Project Description
Condition Based Maintenance (CBM) uses the sensor to collect real-time measurements (ie. pressure, temperature, and vibration). CBM data allows maintenance personnel to perform maintenance at the exact moment it is needed, prior to failure. This can be done by Anomaly Detection to detect machine failures. There are some types of anomaly detection [1-3]. In this project, using Data Mining. This approach can give patterns that are more informative and easier to understand. The patterns can be used further to trigger alerts for alerting machine failures. On the other hand, this approach is also quite light and does not require much time. So, it is considered suitable for solving anomaly detection problems in small datasets, such as the used dataset.
Dataset Description
The dataset is vibration sensor readings from NASA Acoustics and Vibration Database. sensor readings were taken on four bearings that were run to failure under constant load and running conditions. The vibration measurement signals are provided for the datasets over the lifetime of the bearings until failure. Failure occurred after 100 million cycles with a crack in the outer race.
Method
The idea behind this approach is adopted from the proposed framework in [4], but simpler and customizable according to the given dataset. The first step is to discretize the value of each sensor which indicates the sensor is in normal or failure condition. Then the results are clustered to find out the condition of the machine so that in the end a pattern can be obtained to give an early warning when the machine experiences signs of failure.
Results
Rules
Rules obtained for machine failures:
- Bearing 1: Failure, Bearing 2: Failure, Bearing 3: Failure, Bearing 4: Failure -> Machine Failure Alert!
- Bearing 1: Normal, Bearing 2: Failure, Bearing 3: Failure, Bearing 4: Failure -> Machine Failure Alert!
- Bearing 1: Normal, Bearing 2: Failure, Bearing 3: Failure, Bearing 4: Normal -> Machine Failure Alert!
- Bearing 1: Failure, Bearing 2: Failure, Bearing 3: Normal, Bearing 4: Normal -> Machine Failure Alert!
- Bearing 1: Normal, Bearing 2: Failure, Bearing 3: Normal, Bearing 4: Failure -> Machine Failure Alert!
Bearing Status
From Bearing Status:
- It can be seen that Bearing 1 is the first sensor that indicates failure. It doesn’t, making the machine fail immediately. But, it might be considered as the root cause of machine failure.
- Machine start failure when Bearing 1 and Bearing 2 fail.
Time Interval
By using the generated rules, it gives a time interval of 23 hours and 20 minutes (since the first alert) before the machine is completely broken.
References
[1] https://bhanushahi.medium.com/anomaly-detection-a7f28ccd5936
[2] https://medium.com/analytics-vidhya/anomaly-detection-in-python-part-1-basics-code-and-standard-algorithms-37d022cdbcff
[3] https://towardsdatascience.com/how-to-use-machine-learning-for-anomaly-detection-and-condition-monitoring-6742f82900d7
[4] Palembiya R.A., Setiawan M.N., Gultom E.O., Prayitno A.S.D., Kurniati N., Iqbal M. (2021) A Smart Predictive Maintenance Scheme for Classifying Diagnostic and Prognostic Statuses. In: Mohamed A., Yap B.W., Zain J.M., Berry M.W. (eds) Soft Computing in Data Science. SCDS 2021. Communications in Computer and Information Science, vol 1489. Springer, Singapore. https://doi.org/10.1007/978-981-16-7334-4_8