Reliability assessment of solar photovoltaic systems based on fuzzy fault tree analysis

Solar photovoltaic (SPV) systems are widely used as a renewable energy source to produce electricity to endusers. SPV system convert free and unlimited sunlight into electricity without carbon dioxide emission or any other air pollutants. The stochastic nature of SPV system may be responsible for degradation of performance in terms of availability, production, etc. Therefore, the assessment of the performance measures like reliability and quality of its components is vital concern that may open the new vista of research. There are various reliability assessment techniques used in conventional (thermal power plant, hydropower plant, nuclear power plant, etc.) and non-conventional power system (SPV power plant, wind power plant, etc.) like Fault Tree Analysis (FTA), Failure Mode and Effect Analysis (FMEA), etc. FTA is one of the powerful reliability assessment techniques, which gives modes of the failure and its consequences, that are demonstrated on SPV systems. FTA can further be broadly classified into two: qualitative and quantitative analysis. Qualitative FTA can be solved by determining a minimal cut set (MCS) and quantitative FTA can be solved by calculating the failure probability of each the basic event. In conventional FTA failure probability of each event must be known, to assess for quantitative FTA. However, to quantify probabilistic failure, data of all basic events seem to be inadequate. To overcome this problem, conventional FTA may be integrated with fuzzy linguistic scale method. In this research work, failure possibilities in terms of a linguistic variable of unknown failure data for all basic events, which lead the Top Events for SPV systems are assigned by the experts from academics, research, industries and practicing engineers. For Aggregated fuzzy number based on expert’s opinion, a consistency agreements method is used. Fussell-Vesely (F-V) importance measures is also implemented to rank the basic events and MCS to apportion the most critical event.