Capacity Optimization and Hierarchical Robust Control Verification of Wind–Solar–Diesel Hybrid Microgrid Based on Improved Multi‐Objective Grey Wolf Algorithm

Global decarbonization intensifies demand for off‐grid renewable microgrids in extreme environments like the Qinghai‐Tibet Plateau. Low atmospheric pressure, intense radiation, and diurnal temperature swings challenge capacity configuration and dynamic control. Existing hierarchical control lacks robustness against equipment degradation and load mutations, with limited high‐altitude hardware validation. This study proposes a codesign framework integrating enhanced multi‐objective grey wolf optimization and hierarchical robust control. Algorithm improvements include: tent chaotic mapping for population diversity, double clustering for solution distribution, and Gaussian–Cauchy mutation for dynamic constraints. Control innovations feature multistage convergence factors and SOC dynamic models for rapid power tracking and voltage/frequency regulation. This study addresses extreme‐environment challenges for off‐grid microgrids on the Qinghai‐Tibet Plateau through codesigned algorithmic and control innovations. We achieve multi‐objective component capacity optimization via an enhanced algorithm, while hierarchical robust control ensures operational feasibility. The upper control layer executes global scheduling based on algorithm‐optimized configurations; the lower layer enables real‐time power tracking and voltage/frequency regulation. This resolves the disconnect between traditional optimization and actual operation, enhancing adaptability to plateau‐specific disturbances (low voltage/pressure, temperature swings, renewable volatility). The improved algorithm accelerates convergence by 58.9% versus standard grey wolf optimization, maintaining > 97.1% power supply reliability. Using modular HIL platform, we simulated plateau conditions (low pressure, extreme temperatures, dust storms) to validate the algorithm‐control codesign. The results show that the improved algorithm significantly improves the global distribution uniformity of the Pareto solution set, maintains a high power supply reliability of 97.1% in the case of wind and solar energy resource fluctuations and sudden load changes, the probability of power shortage is less than 0.02, and the annual power shortage is less than 50 kW h, which fully meets the stringent reliability requirements of the plateau off‐grid system. The framework advances multi‐objective dynamic optimization codesign, accelerating renewable energy deployment in ecologically fragile regions.