SBIR-STTR Award

This project supports the Nuclear Energy Plant Optimization Program objective to ensure the continued safe and reliable operation of the Nation’s nuclear power plants. To achieve this goal, it is essential that accurate on-line information about the current state of the equipment be available to the operators. Such information is needed to determine the operability of safety and control systems, the condition of active equipment, the necessity of preventative maintenance, and the status of sensory systems. This project will demonstrate new adaptive on-line learning algorithms to enable the automated optimization of equipment-condition-monitoring software for a wide variety of power plant systems. These artificial intelligence techniques will be used to ensure the accurate measurement of key reactor and plant parameters, assess equipment in-service performance, and determine equipment integrity and the need for maintenance. In Phase I, the new algorithms will be incorporated into a proven three-step diagnostic framework consisting of a parameter estimation step, a fault detection step, and a diagnostic decision step.

Commercial Applications and Other Benefits as described by the awardee:
In addition to the application for nuclear power plants, the on-line equipment-condition-monitoring software should be applicable to any process control system where unexpected process interruptions could cause equipment failures, false alarms, or unsafe conditions. The new capability should substantially improve plant uptime, thereby increasing output and operating revenue in a wide range of industries

In order to ensure the continued safe, reliable, and efficient operation of the Nation’s nuclear power plants, improvements are needed in the accuracy and timeliness of information delivered to the operators about the condition of the plant equipment. The information is needed to determine the operability of safety and control systems, the health of active equipment, the necessity of preventative maintenance, and the status of sensory systems. This project will develop adaptive modeling and decision support software techniques to enable more effective life cycle management of aging nuclear plant equipment. The overall approach is to capture and preserve essential equipment diagnostic knowledge from veteran operators, so that the software can use this knowledge to assess equipment performance and integrity automatically and in real-time. In Phase I, adaptive on-line learning algorithms were demonstrated to enable the automated optimization of equipment-monitoring software for a wide variety of power plant systems. New decision support techniques were shown to be effective in capturing diagnostic knowledge in a portable and reusable format and to provide much earlier detection of equipment problems compared to current industry practice. Phase II will develop and implement adaptive modeling procedures to enable the production of more cost-effective equipment-monitoring software; develop and implement an effective on-line decision support capability; deploy and demonstrate the technology at two representative U.S. nuclear power plants; and produce a commercial-ready product for market development.

Commercial Applications and Other Benefits as described by the awardee:
In addition to the application for nuclear power plants, the on-line equipment-monitoring software should be applicable to any process control system where unexpected process interruptions could cause equipment failures, false alarms, or unsafe conditions. The new capability should substantially improve plant uptime, thereby increasing output and operating revenue in a wide range of industries.