To Ease Our Internship Experience When We Created Our First Software

Introduction

My batchmate and I had the opportunity to work on a fascinating problem during our summer internship. We were students of Chemical Engineering at IIT Madras, and our summer internship lasted for about six weeks. At one point, we decided that writing our own software would ease our work. As part of the internship program, we created our first software!

Background

Our internship took place at a power plant, where we were tasked with reconciling the oil quantification process. The plant, located in a remote area of Dharmapuri District in Tamil Nadu, used LSHS (Low Sulphur Heavy Stock) fuel for generating electricity. Diesel generators burned this fuel to produce power.

Problem Statement

Our job was to reconcile the oil calculation. Oil arrived at the plant via tankers and was then burned in diesel engine generators to produce electricity. The power generation company was suffering significant losses due to discrepancies between the oil received and the oil consumed. These discrepancies were costing the company 3-5 crore rupees annually.

Initial Investigation

During our internship, we were shown the entire process. We observed how oil in tankers was measured, how samples were collected to assess fuel quality, and how oil was transferred to large storage tanks. We learned how oil quantities were measured in these tanks, and how it was transported through intricate pipelines before being measured again at the diesel generators.

The first thing we did was check for leaks in the pipeline. It seemed like an obvious solution, and our technical supervisor agreed to let us inspect the entire pipeline with a team. We also increased the number of sample sizes to improve the accuracy of our measurements.

Despite these efforts, we found no definitive improvements in the measurements.

Technical Challenges

Halfway through our internship, we had a review meeting with the technical head of the plant. He informed us that he had already carried out most of our suggested studies. We then asked if other plants faced the same problem, and learned that some government companies were also losing millions due to improper oil quantification.

Seeking Guidance

Frustrated with the remote setting and slow internet, we struggled to find information on addressing these problems. During a weekend break, we returned to IITM and sent a long email with all the details to our professor, hoping for guidance. We impatiently waited for our professor's reply, and when it came after three days, it was a simple one-liner: "Do error analysis."

The Breakthrough

Initially, this response seemed insufficient, given the complexity of our issues. However, we soon realized that error analysis was indeed crucial. Measurements of temperature, density, specific gravity, and other parameters were essential for accurate oil quantification.

Implementing Error Analysis

The tanks stored oil, measured at three different heights for parameters crucial for the mass calculation. Large tank calibration charts were available as Excel sheets were referred to for computing the oil quantities. Managing these calculations and conducting error analysis was challenging. We performed error analysis meticulously, identifying which parameters most affected mass calculation.

Software Development

My batchmate excelled at programming, and I was strong in the domain part. To solidify our recommendations, we decided to write software in VB.net. We developed algorithms to determine the impact of various parameter miscalculations on the overall mass.

Completing the Software

We completed the software in just two weeks. It quickly established that temperature played the most crucial role. We researched more accurate temperature measurement devices and recommended the company purchase them. Our algorithm proved more accurate than the company's existing methods, and they adopted our suggestions.

Conclusion

Creating great software requires expertise from different fields. My domain knowledge and my friend’s programming skills allowed us to conduct a comprehensive study and develop an effective solution. Both of us learned from each other and benefited from the experience, as did our client.

Software Creation Details

We focused intensely on the problem at hand, given the limited time. We read data from multiple Excel files and processed it for quantification, reducing manual calculations. We automated plotting to visualize insights and kept the algorithm flexible for quick adjustments.

Without being experts in the craft, we created an impressive piece of software. We proudly named it DG Software ( Dwij C. and Gaurav Gaur ) and provided knowledge transfer to our technical boss for future use.

Reflection

This was a great first software creation experience. Today, with so many new technologies, design patterns, and algorithms available, it’s easy to over-engineer solutions. Starting simple, as we did, can be more practical.

Have you had any such experience of writing quick software that gave you immense benefits in a short period? Please share in the comments.

#InternshipExperience #SoftwareDevelopment #ChemicalEngineering #IITMadras #PowerPlant #LSHS #ErrorAnalysis #Programming #VBNet #Collaboration #SoftwareEngineering #EngineeringStudents #InnovativeSolutions #Quantification #EngineeringInternship #Teamwork #CGM