PIPESIM-Generated Data for PINN-Based Prediction of Bottomhole Pressure and Temperature in CO₂ Injection Wellbores

This dataset supports the development and validation of a physics-informed neural network-based hybrid approach for predicting bottomhole pressure and bottomhole temperature in CO₂ injection wellbores. It contains two components: (1) a simulator-generated dataset comprising 696 operating cases produced using a PIPESIM model of an offshore CO₂ transportation and injection network and used to train, validate, and test the simulator-trained PINN; (2) a dataset generated using a PIPESIM model configured to represent the Shell Quest CO₂ injection system under varying injection conditions.