Optimizing structural integrity of a pressure vessel via finite element analysis and machine learning based XGBoost...

Pressure vessels are essential in many sectors, notably the oil and gas industry, since they undergo harsh circumstances. Under pressurization, these containers accumulate strain energy, which can cause the material to weaken and ultimately break down, particularly in ductile materials. Accurately estimating the burst pressure of pressure vessels is critical for maintaining structural integrity and safety. This study offers a machine learning (ML) approach using existing literature data for estimating burst pressure that takes crucial variables, including yield strength, ultimate strength, inner diameter, material type, and thickness, as input features. This work emphasizes how integrating Finite Element Analysis (FEA) with sophisticated ML approaches such as XGBoost may improve the precision and reliability of burst pressure estimations based on error metric calculations. The XGBoost model's predictions were extensively tested against data obtained from FEA, a well-known approach for determining structural integrity under extreme circumstances. The results showed that the XGBoost model is more accurate and robust across different materials, proving that it is a good predictive tool when contrasted with conventional approaches that use mathematical equations. This is due to the XGBoost model's capacity to include several characteristics concurrently, which takes into consideration complex interdependencies. This study adds to creating a complete framework for forecasting burst pressure, which is critical for enhancing engineering practices and safeguarding assessments in pressure vessel design and operation. The findings from this study may be applied to various structural safety scenarios, allowing for the possibility of new inventive solutions in engineering visualization and safety evaluation. The developed approach substantially decreases computational capacity relative to traditional FEA simulations while preserving high precision. This establishes the model as a realistic technique for sectors requiring timely and reliable forecasts of burst pressure, including oil and gas, energy, and industries.