This paper suggests a new structure based on pessimistic multi-granulation rough sets and Pythagorean fuzzy sets using soft relations to evaluate and select energy supply projects. A dependable energy supply in times of crisis requires a decision-making strategy that can maintain uncertainty, incomplete information, and high-risk situations. This is because the granularity can be variable through the incorporation of soft relations, which support the varying level of information accuracy, and the pessimistic approach makes the evaluation process risk-averse. This is particularly crucial in the energy sector, where inaccurate decisions can result in significant economic losses or prolonged shortages. The Pythagorean fuzzy sets help refine the model and provide a more precise evaluation of options by specifying hesitation and ambiguity more effectively than existing models. The extent to which the framework can be applied is illustrated in a case study of choosing renewable micro grid projects over diesel generator projects during times of intense power shortages. These alternatives are evaluated based on speed of deployment, operational cost, and carbon emissions by expert panels. The findings suggest that the model can identify the safest and most sustainable option in an unpredictable scenario. By combining rough set theory within the framework of the Pythagorean fuzzy setting, this study proposes a viable, cautious, and flexible framework for energy decision-making in sustainable power system planning and crisis management solutions. Mathematics Subject Classification 03E72. 68T27.68T37. 90B50