The sudden surge in the fuel cost and increase in environmental impact causes a reduction in electricity generation through conventional energy sources. The power system is moving towards the incorporation of distributed energy resources (DERs) into the existing microgrid (MG). Distributed generation (DG), storage system and geographically scattered loads forms a MG. The stochasticity and unpredictable output associated with DGs may create imbalance in supply and demand. Further, the unexpected peak loads on the system mayjeopardize the system reliability. Hence, proper energy management system is obligatory for peak reduction so as to maintain the power balance. One of the emerging and efficient energy management techniques is demand response. While incorporating demand response, a part of curtailed loads i.e., shiftable loads at peak loads should be re-distributed to other off-peak hours where the load demand and electricity offer price is less. The re-distribution of loads at non-peak hours eventually increases the market clearing price (MCP). Increment in MCP mostly effects the inelastic loads as they cannot shift their schedule from one horizon to another since they are must run loads and of on/off in nature. In this context, the present work focuses on optimal rescheduling of the redistributed loads that doesn’t impact much MCP and operational cost (OC) of the MG much and should able to maintain the dynamic balance between supply and demand. A cost-based method was employed in which grey wolf optimization (GWO) algorithm is used to optimize the OC of the MG.
Birdwatching or Birding is observing or watching the birds which is found to be one of the serene activities to do in daily life. But recognizing the bird species is hard for humans as it requires a bit of support from the bird book. Also, in the olden days, many ornithologists and researchers faced hardships in detecting bird species and learning about the different patterns of bird species. We have developed an Arduino-based system that performs automatic bird species recognition. This system will be helpful for ornithologists, researchers, and other enthusiasts to learn about the existence of different bird species in a given geographical region. This system is developed using Arduino Uno, PIR Motion Sensor, and an ESP-32 camera. When the motion is detected, the ESP-32 camera captures the image and uploads the image to Google Drive. The images in Google Drive can be given to the trained deep learning models to predict the name of the bird species. To develop the deep learning model, we have used a Kaggle data set that consists of 450 bird species images of which 70,626 training images, 22500 test images, and 2250 validation images.
This paper discusses the clinical chatbot which could examine the contamination and deliver essential insights regarding the contamination previous to counseling a specialist. To lower the healthcare charges and similarly increase openness to clinical facts the clinical chatbot is constructed. Certain chatbots pass approximately as clinical reference books, which facilitates the affected person to discover approximately their contamination and assists with running on their fitness. The person can accomplish the real gain of a chatbot simply while it is able to examine all forms of infections and deliver essential data. In later sections of the paper, a doctor appointment system software is also discussed that is integrated with the chatbot so that the specialist which is recommended by medical chatbot can be easily visited at the mutually agreed time of both specialist and patient. Such an appointment system has its own advantages like it reduces the waiting time of the patient, patients can choose the appointment time in keeping with their desire additionally to be had and booked slots are proven in powerful graphical person interface.
In this paper, we have implemented a novel unsupervised graph-based algorithm for Hindi Word Sense Disambiguation. For disambiguation, we perform a random walk on graph created for each instance. The nodes in the graph are various senses of the words appearing in the context of ambiguous word. The edge weights are assigned using semantic similarity between pair of nodes. We compare two path-based similarity measures. The experimental investigations suggest that a Leacock-Chodrow similarity measure performs better than Shortest path measure. We observed an accuracy of 72.09% averaged over all the instances of five Polysemous nouns.
Mobility impairments and difficulties are a problem faced by a significant portion of the population. The statistics on differently abled people and the mobility aids can benefit a great deal for governmental institutions in areas such as policy-making and infrastructure development. This paper proposes an automated process for the detection and identification of people who use mobility aids. The YOLO (You Only Look Once) algorithm is used to train the dataset. The performance metrics and the feasibility of the algorithm are being discussed. The primary aim of this paper is to help improve the facilities made for differently people in public places by the detection and analysis of the different kinds of mobility aids that people use and how it affects movement in a public place. The model detects and classifies each person in the image as differently abled or not, and if found to be differently abled, it finds out the type of the mobility aid used and the number of differently abled people in each class as well.
Nowadays in urban environments, autonomous vehicle innovations boost road safety and alleviate traffic congestion. The detection of traffic lights is an important aspect of autonomous vehicle systems because it ensures that vehicles can respond to traffic signals quickly and appropriately. However, despite adopting vision-based technologies, autonomous cars are somehow unable to identify traffic signals reliably. This situation emphasizes the significance of using Vehicle to Infrastructure (V2I) communication technology to detect traffic lights. This paper aims to introduce V2I communication in the context of autonomous vehicle decision-making at traffic lights. The proposed method uses V2I transmission and heading angles of the roadways to make autonomous vehicles accurately detect traffic lights at the intersections using a nRF24L01 transceiver and digital compass as the key components.
This paper describes the robust optimal control algorithm for a submersible autonomous robot (SAR) to achieve the desired yaiv. Yaw rate of SAR is controlled through the design of robust state feedback optimal control law. Semi-definite programming is considered to develop the proposed robust optimal control algorithm. A linear matrix inequality (LMI) considewred in terms of linear quadratic regulator (LQR) is used to solve the control problem. A polytopic approach based robust optimal control laiv is formulated in a steering plane for SAR. For the implementation of the proposed control algorithm, YALMIP tool is used in Matlab/Simulink environment. The proposed control algorithm ensures the robust behavior by tracking the desired yaw.
A comparative analysis of various decoupling techniques is carried out in this paper. As most industries deal with multi-input multi-output (MIMO) systems, decouplers are widely used to reduce loop interactions. The design procedure for the commonly used decouplers is explained briefly along with their advantages and disadvantages. The reduced order first order plus dead time (FOPDT) model is obtained for each of the decoupled subsystems. Further, the PI controller is designed based on the frequency domain specifications to attain the design criteria. The servo and regulatory responses of the decouplers are contrasted with each other. Further, a robustness study is carried out for the designed subsystems.
Low power architectures are more pronounced for different applications that extend from Internet of Things to Quantum computing. Primitive combinational logic circuits induce from bit deletion due to information loss during the processing of input information which results in energy loss. The computations involving reversibility cancel the loss of information by sustaining the input bits from output. In the basic arithmetic and logic units, the combinational circuits play a significant role in determining the performance of the processor. The principles involved in the design of reversibility is an upcoming technology for ultra-low power applications. The reversible logic circuits furnish a thoroughly new way to progress in Quantum computing. In this article, we propose an energy tolerant low power reversible multiplexer with optimum energy loss. The proposed multiplexer also reduces the ancillae, garbage outputs and quantum cost considerably.
In this letter, we present the design of a 5G compact MIMO (multi-input, multi-Multi-Output) antenna with isolation enhancement at frequencies less than 6 GHz. To achieve the required bandwidth, the proposed antenna incorporates a unique defective ground structure (DGS), grounding branches, and a T-shaped parasitic patch, which are used to minimize MC (mutual coupling) for improved isolation and also improve antenna performance. Strapline feeding is used to improve the antenna characteristics, and RT Duroid RO4003 substrate has been used with a relative permittivity $\varepsilon_{r}$ of 3.55. The substrate size of the proposed antenna is 36. 2x44.5xl.524mm3, and it has a band of frequency of 3. 37-3.72GHz (s11 & s22 < -10dB) and isolation is < -15dB (S21& S12). With envelope correlation coefficient (ECC), diversity gain (DG), and total affective reflection coefficient (TARC) values of-15.2 dB, 0.001, 9.99, and-18.2 dB, respectively, the predicted antenna structure provides good isolation. The postulated antenna’s band do not shift in respect to the single monopole antenna, and its impedance matching is optimized.The parametric study of the proposed antenna was carried out using the HFSS simulator. Based on results obtained, the proposed antenna is most suitable for 5G communications at sub-6 GHz.
To survive in the telecommunications industry’s severe competition and to keep existing loyal customers, predicting prospective churn consumers has become a critical task that may be accomplished with efficient predictive models. For many years, churn studies have been utilized to boost profitability and make customer-company relationships more sustainable. Customer churn is estimated using a multi-layer perceptron prediction model based on ANN. Furthermore, the suggested model manages the data’s uneven class distribution using an advanced oversampling strategy called SMOTE-ENN based on K-Nearest Neighbors. The models accuracy was compared with and without SMOTE-ENN and the model using SMOTE-ENN showed better results.
In this work, Energy efficient Binary content addressable memory (BiCAM) is designed using reversible logic which is in high demand in fields of quantum computing, nano technology, data centric computing, software-defined networks and wide variety of high speed applications. BiCAM performs search operation in a single clock cycle. Binary content addressable memory (BiCAM) design based on reversible logic reduces power dissipation. Due to minimum heat dissipation, Reversible logic has gained its interest in recent years. A novel design of BiCAM array using reversible logic gates with transmission gate logic improves speed and reversibility ensures low power. The proposed reversible logic based 4x3 BiCAM array using transmission gates is implemented using mentor graphics at 130nm technology with Vdd=1.2V shows 95.2% efficiency in power compared to CMOS based BiCAM design. All reversible logic gates used in BiCAM design as analysed using transmission gate logic and compared with conventional gates.
Transformers have been extensively employed in various vision issues, particularly visual recognition and detection. Detection transformers are connected to end-to-end networks for object detection. Self-attention modules in the transformer give huge efficiency, making excellent object detection models. The decoder transformer fails to initialize query content properly and also fails to provide specific prior knowledge, which might potentially enhance inductive bias. This paper uses encoder and decoder transformers for object detection in deep foggy conditions. High-Resolution Network (HRNet) has been used in the backbone of this architecture to extract deep feature representation. The proposed method validates and compares with other detection techniques in terms of average precision (AP), the variety of factors, and frames per second (FPS) using the Foggy Cityscapes dataset. The qualitative results indicate that the proposed technique improves detection accuracy in deep foggy conditions.
By applying a constant uninterrupted current via more than two electrodes put inside alternatively close to the tumour, tumours can be treated electrochemically (EChT). To develop a reliable dose-planning system, it is essential to understand the underlying mechanism of destruction of EChT. One method for doing this is mathematical modelling. Through modelling of the anode processes, the importance of chlorine in the underlying EChT degradation mechanism has been made clear. It has been found that the production of hydrogen ions depends significantly on tissue and chlorine activities. The amount of these processes that contribute to the tissue’s acidity near to the anode and the applied current density are highly connected. Throughout the course of the treatment, early deduction time is utilised.
The prominent nature of cognitive radio (CR) technology is to favor the unlicensed users (or secondary users (SUs)) to employ the spectrum holes which are left by the legitimate users (i.e., primary users (PUs)). Primarily, for the clear-cut identification of spectral gaps over any fading/shadowing channel or of the accurate implementation of the CR algorithms in the next generations of wireless communication systems, it is desired to make use of the cooperative spectrum sensing (CSS) technology. In this context, this paper evaluates the role of the proposed cooperative cognitive radio system (CCRS) over Nakagami-n/Rician fading environments by taking into consideration several hard-decision (binary) fusion rules (HDFR)namely OR/AND/MAJORITY rules. Furthermore, in order to analyse the performance of the CCRS over Rician fading channel, this paper derives the analytical closed-form expressions for the performance metrics namely the probabilities of false alarm, miss-detection, total error rate (TER), and throughput by taking into account various network specifications such as the channel error probability (q), fading severity parameters (K), number of samples (M), number of SU nodes (N), average sensing (S) channel signal-to-noise ratio (SNR), and sensing threshold ($\lambda$). In addition, by making use of different HDFS schemes, this work formulates the analytical framework for determining the optimal threshold $(\lambda_{opt})$, optimal voting (fusion) rule $(k_{opt})$, and optimal number of SU nodes $(N_{opt})$ values for attaining the lowest value of TER.
To overcome the effect of impulsive noise on physical systems an algorithm was developed in the past using an improved proportionate maximum correntropy criterion (IPMCC) based on an adaptive filter incorporating l 1 -norm. This approach yielded satisfactory results for the identification of timevarying sparse physical systems but its execution is limited because it tends to draw the active and dominant coefficients to zero irrespective of their magnitude. To overcome the limitation of the l 1 -norm based IPMCC algorithm, this paper presents a l 0 -norm based IPMCC approach that significantly enhances the convergence rate of the inactive coefficients which influence the sparse system. Simulation findings prove that the suggested technique delivers a rapid convergence and lower steady-state error in contrast to the PMCC and l 1 -norm based IPMCC algorithm.
In this work, a rectangular SRR-based unit cell is proposed for 5.2GHz wireless applications. The proposed structure exhibits double negative metamaterial properties. A polymide dielectric substrate material with a dielectric constant of 3.5 and a thickness of 0.1 mm was utilized to construct and validate the proposed structure.A numerical simulation tool, CSTMW studio, which works on the FDTD method is used to simulate the structure. The unit cell works between 4.90 GHz and 5.94 GHz, yielding an overall bandwidth of 1.04 GHz with 11 of -46.32 dB and the unit cell is validated with ADS tool with equivalent circuit diagram.
The purpose of this research is to demonstrate that using the generated synthetic dataset will produce better results than the original teeth x-rays and thus be more efficient for carrying out hierarchical multiclass classification, this will also allow dental caries classification to be automated in accordance with the GV Black Standards. Five distinct models that were trained and tested using both the new and original datasets are compared. As the first step in each of these tasks, simple object detection will be used to identify each tooth, and then hierarchical classification will be used to achieve the desired results for classifying the cavities so that appropriate treatment can be determined based on the class of caries.
In this study, the performance of a Dual gate source drain schottky barrier tunnel field effect transistor (D-G-S-D-STFET) is analysed using a high k dielectric composed of HfO 2 and a low k dielectric composed of SiO 2 , respectively. The DC and the analog /RF performances of the device are examined the in-depth DC performance analysis, like transfer characteristics ($\mathrm{I}_{\mathrm{D}}-\mathrm{V}_{\mathrm{GS}}$), and the RF performances like transmission frequency $(\mathrm{f}_{\mathrm{T}})$, and transconductance generation factor (TGF) are simulated using SILVACO TCAD. It has been observed that the transmission frequency $(\mathrm{f}_{\mathrm{T}})$ is in enhanced in the proposed device in the THz range compared to the conventional structures at $(\mathrm{L}_{\mathrm{G}})$ of 50nm. Further, the proposed device has shown lower power consumption, a maximum $\mathrm{I}_{\mathrm{ON}}/\mathrm{I}_{\mathrm{OFF}}$ ratio and higher ON current are observed. Therefore, proposed device has significant potential for use in applications requiring low power and high frequency.
Autonomous or self-driving vehicles are self- decision systems that must choose their next course of action when an object or other vehicle is spotted on the road. Smooth driving is crucial for self-driving cars to avoid causing other road users to become stranded in a traffic jam. For autonomous vehicles, understanding traffic objects is crucial. The self-driving automobile can perform the required action after identifying an object’s precise class. Self-driving cars should be extra cautious when passing big vehicles like buses or trucks if they are identified. Consequently, one of the key capabilities of self-driving automobiles is object identification. Computer vision tasks like object classification can be useful for object recognition. Deep learning is a key subfield of machine learning and crucial for a number of computer vision applications such object detection, localization, and classification. In this paper, we present a model for real-time object classification using the Convolutional Neural Network (CNN) of AI approach. It includes a thorough discussion of various deep learning techniques for classifying objects in RGB pictures and LiDAR point clouds.
