyasamy, Balamurugan Balusamy, Shitharth Selvarajan, Taher Al-Shehari and Nasser A. Alsadhan have considered this fact in their article in Scientific Reports, Vol.15. The researchers have actually enhanced the network selection process with Multi-Attribute Decision Making (MADM) approach, grounded on fuzzy logic and Fuzzy Analytic Hierarchy Process (FAHP). Rather than considering the network attributes alone, the researchers have used the user preferences to choose a network for carrying out conversational, interactive or streaming services, thereby limiting unnecessary handovers. However, real-time adaptability and the issues related to scalability, computational overhead and energy usage still persisted in their network selection paradigm too. Hence, the researchers recommend future investigations to centre around solving these issues with machine learning techniques, with special consideration on user context and network security as well.
Image courtesy: www.freepik.com
Speed Control in Permanent Magnet Synchronous Machines
Most real-time as well as industrial applications, like robotics, renewable energy systems or electric vehicles demand motors, which impart high power density, dynamic response and efficiency of operation. Permanent Magnet Synchronous Machines (PMSMs) are better known to own these characteristics with additional benefits as less maintenance and noise-free or vibrationless functioning. However, nonlinearities and complex dynamics do always exist in real-time operations, hindering speed control in PMSMs. As speed control is vital in any electrical application for its stable operation and/or minimized energy usage or losses, various speed control techniques have been put forth by the past researches. The vector/ scalar control, Direct Torque Control (DTC), Proportional–Integral–Derivative controller (PID), Fuzzy Logic Control (FLC), Optimization-based Control are to name a few. Satishkumar D, Jayanth D, Harsith Gowda M, Akash Patil and Jayanth C K from New Horizon College of Engineering, Bangalore, India have taken the research to a bit higher-level by optimizing the FLC parameters using Particle Swarm Optimization (PSO). “The specific goals are to increase transient and steady-state performance, reduce overshoot, raise energy efficiency, and provide speedy and accurate speed response”, the researchers state, as they mimic the human-level, fuzzified decision-making with optimized parameters for achieving PMSM speed control. They specially attributed their findings towards electric vehicle applications, which are generally aimed at net-zero emissions. However, novel optimization algorithms, their hybridizations and integration with intelligent predictive models or smart environments are still sought in future to increase the PMSM performance. Image courtesy: www.vecteezy.com
Disclaimer
The discussions in 'Trending Research' are purely based on the already-published articles received from the respective authors for the easy portrayal of trending research themes to young researchers
This magazine does not accept or publish any kind of novel research and it does not create copyright infringements to the discussed journal articles
Page updated
Google Sites
Report abuse