Related Experiment Video
Updated: Oct 29, 2025

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
Identification of sloshing noises using convolutional neural network
Golla Siva Teja1, Chennuri Saurav Vara Prasad2, B Venkatesham1
1Department of Mechanical and Aerospace Engineering, Indian Institute of Technology Hyderabad, Sangareddy, Telangana 502285, India.
Abstract:
Sloshing in fuel tanks has become a new source of noise in hybrid and high-end vehicles in the wake of reduced noise from major sources like the engine. It occurs due to the interactions of fluid inside the tank under various driving conditions of the vehicle. Interactions of fluid with the tank walls cause hit noise, and the fluid-fluid interactions cause splash noise. As the generation mechanism is different, the hit and splash noises demand different noise controlling strategies. Thus, identifying these noises during the design stage is important for implementing effective solutions in designing a quieter fuel tank. This paper presents a convolutional neural network (CNN) based methodology for the identification of sloshing noises under different conditions of fill level, excitation, baffle configuration, etc. Data for training and testing the network are collected using a reciprocating test setup, which facilitates the generation of hit and splash noises in a rectangular tank. The identification accuracy of the features learned by CNN is compared with the hand-crafted features using support vector machines. The applicability of the proposed CNN model is tested for practical scenarios like vehicle braking, where different types of sloshing noises occur in quick succession.
More Related Videos
06:19Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
Published on: August 16, 2024
09:27Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline
Published on: January 30, 2019
Related Concept Videos
Uniform Depth Channel Flow: Problem Solving
Classification of Signals
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
Hydraulic Jump: Problem Solving
Perception of Sound Waves
The pitch of a sound depends on the frequency and the pressure amplitude of the source. Two sounds of the same...
Convolution: Math, Graphics, and Discrete Signals
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
Convolution Properties II
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...