Applying AI and Machine Learning concepts at work can help you power ahead your career. Thus, taking up an upskilling course will help learn the required skills for you to power ahead. Read to learn more about Anuja Namboodiri’s journey with Great Learning’s PGP Artificial Intelligence and Machine Learning Course in his own words!
I am Anuja, and I hold a bachelor’s degree in Power Electronics and a master’s in Electrical Engineering. I have three years of experience in the Electrical Engineering domain, specifically in Electric drives. I am a part of the analytics team that provides varied analysis/insights to predict maintenance needs, operational risks, and possible system faults. It will allow equipment to perform better and last longer-keeping production running as planned.
My role as an R&D Engineer required me to do case studies involving failures/mal operations in electric drives using the real-time signal data and various failure events recorded by the IoT device. The outcome from the case studies is further used to develop algorithms that can help to predict/identify abnormal drive operations in the future surrounding the same kind of faults.
If I highlight a few of the struggles I faced as a domain engineer in an analytics team before joining the course, it will include – visualization of the data in the first place, then understanding the data patterns and their importance in finding a solution to the problem in hand. Additionally, there was this very common fear of ‘python programming’ for someone who is not from a programming/computer science background.
I tried my hands on self-learning, but the lack of a proper structure and the overwhelming resources of information, both reliable and not reliable ones, instigated me to apply for the course. Just a few months into the course, and I could apply all that I learned to one of the case studies that arrived on my table. This time unlike my previous case studies, I could better understand the data and implement the POC by applying the machine learning knowledge I had acquired. The problem statement required me to differentiate the instances when a machine had an abnormal operation than as expected. For the first time, I could go beyond my comfort zone of doing just the case study to create a POC and also contribute to developing an algorithm that is now ready to be deployed.
I have been able to use the concepts of Statistics and Regression that I learned through the course of the program. Additionally, I have been able to ramp up the time required for case study thanks to the improved python programming skills, exhaustive practice from the projects in the EDA (Exploratory Data Analysis), and feature engineering, which was a major part of the pre-study while understanding and preparing the data for algorithm development. Also, I have been able to better visualize my intermediary as well as final analysis results.
Knowing the concepts that I learned during the course helped me to identify the algorithm that would be the best match for the use case I had in hand.
I could successfully apply the algorithm to real-time data and differentiate between normal and abnormal drive operations. I have successfully been able to expand my spectrum of skills in 6 months, and I have been able to apply what I learn by contributing to the analytics team. This is just the beginning, and with continuous learning and the application of what I learn, I am sure I will be able to contribute well in the electrical domain.
