As scientists and engineers improve the durability of photovoltaic (PV) modules, the inverter system, which converts the harvested energy into power usable by the grid, becomes the next emerging challenge in the reliability of the PV system. Addressing these issues represents a difficult task to both the PV and power systems communities for the following reasons:
- The inverter system must continually perform a far greater number of switching operations over its lifetime than other high-power counterparts in other industries.
- There is a large diversity in the operating environment of the inverter system, many of which are much harsher than those associated with traditional power applications.
- The overall system lifetime cost must decrease for photovoltaics to be financially viable. Currently the cost and reliability of PV modules is improving faster than the rest of the system. One way to improve system costs is by improving inverter reliability. While PV modules have been rated for 30 years, inverters are typically rated for 5-10 year lifetimes. In the field, inverter failures and subsequent replacement rates have been as often as every 2 years.
- The inverter is a complex system consisting of control software and many components, each of which can fail.
The objective of the inverter reliability program is to develop testing standards, operating standards, and monitoring techniques for inverters with the goal of developing inverter reliability models to decrease PC system cost and increase energy availability. We achieve this objective by accomplishing the following goals:
- Goal 1: Reliability Characterization – Characterize and study the failure modes, failure mechanisms and degradation on a component or subsystem level through simulation and accelerated testing. Utilize the results to develop testing and operating standards.
- Goal 2: Prognostics and Health Management – Develop and implement prognostics and health management (PHM) techniques through a combination of recognizing degradation signatures, collecting laboratory and field data, and predictive modeling.