IoT-based Machinery Failure Predictive Solution Using Machine Sound Data
Bhagyashri Kantale, Dr. Anjali Raut
ME-Department of Computer Science and Engineering,
HVPM College of Engineering and Technology, Amravati, India
ABSTRACT-
My dissertation is intended to introduce an IoT-based model for real-time condition monitoring of electrical machines, which addresses the challenges of data storage and scalability. The main objective is to correctly classify the acquired machine sound signal into the corresponding machine conditions to be faulty and normal, which is a common multi-class classification problem. This dissertation develops an IoT-based failure prediction solution to reduce maintenance costs. Our proposal allows for highly accurate failure estimates that lead to effective action when it is actually needed. Estimated maintenance is proposed as a measure for the actual maintenance of the machine and hence proper maintenance is done at the right time. This test surveys the different techniques used to classify machine sound data. The aim of the centre is to effectively compare the sound sign of the obtained machine to the conditions of the machine, for example defective and normal, which is usually a multi-category order problem. We conducted a demonstration of the proposed analysis scheme and the system design consisting on recording the sound data of a DC-motor for about 23 minutes with the variation of speed to mimic some failure scenarios. We obtained results that confirmed the effectiveness of our solution in differentiating between the failures signs with no prior learning of the failures and in tracking the slight drift in the machine behaviour. We were able to start predicting failures since day 1. The proposed system design permitted us to limit the payload of data packets which would reduce the cost of the sensor-node data transmission, the power consumption in the sensor node as well as the network traffic.
Keywords - Machine tools condition analysis, Machine learning, IoT, Sound data.
