Effect of more starts 1

[athomas236]

We have a client that runs 100MWe of diesels on the same site as a combined cycle plant. At about 10.00pm each day the client shuts down the combined cycle plant because of the fall off in electrical demand.

I have been asked to study if the clients approach could be improved and would welcome any advice.

Some of the factors I have thought about are:

1. More starts on the GT's will increase the number of operating hours thus bringing forward GT maintenance which increases annual maintenance costs

2. More starts means higher fuel consumption without generation which increases costs

3. More starts means more water used/wasted because of boiler blowdown which again increases costs

Regards

athomas236

 

[JEB66]

do not forget
If they keep running, how much money they will loose

 

[ccfowler]

The part-load efficiency of GT's is usually fairly low, but if the reduced load is in a reasonable range for the GT combined cycle unit, there may be some serious basis to compare the balance of higher fuel costs vs. reduced maintenance costs. Operating the GT at reasonably reduced loads should significantly extend its service life due to reduced temperatures during low load operation.

It seems likely that the diesels can tolerate daily on-off cycling with lesser maintenance burdens than the GT combined cycle plant as long as reasonable operating practices are followed (gentle warm-up, gentle cool-down, and gentle loading rates).

For reliability, it may be wise to keep diesels in operation at very reduced loads, and the effects of this should be thoroughly considered in your analysis.

This is an interesting problem to study, but don't be surprised if the part-load fuel costs of the GT combined cycle unit dominate the entire matter.

 

[rmw]

A place to get some good GT information whether or not the ones at your site are made by this mfg or not is;

http://www.gepower.com/prod_serv/products/tech_docs/en/all_gers.htm

Without opening the files, I think that GER 3620j has the type of information you are using.

I know that GE uses a "factor fired hours" number for determining maintenance, and there by big buck cost, intervals.

Restarts eat up fired hrs, but hot restarts don't eat up as much as cold starts. Your cost analysis is going to have t balance the less efficient operation at night, (assuming they have some place on a grid to put the power) vs what it is going to cost to overhaul these things when the time comes.

I personally thing that there are a lot of new GT owners that are going to be in for some sticker shock when time comes to do majors on these machines, assuming they make it to enough hours to require a major, and don't 'swarm' first.

But, I am entitled to my own stupid opinion, now, aren't I?

Hope you find something that will help you on that site.

rmw

 

[athomas236]

Thank you all for your helpful advice. To get things moving I have requested the information listed below from the client.

For rmw, I did download the GE paper and found it very useful.

Best regards

athomas236

Information requested
1. Combined cycle units
• Number of blocks at the plant
• Number of gas turbines per block
• Number of HRSG's per block
• Net electrical rated output of block (ISO/site)
• Variation in block net electrical output with gas turbine output
• Start up curves from cold, warm and hot conditions with fuel consumption shown

2. Gas turbines
• Net electrical rated output per gas turbine (ISO/site)
• Main fuel type and GCV/NCV/density
• Performance curves showing fuel consumption against output and ambient conditions
• O&M guidelines for inspection and maintenance intervals with indication of shut down periods

3. Diesels
• Number of engines
• Net electrical rated output per engine (ISO/site)
• Main fuel type and GCV/NCV/density
• Performance curves showing fuel consumption against output and ambient conditions
• O&M guidelines for inspection and maintenance intervals with indication of shut down periods

4. Miscellaneous
• Cost of main fuels for gas turbines and diesels
• Profile of electrical demand for typical day to be satisfied by plant and current division between gas turbines and diesels throughout the typical day
• Typical costs for inspection and maintenance shutdowns for both gas turbines and diesels
• Cost of steam cycle make up water and quantity of water lost during start ups from cold, warm and hot conditions

 

[tomatge]

athomas

This is an interesting investigation ... if your client is running 100MW of diesel power, I've probably heard of the site - can you give a clue as to where this is (even if only what part of world if you have confidentiality issues).

I would add that you need to find out what type of diesels, specifically whether slow-speed or medium speed. Slow-speed plant does not respond too well to extended low load running, nor to frequent start/stop operation. What would the be shared duty of the plant if all were to be kept running.

Gas turbne plant also does not respond well to start/stop operation. I have a maintenance contract from a GT supplier who class one start as being equivalent to 10 hours of running for the purpose of maintenance planning.

In some respects, implications on maintenance and reliability are more difficult to quantify that direct costs such as fuel (which are easily calculated from operating curves).

 

[athomas236]

tomatge

Thanks for the response.

I have some information about the GT's, they are Westinghouse W251's. Westinghouse's data shows that inspections/maintenance are scheduled based on equivalent base hours and starts. For example, the minor combustion inspection should be scheduled at 2000 equi hours or 100 equi starts whichever comes first.

Based on this type of information, I am making a spreadsheet that will schedule the future GT inspections/maintenance for different patterns of starts and running hours. From this I hope to be able to calculate maintenance costs.

I have no information on the diesels yet that would allow a similar schedule to be prepared.

athomas236

 

[rmw]

I would be surprised, no shocked, if the inspections or overhaul interval for the diesels was anywhere close to that of the CT's. Ditto for the cost of same.

rmw

 

[athomas236]

rmw,

Frankly speaking I have no idea about diesel inspections but am waiting for information from the client.

I remember doing a tender evaluation for 2x27MWe diesels years ago and then the costs of maintenance between different engines along with the lube oil consumptions were big issues.

I do not have exactly the same problem now but want to do a like for like comparision.

Thanks again

athomas236

 

[ccfowler]

athomas236,

Operating the GT's at reduced loading overnight should greatly extend the maintenance interval hours just by virtue of the reduced hot gas path operating temperatures. Since this would also avoid many start/stop cycles, the reduced maintenance costs and the reduced outage costs (lost generation, replacement power, etc.)could be very signifcant. These costs should be thoroughly considered in your analysis along with reduced fuel usage efficiency.

Since this is a combined cycle plant, the steam unit can still be producing significant amounts of power to mitigate the reduced efficiency of the GT at reduced loading.

 

[rmw]

You have't said what is the output of the GTCC plant. At what point are they stopping the GT's and starting the diesels??

The GT's are much less efficient at low loads, and if the HRSG has to be supplental fired at night to keep the ST on line, then that is a big cost.

Still, the cost of low load, inefficient operation has to be balanced against the starting costs of the diesels, and the accelerated overhaul costs of the GT's due to the start count.

I suspect that it may be that the load drops low enough that it is not possible to keep the GTCC on line, at which time the diesels are started out of necessity. Could I be right. And, what is the capability of keeping the GT's on line, and tripping the ST. Can you bypass to the condenser, or can you bypass the GT exhaust around the HRSG's.

Maybe the solution is a set of diverter dampers in between the GT and the HRSG.

Let us know some more about the installation.

rmw

 

[athomas236]

rmw,

The two GT's are rated at 45.6MWe each and the ST at 48.5

Regards,

athomas236

 

[cub3bead]

Is it possible to go from 2X1 operation to 1X1 operation on the CC plant? Automatic 50% reduction in starts on the GTs.

Are you addressing the needs of the ST in the cycling operation? To minimize startup time in the morning you need to keep steam to the seals and maintain vacuum on the backend. This will help keep you ST metal temperatures elevated and shorten the time back to the line.

A big side benefit by not breaking vacuum is you keep all of that nasty oxygen out of the system which will help to minimize accumulation of corrosion products in the HRSG.

Do you have a condensate polisher?

rmw's comments about a turbine bypass system and diverter dampers are common tools for keeping the GTs fully loaded. A star to rmw for suggesting them. The diverter dampers do tend to be notorious for leaking and tend to eat up lots of maintenance budget.

 

[rmw]

With emphasis on the "lots". But then again, that applies to all aspects of combustion turbines, doesn't it?

rmw

 

[athomas236]

Thanks a lots guys some good advice.

We do have bypass stacks and I do have performance data for 1x1 operation.

I can understand the benefits of keeping the glands sealed to shorten starts up times but not sure what the benefits would be with having condensate polisher. Is it possible to provide more information on this.

Thanks again


athomas236

 

[cub3bead]

A full flow condensate polisher will reduce the ionic and particulate load in the HRSG feedwater significantly. The benefit of this reduction is that the HRSG stays cleaner because all of that "crud" isn't deposited in the HRSG.

Re-starts will be quicker because you will minimize, if not eliminate, holds on pressure ramping because of water chemistry and/or steam purity issues.

 

[Adam01]

This is an interesting question as indicated by the number of answers.

Generally the recip engines will have a flatter heatrate curve and can be turned down with less effect on the efficiency in comparison to the GT's.

I imagine the HRSG will have a low load limit that limits the turndown of the GT's that will further constrain your ability to reduce load.

Loose one GT would be a simple choice as already suggested. If you have diverter valves in teh HRSG inlet duct you can separate the GT and STG giving you further flexibility.

On the GT cycling you will have to reduce the starts significantly to have any real effect. Going from 300 to 250 starts a year will do little benefit to the hot section maintenance cost, but a reduction from 300 to 100 will do a great deal. Our comparison is 300 to 200 hours/year on 1 GT vs 20 on another results in nearly double the maintenance costs.

My experience with recip maintenance is minimal but maintenance cost is quite high and is more independant of the number of starts than are GT's. Maint cost has a high dependance on operating hours so reducing the operating hours by half will also give you a win. It also provides an opportunity to do that more frequent maint overnight making them more available.

As you have stated a comparison of the operating and maintenance cost impacts of a number of options is the way to go.

The HR curves will be needed for GT's and recips, as will costs as you know. The most difficult issue will be identifying the different maintenance costs for the different starts regimes. You may have to approximate these maint costs into a number of bands ie 1 to 100, 100 to 200 etc. You will need to consider the effect of cycling the HRSG's too these do not respond well to cycling and can be very very expensive to repair.

You did'nt mention (or I did not see) the motivation of the Power Plant to shutdown overnight. Is the plant selling into a grid or is this on a remote network (all power contracted) Either way you will need to have some information on this to determine the implications. to explain further, if the plant is on a remote network then you may simply not have the option to have plant on anyway. In this scenario it is which of the shutdown/turndown options is the best, rather than having to consider selling price.
Is the overnight load peak half the daily load or is the HR of a 100 MW block low enough to make a profit?, If this was so I suspect that all plant would be online (unless I have overestimated the efficiency of the recips). I could not find where you stated the size of the recips.

If you are following the load (isolated network) then my gut would tell me to start with keeping teh GTand STG online at max output and taking recips off after reaching min turndown as the load reduces and opposite as the load increases. These can start-up quicker than the GT and or STG giving you a further benefit in start-up costs.

You could also save on water by boxing up the HRSG's. Not sure why they are blowing them down.

Interesting......

Adam01

 

[taylorg]

athomas236,

Do you know what diesels they have yet - manufacturer, model, speed, fuel used, age and hours run. If they are near the end of their life the exercise you are carrying out may be fruitless as you will obviously be looking long-term.
I'm a completely biased diesel man by the way so any theoretical O&M costs you get for the CCGT plant will be double in reality

 

[athomas236]

taylorg

Over the time I have been considering this matter, like you I am becomming convinced of the advantages of diesels. Lower maintenance costs, higher effieciency which is almost constant with load.

So why do consultants recommend such tempramental gas turbines.

athomas236

 

[taylorg]

athomas236,

CCGT's are usually preferred because the captial cost can be around $200/kW cheaper than diesels, the efficiency is generally higher, and because most consultants know very little about diesels (particularly the unit sizes now available). The reality of running GT's versus diesels is conveniently forgotten about when planning a new plant.
Fuel is also a key factor: if gas is available and base loading is required then CCGT tends to be cheaper in the long run.

Anyway, back to your original question...

The general rule of thumb (despite my preference for diesels)is that your clients CCGT plant should NOT be shut down in favour of the diesels, for a few reasons:

1. 2-shifting a CCGT plant is not healthy for it in the long run.
2. The diesels can be quickly started and fully loaded when the demand increases in the morning. A CCGT plant takes much longer to start and load, even when on 'hot standby'.
3. The efficiency of the CCGT plant will probably be higher than the diesels.

The above is, of course, highly dependant on the particular circumstances of the plant you are looking at. Age/condition of the equipment is a major factor, along with any fuel/power supply contracts that are in place.

Do you know why they shut down the CCGT plant? It doesn't make sense unless their are unique circumstances there.

Hope that helps you out - please let us know how you get on and what you decide!