Tiger™
Knowledge Based Gas Turbine
Condition Monitoring Introduction
INTRODUCTION
This is an introduction to the Tiger knowledge based gas turbine
condition monitoring system. Its purpose
is to:
q Give an
overview and explanation of Tiger
q What it is
q What it
isn’t
q Its
benefits
q Some of the
frequently asked questions relating to Tiger
Tiger
is designed to assess the condition of the gas turbine by continuous
interpretation of its behaviour. This
knowledge of the state of the gas turbine is the information needed to make
decisions for condition-based maintenance.
By having Tiger analyse
continuously the state of the gas turbine we can then make decisions about:
what needs to be maintained on the turbine; what doesn’t need to be maintained
on the gas turbine; are problems developing; and do certain problems not exist.
A key aspect of Tiger is to check for anything going
wrong at once per second so that the smallest deviations are picked up right
away at the earliest possible stage.
These are presented through the fault manager to give a day-to-day
summary of all the problems which have occurred on the gas turbine.
Tiger
also performs calculations of the performance of the gas turbine on a
continuous basis so that performance deteriorations can be tracked and the
information needed for making decisions on off-line washing is available. Tiger
also provides diagnostics to assist any time there is a disturbance; a problem;
a gas turbine trip, or a failure to start.
Tiger rapidly analyses the
data available to it to help the user understand why the gas turbine tripped or
why it wouldn’t start.
TIGER OVERVIEW
Tiger
is focussed on helping people get the most out of their gas turbine. Generally in a single cycle or a combined
cycle power generation facility, there are two main goals for companies; one is
they do not want to stop unexpectedly or unnecessarily. If it is run on base load they need
continuous power output, but they also want maximum efficiency and that
reflects not only efficiency from the viewpoint of the compressor blades being
cleaned, but also not having other problems that inhibit the performance of the
gas turbine.
A
key task is to know what needs to be maintained and when it needs to be
maintained. If I am going to stop for an
overhaul, I need to know what the faults and problems are that I need to take
care of, and what parts of the system seem to be performing fine. If a problem is developing it will need
unusual maintenance or actions. I need
to know this as early as possible.
By
continuously assessing the condition of the gas turbine, Tiger is able to record and describe the
problems that need maintenance and need attention so that reports can be
prepared to justify various maintenance actions.
This
addresses a major need, which is, if you just look at a gas turbine you don’t
actually know what is wrong with it; what the problems are; what needs
attention; what state the gas turbine is in.
With Tiger building up a
complete history of all the problems that have been detected, you then have the
information to make these decisions.
What
we really want is a way to understand the condition of the gas turbine by
having this assessment of all the niggles; the problems and the little things
that have occurred, so that we have a complete view of where the gas turbines
performance is.
We
want to detect problems at the nearest possible stage so that the slightest
hint of any developing faults are detected.
It can then be monitored and its actions can be planned around it.
We
also need to understand what happens when problems do arise so that if there is
a trip, or some other incident occurring, we can understand not only what the
cause is, but what the impact and the likely long-term influence will be.
Much
of this is like having a doctor following you around continuously. If you look at a gas turbine at a 10,000-hour
inspection, you don’t know what happened to the gas turbine; what problems have
occurred; and what state it is in. This
is the same as your doctor.
When
you go for an annual check-up, he looks at you and sees what you look like now,
but he doesn’t really know the stress and strains that you have had over the
past year; the problems or faults that you have developed. So imagine your doctor could follow you at
once per second, continuously. He would
then have a complete idea of your health history over the past year; all the
colds and little problems you have had; and all the unusual situations that you
recovered from. So this is the idea of
building a continuous assessment of the gas turbine – to get a picture of its
condition so that you have the information you need to make your condition
based maintenance decisions.
Where
does Tiger fit in the context of
condition monitoring? There are many
types of condition monitoring, and in many peoples minds, condition monitoring
equals vibration based condition monitoring.
That is they think that the type of vibration assessment that Bentley
Nevada does is the only type of condition monitoring. But in fact, there are several techniques
such as visual monitoring, behaviour monitoring, vibration monitoring and oil
or debris analysis. Tiger fits into the category of
behaviour monitoring – we monitor how the turbine behaves, how it operates; how
it changes to load or other conditions, and based on that information, make an
assessment of the condition.
The smart bit of Tiger,
the intelligent bit, is the continuous assessment of the condition based on the
way the gas turbine behaves. Every time
the fuel valves are meant to move in response to a load set-point change, Tiger checks with the fuel valve to see
that it is moving properly; that they reacted the way they should; did the
exhaust temperatures increase?; and did the power increase all the way it
should?
All
this reflects the behaviour, which is a bit like the idea if you want to know
how nimble or capable someone is - seeing them sitting there at a nightclub
doesn’t help, but if you see them getting up and dancing a bit, then you know,
that it is the dynamics and the movement that really shows whether the system
is working properly.
So,
Tiger fits very clearly in
condition monitoring using the behaviour of the gas turbine as the influence,
and not the vibration techniques. Tiger compliments vibration based
condition monitoring techniques such as provided by Bentley Nevada .
It actually provides a lot more ‘bang for the buck’. Tiger
is relatively simple and does not require additional sensors, it is only
software. Tiger can be installed in only half a day, with just an
interface to the control system implemented with the right data communication
so it is a very low cost rapid system to install.
Vibration
based condition monitoring is very expensive and requires special sensors for
the special hardware, or signal processing, and only looks at the mechanical
aspect. Tiger
is much more powerful because it looks at the supporting system such as the
fuel system, any dry low NOx, or NOx reduction control system; the combustion
system; the starter mechanisms; the generator if the data is available through
the controller.
Tiger fits clearly in the context of condition monitoring by
looking at the behaviour and the clever bit is the knowledge-based assessment
of the condition continuously. So the
approach of Tiger is continuous
monitoring and diagnosis at once per second.
The
unique aspect of Tiger is the fact
that it does a diagnosis of the data at once per second. Other systems such as Scada systems will
allow the user to display data and store data, but doesn’t provide any
interpretation at all. They might
provide simple limit checking and alerting capabilities, but nothing dynamic or
sophisticated.
Tiger uniquely and continuously does an analysis and diagnosis of
the data so that, not only does it see if something is unusual, but also it
builds up a diagnostic explanation of what that is. All that is done at once per second,
continuously 365 days a year, or whenever the gas turbine is running.
Tiger combines several layers of diagnostic
analysis. The lowest level is continuous
automated fault detection so it has many ways to detect that something is
unusual or a fault is present. Tiger has multiple levels of diagnostics
that combine what has been detected as faulty into a partial diagnosis, or a
more complete diagnosis, to the best diagnosis conclusion we can reach for a
given situation. It is important to
recognise that Tiger cannot
diagnose everything and cannot diagnose everything perfectly. It does the best it can to put together the
conclusions of the data available. Some
situations we have never seen before, no one else has seen before, we want to
diagnose, or know what to reach as a conclusion. Sometimes the sensors that we have limit the
amount of diagnostic conclusions in Tiger. For example, we might be missing a critical
measurement to separate whether the problem is in the flow or the valve control
itself.
All of the output of Tiger
is presented through an easy to read fault manager, a summary of the
conclusions and the diagnostics that have been developed, which is presented in
a hypertext tree style, so you can read the high level summary first, and then
click with the mouse to browse incrementally through to have the
explanation.
The
fault manager is the real key benefit of Tiger,
all the diagnostics at once per second are summarised into a simple set of
diagnostic messages. They are very short
and concise to read, that way if you want to check 'Is my turbine okay? 'Was
there a problem on the last day? Has anything unusual happened? You can just read through the diagnostics
rather than having to browse lots of data.
This enables the ability to actually check the turbine once per second
realistically. Manually by hand, you
could never do that. You can never go
through all the graphs of all the data 24 hours a day. Tiger’s
fault manager provides that unique summary to let you do that very easily.
All
of the data is kept permanently, and Tiger
has extensive data archives, displays and trending mechanisms so that you can
view things rapidly; investigate incidents and problems, as well as look at
trends from short term to very long term.
All the data is kept permanently so that you can go back and look at any
problems or key aspects at any point.
To
summarise the steps of condition monitoring, you first need to frequently check
the machine at some interval; detect problems at an early stage; then assess
the condition of the machine and use that assessment to build up a picture of
the overall condition of the machine.
That is standard condition monitoring.
Tiger’s approach is to take data, 600 measurements every second,
from the gas turbine controller. To
detect problems we have thousands of fault detection checks that are
automatically tuned to a particular gas turbine. Tiger
looks at data on the gas turbine for a short period and adjusts all the limits
to detect problems for each operating band for that gas turbine.
The
analysis is done by using the knowledge capture from gas turbine experts, and
over 40 years of operating hours that have run through Tiger. A picture of
the condition of the gas turbine is presented through the high-level fault
manager. Tiger
has an extensive set of displays that provide quick views of the state of the
gas turbine at a very high level, and are easy to look at including each
sub-system and identify whether any problems have developed.
The
diagnostics are also continuously available.
The fault manager provides the diagnostic summary. In the diagnostic summary all identified
events are presented three times, first, by the priorities, critical;
indications of faults or just warnings.
Second, by the sub-system of the gas turbine such as the gas fuel; the
liquid fuel; the air inlet system; the combustion system; and the NOx reduction
system. Thirdly, a complete time
sequence.
The
idea is that no matter which way you are trying to look at a problem there are
a set of messages organised to be helpful.
So if you want to investigate what is wrong with a particular
sub-system, you can use the sub-system view.
If you are trying to identify precisely a fault, you can look in the
time sequence. When you are making a
quick daily check of the overview of the turbine, you can look at the highest
priority messages to see.
For
every diagnostic message, by clicking on the mouse in the hypertext tree, you
can get a diagnostic explanation of what the problems and symptoms were that
lead to that particular conclusion. In
that way, you can understand how it reaches certain conclusions and follow through
the logic of the diagnosis yourself.
An
important aspect of Tiger is that
once it reaches a conclusion it has a set of messages that are now ‘accounted
for’. So within the fault manager the
‘accounts for’ takes the big flood of messages that have to be resolved and
organises them into those that are important, and those that you can
effectively ignore. So instead of having
to look at pages of printouts, the Tiger
diagnostics are organised in a tree, saying these are the important messages
and these are messages that are accounted for and you can ignore them.
For selected customers and gas turbines, Tiger also provides links to on-line manuals so that if you
want further information on a particular diagnostic conclusion or sub-system,
you can jump directly to the on-line manuals.
Tiger also has extensive graph facilities providing six windows
with up to eight channels in each. The
graph can hold virtually any time periods selected by the user, and data can be
expanded out to x10 (times ten) for looking at concise detail. The graph has shown itself to be extremely
useful and help people verify faults; troubleshoot the timing and sequencing of
events, and investigate problems that otherwise would be very hard to
troubleshoot.
Very
often, the output of the fault manager and the diagnostic graph are put
together to make reports to either document an incident or document work that
must be done. So for example, if there
has been a repetitive problem on a gas turbine the Tiger diagnostics can be used to show the fault manager
conclusions and identify the faults that arise from when the incident
occurs. Diagnostic graphs can be
attached to the report to show in more detail the exact sequencing.
Tiger also includes a query mechanism so that you can query
across the diagnostic conclusions and find out how often certain faults occur,
or, get a list for example, of all the faults in the gas fuel system that were
critical in the last month, and how many times each one occurred. This is part of the management information to
help you make decisions about the condition of the gas turbine. By having a
summary of all the faults and problems that have occurred, it is now possible
to identify the main faults that need attention so that you can now take the
action to eliminate those.
Tiger continuously performs a performance analysis of the gas
turbine. It does this several ways using
ISO formulas. It calculates dry
corrected power and dry corrected heat rate, assuming the appropriate
instrumentation is available. Tiger watches the data on the turbine
continuously and finds the most stable set of data each day to use in its
calculations, so you get a much more accurate assessment of the daily
performance compared to taking the measurement at 1400hrs for example. Every time there is a crank wash, the base
expected power levels are reset so you can instantly compare on the graphs
where your performance is relative to the output.
Also,
these performance reports can be printed out from Adobe Acrobat reports so that
they can be inserted into notebooks and files.
Tiger calculates dry
corrected power; dry corrected heat rate; compressor efficiency and also does
daily, and 30-minute average of NOx levels, to provide a performance evaluation
against the regulations.
The
diagnostics and the knowledge in Tiger
have been developed by over 50 fired years of gas turbine operation, and well
over 750 incidents and faults. Every
time Tiger detects an incident,
this is logged and analysed by the Tiger
developers to ensure that the diagnostics were adequate and appropriate, and
they are constantly and continuously improved.
With over 50 fired years of operation, we provide a very stable and
reliable set of detected events and diagnostics.
A
key aspect of installing Tiger is
that it uses the existing sensors already in place on the turbine through the
control system and does not require additional instrumentation. It might be that a plant decides to add extra
instrumentation once Tiger is
installed, for example, appropriate performance measurement instruments. But to install Tiger is a rapid software process that does not require
extra sensors and instrumentation to be fitted, this greatly reduces the cost,
as the user sees the benefit and capability of Tiger,
they can then add additional sensors as required.
A
key aspect of Tiger compared to
other condition monitoring systems, is that it has a wide range of sub-systems,
such as the gas fuel; the liquid fuel; the NOx control system; the combustion
system, and the generator. By looking at
a wide range of sub-systems, helps to isolate where the problems are
originating and whether impacts work.
Tiger has extensive data displays; trending and archive
facilities and very importantly, all of Tiger’s
capabilities can be accessed remotely through a modem or other type of network
link. So if a remote site is having a
problem, the head office or engineering staff can use a Tiger to dial in, evaluate first the diagnostics by
downloading diagnostics not data initially; analyse the diagnostics to see the
problem and situation; then pull back the relevant data for that incident to
very rapidly get an understanding.
Tiger has been used many times to help people remotely monitor
gas turbines. For example, during a
problem start-up, or after some maintenance work has been performed to ensure
that everything is working okay. Tiger is set-up already so that if you
have a fleet of gas turbines and you want to centrally monitor them, it
provides the ability to collect diagnostics overnight from each site, so that
in the morning you can quickly assess what happened yesterday on all the gas
turbines by browsing through the diagnostics.
The remote link can also be used by other members of the Tiger team to provide additional
engineering backup and support.
Tiger has provided consistent and substantial benefits for
customers. In every case shortly after Tiger
is installed, we find that Tiger
has identified many faults and problems that were already existing but nobody
was aware off. Some of these then help
sites to understand why they are having
to do maintenance more frequently and earlier than expected.
Tiger has a $150,000 Club, which are sites for which Tiger has produced a saving, or benefit,
of at least $150,000 within a year. In
fact, virtually all the mature and established installations of Tiger are members of the $150,000
Club.
The
faults cover a wide range of problems from:
q Hydraulic system problems affecting
the generator;
q Steam injection problems affecting the
performance of the gas turbine, which were then documented through Tiger, and that part of the
justification of a major refurbishment;
q Liquid fuel being unstable at low-load
levels, where the turbine would oscillate quite severally until it got to a
higher load and this was traced to the liquid fuel;
q Various problems with other
oscillations affecting first and second stage nozzles were tracked back using Tiger to help identify.
In
one situation, they knew that they had a very short life on the blades but
couldn’t understand why. As soon as Tiger was installed it identified that
there were oscillations occurring and Tiger
was rapidly used to track down the cause of these oscillations so that problems
could be detected.
Tiger has found, not only the failed start-up problems, but also
problems during shutdown that would prevent a start-up. In one case, an oil platform was down for two
days until an engineer arrived and used Tiger
to identify the cause of the faults. The
problem was that they did not know how to identify the problems of the liquid
fuel. Tiger
had been installed on the oil platform only a few days before, and had not had
the opportunity to train the staff on the oil platform on how to use Tiger.
So as soon as a competent engineer on Tiger
was flown out, he used Tiger to
isolate the fault. After two days of
shutdown, he had the system back up and running.
In
another example, a site was able to delay its shutdown until the weekend when
they had the appropriate repair equipment and staff on hand. Rather than shutting down mid-week which
would have been inconvenient and also not prepared. That greatly reduced the cost and avoided an
unplanned shutdown.
For
one of the gas turbines, Tiger was
used to catalog the wide range of problems that existed. This provided a good foundation on how good
the turbine was, and from that case, they were able to make a warranty claim to
have many problems fixed and repaired that the vendor had claimed were not
existing and not a problem. So everybody
benefited by identifying problems and then under warranty being able to claim
against those.
Tiger is also ideal for cataloguing the state and problems of the
gas turbine at the start of any type of contract, so that when you take over
your maintenance contract you can verify how good the turbine is; whether any
problems are there that can affect the maintenance commitments.
Tiger has also been used very
successfully to underwriting and guarantee commitments of high
availability. By identifying small
problems at the earliest stage, by way of rapidly troubleshooting causes of
trips or failed start, companies have achieved much higher availabilities
through the information collected on Tiger. So it provides a good foundation of
underwriting any fixed commitments that one has. In fact, once you see the information Tiger gives you about the state of the
gas turbine, you would wonder how you were ever brave enough to run a gas
turbine without it!
Tiger provides consistent benefits to the user. Generally the pricing model is on an annual
lease, although this will depend on the situation. It is consistently providing a payback within
a year for each customer and it is designed to help minimise the downtime,
either by reducing the analysis required to understand why a fault failed; identifying
problems that would prevent a restart; helping to plan effectively for the work
that would be required when it does shut-down so that when you do go down for
maintenance then you have all the right equipment and repair information
available; it helps to maximise the service intervals because Tiger provides a good assessment and
explanation of whether the turbine has performed well and smoothly, or had a
lot of disruptions or problems. You have
a good view of whether the turbine is likely to be in good enough condition to
extend the service or whether there is a problem. We help sites to optimise cleaning
cycles.
Finding
the best data to use in the performance calculations and continuously making
performance calculations, we provide the information needed so that the site
can be decide whether it is beneficial to perform an off-line wash at the
current time or to wait a bit longer.
Tiger helps with the advanced prediction of faults through the
detection of faults at the earliest possible stage. The concept is to detect faults as early as
they appear and then track how they change and how severe they are so that you
have the information you need to make your own prediction. Tiger
does not actually directly predict the faults, but provides the data needed so
that you can make an assessment.
Tiger provides a good mechanism to track the history of the
faults and problems. The query mechanism
can tell you all the times a certain diagnostic message appears, or how many
times it’s appeared. So for example, you
can rapidly get a report of all the times you had a sticking liquid fuel valve
and see whether the frequency is increasing.
The diagnostics within Tiger
can be used to replay an incident at any time, so that you can go back and post
investigate incidents time and time again, or from different viewpoints.
Very
often the history of Tiger has
been used to go back one or two years, to recheck certain incidents for a sign
of other problems. All these give a very
consistent payback to the sites where we have Tiger
installed.
In
summary, the benefit of Tiger
is
that we know everything that happens to the gas turbine. The once per second analysis is like your
doctor following you – it keeps track of all the problems that you have had, so
that now you have the information needed to make a good assessment of the
condition.
Tiger helps to detect the first symptoms of a developing problem,
so that you get the earliest indication of any fault or problem that comes, and
then you can monitor it over time to see how rapidly, and how severe it is
becoming. Tiger can capture transient or short events with its once
per second sampling time and rather than having to spot check a turbine, you
are able to focus very rapidly on the rare times that sort one transient events
occur.
Tiger has all the tools starting with the diagnostic summary to
troubleshoot or investigate major incidents; to analyse the cause or
sequencing; what happened first, to make an assessment of the damage or the
strain on the turbine that this has caused, and then to write reports and
document it through the displays available in Tiger.
All
that enables you to understand the condition of the gas turbine so that you can
make appropriate decisions, and if we think of the problem of ‘What state is my
gas turbine? Is it okay? Or not okay?’ for maintenance actions, through the
data available from Tiger, you
know it’s precisely state. All this
enables condition-based maintenance which is the goal of minimising the costs
on a site.
HISTORY OF TIGER
A
short explanation of the development history of Tiger. Dr. Robert Milne, Managing Director of
Intelligent Applications in Scotland ,
first had the idea of the concept of Tiger
about 7 years ago. The company was
working in the application of artificial intelligence to industrial problems,
and had developed a very successful knowledge based diagnostic system for
vibration based condition monitoring.
The system was called Amethyst, and was sold by IRD Mechanalysis – at
the time, the world’s largest supplier of hand-held vibration data
collectors.
The
Amethyst system was proving successful in the use of artificial intelligence
diagnostic techniques for condition monitoring.
Gas turbines were then identified as a major area where condition monitoring
would be of great benefit. Rob Milne and
Intelligent Applications had unique experience to combine the techniques of
artificial intelligence; the knowledge of diagnostics and on-line continuous
diagnostics; with the area of condition monitoring and the needs of condition
monitoring.
After
some background research and analysis, Exxon Chemical in Fife ,
just North of Edinburgh, at their Fife Ethylene Plant, agreed to be the initial
customer for Tiger. A relatively simple system was developed to
perform continuous knowledge based diagnostics of the data available on their
General Electric Frame 5 two-shaft gas turbine.
Roughly a year after that system was first developed, Intelligent
Applications lead a consortium of six companies, with funding from the European
Union, under what is known in Europe as the
ESPRIT Program. This project had
Intelligent Applications as the lead; Exxon Chemical as the prime user; John
Brown Engineering as a manufacturer, who had installed the turbine at Exxon
Chemical; Dassault Aviation in Paris ,
looking at smaller aero-turbines; as well as two research groups: CNRS-LAAS in Toulouse and the
Polytechnic University of Catalonia in Barcelona .
These
partners were picked because they had the relevant technical knowledge for the
problems that need to be addressed. That
three-year project, with a total budget of almost $5 million, developed the
prototype Tiger system. Tiger
was then able to develop it over the next year from a prototype to a ready
commercial product and the first installations of Tiger began.
Over
the next few years, the Tiger
installations have steadily increased to well over 25 locations, and 50-fired
years of operation. The product is now
well developed and stable, and even the diagnostics are stable in the sense
that very rarely do major new faults or problems occur, that are not covered by
the diagnostics.
TIGER INSTALLATION
So
what does a Tiger installation
mean, or what does a user get with Tiger? The Tiger
software runs on a Pentium computer and is interfaced to the gas turbine
controller through a communications link.
For example, for a General Electric Mark V controller, Tiger communicates through a modbus
serial link with an Inode supplied by General Electric. The software also includes a backup tape
drive; a printer, and a modem to enable the remote communications.
The
software is installed and provided as part of the initial purchase of Tiger.
Depending on the contract that is agreed, continuous support is provided
for diagnostic backup and analysis of any problems that might occur. In addition, the Tiger diagnostics and the Tiger
software, are being continuously upgraded based on the operational experience
in the field so that a user gets new diagnostics several times per year, and
new faults are analysed and improvements made to the system.
FREQUENTLY ASKED QUESTIONS
In
this section, I want to address a number of questions that come up very often
in presentations to give some background to the questions asked.
Are new sensors needed?
People often ask this question because
they are worried about the cost of having to fit new sensors. The answer is no. Tiger
works with the data available through the gas turbine controller, and hence,
uses what is available. If the right
sensors are not available certain diagnostics are not effective or cannot
isolate the faults to certain levels. So
we will work with whatever they have installed, but the quality of the
diagnostics and their ability to isolate faults might be limited if they do not
have as much as we normally like.
Also in the performance area, the
performance module requires certain inputs.
If those inputs are not available, it cannot perform the performance
analysis.
How does Tiger compare with vibration based condition
monitoring, such as supplied by Bentley Nevada ?
The Bentley Nevada
system requires additional sensors to be drilled into the casing of the gas
turbine; it requires special cabling and some processing equipment that is
generally extremely expensive to supply and fit. Well over $100,000 for installation
itself. In addition, the gas turbine has
to be stopped for a substantial period of time for all these installation
activities. The vibration systems then
look at the mechanical aspects of the shaft, such as, unbalanced, problems with
the bearings themselves, or any kind of rub or bending of the shaft. Tiger
looks at much more with a much simpler installation cost. So Tiger
is much more ‘bang for the buck’ because it doesn’t require additional sensors,
it is easy and rapid to fit; it looks at a wide range of sub-systems and not
just the mechanical aspects of the bearing.
What if I have a vibration-based condition monitoring system and I want to integrate that with Tiger?
Theoretically this
is possible. Intelligent Applications
have had discussions with people with Bentley Nevada on how we can incorporate the
analysis done by the vibration based monitoring system with the Tiger system. Theoretically it is no problem, however, to
date, none of our customers have had a full Bentley Nevada system installed. In fact, the people we help, tend to find Tiger much higher value than buying a
Bentley system anyway.
If I have several gas turbines on my site, can Tiger analyse all of those, and how many computers do I need?
The Tigers are networked for any site, so
that from one Tiger workstation,
you can see any of the gas turbines on the network just by changing from a menu
item. So if there are four gas turbines,
from any one of the display stations for Tiger,
you can see all the other gas turbines.
You can even graph the data across any particular gas turbine.
What do I need if I have several gas turbines?
The pricing of Tiger is based on a software licence,
per gas turbine. So the number of Tigers you pay for is effectively the
same as the number of gas turbines you have.
The number of computers depends on two factors: the physical closeness and location of the
gas turbines and their control rooms.
Because we have the serial links, the limitation is generally how many
serial links we can provide into a Pentium computer and how rapid that can go.
In many cases, a Pentium could process data from several gas turbines, but the
serial links could really only manage two of them just from the physical way
PC’s are made.
Also, you can have
display nodes connected on a network. So
for four gas turbines, we might provide three computers: one computer analysing
the data from two gas turbines; another computer analysing the data from the
other two gas turbines; and the third one being a view node or display station
in the engineers office or control room, with the ability of all three of them
to display data on the systems. Users
can have any number of display stations and remote stations (for a cost).
What if I have a Windows NT network. Can I view the results of Tiger?
Yes. We have software available so that the
diagnostics and displays of Tiger
can be displayed on a Windows NT network.
Does Tiger really keep all the data, or how much data does Tiger keep?
Yes, we keep all
the data permanently. The local disk
drive will typically hold 4-6 months worth of once per second data, all the
diagnostics and all the 24-hour trends that are created. That data is available on-line. We then back-up the data through a tape-drive
and write a CD-Rom containing all the diagnostics data and trends for each
month. That CD-Rom provides a very rapid
and easy access to any of the historical data and can be reloaded onto a hard
disk at any point. Currently the backup
system is through a DAT tape.
Rather than using a DAT tape, why not put a CD-Writer or DVD-Writer into the Tiger system?
This is being
investigated, but the technology and the products are changing rapidly and we
are waiting a few months for everything to stabilise, it will probably be
available in the future.
Can I access Tiger remotely?
Yes, everything
you can do locally in Tiger can be
done remotely. A remote view node or
display system can be programmed to automatically connect across the network
and collect data.
What turbines are available?
Currently, Tiger is installed on a General Electric
Frame 5 and a General Electric two-shaft; a General Electric Frame 6; a General
Electric Frame 9b. A General Electric
Frame 7e has been installed in the US and work is being done to
prepare General Electric Frame 9 as well.
We are also in the
process of developing a Westinghouse 501 version and Pratt and Whitney version
of Tiger.
Can I take data from several sources such as the distributed control system, or other existing systems?
Theoretically yes,
the problem is the drivers from the distributed control system for
communication with other computer systems.
If it is a modern system with a good control system, then it is easy to
interface Tiger, assuming the
software information or drivers are available.
Sometimes the DCS vendors don’t like to co-operate and hence, make it
expensive or inhibit us getting data fed into Tiger.
What about covering the balance of plan, the heat recovery steam generators or the steam turbines?
Tiger is technically
capable of this. We have not done it in
the past because of the problems of getting data from other data sources and
those problems are because the DCS provider made it more expensive to provide
data to us. In the future, Tiger will be extended to cover the
balance of plans and a steam turbine.
Currently it just covers the gas turbine. We have Tiger
installed on a steam turbine with a Mark V controller.
How does Tiger adjust to an individual gas turbine?
When Tiger is first installed it takes a week
or two of data and analyses how the gas turbine varies, identifies a small
operating range, and then sets the limits within Tiger to the variation of that gas turbine, rather than to
control limits and machine protection limits.
In that way, Tiger
customises itself to how each gas turbine individually behaves.
Can Tiger help me make reports to document problems or incidents?
Yes.
Particularly if Tiger is
displayed off a Windows NT network. It
is simple to cut and paste the fault manager displays and graphs into Microsoft
Word documents to then prepare a report.
We have used Tiger output extensively to document faults and problems,
either to justify what maintenance work has done or could provide an
explanation of the incident to provide clear information to the user.
What about the fleet of gas turbines I have.
Tiger is designed to be accessed remotely
from a variety of locations, so if you have a fleet of 6-8 power stations all
with several gas turbines, all the software is in place within Tiger to provide a central
monitoring facility or distributed
monitoring facility to see each of these gas turbines from the new location.
Already there is an installation set-up
that from one office they can monitor several remote gas turbines in a very
simple way. If a company was interested
in central monitoring of the remote fleet, it is ideal.
Can Tiger help me before I send a Field Engineer?
Yes.
Tiger is able to provide an
analysis of the problem through its diagnostics so that before a field engineer
is sent to site, he has got a good understanding of what the probable cause and
area is so that he doesn’t have to arrive and discover that he is missing a
piece of equipment.
Can Tiger work with any control system?
No.
Tiger needs a modern
digital control system such as a General Electric Mark V or even Mark IV
controller. For example, Tiger cannot be used with the old
General Electric Mark II controllers.
How can I get a summary of my most common problems?
Tiger’s query mechanism searches through all
the diagnostic messages and provides either a summary of all the times a
particular fault or problem occurred with all the date and times. Or provides a summary of the number of times
that fault occurs.
For more information, contact:
Sermatech Intelligent Applications
Tel: +44 (0)1506 472047 Fax: +44 (0)1506 472282
Email: ia@intapp.co.uk WWW:
www.intapp.co.uk
|
No comments:
Post a Comment