This study proposes a hybrid metaheuristic that combines the Coyote Optimization Algorithm (COA) and the Water Cycle Algorithm (WCA) to improve multi-objective optimization of a hybrid renewable energy microgrid (HRES) for Nusa Penida Island. Such islanded systems must balance investment cost, operational reliability, and renewable curtailment while facing stochastic weather and demand. The optimization therefore targets simultaneous minimization of cost of energy (COE), loss of power supply probability (LPSP), and dummy load (DL) as key indicators of affordability, adequacy, and energy-utilization efficiency. A sequential Hybrid COA–WCA framework is implemented using an annual time-series of electrical demand and local renewable-resource profiles. Candidate solutions encode the main HRES components, including photovoltaic generation, wind generation, battery storage, and conventional backup biodiesel generation, while respecting practical operating limits. Multi-objective optimization is handled using a weighted-sum formulation, subject to standard power-balance, component-operating, and reliability constraints. The proposed approach is benchmarked against standalone COA, WCA, and WOA under multiple uncertainty scenarios that perturb techno-economic parameters and resource–load conditions. The Hybrid COA–WCA achieved the lowest mean objective value (mean f = 0.87274; min = 0.82136; max = 0.92409) and the best final-iteration mean of 0.87258 compared with COA (0.87303), WCA (0.87367), and WOA (0.87756). The optimized design delivered COE = 1.388, LPSP = 0.03180, and DL = 6,609.025 on average across scenarios. Robustness analysis also indicates faster stabilization and the smallest end-of-run fluctuation range (0.82136–0.92409), confirming improved convergence stability relative to the benchmark algorithms. Overall, the Hybrid COA–WCA provides a stable and competitive optimization approach for HRES sizing under uncertainty, yielding consistently high-quality solutions that support planning and decision-making for island microgrids.