<p>Development of complex aerospace systems often takes decades of research and testing. High performing propellants are important to the success of rocket propulsion systems. Development and testing of new propellants can be expensive and dangerous. Full scale tests are often required to understand the performance of new propellants. Many industries have started using data science tools to learn from previous work and conduct smarter tests. Material scientists have started using these tools to speed up the development of new materials. These data science tools can be used to speed up the development and design better propellants. I approach the development of new solid propellants through two steps: Prediction of delivered performance from available literature tests, prediction of ideal performance using physics-based models. Random Forest models are used to correlate the ideal performance to delivered performance of a propellant based on the composition and motor properties. I use Parsimonious Neural Networks (PNNs) to learn interpretable models for the ideal performance of propellants. I find that the available open literature data is too biased for the models to learn from and discover families of interpretable models to predict the ideal performance of propellants. </p>
| Identifer | oai:union.ndltd.org:purdue.edu/oai:figshare.com:article/19651167 |
| Date | 29 April 2022 |
| Creators | Peter Joseph Salek (12455760) |
| Source Sets | Purdue University |
| Detected Language | English |
| Type | Text, Thesis |
| Rights | CC BY 4.0 |
| Relation | https://figshare.com/articles/thesis/Prediction_of_Delivered_and_Ideal_Specific_Impulse_using_Random_Forest_Models_and_Parsimonious_Neural_Networks/19651167 |
Page generated in 0.0027 seconds