During our talk, we addressed common challenges associated with navigating real-world gas turbine operations data. We introduced fundamental strategies from data science that are applicable to performance optimization and root cause analysis.

By leveraging these strategies, the audience can improve operational efficiency and more effectively identify the underlying factors contributing to performance issues. This approach enables them to make data-informed decisions and drive targeted improvements in their operations. We explored practical applications, helping them unlock the untapped potential of their organization’s data. The talk was broken into four main sections, with Python examples throughout:

• What is Data Science?

Data scientists work with subject matter experts to answer important questions. We discussed what data scientists are and how they can assist an organization.

• Obtaining and Storing Data

The first step in any data-informed analysis is collecting and storing data. In this section, we discuss how this is often done and how to avoid some common pitfalls.

• Preparing Data

Here we explored how to prepare data for analysis, a process often done by labeling and filtering the data. We also introduced code that can generate standard features for gas turbine data, such as steady-state vs transient and baseload vs part-load operation periods.

• Analyzing Data

In this section, we discussed strategies to utilize properly-prepared data and efficiently create figures for various stakeholders using Python.

How can you access the materials?

The slides and sample code are publicly available. Please Contact Us for the files.