Abstract:In the current wave of accelerated digitalization and intelligent transformation within the power industry, the demand for low-cost operation and refined management of thermal power units has grown significantly. This paper analyzes the necessity of constructing an intelligent wind and smoke system for power plants and provides a detailed design of an integrated cloud platform for such a system. At the same time, an online intelligent early warning and diagnosis system for wind and smoke system equipment faults in power station is studied and applied based on machine learning and statistical methods, which provides a useful reference for the construction of intelligent wind-smoke system in power stations.