Embedded Files
(Volume: 2, Issue: 5)
Control IC Heat Dissipation with Optimally-Designed Microchannel Heat Sinks
Smart electronic devices are a great craze in the modern world and they are nothing without Integrated Chips (IC). By IC, the more will be the transistors, resistors, capacitors and other electronic components being embedded on to a single chip, ending up in more compact electronic devices with lower power consumption and cost. However, all these densely-interconnected components will undoubtedly dissipate heat that either ruin the expected performance or even damage the IC. So, how the temperature rise in the ICs be mitigated? By employing heat sinks... Heat sinks are heat exchangers placed on the electronic components, busily engaged in transferring heat from the components to the surrounding fluid medium through conduction, convection and radiation to achieve cooling. However, larger heat sink surface area with lower impedance is mandate for providing incessant cooling to the IC components. The microchannel heat sink serves this purpose with its numerous, finned miniature channels, which generally pass liquid coolants through it. However, to yield maximum IC performance with increased heat transfer coefficient, few questions emerge in our minds and they are: (i) What can be the size and the number of fins or the channels? (ii) What can be the material or the coolant used? Hence, research on optimally designing the microchannel heat sinks becomes necessary. One such research is the numerical simulation of microchannel heat sink, authored by G. Anjaneya, S. Sunil, Shrishail Kakkeri, Mahantesh M. Math, M. N. Vaibhav and C. Solaimuthu from Karnataka, India, in International Journal on Interactive Design and Manufacturing, Springer. Utilizing ANSYS Fluent, a commercial Computational Fluid Dynamics (CFD) software, the authors have examined the effect of hydraulic diameter, surface area and number of channels on the heat dissipation ability of the microchannel heat sink. As per the authors, “the surface temperature decreased with an increase in cross-sectional area, number of channels and hydraulic diameter from 361 K for a simple rectangular model to 332 K, 326 K, and 324 K for a 5-channel fin, 8-channel fin and 11-channel fin model, respectively”. Since the microchannel heat exchangers play a vital role in the design of Heating, Ventilation and Air Conditioning (HVAC) systems, aircrafts, microelectronic devices, renewable energy systems, rockets, smart appliances and similar other electronic systems, there is a boundless scope for the future researchers involving in heat sink design optimization.
Image Courtesy: www.freepik.com
Intelligent control of motor speed and rapid charging in 4-wheeler EVs
The desire to find new means of transportation, which can simultaneously cut down greenhouse gas emissions and reduce non-renewable resource usage for achieving a pollution-free and eco-friendly environment, has resulted in the invention of Electric Vehicles (EV). As the name suggests, it operates on electricity using key components like, electric motors, battery pack, charging port and related power electronics as well as control systems, rather than burning petrol, diesel or gas as in Internal Combustion (IC) engines. But, can the EVs really replace the IC engine- based vehicles? The answer is uncertain because there are few shortcomings with the EVs too. To name a few: (i) limited driving range due to the limited battery storage; (ii) longer charging period and less availability of charging stations and (iii) reduced torque and acceleration than their IC engine counterparts. Since EVs are at the verge of development, research on improving battery life as well as storage using better cost-efficient materials for battery manufacturing and adaptive controlling of vehicle dynamics and motor parameters to enhance motor speed is essential. A. C. Mulla et al. from Pune, Maharashtra, have shown importance in designing and simulating DC fast chargers and adaptive intelligent controllers of motor speed for four-wheeler electric cars. Their research article in the 2023 IEEE 8th International Conference for Convergence in Technology (I2CT) was specially intended on tracking the driver reference speed with the actual speed of motor in various road conditions (for instance, on-road, off-road and sporty), by considering vehicle dynamics like, aerodynamic drag, rolling resistance, uphill resistance and acceleration. Since four-wheeler utilization in the future is never going to dwindle, research to design EVs with better operability in terms of speed, energy storage and emission control is highly sought.
Image Courtesy: www.freepik.com
Transformation to I4.0 Enforces Risk Assessment and Management
A ‘smart’ industrial manufacturing sector with increased productivity and efficiency is what the notion Industry 4.0 (I4.0) is aimed at. With this industrial revolution, the man-to-machine communication is drastically reduced, enabling machine-to-machine communication through the incorporation of information technologies like, Artificial Intelligence (AI), Internet of Things (IoT), cloud computing, big data analytics, cyber-physical systems and other progressing technologies. A development always has its associated risks. Here too, the productivity improvement is attained at the cost of ecological, social and related security as well as industrial risks. Bhaveshkumar Nandanram Pasi, Subhash K. Mahajan and Santosh B. Rane in their article in Journal of Science and Technology Policy Management, Emerald Publishing, vol. 14, issue: 3 have identified that economic risks, ecological risks, social risks, technical risks, information technology risks and legal as well as political risks as the major risks, which the industries have to face upon I4.0 transformation. Employing the Decision-Making Trial and Evaluation Laboratory (DEMATEL) approach for its ability to identify the relative importance among the various risk factors with minimum data, the researchers have framed a I4.0 risk management procedure for the managers in manufacturing industries. As per the framework, the potential I4.0 risks has to be identified, assessed and prioritized for perfect planning of their control. Prioritizing the economic risks over the others, the researchers impose that research in future is still needed. This is because their study was made with limited number of experts and participants with time-varying opinions and knowledge. There is yet another direction for future research in this context, wherein the researchers can contribute in automating the risk assessment and management procedures.
Image Courtesy: www.unsplash.com
Automatic roughness prediction for flexible pavement management
Nowadays, flexible pavements are more prevalent to be seen in highways and streets. The layered construction, which offers improved strength, flexibility and durability to diverse traffic load and weather conditions because of stress distribution across different layers, might be cited as the reason. However, the deflection property of these pavements is also accountable for the rapid wear down in the form of rutting or alligator cracks, affecting the ride quality. Hence, the pavement’s roughness needs to be monitored for ensuring its strength and the proper distribution of the load’s stress, so as to guarantee a better ride. There is a measure for estimating the pavement’s roughness in ‘inches per mile’ and it is the International Roughness Index (IRI). The higher the IRI, the higher is the roughness of the pavement and it indicates that the pavement will be rapidly prone to wearing defects. Normally, the flexible pavements need repair more frequently than the rigid ones with huge renovation cost. Hence, regular IRI measurements are needed to assess the severity of repair beforehand and to cut down any additional expenses. But, manual estimation of IRI on pavements with heavy traffic volume is arduous, prompting the utilization of automatic IRI prediction systems. Four researchers, two each from South Korea and Egypt, have explored the effectiveness of novel machine learning techniques in predicting the IRI. Employing the Long-Term Pavement Performance (LTPP) datasets of pavement age, initial IRI, alligator, longitudinal and transverse cracks, standard deviation of rutting and subgrade plasticity index to model the IRI prediction, the researchers recommend future research in ensuring robustness of prediction across different datasets and novel optimization models. The present road network experiences heavy traffic volume already. Hence, automatic IRI prediction turns out to be the only solution to monitor the pavement condition in the abruptly-changing weather conditions. As an effect, the responsibility to impart accurate and robust IRI prediction systems falls on future researchers…
Image Courtesy: www.freepik.com
Sentiment Detection with Emotion Recognition for Language Processing Code- Mixed Data
Human-Computer Interaction (HCI) through speech has gained popularity nowadays and the key aspect governing it is the Natural Language Processing (NLP). However, in a multilingual society like ours, NLP is greatly challenged by code-mixing. Code-mixing is a term that signifies the usage of two different languages by a person in his/ her speech, while trying to express an opinion. The natural language usually has subtle differences in its semantics and contextual meaning. Hence, it is not at all a surprise to find difficulty in achieving HCI using code-mixed data. The process that can help to some extent in this regard is the sentiment detection, otherwise known as opinion mining. Using sentiment detection, the words or phrases in a speech are searched for an opinion as positive, negative or neutral. But how far the detected opinion is true can be discovered only from the emotions that the words or phrases carry. For instance, a person satisfied with a service will provide a positive opinion with happiness and not by anger or disgust. Hence, sentiment detection and emotion recognition on working together can effectively handle code-mixed NLP. Researchers from Indian Institute of Technology (IIT) Patna and IIT Bombay do support this idea in their publication in Knowledge-Based Systems, Elsevier, while they tried to observe the semantic and contextual meaning from Hinglish (Hindi+ English) data. Annotating emotions like, happiness, sadness, anger, surprise, fear, disgust and others classes on the benchmark SentiMix dataset, the researchers have performed multitask learning to simultaneously model sentiment and emotions in the code-mixed data. The researchers state that NLP of Hinglish data is new to research and confirm that sentiment detection from the data improved with emotion annotations. NLP of code-mixed data plays a vital role in language understanding, while acquiring political news or an individual’s opinion on some hot issue and the customer reviews for a product or during the establishment of HCI, irrespective of the geographical location. Since the language can very often undergo changes and the opinions have major impact on transforming the society, research on sentiment and emotion-based NLP across code-mixed data is highly encouraged.
Image Courtesy: www.freepik.com
Page updated
Google Sites
Report abuse