Berlin, June 11, 2012 — Solar power will play a key role in tomorrow's energy mix in Germany. In order for the sector to grow and to ensure a good return on investment, the reliability of photovoltaic power plants is critical. Further development of PV power electronics will help improve overall plant reliability.
Constant load fluctuations burden the power electronics of solar inverters in PV plants. This increases the risk of sudden plant failure. Up to now it was almost impossible to safely predict the service life of expensive power semiconductors.
GE, together with Fraunhofer Institute for Reliability and Microintegration IZM, a method to forecast the remaining lifetime of power electronics is being developed. The first prototype of this measurement system will be presented in Berlin in September 2012.
The achievements so far of the still ongoing joint CoMoLeFo project (Condition Monitoring for Power Electronics in Photovoltaics) are identification and detection of relevant aging mechanisms of Insulated Gate Bipolar Transistor power modules.
IGBTs are switch elements designed for high-power levels and constitute the central component of modern inverters. Mathematical algorithms and measuring data are now able to make precise assessments concerning the condition of semiconductors during operation.
This enables a forecast of the remaining lifetime of an IGBT and therefore conditions based maintenance. Reliability is improved and hence, the operating time can be increased significantly while expenses caused by unscheduled downtimes and consequences thereof are reduced.
CoMoLeFo is based on two complementary processes: One process indirectly determines the chip temperature by measuring relevant electrical parameters of the IGBT during operation. In order to evaluate the resulting load exposure to the system, the temperature data is applied to a physics of failure model.
The other process monitors the parameter shift due to aging. In this way, the remaining lifetime of the IGBTs can be determined by a software based remote diagnostic system. The information can be forwarded to the customer and the service company.