Ele 02 Ncrenb
Ele 02 Ncrenb
Ele 02 Ncrenb
Abstract: As we all know that Condition monitoring together with predictive maintenance of electric
motors and other equipment used by the industry avoids severe economic losses resulting from
unexpected motor failures and greatly improves the system reliability. This paper describes a Machine
Learning architecture for Predictive Maintenance, based on Machine Learning approach. The system was
tested on a real industry example, by developing the data collection and data system analysis, applying
the Machine Learning approach and comparing it to the simulation tool analysis. Data has been collected
by various sensors. With the help of this paper, we want to monitor and increase the life span of Electric
motor and other equipment's.
Keywords - Predictive maintenance, Fault Diagnosis, Anomaly Detection, Deep Learning,
I. INTRODUCTION
Recently, with the emergence of Industry 4.0 (I4.0), smart systems, machine learning (ML) within artificial
intelligence (AI),predictive maintenance approaches have been extensively applied in industries for handling
the health status of industrial equipment. Due to digital transformation towards I4.0, information techniques,
computerized control, and communication networks, it is possible to collect massive amounts of operational
and processes conditions data generated form several pieces of equipment and harvest data formaking an
automated fault detection and diagnosis with the aimto minimize downtime and increase utilization rate of
the components and increase their remaining useful lives. Predictivemaintenance is inevitable for sustainable smart
manufacturing in I4.0. Machine learning (ML) techniques have emerged as a promising tool in Predictive
maintenance applications for smartmanufacturing in I4.0, thus it has increased attraction of authorsduring
recent years. This project aims to provide a comprehensive review of the recent advancements of ML
techniques widely applied to Predictive maintenance for smart manufacturing in I4.0 by classifying the research
according to theML algorithms, ML category, machinery, and equipment used, device used in data
acquisition, classification of data, size and type etc. This project proposes predictive maintenance of electric
motor based on sensors that monitor various reasons for failures such as Transient voltage (“surges” or
“spikes”), Voltage imbalance, Current imbalance, Vibration, High operating temperature, Motor overload,
Misalignment, Moisture etc. Machine learning prediction models will be used for predictive maintenance of
motor and to predict failures in motor.
1
www.viva-technology.org/New/IJRI
II. PROBLEM IDENTIFICATION
In Predictive Maintenance is a strategy viable adopted when dealing with maintenance issues given the
increasing need to minimize downtime and associated costs. The methodology has been implemented in the
experimental environment on the example of a real industrial group, producing accurate estimations. this is
no longer the case. The advantages of predictive maintenance are accepted in industry today, because the
tangible benefits in terms of early warnings about mechanical and structural problems in machinery are clear
These are the problems we will receive every 2 minutes from all the 4 sensors. Continuous monitoring of the
Electrical equipment will take place with the help of this Project.
3.1 Objective
The design Machine learning approach for predictive maintenance based on the csv file obtained To Collects
data using current sensors, voltage sensor, humidity sensor, temperature sensor and vibration sensor and Logs
data to ThingSpeak cloud server To process Data logged can be exported to csv file.
To design and develop advantages of predictive maintenance are accepted in industry today, because the
tangible benefits in terms of early warnings about mechanical and structural problems in machinery are clear.
The method is now seen as an essential detection and diagnosis too that has a certain impact in reducing
maintenance costs, operational vs. repair downtime and inventory hold-up.
2
www.viva-technology.org/New/IJRI
3.3 Operating Principle
The system uses sensors will be used to monitor and collect data.The data collected will be passed to NodeMCU
Microcontroller.NodeMCU microcontroller is chosen because it supports inbuilt Wi-Fi. Hence data can be
directly stored to cloud storage over the internet.Cloud Storage that will be used for storing data is
ThingSpeak Cloud. It supports data storage along with visualizations. Once sufficient amount of data is
collected it can be exported into Excel .csv format.The data exported from ThingSpeak Cloud is the dataset
which will be used as a dataset for Machine Learning.Pre-processing steps such as exploration, cleaning and
transformation of dataset will be done using Python.Machine learning prediction models such as Random
Forest, Decision Tree, Naïve Bayes etc. can be used for predictive maintenance of motor and to predict
failures in motor.Visualisation of the data can be done using Python Modules such as matplotlib, seaborn,
ggplot etc.
IV. METHODOLOGY
A proposed module is attached to the Sensors will be used to monitor and collect data The data collected will
be passed to NodeMCU Microcontroller NodeMCU microcontroller is chosen because it supports inbuilt Wi-
Fi. Hence data can be directly stored to cloud storage over the internet Cloud Storage that will be used for
storing data is ThingSpeak Cloud. It supports data storage along with visualizations. Once sufficient amount
of data is collected it can be exported into Excel .csv format. The data exported from ThingSpeak Cloud is
the dataset which will be used as a dataset for Machine Learning Pre-processing steps such as exploration,
cleaning and transformation of dataset will be done using Python Machine learning prediction models such as
Random Forest, Decision Tree, Naïve Bayes etc. can be used for predictive maintenance of motor and to
predict failures in motor.
3
www.viva-technology.org/New/IJRI
V. FIGURES AND TABLES
VI. CONCLUSION
This Project was to explore predictive maintenance and fault diagnosis methodologies using different deep
learning data driven techniques that can incorporate the feature selection and diagnosis in a single step,
eliminating the need of field specific expert knowledge. Machine learning methods can leverage the use of
historical data and data acquired by sensors in smart industries and allow for prediction models using raw data
without the need of industry specific knowledge and feature engineering. the development of accelerators that
would allow running machine learning natively in microcontrollers, these machine learning methods will offer
a good solution to fault detection and predictive maintenance.
Acknowledgements
We shall be failing in our duty, if we will not express our sincere gratitude to all those distinguished
personalities with the help of whom we have successfully completed our project. My deep gratitude to Dr.
Arun Kumar, PRINCIPAL, VIVA INSTITUTE OF TECHNOLOGY, who always been playing a great role
in all round development of the student. My deep gratitude to Prof. Bhushan Save, THE HEAD OF
ELECTRICAL DEPARTMENT and also our project guide Dr. Deepak Sajnekar and our project
coordinator Prof. Rahul Abhyankar for her valuable guidance, advice and constant aspiration to our work,
teaching and non-teaching staff for their kind support, help and assistance, which they extended as and when
required. Last but not the least we wish to thank my friends for providing technical and moral support. We
hope that this project report would meet the high standards of all concerned people and for their continuous
co-operation during the whole period of period of project that helped us in enhancement of this project.
REFERENCES
[1] H. M. Hashemian and W. C. Bean, “State-of-theart predictive maintenance techniques,” IEEE Transactions
on Instrumentation and measurement, vol. 60, no. 10, pp. 3480–3492, 2019.
[2] S.-j. Wu, N. Gebraeel, M. A. Lawley, and Y. Yih, “A neural network integrated decision support system
for condition-based optimal predictive maintenance policy,” IEEE Transactions on Systems, Man, and
Cybernetics- Part A: Systems and Humans, vol. 37, no. 2, pp. 226–236, 2017.
[3] E. Frontoni, R. Pollini, P. Russo, P. Zingaretti, and G. Cerri, “Hdomo: Smart sensor integration for an
active and independent longevity of the elderly,” Sensors, vol. 17, no. 11, p. 2610, 2017.
[4] B. Lu, D. B. Durocher, and P. Stemper, “Predictive maintenance techniques,” IEEE Industry Applications
Magazine, vol. 15, no. 6, 2019.
[5] B. Lu, T. G. Habetler, and R. G. Harley, “A survey of efficiencyestimation methods for inservice induction
motors,” IEEE Transactions on Industry Applications, vol. 42, no. 4, pp. 924–933, 2020. [6] W. T. Thomson
and M. Fenger, “Current signature analysis to detect induction motor faults,” IEEE Industry Applications
Magazine, vol. 7, no. 4, pp. 26–34, 2019.
[7] R. Yam, P. Tse, L. Li, and P. Tu, “Intelligent predictive decision support system for conditionbased
maintenance,” The International Journal of Advanced Manufacturing Technology, vol. 17, no. 5, pp. 383–391,
2018.
[8] J.-H. Shin and H.-B. Jun, “On condition based maintenance policy,” Journal of Computational Design and
4
www.viva-technology.org/New/IJRI
Engineering, vol. 2, no. 2, pp. 119–127, 2021.
[9] G. A. Susto, A. Schirru, S. Pampuri, S. McLoone, and A. Beghi, “Machine learning for predictive
maintenance: A multiple classifier approach,” IEEE Transactions on Industrial Informatics, vol. 11, no. 3, pp.
812–820, 2020.
[10] K. Wang, “Intelligent predictive maintenance (ipdm) system–industry 4.0 scenario,” WIT Transactions
on Engineering Sciences, vol. 113, pp. 259–268, 2016.
5
www.viva-technology.org/New/IJRI