INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 9, ISSUE 02, FEBRUARY 2020 ISSN 2277-8616 4683 IJSTR©2020 www.ijstr.org Development Of Iiot Based Condition Monitoring System For Rotating Machine Elements Abhinav Gautam, Deepam Goyal, B.S. Pabla Abstract: The condition monitoring and fault detection have gained popularity due to their potential benefits including improved productivity, durable costs, and increased system efficiency. In this paper, the application of the industrial internet of things (IIoT) based system has been discussed to diagnose the gear faults. The vibration and thermal signatures were acquired from the healthy and defective gears. The experimental set-up incorporates rack-pinion gears, one electric motor (for providing input torque), one vibration sensor, one temperature sensor and one mobile (as receiver for receiving the warning or precautionary maintenance schedule-based message). The acquired signals were then sent to thingspeak.com based IIoT system for online data monitoring and transmission. In the event of any vibration response going away from the permissible range, it is considered as a defect in the system. In such cases of possible failure, the system offers an early warning to the operator on their mobile device regarding high vibration levels. The outcomes reveal that the proposed methodology can help in avoiding undesired and unplanned system shutdowns due to gear failure. Index Terms: Gears, vibration signatures, thermal signatures, industrial internet of things ———————————————————— 1 INTRODUCTION The gearbox is one of the most important transmission mechanisms in rotating machinery. Its health directly affects the normal operation of mechanical equipment. In the unrelenting working environment, the gearbox is prone to failure, and rotating machinery often does not stop to check at the initial stage of failure, so a single fault may induce other faults, resulting in complicated faults, serious economic losses, and casualties. The importance of real-time online monitoring is strengthening persistently as the industries try to enhance the machine availability and have a premature alarm of incipient breakdown. Due to recent advances in the enhancement of wireless network technology, the uses of internet-connected devices viz. smartphones and laptops have made the realization of a smart sensor in combination with signal processing circuitry a possibility. Therefore, it is required to implement the Internet of Things (IoT) which can recognize and deal with different types of failures within a machine for condition monitoring and aiding timely maintenance resolutions. Kannan et al. [3] designed a digital twin for grinding wheel as an information-sharing tool to connect, organize and collaborate with the production process to increase productivity and efficiency. The design includes product pre-knowledge component, Radio Frequency Identification (RFID) digital emulation and IoT powered wheel end-of-life prediction and service channel to be built and integrated. Theorin et al. [4] applied Line Information System Architecture (LISA) software used in the automotive industry, where processes are intermittent and the product flow is non- linear. LISA offers flexibility and scalability both for control of low-level applications and aggregation of higher-level information. In future LISA can be implemented to visualize decision support and integration of online optimization. Demetgul [5] demonstrated a fault diagnostic approach based on a multilayer artificial neural network to determine the worm gear condition. It was found that the proposed strategy can be used to predict the gearbox's oil level and speed as well as the heating patterns for all these operating conditions. Garcia- Ramirez et al. [6] suggested an approach based on thermographic image segmentation for fault diagnosis of the rotating machine. This approach can detect bearings faults, rotor broken bars, mechanical unbalance, misalignment, and also voltage destabilization in an induction motor's early stage of failure. Goyal et al. [7] built a low-cost non-contact vibration sensor for detecting the faults in bearings. The supervised method of learning, Support Vector Machine (SVM), was used as a tool to verify the sensor's effectiveness. To develop a framework for the diagnosis of faults for machine health monitoring, experimental vibration data collected for various bearing defects under different loading and running conditions were analyzed. Fault diagnosis has been accomplished using a discrete wavelet transform for denoising the signal. Mahalanobis distance parameters were used to choose the strongest characteristic of the related characteristics extracted. After that selected parameters have been passed to the SVM for the detection of various bearing defects. The results show that the vibration signatures acquired from the established non-contact sensor correlate well with the data obtained under the same conditions from the accelerometer. Saez et al. [8] developed a framework for assessing the performance of real- time hybrid simulation manufacturing networks. Continuous and discrete variables of different machines are tracked for performance analysis using a virtual environment that works synchronously with plant floor equipment as a guide. Data are collected from machines using IoT solutions. Leitao et al. [9] introduced a service-oriented architecture model and integrated different types of digital software tools to construct an engineering framework for manufacturing system acceptance, setup, simulation, a basic outline, control, and monitoring. Xiaoli et al. [10] developed an intelligent web- based fault prediction framework. The program aims to improve the quality of work and the smart level of detection of faults. First, the system's characteristics are examined, and then the system's functional architecture is configured to perform detailed condition monitoring, reliable information transmission, and computer-intelligent fault prediction information processing. Civerchia et al. [11] designed the ———————————————— • Abhinav Gautam, Deepam Goyal and B. S. Pabla, • Department of Mechanical engineering, National Institute of Technical Teachers Training and Research, Chandigarh, India. • E-mail: abhinav596@gmail.com • E-mail: pablabs@nitttrchd.ac.in • E-mail: bkdeepamgoyal@outlook.com