For a long time

constrained by technological development

power enterprises still adopt regular maintenance and accident elimination mode for equipment maintenance

but this way causes a great deal of waste and unnecessary loss. Condition based maintenance for power plants is always a maintenance mode expected by electric power enterprises. With the development of artificial intelligence technology

especially the application of in-depth learning in industrial field

the application of artificial intelligence technology in the field of power generation is gradually increasing

and then the characteristic values representing the essential attributes of industrial equipments are obtained

which can be used as the criteria for judging the health status of equipments

so that condition-based maintenance can be implemented in power plants. In this paper

the characteristic value was used to evaluate the turbine state of a hydropower station

and the fault state of the equipment was found accurately and timely

indicating the application effect is good.