Massive amounts of data generated by fusion machines (such as JET) require developing automatic methods for data analysis. Edge-localized modes (ELMs) are instabilities occurring in the edge of H-mode plasmas. The aim of this work is to develop an automatic off-line method for identifying and locating ELMs. This method uses Universal Multi-Event Locator (UMEL) as the event locator. The combination of information from D emission and diamagnetic energy allows the recognition of single ELMs. This paper shows the way in which waveforms of a wide range of discharges can be treated and how UMEL is applied in order to identify and locate ELMs independently of the signal amplitudes. A large database of more than 1200 discharges has been used to test the performance of the method obtaining 226 751 ELMs.