RAM STUDY METHODOLOGY PART 3 of 3

The use of RAM (Reliability, Availability, and Maintainability) studies in the oil and gas industry has a long history and has become a standard tool to quantitatively predict a production and/or processing facility's performance. Such studies are generally performed in the early design phase through to the detailed design phase and into the operational phase to verify that a facility configuration and design meet the required availability, in turn providing operators and their partners confidence that contracts or production outputs can be achieved.

Renewable Energy Production Challenges

Most renewable energy sources face a particular challenge compared to conventional power production; they typically rely on a natural energy source (i.e. wind, sun, wave height, etc.) which are highly unreliable depending on meteorological conditions. When performing RAM simulation for Renewable Energy sources such unreliability must be included in the model.

In most cases this should be done by applying production profiles of the natural renewable energy sources to a RAM model that contains the production critical assets required to translate the natural renewable energy source into electricity being transmitted into the power grid and/or volume of produced hydrogen. These natural renewable energy production profiles, for example windspeed or sun intensity, will have the biggest impact on production availability over the facility’s system life and need to be as accurate as possible. Since such natural renewable energy sources can change rapidly, it is important to use production profiles with a suitable resolution (i.e. hourly). That may result in very long RAM simulations run times, depending on computer performance and chosen production profile resolution.

 

Typical WindFarm Performance Assessement

The Wind Turbine Generators (WTG) within a wind farm are typically of the same design and capacity. The operational availability of each WTG would be the same, but the production availability can vary depending on their individual production profile (i.e. wind speed profile) and the configuration how the WTGs are connected to the high voltage bus connection. Typically, WTGs are grouped together in a daisy chain configuration passing the power from one WTG Ring Main Unit to another, which causes all upstream WTGs to loss production during downstream Ring Main Unit outages.

Using a flow network is essential for wind farm RAM modelling to be able to model the daisy chain configuration and to assign specific wind profiles to each individual WTG. The Electrical Balance of Plant (E-BOP) is likely to contain the largest loss contributors since failures in the high voltage transformer or switchboard could cause partial or even total loss of the power transmission into the grid. An individual WTG may only show very low loss contribution, since one failing WTG will have an insignificant impact on total power production. Therefore, element critically analyse of wind farms need to be done on the WTG component level. This can be done through post-process calculation where, for example, the combined loss of all WTG generators, gearboxes, and blades is analysed.
 

Maintenance Constains

Renewable energy facilities are likely to experience more varying maintenance constraints than typical oil and gas facilities. These varying constraints are related to varying distance of the power generating facility (i.e. WTG, solar array) to the maintenance hub. Different failure modes could also require different maintenance equipment (i.e. heavy lift cranes) with different mobilisation times.
 

End of Life Scenarios

When renewable energy facilities approach end of life the operator may choose a run-to-failure strategy. This is typically not possible for conventional hydrocarbon power plants due to their Safety Critical Elements required to always be operational. For wind farms a single WTG may only produce a fraction of the total power production and a repair/ replacement of major equipment in WTG may financially more costly then permanently shutting down the WTG and losing that power production.

Simulating such a run-to-failure scenario is not typically what RAM does since RAM simulation is generally based on using Mean Time to Failure (MTTF) parameters that are based on a constant failure rate. Simulating a run-to-failure scenario close to the facility’s end of life requires changing the model assets from an exponential failure distribution to a Weibull distribution function. Furthermore, the asset repair time needs to become larger than the remaining life and such failure should not require any maintenance crew.