Sustainable Computing for an AI-Driven Future is a mindset for computing and the environment. It's all about ensuring that computers and AI-based technologies do not harm the planet. The size and power of the Artificial Intelligence systems are increasing. They are used in many areas like healthcare, finance and education. These systems require a tremendous amount of energy to operate and energy has a detrimental impact on the environment. So Sustainable Computing is about creating computers and Artificial Intelligence systems in a way that is not harmful to the environment. The basic premise of Sustainable Computing is to use energy in computers. The old computers are not designed for Ai, so they consume a lot of energy. New computers are being invented that use energy and they have special chips that can assist in this. These chips are a bit similar to the brain of a computer. They are able to do multiple things at once without expending much effort.It is also very significant to make energy utilizing software.
This paper presents a new design of memtranstor emulator using operational transconductance amplifier, voltage differencing differential input buffered amplifier, dual-output second-generation current conveyor (DO-CCII), and three grounded capacitors. The fingerprints of memtranstor have been obtained on LTspice using 180 nm CMOS technology parameters. The pinched hysteresis loops have been obtained for different ranges of frequencies. The inclined pinched hysteresis loops and butterfly responses have been achieved by varying the DC voltage. The transient analysis and Monte Carlo analysis have also been obtained for the proposed memtranstor emulator. The custom layout of proposed memtranstor emulator has been developed. The proposed design of memtranstor emulator has also been verified through the macro models of available ICs on SPICE. Further, the experimental results have been validated through hardware implementation of proposed memtranstor emulator using LM13700, AD830, AD633, and AD844 ICs. The obtained results clearly indicate the workability of proposed design of memtranstor emulator. The performance of proposed memtranstor emulator has also been verified through its application in chaotic circuit and synaptic plasticity.
Skin cancer is among the most frequent and fatal illnesses in the world, but early and correct diagnosis is one of the main challenges because of the complicated visual patterns of skin lesions and the absence of interpretable diagnostic devices. Conventional techniques are largely based on experienced dermatologists, but manual inspection is time-consuming, subjective, and liable to misinterpretation, with high false positive outputs. To address these issues, we present a unique DCNN architecture that employs the Swish activation function to effectively identify complex patterns in the skin lesion dataset. This model has remarkable performance in diagnosing skin cancer, as evidenced by a 98.31% accuracy rate, a 98.12% precision rate, a 98.01% recall rate, and an F1-score of 98.09%. Utilising several localised and global explainable artificial intelligence (XAI) approaches, we evaluate the model's predictions to ensure transparency and reliability in medical contexts. To reconcile the disparity between AI research and its use in healthcare, our findings underscore the necessity of integrating deep learning with explainability. These XAI solutions tackle critical concerns such as inclusiveness, transparency, and error control, providing medical practitioners with a comprehensible and reliable framework for assessing the model's reasoning process. The suggested technique establishes a reliable mechanism to assist physicians in the early and precise identification of skin cancer, while enhancing diagnostic accuracy. Future studies will focus on enhancing the model's computational efficiency and incorporating more datasets to ensure its durability and fairness across various demographic groupings.
The search for improved solar energy conversion efficiency consistently motivates innovation beyond the scope of traditional photovoltaic (PV) technologies. A major drawback of such active dual-axis solar trackers is their inherent parasitic power consumption, i.e., some of the produced electricity is used to operate their own motors at the cost of decreasing net energy output. The present work presents and assesses a new solution to these circumstances: an autonomous solar tracker hybrid that allows self-generated operation. The system combined a thermoelectric generator (TEG) with the PV panel such that the plant uses, at advantage, the waste heat of its own surface to produce electric power itself and get continuous sun- tracking activation. This breakthrough feature annuls the need for external power sources to keep the tracker running. Energy pricing for a scaled plant operation was designed with PVsyst thus showing that annual grid injection 16.436 MWh is feasible. Secondly, the electrical characteristics of the integrated PV-TEG system and standalone power management circuit have been modelled and optimized in MATLAB/Simulink. Third, the structural robustness of the dual-axis tracking set-up was analytically confirmed through a finite element analysis. It has been prototyped in Durgapur, India and pilot tested extensively. The system's autonomy and its performance were experimentally validated. The hybrid (viz., PV+TEG) tracker was always superior to the fixed-tilt and common tracking configurations for all seasons. In winter, it provided an average power gain of 3.03 kW (i.e., 53.25 %) over the fixed system and a 0.23 kW (i.e., 2.68 %) gain over a tracker without TEG integration. Corresponding summer gains were 1.45 kW (i.e., 23.60 %) and 0.22 kW (i.e., 3.65 %), respectively. This work successfully transitions the concept of a solar tracker from an energy consumer to a fully self-sufficient energy harvester, presenting a practical model for enhancing the net yield and operational independence of renewable energy systems.
