The gas turbine is the engine at the heart of the power plant that produces electric current. A gas turbine is a combustion engine that can convert natural gas or other liquid fuels to mechanical energy. This energy then drives a generator that produces electrical energy.

This is a Machine Learning project on Gas Turbine Compressors and Turbine decay evaluation for Propulsion Plants.

To build Machine learning models that can predict the GT Compressor decay state coefficient.

Achieved an RMSE(root-mean squared error) score of 0.000844 for our Compressor model.

• report.html: The Exploratory Data Analysis report generated with Pandas-Profiling. Result of which was used on narrowing down on model.

• siem_bijon_guha_submission.ipynb: The Jupyter notebook where 7 different ML models were evaluated for the GT Compressor decay state coefficient.

• CatBoostRegressor_GT_Compressor.pickle: Pickle file of the CatBoost model for Compressor

• Model_Comparison_GT_Compressor.csv: CSV file that contains the comparison chart of all the models tested for the Compressor decay state coefficient

Python version : 3.7 Packages Used: pandas, numpy, matplotlib, seaborn, sklearn, xgboost, lightgbm, catboost

Seven different Machine Learning models were used for this project.

• Linear regression

• Random Forest

• K Nearest neighbours

• Support vector machine

• XGBoost

• Light GBM(LGB)

• CatBoost

Top 5 features :

• Gas Turbine shaft torque (GTT) [kN m]
• GT Compressor outlet air temperature (T2) [C]
• Gas Generator rate of revolutions (GGn) [rpm]
• HP Turbine exit pressure (P48) [bar]
• Gas Turbine exhaust gas pressure (Pexh) [bar]

CatBoost model performs exceptionally well. It is fast, efficient and simple to implement as well.