Sri Sairam College of Engineering

The increasing integration of distributed renewable energy sources (RES) into modern utility grids has intensified power quality (PQ) challenges, particularly under nonlinear and dynamically varying operating conditions. To overcome these challenges, this paper presents an intelligent method for developing PQ in grid-connected systems with a Tilted Integral Fractional Derivative with Filter plus Fractional Derivative (TIFDNFD) controller. The proposed method combines the Humboldt Squid Optimization Algorithm (HSOA) and Temporal Inductive Path Neural Network (TIPNN), which is termed the HSOA-TIPNN technique. The main objective of the proposed method is to minimize Total Harmonic Distortion (THD), thereby enhancing PQ. The HSOA is utilized to optimize the gain parameters of the TIFDNFD controller, while TIPNN is used to predict the performance of the controller. The proposed technique is tested in MATLAB and compared with Artificial Neural Network (ANN), Convolutional Neural Network (CNN) and Deep Recurrent Neural Network (DRNN). The proposed HSOA-TIPNN is able to get the THD of 0.8% while ANN, CNN, and DRNN are 0.9%, 1.0%, and 1.1%, respectively. It also achieves a prediction accuracy of 99.90%, an MSE of 0.0032, and a settling time of 0.15 s, demonstrating improved harmonic suppression, faster dynamic response, and enhanced PQ under nonlinear and disturbed grid conditions.

Predicting clinically significant drug–drug interactions (DDIs) continues to be an unresolved challenge in contemporary pharmacovigilance, primarily due to the inadequacy of current computational frameworks in addressing the nonlinear, multi-scale characteristics of simultaneous drug metabolism. This paper presents the Quantum Graph-Differential (QGD) model an exact mathematical framework that combines quantum-inspired graph theory with a set of interconnected fractional differential equations to describe and forecast pairwise drug–drug interactions (DDIs). The principal component of our construction is the quantum interaction graph 𝒢_Q = (V, E, W_Q) , wherein the vertex set represents distinct drug molecules as quantum states within a finite-dimensional Hilbert space, and the complex-valued edge weights are obtained from the overlap of shared metabolic pathways and transporter affinity profiles.A Schrödinger-type equation on 𝒢_Q governs drug–drug coupling, and the graph Hamiltonian H is constructed from a novel fractional quantum graph Laplacian ℒ_Q^α , α∈ (0,1] . A hybrid quantum-classical dynamical model is created by coupling the time evolution of the interaction wavefunction Ψ _s(t) to a compartmental pharmacokinetic/pharmacodynamic (PK/PD) ordinary differential equation system. Using Banach fixed-point and semigroup theory, we prove existence, uniqueness, and long-time asymptotic stability of solutions. Using the QGD framework on a selected dataset of 7,428 clinically confirmed DDI pairs from DrugBank v5.1, TWOSIDES, and FAERS, our model outperforms five established baselines by 1.5–13.9 percentage points in AUC, with an average precision of 0.948 and an AUC of 0.962. Quantum edge weighting alone explains a 3.7

The main objective of the present analysis is to investigate the effect of incorporating silane-treated Helianthus annuus waste derived cellulose into sugarcane leaf fiber-reinforced vinyl ester composites. Both the fiber and filler materials were effectively surface treated using 3-aminopropyltrimethoxysilant. The composites were prepared and their interlaminar shear strength (ILSS), wear resistance, water absorption, and flammability properties were evaluated in accordance with ASTM standards. The test results revealed that silane treatment significantly influenced the overall performance of the composites. Among the tested specimens, the BSC5 composite, which was treated with silane and reinforced with 1 vol

The creation of effective healthcare artificial intelligence (AI) systems demands access to large and diverse datasets dispersed in various healthcare institutions. Nevertheless, strict privacy policies in the United States such as HIPAA, in Europe such as GDPR, and other laws in other countries pose a tremendous hindrance to centralized aggregation of data. Although Federated Learning (FL) allows all involved parties to jointly train models without centrally storing sensitive patient information, gradient updates exchanged in the training process may reveal much valuable personal information via more advanced reconstruction attacks. The purpose of the study is to create and fully test a privacy-guaranteed federated learning model that facilitates and federates diagnostic AI training among multiple healthcare facilities and formally guarantees some differential privacy and retains clinic-quality model performance. According to the research, we introduce PP-FL-DP (Privacy-Preserving Federated Learning with Differential Privacy), a unified system introducing three innovations: (1) Hierarchical Privacy-Preserving Architecture (HPPA) with defense-in-depth, i.e., local differential privacy, secure aggregation, and Byzantine-robust features; (2) Sensitivity-Aware Gradient Perturbation (SAGP) with layer-wise adaptive clipping, based on empirical gradient sensitivity analysis; and (3) Three different healthcare datasets, including breast cancer mammography (N = 12,500 images, 5 federated clients), diabetic retinopathy fundus imaging (N = 35,000 images, 8 clients), and ECG arrhythmia detection (N = 45,000 recordings, 6 clients) were thoroughly evaluated. Hyperparameters (batch size, 16,32, 64, 128, learning rate, 0.001-0.1 and aggregation threshold, τ = 0.1–0.9) were fully analyzed. PP-FL-DP had a diagnostic accuracy of 96.2 ± 0.4