In the context of the European project Bioinspired Electroactive Aeronautical Multiscale Live-Skin “BEALIVE” (http://smartwing.org/BEALIVE/EU/) and the national French research project ANR-EMBIA, a ROM is proposed for turbulent flows surrounding morphing wings. A Reduced Scale A320 wing prototype has been employed in this work at a Reynolds number and high incidence angle of 10 degrees. The wing is deformed during time using a traveling wave concept, trailing-edge vibrations [1,2], or cambering [3]. This leads to a displacement of the coordinates as a function of time and requires an adaptation of the mesh. As consequence, large parametric studies with various parameters require time and resources. To tackle this issue, a ROM is developed based on the Proper Orthogonal Decomposition and Machine Learning. The ROM is constructed using an enriched POD of multiple simulations, an interpolation is conducted using a trained Machine Learning algorithm for better prediction. The new model not only accounts mean solutions but predicts transient dynamics like boundary layer separation, shear layer, and coherent wake instabilities. This is achieved while maintaining and capturing correct energy distribution of turbulent structures and considering smaller energy chaotic structures by using spatial modes.