<p>Accurate wind speed (WS) predictions are essential for the successful integration of wind energy into microgrids (MGs), ensuring MG stability, improving energy efficiency, and reducing costs. This paper introduces a new hybrid model, the Symbolic Bivariate Weibull Distribution (SBWD) Model, for predicting WS at a target location using wind data from a nearby location. The proposed model combines the symbolic approach with the BWD model, offering both flexibility and simplicity in implementation. The SBWD model achieves high forecast accuracy, even when the correlation between locations is relatively low, addressing a major limitation of the BWD model. This approach enables wind speed prediction in areas without permanent weather stations, significantly reducing infrastructure costs. Additionally, the model is adaptable to diverse environmental conditions, incorporating factors such as geographic coordinates and temperature differences. By enhancing the short-term forecast performance of the BWD model in low-correlation scenarios, the SBWD model is particularly useful in regions with limited meteorological data. Genetic optimization is employed to determine the optimal model parameters. The approach is applied to four distinct locations in Egypt, considering varying distances and correlation factors. The SBWD model outperforms the BWD model, reducing the RMSE from 0.7444 to 0.66576 (an 11% improvement) at high correlation, and from 1.4386 to 1.1571 (a 19.5% improvement) at low correlation, thus significantly improving wind speed prediction accuracy. The results are also compared with other models like The Auto Regressive Moving Average (ARMA), ELMAN Neural Network (ENN) and Multi-Layer Perceptron (MLP).</p>