Many techniques deal with life forecast and failure detection of aluminum electrolytic capacitors which are utilized as a part of power electronic converters. The main idea of these techniques is to estimate the values of Equivalent Series Resistance (ESR) and Capacitance (C). Observing the ESR and C values changes of an electrolytic capacitor can provide its remaining useful life. The drawback of these methods is that they all need a prior offline phase to construct the ageing model before being able to estimate the RUL (Remaining Useful Lifetime) of the capacitor. We are working on a method that gives an estimate of the RUL without the need of the offline phase. This, nonetheless, requires the knowledge of the shape of aging laws. The parameters of the laws will be determined automatically online. In order to know the shape of these laws, accelerated aging tests are set up to test the effect of the operational conditions (temperature, voltage, current) on the aging of the capacitor. Early results show that a cubic regression has the best fit with the experimental aging data.