<p>This paper introduces an uncertain energy spot price model to characterize the dynamic evolution of electricity and coal prices in markets with limited data availability. By integrating seasonal components and mean-reverting uncertain differential equations, the proposed model effectively captures the non-storage nature and price spikes inherent in energy commodities. Unlike traditional stochastic models reliant on large historical datasets, our framework leverages uncertainty theory to address belief degrees under insufficient observations. The model is further applied to price dark-spread options, which derive value from the spread between electricity and coal spot prices. Numerical experiments using real market data demonstrate the models superior performance over stochastic counterparts in capturing price volatility and generating reliable confidence intervals. This work bridges a critical gap in energy derivatives valuation by providing a robust tool for investors and risk managers operating in data-scarce environments.</p>