COMPARISON OF TIGER™ WITH OTHER TELE-DIAGNOSTIC SYSTEMS

Comparison of Tiger™ with Other Tele-Diagnostic Systems

Intelligent Applications have developed Tiger knowledge based continuous on-line condition monitoring and diagnostic software system for gas turbines.  Tiger is the most advanced condition monitoring and diagnostic software system available for gas turbines.

This document appraises Tiger's features and benefits against those of its competitors.

The undernoted columns tabulate only some of Tiger's features against those of its competitors.  The listed Tiger features are by no means exhaustive.

The paragraph reference column is a guide to the sections of the attached document which describes in more detail the differences between features of Tiger and its competitors.

TIGER Features	Competitors Features	Paragraph Ref.
1 Second Scanrate.	3 second Scanrate	4, 5, 9, & 10
Automatic Knowledge Based Diagnostic Interpretation.	Manual Diagnostic Interpretation of logged data.	6, 7, 8, 11, 12 & 15
Unlimited Historical Data Storage with Replay Facility.	Limited data storage period	12, 13 & 14
Scheduled Engineering Analysis Back-up with Recommendation Reports.	Call up as required	18.
Trends displayed over unlimited periods for various areas of GT operation.	Trend displays over limited periods.	23
On site display and full client use of installed TIGER system with remote stations where required.	Remote MKV access only.	16
24 Hour 7 Days call out.  H.Q. Back up Facility.	Call up as required.	18
Hypertext link to Operation/Maintenance Manuals.	Not provided.	19
Live displays of Schematic Piping Diagrams for each GT system.	Not provided.	19
Identifies priorities relative to maintenance.	Not provided.	20

1.     Tiger uniquely provides continuous on-line monitoring and diagnostic interpretation of the key information of a gas turbine based on what the gas turbine controller has as information available.  This unique and very important capability, has considerable benefit, but requires a large amount of technological expertise to perform the continuous automated interpretation of the information available without relying heavily on human experts. 

2.     Because the analysis of the operating parameters of the turbine is so important and useful, several of Tiger's competitors are attempting to offer a monitoring package that can be used by experts remotely to troubleshoot problems on the gas turbine. 

3.     This document gives a short summary of these competitive offerings and explains why they are considered inadequate for proper monitoring and troubleshooting of a gas turbine. Presentations on these systems all have the same high-level objectives, and offer similar benefits.  The high level objectives of a better assessment of the condition of a gas turbine leading to: more efficient maintenance, increased reliability, and availability, are common goals of any monitoring system.  Analysis of what these systems provide and what Tiger's provides shows a large difference in Tiger's favour.

4.     The presentations of Tiger's competitors are somewhat misleading, KEL believing that they fall considerably short of the goal of achieving maximum plant availability and reliability. As a specific example, the analysis cycle of Tiger is once every second for the data and all analysis is performed at this once per second cycle.  Extensive experience of Tiger, for more than several years, across more than 20 turbines, has demonstrated that once per second is generally okay for troubleshooting, but only just. Because of the speed of the reaction of the turbine, there are many problems that develop and propagate within only one or two seconds.  Hence, the once per second analysis is critical. Tiger's one-second scanning rate is the shortest scanning period available.

5.     Tiger's nearest competitors scanning rate is once every 3 seconds.  This is too wide a scan period and will miss many incidents, and more importantly, will not identify the sequence of events leading to many problems and trips.  No matter how good their expert analysis is, at this time interval they will not be able to troubleshoot many very critical events.  

6.     Tiger's competitors systems all offer a data logging facility located near the gas turbine. This records a limited amount of data, generally of a duration no longer than 3 hours and commonly much shorter, or a longer period with only broad sampling intervals such as once every 10 minutes.

7.     When the plant calls GE to indicate that there has been a problem, GE will dial into the data collection machine and download the data to their own analysis stations at their offices.  They will then manually perform an analysis of the data and report back to the plant.   In summary, they are offering a data logger located near the PC, that they remotely access when requested to do so and manually analyse only small amounts of data related to the time periods close to the specific incidents.  Others service is limited to this type of capability because it does not have any form of automated continuous expert interpretation and diagnostics. Tiger does! 

8.     If a system were to record all data without continuously interpreting it, the backlog of analysis would be so great and the amount of data to transfer remotely would be so great that it would become impractical for them. Tiger's philosophy is continuous once per second data acquisition and diagnostic interpretation - it is like a doctor following you every second.  Because Tiger is continuously interpreting the data, it can retain the analysis of every second of input forever, rather than being restricted to a small number of incidents. 

9.     The number of inputs received by Tiger and competitor systems, are broadly similar, roughly 200 analogue and 400 digital measurements. Tiger examines all of these inputs at once per second. Tiger's competitors systems do not!   

10.   There are some events that would require analysis at the same speed as the reaction of the controller, such as detailed problems related to, for example, the fuel valves. Currently, no system continuously analyses and diagnoses based at this data acquisition speed.  The only reason Tiger does not do it, is the communications capability of the I node, which is not adequate to provide Tiger with that much data that quickly. The experience of the currently installed fleet of Tiger's shows that once per second is adequate for over 99% of incidents, and hence, the once per second analysis is the optimum balance between on line continuous diagnostic interpretation and the speed of reaction of events. 

11.   A key element of others monitoring philosophy is that the diagnostic analysis is only performed on known incidents.  For example, the plant has to telephone the monitoring service provider and say ‘We have a problem, can you look at the data’.  Although the plant will identify major problems, the time to contact them, download the data, and perform analysis, leads to an unacceptable delay with, no doubt, substantial costs for lost production.   Because Tiger on its own is interpreting the data once per second, it will already have an analysis of the incident, possibly before the machines operators are even aware that an incident has taken place.

12.   Others monitoring philosophy has another serious flaw.  There are many minor incidents on the turbine that affect its overall operation and performance, and also are early indications of bigger problems developing.  Due to the other systems longer scan period these are often not identified at the time they take place. The other systems will also throw away data after a few hours, and so, if they were not aware of the incident, this information will be lost forever. Tiger's diagnostics will pick up the incidents, even if their duration is only one second, and keep a permanent record. 

13.   When certain problems occur, it is desirable to look back over time for other indications or cross correlations of operating conditions and problems.  With others systems the data will be lost.  With Tiger, all data is retained forever, so it is possible at any time to go back, even several years, to look at old data.  For example, at Exxon Chemical there was an ongoing problem with their steam helper turbine, and they decided to check all specific incidents over the past two years to check whether evidence of the problem existed or not.  Using Tiger they were able to easily identify all these incidents over the two-year period, go back and look specifically at that data with a new outlook.  This very important capability is not possible with others systems.

14.   Other systems also have a finite pre-window, possibly only as long as one to two hours before an incident, but critical events could take place over an even longer period, so they are making a compromise and hoping that three hours will be enough.  In Tiger all the data is permanently available.

15.   Other systems will keep a small amount of information on major events, such as the last three trips, the last start-up and the last shutdown.  This is not enough information to keep locally.  For example, there may be problems that will require two or three attempted restarts which would rapidly fill up their data storage capacity, unless data was off loaded in between efforts to get the turbine restarted as quickly as possible.    If there is a failed start or an unexpected trip, the diagnostic would not be available until the data had been downloaded and manually analysed.  Such a scenario would greatly increase the downtime of the GT.  In this same situation Tiger would perform the diagnostic analysis in real-time and the information would be immediately available.  This would be critical in the case where there was a failed start and where it would be necessary to quickly understand the cause and initiate a restart. 

16.   With other's systems the communication between the Plant and the OEM’s office is to have them advise on any incidents will be initiated by the Plant Operator. This means that the system is fundamentally a reactive process.  The Tiger service is proactive. The analysis being available immediately at the Plant on screen for Operational staff to view and action and also at remote stations.

17.   Others systems provides a manual analysis by whoever is on call at the time.  It may happen that faults and problems will exist that the individual on call has not experienced them before.  This could result in an excessive time being taken to resolve the problem.  The analysis of the data collected is performed automatically by the Tiger diagnostic system.   Should it be necessary to contact TTS for assistance Tiger's already available analysis backed up with TTS's support will almost certainly resolve the problem.

18.   If required TTS will provide a “hot line” facility with engineers being available on call 24 hours per day to liase from head office with the client's on site engineers to simultaneously view the GT condition during start up and/or normal load running condition. As well as this hot line service Tiger can be supported from TTS's head office with engineers carrying out a weekly, monthly or ad-hoc review of the gas turbine running condition and trends.  A report on each gas turbine is issued following these reviews.

19.   Tiger has the facility of linking via hypertext into the gas turbine Operation and Maintenance manuals and will take the user through the manuals to the particular device that its analysis identifies as being a source requiring investigation. Tiger also displays live mechanical schematics which show the running condition of all devices which report to the gas turbine control system.

20.   Tiger analyses and identifies priorities relative to maintenance advising and recommending when replacement or checks should be carried out to gas turbine parts and devices e.g. immediately or next scheduled outage. 

21.   The objectives of gas turbine condition monitoring systems are achieved through expert analysis of data. Tiger's competitor offering suggests that this “expert analysis” is a result of their tele-diagnostic system, but it is not, it is only a result of their experts looking manually at the data. Tiger's competitor's systems only have a data replay capability and simple mimic displays to show the status of some parts of the turbine.  This is quite primitive compared to Tiger’s extensive display system. 

22.   For long-term analysis Tiger's competitors offers to look at 7 days of data at intervals of 10 minutes or 30 days of data at intervals of 30 minutes.  The idea of this is to capture any deterioration taking place.  Our experience has shown that this is very inadequate. 

23.   The Tiger trending system can display data down to once every second. Rather than an isolated sample taken at, say, a 4 minute interval, Tiger's trends make a summary of the data over that period. This means that the Tiger analysis encapsulates all events that have taken place, and more importantly, the user is able to identify something interesting has happened and go direct to the once per second detailed data to analyse the incident.  If you only have once every 10 minute data available, you will not be able to go back and see exactly what the problem was.

24.   More concerning is that if their data is only a sample taken once every 10 minutes, this is totally inadequate for troubleshooting any type of problem, except very obvious deterioration.  The Tiger experience has shown that the gas turbine is a very dynamic system constantly changing its operating set points and conditions.  At long time intervals, such as once every 10 minutes, almost all problems will be masked by the dynamics of the gas turbine and hence, invisible.  This analysis at long time intervals that is offered by the competitive systems is then probably useless except for a small number of very obvious problems.

 

Summary

Although the objectives of Tiger and its competitors are similar, because they are the objectives of good continuous monitoring, the competitive offerings to Tiger are very different.

Tiger's competitors fundamentally offer a data logger that is used in a reactive way by human experts to manually produce an analysis of known incidents. Tiger operates continuously at once per second in a pro-active way constantly interpreting and diagnosing incidents, whether they are known at the time or not. 

Tiger does automatically what its competing systems only do by manual support, that is, it continuously observes the GT recognising and analysing abnormalities and reporting these directly to the plant operator and remote monitoring stations.

It enables storage of all running data for an unlimited period of time and has the facility to replay historical running data to observe actual past occurrences which effected the running of the gas turbine.

Tiger will maintain its considerable advantages over its competitors through continuous software updates incorporating experiences gathered and passed on from the numerous users of the GE heavy duty gas turbines.

The unique facilities of Tiger together with the engineering back up available from TTS in Orlando make Tiger the most sensible choice for a client seeking a tool which will greatly assist in achieving minimum unscheduled machine downtime and maximum machine availability.

For more information, contact: Sermatech Intelligent Applications 1 Michaelson Square, Livingston, West Lothian, Scotland UK, EH54 7DP Tel: +44 (0)1506 472047              Fax: +44 (0)1506 472282 Email: ia@intapp.co.uk                                WWW: www.intapp.co.uk

Tiger is trademark of Intelligent Applications Ltd

21 August 2001