The increasing integration of renewable energy sources and the dynamic nature of electricity demand necessitates the adoption of smart energy management systems (SEMS) to ensure efficient, reliable, and sustainable power distribution. This paper provides a systematic and comprehensive review of computational intelligent algorithms applied in smart energy management systems, focusing on load forecasting strategies. Load forecasting is critical for predicting future energy consumption patterns, enabling proactive decision-making and optimized resource allocation. The paper examines a variety of computational intelligence techniques, including deep learning, and hybrid models, highlighting their capabilities, performance metrics, and limitations. By synthesizing current research findings and identifying future research directions, this paper aims to provide valuable insights for researchers, practitioners, and policymakers involved in the development and deployment of smart energy management systems.