Priyadharshini Ponraj

Completed a 5-month site training experience as an Instrumentation Engineer at the PTA Project, Paradip, IOCL. Gained hands-on exposure to field instrumentation activities including loop checking (cold & hot) for PT, FT, and control valves using a 475 communicator, and worked with both HART and Foundation Fieldbus communication protocols. Involved in hydrotesting of impulse lines, ensuring proper pressure testing and leak checks, and participated in clearing punch points raised by IOCL during inspections. Developed practical understanding of calibration techniques for pressure gauges and transmitters using pressure comparators and multifunction calibrators. Observed installation practices such as impulse line fit-up, cable laying, glanding, termination, and verification of instruments as per P&ID. Explored a wide range of field instruments including DP transmitters, radar and ultrasonic level transmitters, vortex and magnetic flow meters, diaphragm-type pressure transmitters, flame detectors, and control valves. Gained exposure to control systems by visiting the Central Control Room (CCR), understanding engineering and operator workstations, interface cabinets, and panel erection activities. Also learned about cable management systems, including FO and power cables. Additionally, observed safety systems such as flame detection and deluge valve operation, and understood real-time industrial practices followed in site environments. This experience significantly strengthened my practical knowledge and provided a strong foundation in industrial instrumentation and control systems.

Designed and developed an IoT-based system to monitor crowd density and movement in real time, aimed at improving public safety and proactive crowd management. The project utilized passive infrared (PIR) sensors to accurately detect the presence and flow of individuals in a given area. By capturing and processing real-time data, the system provided meaningful insights into crowd behavior and congestion levels. The collected data was analyzed using basic data analytics techniques to identify potential safety risks and support timely decision-making. This low-cost, scalable solution is well-suited for applications in public gatherings, events, and urban spaces, highlighting the effectiveness of integrating IoT and embedded systems for smart monitoring and control.

The Piping and Instrumentation Diagram (P&ID) project involved creating detailed schematics for an industrial process plant using Siemens COMOS software. The diagrams accurately represented piping layouts, actuator placements, controller configurations, and valve arrangements to ensure optimal control and safety. This project deepened my understanding of industrial process flow, instrumentation standards, and the critical role of P&IDs in plant design and operations. Through this experience, I gained hands-on skills in using COMOS for engineering documentation and learned how to interpret and draft industry-standard diagrams. I also developed a stronger grasp of how instrumentation components interact within a system and how precise documentation supports safe and efficient plant maintenance and troubleshooting. This project significantly strengthened my technical knowledge in process automation.

Completed a Machine Learning with Python internship where I worked on real-world datasets, applied supervised and unsupervised learning, and performed data preprocessing and analysis. Gained hands-on experience with Python libraries like NumPy, Pandas, and Scikit-learn, and learned to evaluate and optimize model performance effectively.

Developed a predictive model to estimate brine evaporation rates in salt production ponds using advanced machine learning techniques. The project involved collecting and preprocessing environmental and process data, selecting suitable ML algorithms, and optimizing the model to enhance accuracy and reliability. Key Contributions: Built and validated a data-driven model that significantly improved the prediction of brine evaporation performance. Applied supervised learning techniques to analyze complex environmental variables affecting evaporation. Demonstrated the effectiveness of integrating domain knowledge with data science for real-world industrial applications.