Abstract: The system of the present invention uses the importance of a piece of malfunctioning equipment, the severity and the confidence level in a diagnosis to determine repair priority of the equipment by obtaining the product (46) of the confidence level (CF), importance (IMP) and severity (SEV). The these variables are determined by expert system rules. The severity is the reciprocal (88) of the time to failure. The importance (44) is the cost to repair the maximum damage when the malfunction continues. A malfunction can affect several pieces of equipment in combination, the severity and importance associated with each piece is combined (70) with the confidence and used to determine the repair priority. When the diagnosis of a malfunction is by malfunctioning sensors, the expected equipment life (40) and the availability (86) of sensors that provide a partial backup are considered in prioritizing sensor as well as equipment repair.

Abstract: Diagnostic apparatus utilized to monitor a steam turbine-generator power plant. A plurality of sensors around the plant provide data to the diagnostic apparatus which utilizes a digital computer to come to some conclusion relative to actual or impending malfunctions in the plant components. The computer utilizes an expert system diagnosis which establishes a plurality of subsystems which are generic and can be applied to any sensor to obtain indications of (a) sensor output movement, (b) validated sensor readings and (c) malfunction indications based on conclusions reached in (a) and (b).

Abstract: Sensor based diagnostic apparatus for performing on-line realtime monitoring of an industrial or other operating system. The diagnostic apparatus has a rule network for propagation of belief leading to one or more possible malfunctions of equipment in the operating system. The propagation of belief is dynamically altered as a function of the operating conditions of the sensors themselves.

Abstract: A diagnostic system wherein a central diagnostic center receives sensor data relative to the operating condition of a plurality of remote plants. Each plant records just certain sensor signals and transmits them at respective prescheduled transmisssion times. If certain activation limits are exceeded, a data link with the diagnostic center is immediately established so that the data may be transmitted ahead of its normally scheduled transmission time in order that an immediate diagnostic analysis be made on the plant. A communication link between the plant and diagnostic center is kept open for a predetermined period of time in which more data may be sent from the plant to the diagnostic center, and results and instructions, sent from the diagnostic center back to the plant. Such instructions may include the changing of certain parameters in the examination of the sensor signals.