Demand Side Management (DSM) is a well-recognized concept that seeks to optimize the efficiency and effectiveness of a distribution system. DSM encompasses load shifting and load curtailment strategies, both designed to mitigate the system's peak demand. The former is an optimization-based method that repositions the elastic loads to hours with low tariff rates, thereby filling the gaps and reducing the peak. The latter offers incentives to consumers to encourage their participation and reduce energy consumption during periods of high demand. In order to minimize the overall operating cost, the unique work done in this study intends to analyze ten exhaustive cases on a low voltage (LV) microgrid (MG) system and optimally schedule the distributed energy resources (DERs). The complexity of the work is further enhanced by the inclusion of plug-in hybrid electric vehicle (PHEV) which integrates Grid to Vehicle (G2V) and Vehicle to Grid (V2G) technologies to charge and discharge itself using the utility assigned electricity market price. The work used the Differential Evolution (DE) method as an optimisation framework. The ten exhaustive scenarios were analyzed to acknowledge the impact of the involvement and pricing of the grid, PHEV and DSM strategies mentioned above. Numerical study confirms that the load shifting policy and the type of load curtailing policy which rewarded the customer based on their willingness to curtail loads along with delivering benefit to the DISCOM were more cost-effective compared to the rest of the cases. Furthermore, it was also noted that time of usage (TOU) based electricity pricing was economical for entities which participated in bidirectional flow of power like grid and PHEV.

Nowadays, optimal irrigation is essential to maintain global food security because it increases agricultural productivity in areas with limited water availability and makes it possible to grow crops in dry and semi-arid climates. Smart irrigation systems integrate conventional and intelligent control techniques to overcome water scarcity problems. This study reports advancements in intelligent control methods and smart irrigation systems using a hybrid review methodology that combines bibliometric analysis and narrative literature review. The analysis was carried out using VOSviewer software, based on 3,003 publications from 2010 to 2024 that were obtained from the Scopus database using keywords like irrigation system, Internet of Things, artificial intelligence, and sustainability. In conventional controllers such as PID controllers, we emphasize the systems or models with controlled parameters such as soil moisture, climatic factors (like temperature, humidity, and atmospheric pressure), and Soil factors (like fertilizer, pH, and salinity). In addition to this, the intelligent controllers have a learning mechanism, which makes them adaptive, indispensable, and different from conventional controllers. The paper deals with conventional controllers as well as intelligent controllers developed for irrigation to make the system smart, which reduces the wastage of water, improves production, and environmental sustainability in terms of their impact, advantages, key limitations, and future directions.

Automatic identification of emotions is important in human-centered computing. It allows machines to better understand user emotions. Identifying emotions via neural sensing techniques such as electroencephalogram (EEG) is a promising approach. In this paper, we aim to identify the emotions class from EEG signals. We frame emotion identification as a classification task and apply spectral and statistical encoders to extract the relevant features. We validate our approach on EmoNeuroDB dataset. Our method outperforms the EmoNeuroDB baseline, achieving a 42.10% increase in class prediction accuracy.

Abstract This paper introduces a nonlinear PI controller for improved speed regulation in permanent magnet direct current (PMDC) motor drive systems. The nonlinearity comes from the exponential (Exp) block placed in front of the classical PI controller, which uses a tunable exponential function to map the speed error nonlinearly. Such a configuration has not been studied till now, thus meriting further investigation. We consider an exponential PI (EXP-PI) controller and to attain the best performance from this controller, its parameters are optimized offline using salp swarm algorithm (SSA), which borrows its inspiration from the way of forage and navigation of salps living in deep oceans. To indicate the credibility of SSA tuned EXP-PI controller convincingly, numerous experiments on speed regulation in PMDC motor have been implemented using DSP of TMS320F28335. The results obtained are also compared to similar results in the literature. It is shown that the proposed approach performs well in practice by ensuring tight tracking of the speed reference and superb torque disturbance rejection for the closed loop control. Furthermore, superior performance is achieved by the proposed nonlinear PI controller with respect to a fixed-gain PI controller.

This paper presents a framework for generating appropriate facial expressions for a listener engaged in a dyadic conversation. The ability to produce contextually suitable facial gestures in response to user interactions may enhance the user experience for avatars and social robots interaction. We propose a Transformer and Siamese architecture-based approach for generating appropriate facial expressions. Positive and negative Speaker-Listener pairs are created, applying a contrastive loss to facilitate learning. Furthermore, an ensemble of reconstruction quality sensitive loss functions is added to the network for learning discriminative features. The listener's facial reactions are represented with a combination of the 3D Morphable Model's coefficients and affect-related attributes (facial action units). The inputs to the network are pre-trained Transformer-based feature MARLIN and affect-related features. Experimental analysis demonstrate the effectiveness of the proposed method across various metrics in the form of an increase in performance compared to a variational auto-encoder-based baseline.

Bodily behavioral language is an important social cue, and its automated analysis helps in enhancing the understanding of artificial intelligence systems. Furthermore, behavioral language cues are essential for active engagement in social agent-based user interactions. Despite the progress made in computer vision for tasks like head and body pose estimation, there is still a need to explore the detection of finer behaviors such as gesturing, grooming, or fumbling. This paper proposes a multiview attention fusion method named MAGIC-TBR that combines features extracted from videos and their corresponding Discrete Cosine Transform coefficients via a transformer-based approach. The experiments are conducted on the BBSI dataset and the results demonstrate the effectiveness of the proposed feature fusion with multiview attention. The code is available at: https://github.com/surbhimadan92/MAGIC-TBR

Naturalistic music typically contains repetitive musical patterns that are present throughout the song. These patterns form a signature, enabling effortless song recognition. We investigate whether neural responses corresponding to these repetitive patterns also serve as a signature, enabling recognition of later song segments on learning initial segments. We examine EEG encoding of naturalistic musical patterns employing the NMED-T and MUSIN-G datasets. Experiments reveal that (a) training machine learning classifiers on the initial 20s song segment enables accurate prediction of the song from the remaining segments; (b) β and γ band power spectra achieve optimal song classification, and (c) listener-specific EEG responses are observed for the same stimulus, characterizing individual differences in music perception.

Wearable sensors have the intriguing potential to continuously evaluate human physiological characteristics in real-time without being obtrusive. This thesis aims to incorporate physiological sensors data to investigate the Media Perception and Activity Recognition. Our primary research goals include (a) neural encoding-based psycho-acoustic attribute analysis for data sonification, (b) empirical evidence for perceptual subjectivity in neural encoding during human-media interactions, the impact of incorporating behavioral ratings, and (c) the efficacy of attention-based transformer models on physiological data on human activity recognition problems.

We examine user and song identification from neural (EEG) signals. Owing to perceptual subjectivity in human-media interaction, music identification from brain signals is a challenging task. We demonstrate that subjective differences in music perception aid user identification, but hinder song identification. In an attempt to address intrinsic complexities in music identification, we provide empirical evidence on the role of enjoyment in song recognition. Our findings reveal that considering song enjoyment as an additional factor can improve EEG-based song recognition.

Non-invasive physiological sensors allow for the collection of user-specific data in realistic environments. In this paper, using physiological data, we investigate the effectiveness of Convolutional Neural Network (CNN) based feature embeddings and Transformer architecture for the human activity recognition task. 1D-CNN representation is used for the heart rate, and 2D-CNN is used for short-term Fourier transformation of the accelerometer data. Post fusion, the feature is input into a transformer. The experiments are performed on the harAGE dataset. The findings indicate the discriminative ability of the feature-fusion on transformer-based architecture, and the method outperforms the harAGE baseline by an absolute 3.7%.