Abstract
This paper presents a systems level approach to reliability analyses for solar photovoltaic systems and illustrates how these analyses can be useful in photovoltaic systems design even though detailed reliability data are not yet available. Traditional reliability techniques and trade-offs are discussed and relationships between reliability, maintainability and availability are reviewed using minimum life cycle cost as the parameter to be optimized. A list of available techniques which are useful today is presented, as well as a list which provides a series of techniques that will be useful when more field and test reliability data on photovoltaic systems are available. Three typical solar photovoltaic systems are illustrated in block diagram form and discussed. These are (1) a stand-alone photovoltaic system, (2) a stand-alone system with battery energy storage and (3) a system with a utility back-up. Each system is analyzed to the subsystem level using two reliability analysis techniques: the failure modes and effects analyses (FMEA) and the fault tree approach. The FMEA is an approach which starts at the lowest level of the system being analyzed and considers each subsystem failure one at a time and two at a time, and traces upward through the block diagram to identify the effect these failures have on the total system performance. A summary of these results gives the designer an indication of the critical subsystems. The fault tree analysis is a logic diagram approach which begins with a failure of the system at its highest level to be the primary output of the diagram. Inputs to the analysis are made from those levels which can cause failure at the highest level. Thus, only subsystems which can cause system failure are included in the diagram. If probabilities of failure are available for the various subsystems, fault trees may be used to construct a probability analysis of the system to predict the probability of success or reliability of the photovoltaic system involved. In this paper, this analysis is performed using assumed values of probability of success for the subsystems considered. This discussion presents the view that taking reliability and availability into consideration early in the design at the system level will prevent some major field problems. It will also favorably influence the subsystem design goals, especially the control subsystems. In addition, this approach should encourage the early development of complete photovoltaic systems that satisfy the demanding availability precedent set by the electric utilities.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have