Artificial intelligence (AI) has been successfully applied to find solutions to problems in a large array of fields. At its core, machine learning serves to develop statistical models that are able to analyze or predict the behavior of a given system. This study aims to make use of AI techniques to monitor the health condition of the cooling system of a variable speed drive. To do so, measurements are collected from a physical testing setup at different operating conditions, and these data samples are then used to train artificial neural networks. Two main approaches to this condition monitoring are detailed in this thesis, developing models to predict the behavior of the system at future times and to directly derive the value of a health indicator. This thesis also aims to detail the modeling of the drive and its cooling system and to provide a theoretical background on artificial neural networks.