Researchers from Spain’s Universidad Politécnica de Madrid (UPM) and the Massachusetts Institute of Technology (MIT) have created a probabilistic model to assess the reliability of PV modules and their warranties.
They described their findings in “Warranty assessment of photovoltaic modules based on a degradation probabilistic model,” which was recently published in Progress in Photovoltaics. The model can calculate the probability of failure in a product during the warranty period based on degradation data.
The model evaluates the number of products that fail within the warranty period and the failure times. The scientists said the distribution of returned modules in the warranty period tells how many elements fail, and also when they fail.
“To know when the modules fail in the warranty period is very important for logistic purposes,” researcher Manuel Vazquez Lopez told pv magazine.
The scientists explained that the exact degradation failure times and the associated failure rate function covered by warranties usually depends on the specific panel degradation parameters and the sales function, which corresponds to the time in which the solar module begins to operate.
“As it is known the main failure mechanism of PV modules is degradation, PV power module decays below a failure limit,” Vazquez Lopez explained. “The model assesses the number of products that fail during the warranty period and when they failed, based on PV modules degradation parameters in the field.”
The U.S.-Spanish group was able to determine that it is impossible to fulfill a 25-year warranty if the threshold of returned modules exceeds 5% and the annual degradation rate is over 0.73%. Under such a scenario, around 50% of the modules fail during the warranty period, with power yield dropping in average by approximately 20% – a situation that the academics describe as unacceptable for any kind of product.
They also said that, depending on the portion of failures allowable during warranty periods, a degradation rate below 0.57% per year is the minimum required to bring the threshold of returned modules under 5%, while a value under 1% can only be reached with a rate of 0.47% per year.
The proposed model can be applied to any real case, the researchers concluded.
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: firstname.lastname@example.org.