S Natarajan

S Natarajan, PM Baiz, S Bordas et al.

In this paper, the linear free flexural vibration of cracked functionally graded material plates is studied using the extended finite element method. A 4-noded quadrilateral plate bending element based on field and edge consistency requirement with 20 degrees of freedom per element is used for this study. The natural frequencies and mode shapes of simply supported and clamped square and rectangular plates are computed as a function of gradient index, crack length, crack orientation and crack location. The effect of thickness and influence of multiple cracks is also studied.

S Natarajan, PM Baiz, M Ganapathi et al.

In this paper, the linear free flexural vibrations of functionally graded material plates with a through center crack is studied using an 8-noded shear flexible element. The material properties are assumed to be temperature dependent and graded in the thickness direction. The effective material properties are estimated using the Mori-Tanaka homogenization scheme. The formulation is developed based on first-order shear deformation theory. The shear correction factors are evaluated employing the energy equivalence principle. The variation of the plates natural frequency is studied considering various parameters such as the crack length, plate aspect ratio, skew angle, temperature, thickness and boundary conditions. The results obtained here reveal that the natural frequency of the plate decreases with increase in temperature gradient, crack length and gradient index.

M Surendran, S Natarajan, SPA Bordas et al.

The extended finite element method (XFEM) was introduced in 1999 to treat problems involving discontinuities with no or minimal remeshing through appropriate enrichment functions. This enables elements to be split by a discontinuity, strong or weak and hence requires the integration of discontinuous functions or functions with discontinuous derivatives over elementary volumes. Moreover, in the case of open surfaces and singularities, special, usually non-polynomial functions must also be integrated. A variety of approaches have been proposed to facilitate these special types of numerical integration, which have been shown to have a large impact on the accuracy and the convergence of the numerical solution. The smoothed extended finite element method (SmXFEM) [1], for example, makes numerical integration elegant and simple by transforming volume integrals into surface integrals. However, it was reported in [1, 2] that the strain smoothing is inaccurate when non-polynomial functions are in the basis. This is due to the constant smoothing function used over the smoothing domains which destroys the effect of the singularity. In this paper, we investigate the benefits of a recently developed Linear smoothing procedure [3] which provides better approximation to higher order polynomial fields in the basis. Some benchmark problems in the context of linear elastic fracture mechanics (LEFM) are solved to compare the standard XFEM, the constant-smoothed XFEM (Sm-XFEM) and the linear-smoothed XFEM (LSm-XFEM). We observe that the convergence rates of all three methods are the same. The stress intensity factors (SIFs) computed through the proposed LSm-XFEM are however more accurate than that obtained through Sm-XFEM. To conclude, compared to the conventional XFEM, the same order of accuracy is achieved at a relatively low computational effort.

S Natarajan, M Ganapathi, S Bordas

In this paper, the flutter behaviour of simply supported square functionally graded material plates immersed in a supersonic flow is studied. An enriched 4-noded quadrilateral element based on field consistency approach is used for this study and the crack is modelled independent of the underlying mesh. The material properties are assumed to be temperature dependent and graded only in the thickness direction. The effective material properties are estimated using the rule of mixtures. The formulation is based on the first order shear deformation theory and the shear correction factors are evaluated employing the energy equivalence principle. The influence of the crack length, the crack orientation, the flow angle and the gradient index on the aerodynamic pressure and the frequency are numerically studied. The results obtained here reveal that the critical frequency and the critical pressure decreases with increase in crack length and it is minimum when the crack is aligned to the flow angle.

Saarth Vardhan, Pavani R Acharya, Samarth S Rao et al.

Music source separation (MSS) is a task that involves isolating individual sound sources, or stems, from mixed audio signals. This paper presents an ensemble approach to MSS, combining several state-of-the-art architectures to achieve superior separation performance across traditional Vocal, Drum, and Bass (VDB) stems, as well as expanding into second-level hierarchical separation for sub-stems like kick, snare, lead vocals, and background vocals. Our method addresses the limitations of relying on a single model by utilising the complementary strengths of various models, leading to more balanced results across stems. For stem selection, we used the harmonic mean of Signal-to-Noise Ratio (SNR) and Signal-to-Distortion Ratio (SDR), ensuring that extreme values do not skew the results and that both metrics are weighted effectively. In addition to consistently high performance across the VDB stems, we also explored second-level hierarchical separation, revealing important insights into the complexities of MSS and how factors like genre and instrumentation can influence model performance. While the second-level separation results show room for improvement, the ability to isolate sub-stems marks a significant advancement. Our findings pave the way for further research in MSS, particularly in expanding model capabilities beyond VDB and improving niche stem separations such as guitar and piano.

Hrishit Chaudhuri, Vibha Masti, Vishruth Veerendranath et al.

In this paper, methods have been explored to effectively optimise traffic signal control to minimise waiting times and queue lengths, thereby increasing traffic flow. The traffic intersection was first defined as a Markov Decision Process, and a state representation, actions and rewards were chosen. Simulation of Urban MObility (SUMO) was used to simulate an intersection and then compare a Round Robin Scheduler, a Feedback Control mechanism and two Reinforcement Learning techniques - Deep Q Network (DQN) and Advantage Actor-Critic (A2C), as the policy for the traffic signal in the simulation under different scenarios. Finally, the methods were tested on a simulation of a real-world intersection in Bengaluru, India.

Madhav Mahesh Kashyap, Anuj Tambwekar, Krishnamoorthy Manohara et al.

This paper tackles the problem of the heavy dependence of clean speech data required by deep learning based audio-denoising methods by showing that it is possible to train deep speech denoising networks using only noisy speech samples. Conventional wisdom dictates that in order to achieve good speech denoising performance, there is a requirement for a large quantity of both noisy speech samples and perfectly clean speech samples, resulting in a need for expensive audio recording equipment and extremely controlled soundproof recording studios. These requirements pose significant challenges in data collection, especially in economically disadvantaged regions and for low resource languages. This work shows that speech denoising deep neural networks can be successfully trained utilizing only noisy training audio. Furthermore it is revealed that such training regimes achieve superior denoising performance over conventional training regimes utilizing clean training audio targets, in cases involving complex noise distributions and low Signal-to-Noise ratios (high noise environments). This is demonstrated through experiments studying the efficacy of our proposed approach over both real-world noises and synthetic noises using the 20 layered Deep Complex U-Net architecture.

A Vinay, Aviral Joshi, Hardik Mahipal Surana et al.

The paper posits a computationally-efficient algorithm for multi-class facial image classification in which images are constrained with translation, rotation, scale, color, illumination and affine distortion. The proposed method is divided into five main building blocks including Haar-Cascade for face detection, Bilateral Filter for image preprocessing to remove unwanted noise, Affine Speeded-Up Robust Features (ASURF) for keypoint detection and description, Vector of Locally Aggregated Descriptors (VLAD) for feature quantization and Cloud Forest for image classification. The proposed method aims at improving the accuracy and the time taken for face recognition systems. The usage of the Cloud Forest algorithm as a classifier on three benchmark datasets, namely the FACES95, FACES96 and ORL facial datasets, showed promising results. The proposed methodology using Cloud Forest algorithm successfully improves the recognition model by 2-12\% when differentiated against other ensemble techniques like the Random Forest classifier depending upon the dataset used.

Arvind Srinivasan, Aprameya Bharadwaj, Manasa Sathyan et al.

In this paper we improve the image embeddings generated in the graph neural network solution for few shot learning. We propose alternate architectures for existing networks such as Inception-Net, U-Net, Attention U-Net, and Squeeze-Net to generate embeddings and increase the accuracy of the models. We improve the quality of embeddings created at the cost of the time taken to generate them. The proposed implementations outperform the existing state of the art methods for 1-shot and 5-shot learning on the Omniglot dataset. The experiments involved a testing set and training set which had no common classes between them. The results for 5-way and 10-way/20-way tests have been tabulated.

Hirshikesh, C Jansari, K Kannan et al.

An adaptive phase field method is proposed for crack propagation in brittle materials under quasi-static loading. The adaptive refinement is based on the recovery type error indicator, which is combined with the quadtree decomposition. Such a decomposition leads to elements with hanging nodes. Thanks to the polygonal finite element method, the elements with hanging nodes are treated as polygonal elements and do not require any special treatment. The mean value coordinates are used to approximate the unknown field variables and a staggered solution scheme is adopted to compute the displacement and the phase field variable. A few standard benchmark problems are solved to show the efficiency of the proposed framework. It is seen that the proposed framework yields comparable results at a fraction of the computational cost when compared to standard approaches reported in the literature.