A substation monitoring and warning system based on infrared technology and image separating

To find the defects of the apparatus in a substation in the early stage, an infrared temperature measurement based monitoring system is established. An infrared-visible hybrid detector consists of an uncooled infrared focal plane array detector and a visible light camera, the lenses of both are of the same specification. As a result, the graphs are exactly matched. The hybrid detector is implanted on a PAN/TILT, which is scanning in a pre-programmed manner. An image separating approach is introduced to identify each apparatus or each component of an apparatus (they are called as objects), with which, the temperature information can be collected in an object-oriented basis. A database of the temperature information of apparatus and their components are established, based on which, a knowledge-based system is established. With the proposed system, there is no need to transfer bulk infrared images all the time, instead, only the collected temperature information of the objects are transferred to the monitoring center. Not only the efficiency can be remarkably increased, but the defeats can also be automatically detected by the knowledge-based rules without need of experienced staffs. The practical application shows the feasibility and advantage of the proposed approach.