As global climate change intensifies, the power industry—a major source of carbon emissions—plays a pivotal role in achieving carbon peaking and neutrality goals through its low-carbon transition. Traditional power plants’ carbon management systems can no longer meet the demands of high-precision, real-time monitoring. Smart power plants now offer innovative solutions for carbon emission tracking and intelligent analysis by integrating IoT, big data, and AI technologies. Current research predominantly focuses on optimizing individual processes, lacking systematic exploration of comprehensive dynamic monitoring and intelligent decision-making across the entire workflow. To address this gap, we propose a smart carbon emission monitoring and analysis platform for power plants that integrates IoT sensing, multimodal data analytics, and AI-driven decision-making. The platform establishes a multi-source sensor network to collect emissions data throughout the fuel combustion, auxiliary equipment operation, and waste treatment processes. Combining carbon emission factor analysis with machine learning models enables real-time emission calculations and utilizes long short-term memory networks to predict future emission trends.