The Morton Effect (ME) is a thermal-fluid–structure interaction instability occurring in rotating machinery supported by hydrodynamic journal bearings. The mechanism of ME consists of a bowed rotor or mass imbalance induced shaft synchronous whirl vibration in the bearing, which causes local, asymmetric heating of the journal, which causes shaft bending, potentially leading to increasing vibration. This study presents an original ME simulation approach that includes a CFD (Computational Fluid Dynamics) bearing groove model, embedded in a deep learning algorithm for computational efficiency and non-expert usage. The groove model provides a 2D oil temperature distribution at the leading edge of the bearing pads, yielding a more accurate journal axisymmetric temperature distribution which is the source of the ME. The paper provides validation of the approach by test result correlation, and illustrates the effects of parameter variation and configuration variation, by examining various oil injection types. The approach may be used for correcting ME occurring in existing machinery, or for designing machinery to avoid the ME.