What I Learned from Prof. Md. Jalil Piran’s Class on Writing a Research Paper

Before taking Prof. Md. Jalil Piran’s class, I thought writing a research paper mainly meant collecting papers, summarizing them, and putting citations together. The class made me realize that a serious paper is not just a collection of summaries. It needs a research gap, a clear structure, honest limitations, strong comparison, and a reason why the paper should exist.

This changed how I looked at academic writing. A paper is not only about showing that I read many references. It is about building a clear argument that helps the reader understand a field better.

First page of the manuscript: Learning-Based Forecasting, Control, Optimization, and Security in Smart Energy Systems — First page of the manuscript — click to view the full PDF.

The Biggest Lesson

The biggest lesson I learned is that every research paper must answer one question clearly: why does this paper need to exist?

A weak paper only collects information. A stronger paper explains what is missing, why that gap matters, and how the paper helps the reader see the topic more clearly.

For a survey paper, this is especially important. A survey should not just summarize papers one by one. It should compare, classify, synthesize, and explain patterns across the field. Without that, it becomes a long bibliography, not a real survey.

How the Class Changed My View of Paper Writing

Start from the research problem, not from random papers.
Read papers with a purpose.
Compare existing surveys before writing a new one.
Find the gap before claiming contribution.
Build a taxonomy instead of listing papers.
Use tables and figures to clarify the argument.
Make claims only when the references support them.
Be honest about limitations.
Do not pretend a draft is publication-ready too early.

My Manuscript Example

As part of this learning process, I worked on a structured narrative survey titled “Learning-Based Forecasting, Control, Optimization, and Security in Smart Energy Systems.” The manuscript uses smart-energy systems as the domain and studies how AI methods connect across forecasting, graph inference, control, optimization, cybersecurity, explainability, and deployment readiness.

The goal was not only to collect papers, but to organize the field through an operational-pipeline view. Instead of treating forecasting, reinforcement learning, graph neural networks, optimization, and cybersecurity as isolated topics, the manuscript asks how these methods connect inside real smart-energy systems.

What the Manuscript Covers

Forecasting

Load, photovoltaic, wind, price, and probabilistic forecasting.

Graph Learning

Topology-aware inference, graph neural networks, PMU/state estimation, and graph transfer.

Reinforcement Learning

Energy management, demand response, storage control, microgrid control, and policy learning.

Optimization

Scheduling, unit commitment, model predictive control, robust optimization, stochastic optimization, and EMS.

Cybersecurity

False-data-injection attacks, anomaly detection, fault detection, and cyber-physical monitoring.

Trustworthy AI

Explainability, uncertainty, physics-informed learning, latency, reproducibility, and deployment readiness.

“

A good research paper is not just about adding more citations. It is about creating a clear argument that helps the reader understand the field better.

The main weakness in current learning-based smart-energy research is not lack of model sophistication. The real weakness is weak operational evidence.

Mistakes I Learned to Avoid

Choosing a broad topic without a clear angle.
Writing summaries without synthesis.
Using citations without understanding them.
Adding tables that are too dense to read.
Making claims that the evidence does not support.
Treating AI-generated text as finished research writing.
Calling a draft “publication-ready” too early.
Ignoring limitations because they feel uncomfortable.
Using weak or unrelated references just to increase citation count.
Making the paper look more serious than the evidence allows.

Why Figures and Tables Matter

One thing I learned is that figures and tables are not decoration. In a strong survey paper, they should help the reader understand the structure of the field.

For my manuscript, the most useful visuals were:

A survey roadmap showing how the paper moves from scope to research requirements.
A reference audit workflow showing how papers were selected and classified.
An operational pipeline showing how data, forecasting, graph inference, decisions, physical assets, cybersecurity, and trust connect.
A task-oriented taxonomy showing the main method families.
Dataset and metric tables showing how different studies are evaluated.

These visuals made the paper easier to understand than long paragraphs alone.

The Importance of Honest Limitations

Another important lesson was that limitations do not make a paper weaker. Bad limitations make a paper weaker. Honest limitations make the paper more credible.

For example, my manuscript is a structured narrative survey, not a formal PRISMA systematic review. That distinction matters. It means the paper can still be useful, but it should not pretend to have the same completeness as a fully systematic review with exact database logs, screening counts, query dates, and exclusion records.

This kind of honesty is important because academic writing is not about sounding perfect. It is about being accurate.

Manuscript Preview

Learning-Based Forecasting, Control, Optimization, and Security in Smart Energy Systems: A Structured Narrative Survey

M. Usmanov and Md. Jalil Piran

Department of Computer Science and Engineering, Sejong University, Seoul, South Korea

Manuscript

A structured narrative survey on learning-based methods for smart-energy systems, covering forecasting, graph learning, reinforcement learning, optimization, cybersecurity, explainable AI, physics-informed learning, and deployment readiness.

View manuscript PDF Back to Writing

The most valuable part of this process was not only producing a manuscript, but learning how to think more critically about research. I learned that paper writing requires structure, patience, comparison, evidence, and intellectual honesty.

That is the skill I want to keep improving.