Some key techno-economic aspects are often overlooked when trying to decide on the right technology for a gas-fired power project, yet they can play a significant part in the overall long-term lifecycle analysis and final decision making process, writes Mark Stevens
When it comes to building a new modern gas-fired power plant, it is important from the customers’ standpoint to ensure that the project specification addresses all of their specific needs relating to the project, and that in doing so it will enable them to be able to make a proper evaluation of the offers they receive – ideally on as much of a level playing field as possible.
To ensure that this is met, at least from the EPC side, the customers (and their consultants) go to lengths to define and spell out in great detail all aspects pertaining to how the plant is to be designed, engineered, manufactured, supplied and commissioned.
They will take great care to ensure that the bidders ensure that all pertinent standards and permits that the contractor(s) must observe and comply with are adequately drawn out in the specification, and rightly so.
As is the norm, the customers will then seek from the EPC bidders certain commercial guarantees against which liquidated damages or penalties for failure to achieve will be levied upon the contractor(s).
These main EPC commercial guarantees are:
- Fixed & firm contract price: to be binding upon the EPC bidder for the validity period of its offer;
- Payment terms;
- Delivery guarantee: namely the contractual time from NTP (notice to proceed) to COD (commercial operation date);
- Performance guarantee: the 100 per cent baseload power output and heat rate for the plant (and possibly certain part-load performances) for specific design condition(s).
In addition, the EPC bidders will also have to provide emission guarantee(s) as required by the specific environment permit for the project, covering the gaseous (e.g. ,NOx, CO) and liquid emissions.
The EPC price, payment terms, delivery and performance guarantees are key inputs needed for the techno-commercial (bid) evaluation.
However, in order for the customers and consultants to make a complete and proper evaluation, there is additional key information required from the EPC bidders, namely:
- Plant maintenance costs: typically customers will focus on the gas turbine unit(s), being the most cost-intensive component in such gas-fired power plants;
- Availability assurance: to allow the customer to make some allowance/prediction for forced and scheduled outages;
- Plant performance degradation: typically being with respect to the plant power output and heat rate deterioration versus operational time due to plant wear and fouling.
These last three items, however, are typically covered under some sort of separate service agreement offering from the bidders rather than being part of the EPC bids, as these guarantees pertain specifically to the operational phase rather than the construction phase.
To these key inputs we can also add the customer’s own project development costs, the project’s financing costs for the EPC phase, fuel and water costs, electricity sales revenue, steam/water/heat production revenues (if a cogeneration project) and realistic operating regimes.
With all of these factors together, the customer can now undertake a lifecycle analysis in order to determine which offer provides the best cost of electricity (CoE) and/or net present value (NPV)/internal rate of return (IRR).
This all sounds fairly straightforward until the customers come to try and make their comparison/evaluation and establish that each of the EPC bidders and/or OEMs have their own maintenance philosophies and/or performance testing methodologies.
Indeed, the following important aspects applicable to all gas-fired power plants – which can have significant impact on lifecycle analyses – are often not given the due attention they deserve.
|Riyadh gas fired plant
‘New & clean’ definition
From the moment of first-firing, the performance of the gas turbine/plant starts to degrade, with the greatest performance loss being seen in the first few thousand operating hours.
So, if the EPC contractor’s performance guarantees are based on a new & clean definition that presumes something possibly in the order of 500-1000 equivalent operating hours, but then finally during the construction phase maybe something on the order of several thousand commissioning equivalent operating hours is actually accumulated, the EPC contractor will adjust the guaranteed performance downwards.
If customers instead take the lead by actually stipulating in their project specifications/request for quotations (RfQs) the “to be presumed” number of commissioning hours, number of ‘commissioning’ starts and number of commissioning trips based on industry averages/experience, then in this way the bidders would have a common basis (to go with the other ‘design conditions’) on which to calculate their respective commissioning equivalent operating hours and, in turn, their respective ‘new & clean’ performance guarantees.
The benefits of this approach are threefold:
- All bidders would have the ‘same conditions’ for the purpose of calculating their respective ‘new & clean’ plant performance guarantees, thereby putting them more on a like-for-like basis for comparison purposes;
- Customers would face fewer instances of having to come to terms with the fact that, due to one reason or another, the actual commissioning EOH ends up much higher than ‘considered’ in the contract, resulting in the performance guarantees being corrected downwards, i.e., worse than expected – which plays out not in customers’ favour;
- If the bidders are, in the end, able to actually complete the commissioning phase with less EOH than presumed by customers’ specifications/RfQs, then this would just mean that the performance basis for the purpose of the guarantees would be higher (better) – which this time plays out in the customers’ favour.
Realistic plant degradation
It is also important for the purpose of carrying out a sensible lifecycle analysis to consider the ‘realistic’ performance degradation to be expected over the financial lifespan of the project depending on technology choice, site environmental/climatic conditions, fuel choice, and probable operating regime(s) that consider the number of start/stops, seasonal load patterns, etc.
Here, customers should request that the bidders provide, as part of their offers, plant output and heat rate degradation plots against EOH.
These plots should be such that the bidders would be prepared to stand behind them from a guarantee viewpoint if required, linked to some form of suitable long-term service contract.
The extent of performance dropoff (degradation) as well as the degree of performance recovery expected at each major inspection would vary from one bidder/gas turbine technology to another – based on each bidder’s assessment of realistic degradation for the specific conditions pertaining to a specific project.
This degradation plot will also have a ‘commercial’ assessment on the part of the bidders, who will be wanting to project something that, on one hand, could be considered sensible/realistic, but at same time is likely to still be competitive versus other bidders/technologies.
Customers would therefore be wise to request that bidders submit their performance degradation plots over a reasonable time frame, such as the financial or technical design life, and not just for the first or second major inspection time frames.
The benefits of this approach are threefold: customers have something that they can use in their evaluation/life-cycle analysis to compare the different bidders/gas turbine technologies; customers are considering a plant performance forecast that is considered to be realistic by the bidders for their respective gas turbine technologies; and, if required, customers have a plant performance forecast that could be used for the purposes of establishing guarantees, if required, linked with some form of long-term service contract.
True cost risks
It is during the bid phase that customers are in the best position to determine whatever they need to know, have tied down and agreed with their selected contractor(s).
As mentioned at the start of this article, too often customers do not give the right amount of attention to the long-term operational phase.
They may make some very simple presumptions regarding the expected operating regime, or presume that all of the gas turbine technologies are effectively the same when it comes to considering and comparing running costs and maintenance regimes.
Again, if customers only look as far out as the first or second major gas turbine inspection interval regarding turbine/plant running costs, the customers may not pick up some specific maintenance requirement that could have financial impacts over and above those considered.
On the other hand, it is also understandable that customers would like to limit any long-term service contract to a reasonable time frame that, on one side, covers the early operational phase when a plant is expected to have the highest operational issues, and, on the other side, is short enough to allow them the ability to reconsider their O&M positions once they have accumulated some operational experience with the selected turbine/plant technology.
However, the bid phase is the best period for customers to ask and find out. It therefore makes sense, at least during the bid phase, to request that the bidders provide their ‘running (maintenance) cost’ projections for their respective turbine/plant technologies over a long time period, say 20-25 years from the start of commercial service, so that, again, a proper lifecycle analysis can be undertaken for comparison purposes.
Then they would finally decide on what is considered from their standpoint to be the preferred, most probably shorter (six to ten years) time frame for any actual long-term service agreement with the selected contractors/OEMs.
There are three key benefits of this approach.
Firstly, by looking to a long time period of, say, 20-25 years, then any and all additional’ major service work that might be required on a specific turbine or plant technology, which could have additional cost and/or outage time impacts, should be picked up on the pre-contract phase radar (examples could be compressor overhauls, rotor overhauls and lifetime extensions).
Secondly, customers are able to undertake more realistic/sensible lifecycle analysis in order to better compare the respective bidders and technologies before making a final decision.
Thirdly, customers have a more comprehensive picture of the long-term running costs pertaining to the different technologies under consideration for their projects.
Even if any possible additional major service work is projected or expected to take place a long way out, such that on a PV basis the financial impact may be considered unimportant, again, it is better on the customer side to ask the questions at the bid phase and be informed than to find out only later during the operational phase, when it may be too late to argue the case.
It is standard industry practice for the bidders to provide the plant reliability and/or availability data for their respective turbine/plant technologies on the basis of 8760 period hours (8784 in a leap year).
Although this can make sense when it comes to availability, it can pose an issue when considering reliability.
Firstly, let us recap the normal definition or understanding of plant reliability and availability. Reliability is generally taken as being the difference between the period hours considered and the actual delivered operational hours for the covered scope in that period, resulting from unscheduled curtailments.
Availability, on the other hand, considers not just forced outages and de-rates, but also any and all scheduled (planned) outages in the same period, namely:
A power plant can, by definition, be 100 per cent available, even if not actually operating, simply by way of the power company declaring it as being available. So to consider the ‘8760 period hours in a year’ time frame is both reasonable and understandable. Reliability, on the other hand, is something that is more usually associated with actual operational performance.
Although the OEMs/bidders consider the 8760 period hours for setting their reliability values, reliability can change dramatically for differing service factors, as shown in Figure 3 (left).
|Figure 3: Relative Plant Reliability (for 175.2 hours forced non-availability)|
The ‘8760 period hours’ time frame represents an ideal case.
In reality, a power plant would not operate constantly during this time, but would instead be shut down when not required by the grid and/or when it is time for scheduled maintenance work to be undertaken.
In Figure 3, we see that for the annual 8760 hours, a 98 per cent reliability factor equates to around 175 hours of forced unavailability.
If this same 175 hours of forced unavailability is witnessed on a power plant that has, perhaps, a service factor of only around 70 per cent (equating to around 6130 period hours) then the plant’s reliability would in fact be around 97 per cent (i.e., 1 per cent less), and for a service factor of just around 50 per cent (equating to around 4380 period hours) this would be around 96 per cent (i.e., 2 cer cent less).
For such scenarios, customers might wish to consider applying a ‘weighting factor’, whereby they require high reliability/availability factors from the plant during high-demand periods in the year, but are prepared to consider some reasonable relaxation during the non-critical periods.
By paying attention to the important techno-economic matters outlined in this article at the specification/request for quotation stage, customers can make a much more informed and comprehensive assessment and evaluation at the bid phase.
Power Engineering International Archives
View Power Generation Articles on PennEnergy.com