Noise Detection System in Machines

In this context proposes a fault detection system based on a noise signal which is obtained through a microphone from electrical machinery. During the working of the Machine, if the Machine gets damage and produces a noisy sound or not working smoothly which shows some issues and that sound will hav

Project Title

Noise Detection System in Machines

Project Area of Specialization

Internet of Things

Project Summary

In this context proposes a fault detection system based on a noise signal which is obtained through a microphone from electrical machinery. During the working of the Machine, if the Machine gets damage and produces a noisy sound or not working smoothly which shows some issues and that sound will have detected by a sensor or a Detector. When the Detector detects the noisy sound (which shows the presence of some problem in machine working) it will alert the system about the problem (or issue) in the Machine. On receiving an alert about the problem, the crew of the organization takes safety measures and will solve the problem so that the effects because of the problem can be minimized and probabilities of big disasters can be overcome earlier.

Project Objectives

The basic goal of this project is to utilize the extra noise of the machine and to minimize the cost burden from the company so that the company grows up and by early prevention, the machine can be secured.

To achieve this goal, many tasks have to be performed sequentially like Handover the Power System of the Machine to the Noise Detection System so that when the detector detects some issue it will power off the machine.

And the second thing is that the detector will only work when the machine starts working. At very starting point of the working of the Machine, the detector (audio sensor) start working (recording the sound of the Machine), During the sound recording, the recorded sound is broken down into shortest intervals (may be 10 or 15 secs) so that it can easily match to the data sets which are recorded already from previous recordings in which the machine is working smoothly and safely.

The firm or company does not afford the expensive machinery which gets damaged due to some issues which can be handled early and resolving that issues can maximize the lifetime of the machines.

Project Implementation Method

The system uses pre-recorded sounds in wave format. Utilizing pre-recorded sounds is only used in the development stage when changes to the system have to be tested against the reference set. Additionally, the system can be easily adapted to realtime sound acquisition through a microphone connected to a computer. The first preprocessing stage consists of converting the audio to a mono signal.
We do not utilize the information that can be acquired from several channels and the mono conversion simplifies the feature extraction process. We assume that all used reference sounds that are recorded in stereo, have microphones placed so close to each other that there is no noticeable time difference between channels.
In the second step, we resample the signal to have a 48 kHz sampling rate, in order to efficiently obtain the frequency spectrum.
 

Benefits of the Project

The proposed fault detection method is based on several low-level features extracted from the sound. Features are extracted both from time and frequency domain.
Utilizing different features in the feature vector describes the sound from many aspects, thus providing a way to detect different faults. For example, the sound of the misaligned drill was only detectable from time-domain features
 

Technical Details of Final Deliverable

The basis for the method is to detect differences in the sound and to detect potentially faulty segments. This approach has advantages. Firstly, we do not need a set of known sounds of the machine in different operating conditions.

As the system is intended to be later implemented on a mobile embedded platform, we charted potential platforms and what kind of limitations they might set for the method. There are many mobile device platforms, on which fault detection could be implemented.
 

Final Deliverable of the Project

HW/SW integrated system

Core Industry

IT

Other Industries

Manufacturing , Security

Core Technology

Internet of Things (IoT)

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Industry, Innovation and Infrastructure, Life on Land

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com