You are here

Online Monitoring Technical Basis and Analysis Framework for Large Power Transformers; Interim Report for FY 2012

The Light Water Reactor Sustainability Program is a research, development, and deployment program sponsored by the U.S. Department of Energy Office of Nuclear Energy. The program is operated in collaboration with the Electric Power Research Institute’s (EPRI’s) research and development efforts in the Long-Term Operations (LTO) Program. The LTO Program is managed as a separate technical program operating in the Plant Technology Department of the EPRI Nuclear Power Sector with the guidance of an industry advisory Integration Committee. Because both the Department of Energy Office of Nuclear Energy and EPRI conduct research and development in technologies that have application to establishing the feasibility of operating commercial light water reactors (LWRs) beyond the current 60-year license limits, it is important that the work be coordinated to the benefit of both organizations.

The Light Water Reactor Sustainability and LTO Programs are working closely with nuclear utilities to develop instrumentation and control technologies and solutions to help ensure the safe life extension of current reactors. One of the main areas of focus is centralized online monitoring (OLM), which has two subprojects: online monitoring of active components and online monitoring of passive components. The research activities associated with online monitoring of active components are presented here. The current fleet of nuclear power plants (NPPs) performs periodic or condition-based maintenance of their active assets/components. The objective of centralized OLM is to implement predictive online monitoring techniques that would enable NPPs to diagnose incipient faults, perform proactive maintenance, and estimate the remaining useful life (RUL) of the asset.

To demonstrate the value of predictive online monitoring, EPRI has developed a Web-based Fleet-wide Prognostic and Health Management (FW- PHM) Software Suite (Beta Version 1.1). The framework of the FW-PHM software consists of four main components: Diagnostic Advisor; Asset Fault Signature (AFS) Database; RUL Advisor; and RUL Signature Database. Idaho National Laboratory (INL) is responsible for performing beta testing of the software. This work includes installation and configuration process evaluation; content-based testing; data synchronization; and a human factors evaluation.

Part of the long-term strategic goal of centralized OLM of active components is to enable industry to implement online monitoring using the FW-PHM software on selected active components. Generator Step-Up Transformers (GSUs) and Emergency Diesel Generators (EDGs) are two specified active components for which monitoring techniques, diagnostic and prognostic models will be developed in the software. INL and EPRI have identified a partner utility for each active component. Braidwood Generating Station (owned by Exelon Corporation) and Shearon Harris Nuclear Generating Station (owned by Duke Energy Progress) are partner utilities for EDGs and GSUs respectively.

Along with beta testing of the FW-PHM software, INL is working with the partner utilities to identify and characterize critical faults that lead to catastrophic failures in both GSUs and EDGs. This will allow INL to populate the AFS database of the FW-PHM software. The AFS database captures details about asset type, source of the fault information, different fault signatures, causes, remedies, and consequences. Based on the identified fault signatures and failure modes, the Diagnostic Advisor is used to diagnose fault conditions.

INL will research diagnostic and prognostic models for GSUs and EDGs over the next two years. These models will be used to populate the RUL database and to make component life predictions using the RUL advisor. The resulting models will be used with data from the utility partners to demonstrate the use of predictive OLM in NPPs. The FW-PHM software is unique in the sense that it standardizes the diagnostic and prognostic approach across assets based on fault signatures and fault features, generates a comprehensive diagnosis report, and allows information sharing between different NPPs via a master database. These capabilities do not currently exist in NPPs, and are expected to support safer long term operation of the NPPs.