Design and Development of Techniques for Equipment Health Monitoring System
Vasireddy Prabha Kiranmai1, Sharmitha S Bysani2, Vijaya Kumar B P3, Kusuma S M4
1Vasireddy Prabha Kiranmai, Department of Information Science and Engineering, Ramaiah Institute of Technology, Bangalore (Karnataka), India.
2Sharmitha S Bysani, Department of Information Science and Engineering, Ramaiah Institute of Technology, Bangalore (Karnataka), India.
3Vijaya Kumar B P, Department of Information Science and Engineering, Ramaiah Institute of Technology, Bangalore (Karnataka), India.
3Kusuma S M, Department of Telecommunication Engineering, Ramaiah Institute of Technology, Bangalore (Karnataka), India.
Manuscript received on 04 December 2019 | Revised Manuscript received on 12 December 2019 | Manuscript Published on 31 December 2019 | PP: 277-282 | Volume-9 Issue-2S December 2019 | Retrieval Number: B10121292S19/2019©BEIESP | DOI: 10.35940/ijitee.B1012.1292S19
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Machines in Industries are often subjected to enormous wear and tear, which if unnoticed, may lead to production delays and increased maintenance cost. Machines must be able to analyse and provide statistics about its health, so that preventive measures can be taken to avoid catastrophes in the industries. Thus, there is a need of automated fault detection and prediction of system’s condition. The concept of equipment health monitoring is a crucial step in the field of research and development in the manufacturing industries. This equipment makes it handy in situations where machines require continuous monitoring and is difficult for humans to provide such attention , especially in the case of unmanned vehicles. Prediction of the status of equipment by acquisition of data from industrial machinery is the critical step in building such a system. Health of machines can be estimated by the data collected by the sensors-temperature, accelerometer, etc integrated with an embedded computing system, like a Raspberry Pi. This IoT model consisting of embedded system with wireless connectivity collects real time data from the equipment/machinery used in industries. This data is used to analyse and predict the health of the equipment, examine the steady-state characteristics using Machine Learning technique, Hidden Markovian Model. The concept of the proposed IoT model is evaluated over a conveyor belt test rig under various conditions, like different loads placed on various locations of conveyor belt and the belt is made to run at different speeds and data is collected over all these conditions. Then, a data model is created using Hidden Markovian Model which is further used in predicting the state of the belt based on the sequential data, here it is the sensor data. Given a state of the belt, this model can predict whether the belt is in proper condition or not, and if human intervention is required. Thus, at any point of time, having this setup on the machinery which needs to be monitored can be used in predicting the faults and notifying the user in case of any faulty behaviour or malfunctioning of machines. This setup can be used for any machines which are subjected to any motion, vibration and thermal changes. This helps in creating a completely automated fault detection systems in the present Industries.
Keywords: Accelerometer, Automated Fault Detection, Condition of Machine, Equipment Health Monitoring, IoT, Hidden Markovian Model, Sensor.
Scope of the Article: Health Monitoring and Life Prediction of Structures