Referring to the Guide to the Expression of Uncertainty in Measurement (GUM), the paper proposes a theoretical contribution to assess the uncertainty interval, with relative confidence level, in the case of n successive observations. The approach is based on the Chi-square and Fisher distributions and the validity is proved by a numerical example. For a more detailed study of the uncertainty evaluation, a model for the process variability has been also developed.

Photovoltaic (PV) power optimizers are introduced in PV systems to improve their energetic productivity in presence of mismatching phenomena and not uniform operating conditions. Commercially available converters are characterized by different DC-DC topologies. A promising one is the boost topology with its different versions. It is characterized by its circuital simplicity, few devices and high efficiency values - necessary features for a Distributed Maximum Power Point Tracking (DMPPT) converter. PV power optimizer designs represent a challenging task since they operate in continuously changing operating conditions which strongly influence electronic component properties and thus the performance of complete converters. An aspect to carefully analyze in such applications is the thermal factor. In this paper, a necessity to have a suitable temperature monitoring system to avoid dangerous conditions is underlined In addition, another important requirement for a PV power optimizer is its reliability, since it can suggest a useful information on its diagnostic aspects, maintenance and investments. In fact, a reliable device requires less maintenance services, also improving the economic aspect. The evaluation of the electronic system reliability can be carried out using different reliability prediction models. In this paper, reliability indices, such as the Mean Time Between Failure (MTBF) or the Failure Rate of a Diode Rectification (DR) boost, are calculated using the evaluation of the Military Handbook 217F and Siemens SN29500 prediction models. With the reliability prediction results it has been possible to identify the most critical components of a DMPPT converter and a measurement setup has been developed in order to monitor the component stress level on the temperature, power, voltage, current, and energy in the DMPPT design phase avoiding the occurrence of a failure that might decrease the service life of the equipment.