Over the past year there has been a lot of activity in the
area of on-line equipment monitoring, diagnostics, and prognostics and this
trend is expected to continue. On-line
diagnostic and prognostic information regarding equipment/asset health is
extremely valuable for scheduling and carrying out maintenance and preventing
unplanned down-time. The net effect is
an overall increase in the efficiency, safety, and profitability of power
industry processes. There are different
techniques and approaches for deducing and acquiring this information including
but not limited to: empirical modeling techniques, statistical modeling,
physical or first principles modeling, and the application of additional
sensors, including wireless sensors.
This session focuses on new ideas and applications in the area of
monitoring, diagnostics and prognostics, as well as the application of these
techniques in the power industry.
P009-“Early
Discovery of Failing Equipment and Sensors in Power Plants using Advanced
Pattern Recognition”
Elmer Hansen, Performance Consulting Services
This study
describes the application of advance Pattern Recognition on-line Monitoring in
a power plant fleet environment.
Advanced Pattern Recognition has failures not typically detected by
traditional control systems. Typical failure identification and alarming is
carried out applying high and low limits to the range of a single instrument
measurement. This results in a rather wide fixed window of operation. Advanced Pattern Recognition technology has
been employed to monitor multiple sensor systems using a moving window to
identify normal and abnormal operation.
Using a moving window takes into account the current operating
conditions of the equipment and allows the early detection of many types of
failures. The moving window is defined
by a pattern recognition generated expected value and the threshold of normal
range about the expected value. The
abnormalities can then compare to typical failure signatures to suggest the
failure origin. The expected value can
also be used to validate the signal and provide a replacement signal
value. Practical applications of
advanced pattern recognition to detect plant equipment and sensor failures will
be described. The variety of incipient
equipment failures, which have been identified in real plant environments will
be categorized. Several failures will be profiled. These will also be illustrated by case
studies. Advanced pattern recognition has been used successfully in power plant
environments to provide the early detection of equipment and sensor failures.
P016-“Sensor Fault Detectability
Measures for Autoassociative Empirical Models”
Wes Hines,
Traditionally,
the calibration of safety critical nuclear instrumentation has been performed
during each refueling cycle. However,
many nuclear plants have moved toward condition-directed rather than
time-directed calibration. This
condition-directed calibration is accomplished through the use of on-line
monitoring. On-line monitoring (OLM) commonly uses an autoassociative empirical
modeling architecture to assess instrument channel performance. An autoassociative architecture predicts a
group of correct sensor values when supplied a group of sensor values that is
usually corrupted with process and instrument noise, and could also contain
faults such as sensor drift or complete failure. This paper describes one such autoassociative
model architecture, specifically autoassociative kernel regression (AAKR), and
presents five metrics that may be used to evaluate performance. These metrics include the previously
developed accuracy, auto-sensitivity, cross-sensitivity, metrics along with a
description of two new fault detectability performance metrics for application
to instrument calibration verification (ICV) and anomaly detection. These parameters are calculated for an AAKR
model of an operating nuclear power plant steam system and were used to
describe the effects of model architecture on performance. It is shown that the ability of an empirical
model to detect sensor faults in ICV systems is largely dependent on the model
uncertainty and to a lesser degree its robustness. It is also shown that the ability of an
empirical model to detect anomalies via the Sequential Probability Ratio Test
(SPRT) is also related to uncertainty and the SPRT detectability is on the
order of 32% smaller than the ICV detectability. These guidelines provide a framework for
developing various models, in that models intended to be applied to ICV and
anomaly detection tasks should focus on the minimization of uncertainty. Furthermore, the ICV and anomaly detection
performance metrics are shown to be within the traditional ±1% calibration
tolerance and their performance under artificially faulted conditions are shown
to be in direct agreement with their theoretical foundations.
P017-“Low Cost Wireless Condition
Monitoring”
Lloyd
Pentecost, Southern California Edison Company
At the San Onofre Nuclear
Generating Station, we have implemented two different wireless condition
monitoring systems. The first is a WiFi (802.11b) system that allows for easily
and quickly establishing a new monitoring system based on emergent plant needs.
The second is based on a Zigbee (802.15.4) like system that allows for very low
cost condition monitoring for low data rate applications. For low data rate applications, Zigbee and
Zigbee-like systems have the potential to reduce the cost of data collection by
a factor of 10 or better. This reduction
in cost will change the cost/benefit ratio and dramatically expand the number
of data points available for collection. The purpose of this presentation is to
describe some early applications of both WiFi and Zigbee-like systems, success stories, cost
benefits and potential future applications.
P049-“Performance Improvement for
Cooling Water Systems; Corrosion – The New Process Variable”
Dawn C
Eden, David A Eden & Russell D Kane, Honeywell International, Inc.
Availability
of the cooling water system is critical to the operation of power plant. Under inadequate control, the cooling water
system can present significant difficulty to the plant in loss of production
capacity, increased cost of cleaning and protective chemicals, increased energy
and maintenance costs, and a reduction in service life. Although regular checks are made to determine
water quality and compliance with prescribed operating conditions, these checks
can be infrequent enough to allow corrosion, fouling and scaling to get out of control. This inevitably leads to outages that can be
costly and severely affect production.
Technology exists that enables a continuous, rapid update of the key
general and localized (pitting) corrosion, scaling and fouling information that
can assist the plant operator in
real-time to ensure cooling water system availability.
This paper
reviews a field proven on-line, real-time process control technology with a
heat transfer efficiency and corrosion measurement to provide a comprehensive
understanding of unit operating condition and fouling/scaling activity in
cooling water and heat exchange equipment.
Data from field-installed systems are presented, with reference to heat
exchange operations in a petrochemical plant and a nuclear power plant. Recent advances in the technology and its
implementation are also discussed, specifically with regard to wireless
communications and direct-to-
P042-“Integrated Wireless Systems for
Power Applications”
Peter
Fuhr, RAE Systems
As wireless devices become more prevalent in the power industries,
the need for a managed solution increases. Viewed another way, wireless sensor
networks, RFID asset tracking systems, voice over wireless IP systems, WiFi,
WiMAX, (etc) all use the limited frequency space and without some coherent
strategy for coexistence, systems failures will undoubtedly occur. A strategy for deployment of integrated
wireless systems in industrial settings is identified and example case studies
are presented.
P053-“Finally,
Dr. Eyal Brill and James Hawkins, P.E.
The speed at which state-of-the-art advances in automation and
control are applied within the utility industries core business is as fast as
waiting for maple sap to drip from a tree in January. Advances in IT have
become commonplace in process and manufacturing industries and the yield in
efficiency and reliability has been striking.
Utilities are spending millions on enterprise software to integrate and
streamline the activities on the business side of their organizations, however,
the industry has been slow to integrate and adopt recent advances in IT that
can squeeze efficiency, reliability and substantial operating savings from its
core business. A new, proven advanced statistically driven enterprise approach,
that pushes the envelop for reliability and optimization of the operating side
of a utility’s core business is discussed. This new technology has many
advances over current solutions including:
·
On-line adaptation of multivariate models based on real-time data.
This requires the ability to differentiate between "abnormal"
situations and "normal" drift of the process.
·
Ability to combine human expert knowledge with statistical models.
·
Open architecture that enables users, using Microsoft standards,
to encapsulate proprietary modeling techniques into the enterprise FDC
platform.
·
Pattern recognition for long term phenomenon
This paper will demonstrate the technology principles with actual
examples from applications in the field.
P056-“On Analysis of
Control-Communication Integration in Distributed Power Grid’s Electric,
Information & Physical Domains”
Teja Kuruganti, Mallikarjun Shankar, Glenn Allgood,
The electric grid is infact two different infrastructures, power
grid and the communication grid. As the grid evolves into a network of highly
distributed generation, transmission and loads, a more complex communication
network follows it. This paper analyzes the dependency of the controls on
communication networks and establishes a roadmap for controls and
communications in power systems.