Analyzing The Impact of Design Factors on Solar Module Thermomechanical Durability Using Interpretable Machine Learning

Solar modules in utility-scale PV systems are anticipated to maintain a prolonged lifetime to effectively rival conventional energy sources. However, the durability of these modules is often compromised by cyclic thermomechanical loading, emphasizing the need for a proper module design to counteract the detrimental effects of thermal expansion incompatibility. Given the complex composition of solar modules, isolating the impact of individual components on overall durability remains a challenging task. In this work, we constructed a comprehensive data set, capturing bill-of-materials and post-thermal cycling power loss from over 250 distinct module designs. Utilizing the data set, we developed a machine learning model to correlate the design factors with the degradation and applied the Shapley additive explanation to provide interpretative insights into the model's decision-making, investigating the impacts of design factors like the busbar number, wafer thickness, and others. Our analysis reveals that the type of silicon solar cell, whether monocrystalline or polycrystalline, predominantly influences the degradation, and monocrystalline cells present better durability. This finding was further substantiated by statistical testing on our raw data set. We also demonstrate that the thickness of the encapsulant remains another important factor, with thicker encapsulants correlated with reduced power loss. The study here provides a blueprint for utilizing explainable machine learning in an intricate material system and can potentially steer future research on optimizing solar module design.