Abstract:The power plant has many equipment and complex process. The traditional equipment status inspection can not predict the hidden danger of equipment. The production operation and management personnel need to browse or inspect the main parameters of the equipment in turn or on-site inspections.It is time-consuming and laborious, and the state of the state can only be judged by personal experience. It is easy to form the non benign phenomenon of the subjective and even emotional equipment management. Through the establishment of big data platform, the operation data of the historical normal operating condition of the equipment are screened, and the key or all equipment models of the power plant are constructed with big data fusion technology.When real time production data is introduced in the model, the platform is analyzed by large data. On the one hand, the real time status of the equipment is removed from the historical situation of the normal state of the state of the small movement to alarm, on the other hand, the hidden parts of the equipment and the expected failure time are predetermined.The equipment management can make the key inspection, the early disposal and the cancellation, so that the failure of the equipment can be prevented, thus the reliability of the whole equipment of the power plant is improved, and the data and technical support are provided for the equipment management to the direction of the state maintenance.