​Electricity feed-in tariffs are decreasing, cutting into solar PV plant operators' margins. In this context, the ability to detect and identify loss-generating faults on the plants' equipment has become critical. To address this issue, and at the request of certain operators, Liten, a CEA Tech institute, is developing an algorithm capable of identifying and measuring sources of losses, from the collection of sunlight to the injection of electricity into the grid.

The researchers started by taking measurements on around 20 instrumented power plants. They are using the measurements to develop models and learning methods to assess as accurately as possible any losses likely to be generated by the PV modules, cabling, converters, and other components. The software will be tested on PV plants with different configurations to ensure that it is robust before it is made available to plant operators. Once it is installed, the software will display "loss diagrams" that can be used to provide an accurate assessment of plant operation.

Changes in the loss diagrams over time will also give an indication of degradation to plant equipment beyond the expected natural aging. Ultimately, the software will help improve plant LCOE (levelized cost of energy), the technical and financial indicator used to determine the economic viability of a plant.