(345p) Machine Learning Assisted Prediction Modelling for Pressure Vessel Design and Stress Analysis.

The importance of Pressure Vessel (PV) to industries is one of the reasons why the design and structural integrity should be fully understood and considered when deploring it in under different conditions. The design of such vessels needs to be broadened with a detailed thermal stress due to its time-dependent different behaviours experienced under load. Therefore, this study aimed at investigation of transient analysis of PV when subjected to different operating condition.

The PV used for this simulation was designed based on standards and subjected to transient-stress analysis (transient thermal and structural) using ANSYS software. A complete evaluation of temperature, heat flux and resulting stress distribution across the vessel was estimated at four different locations within the designed PV and the obtained result was compared with analytically obtained results from appropriate standards. To evaluate machine learning techniques in prediction of Transient and heat flux analysis of pressure vessel designs, in this paper, a new model based on Prophet Forecast Model (PFM); a time series analysis tool has been proposed and developed. Input parameters of the model include transient thermal values through the whole vessel, total heat flux values through the whole vessel. The obtained results showed that the constructed model has a good potential to be used as more applicable model compared to current models in design of pressure vessels.