abstract: The main objective of this research is to develop reliability assessment methodologies to quantify the effect of various environmental factors on photovoltaic (PV) module performance degradation. The manufacturers of these photovoltaic modules typically provide a warranty level of about 25 years for 20% power degradation from the initial specified power rating. To quantify the reliability of such PV modules, the Accelerated Life Testing (ALT) plays an important role. But there are several obstacles that needs to be tackled to conduct such experiments, since there has not been enough historical field data available. Even if some time-series performance data of maximum output power (Pmax) is available, it may not be useful to develop failure/degradation mode-specific accelerated tests. This is because, to study the specific failure modes, it is essential to use failure mode-specific performance variable (like short circuit current, open circuit voltage or fill factor) that is directly affected by the failure mode, instead of overall power which would be affected by one or more of the performance variables. Hence, to address several of the above-mentioned issues, this research is divided into three phases. The first phase deals with developing models to study climate specific failure modes using failure mode specific parameters instead of power degradation. The limited field data collected after a long time (say 18-21 years), is utilized to model the degradation rate and the developed model is then calibrated to account for several unknown environmental effects using the available qualification testing data. The second phase discusses the cumulative damage modeling method to quantify the effects of various environmental variables on the overall power production of the photovoltaic module. Mainly, this cumulative degradation modeling approach is used to model the power degradation path and quantify the effects of high frequency multiple environmental input data (like temperature, humidity measured every minute or hour) with very sparse response data (power measurements taken quarterly or annually). The third phase deals with optimal planning and inference framework using Iterative-Accelerated Life Testing (I-ALT) methodology. All the proposed methodologies are demonstrated and validated using appropriate case studies. / Dissertation/Thesis / Doctoral Dissertation Industrial Engineering 2020
| Identifer | oai:union.ndltd.org:asu.edu/item:57346 |
| Date | January 2020 |
| Contributors | Bala Subramaniyan, Arun (Author), Pan, Rong (Advisor), TamizhMani, GovindaSamy (Advisor), Montgomery, Douglas (Committee member), Wu, Teresa (Committee member), Kuitche, Joseph (Committee member), Arizona State University (Publisher) |
| Source Sets | Arizona State University |
| Language | English |
| Detected Language | English |
| Type | Doctoral Dissertation |
| Format | 106 pages |
| Rights | http://rightsstatements.org/vocab/InC/1.0/ |
Page generated in 0.0063 seconds