Power Modeling More Complex than Refining Modeling

Power Modeling More Complex than Refining Modeling

With my background in refining modeling, I can say power modeling is more complex than modeling most other markets.   The process of producing power is not as complex as refinery production, but the final product pricing and market is significantly more complex.   The refining modeling focus is typically on optimizing yield value while changing the crude input quality within the refineries capabilities.   The products can be stored and transported to the highest paying markets.  With power you just do not have that luxury of storing or even watching where it goes.   I will admit there are certainly more dollars in the Oil & Gas world than the power world.

The power market is highly reliant on what the neighboring power plants are doing, and now potentially, your own customers.  The other power producers may not be using the same feedstock as you.   They also may be operating the plant differently, nonetheless, their actions impact the price and flow of your power in that instant.  The amount of information that must be processed in the power markets, to say the least, is voluminous.   To do proper power modeling, an hourly model is recommend.   If you do not do an hourly calculation you can be misled, for example as in LCOE calculations (http://allenergyconsulting.com/blog/2012/06/05/levelized-cost-of-electricity-lcoe-analysis-potentially-misguides-you-in-the-power-markets/

There are really two types of modeling that one can do.  One is to try to understand market pricing and market dynamics, such as coal and gas demand.   In this setup, you draw your circle quite large and potentially model the entire N. America, which I have done numerous of times.  In this setup you will model over twenty thousand units.   The alternative is to model one’s own system for near-term operations or long-term resource planning, or both.   The market prices become exogenous inputs to the model.   I would caution against this setup in isolation due to the potential impact you may have on the exogenous inputs.   However, in most cases, one typically runs a series of exogenous price inputs to capture the variability.

Both alternatives still deal with the commodity inputs and the load forecast, with the latter being the more potent of the two, assuming you use commodity outlooks across a reasonable range.  Whether you are trying to understand market fundamentals or your system budget, you will have to deal with significant uncertainties.   Market based modeling compounds your uncertainty multiple times.   Modeling your own system, you compress a good portion of the uncertainty into the market price.   However, you still are left with dealing with policy issues (to Carbon Price or not – Renewable Portfolios – Demand Side Management – etc…).  Of the many moving pieces noted in the check list below, most change frequently over time.

Power Modeling Check List

  •          Load
  •          Resources and their Fuel Options
    • o   Coal – by unit, as coal type and transportation is individual (~1300 units)
    • o   Gas – by unit, as each plant could have multiple fuel pipeline access (~4600 units)
  •          Resources and their Performance Attributes
    • o   Uptimes and downtimes
    • o   Ramp rates
    • o   Maintenance
    • o   Cycling
    • o   Start-Up Cost
    • o   Minimum Capacity
    • o   Variable Cost
    • o   Fixed Cost
    • o   Emission Rates
    • o   Secondary Fuels
    • o   Heat Rate – Heat Rate Minimum and Curves
    • o   Storage, Recharge and Initial Contents – for battery & pump storage
    • o   Bidding – fuel adders to multipliers
    • o   Performance Shapes (Solar & Wind)
  •          Emission Pricing
  •          Federal, Regional, and State Policies and Standards (Renewable & Carbon)  

I have been power modeling for over 10 years now – both hands on and managing a team.   Recently, I have spent two years modeling for a client essentially being their outsourced modeler.   The experience has proven to me the viability and value power modeling outsourcing can have.   I would not recommend a complete outsourcing, but I can certainly see outsourcing the functionality pieces.   As Managing Director of Strategic Planning at AEP, I would have loved to maintain some continuity with my power fundamentals team and my modelers as they left for bigger and better jobs.  Each modeler when left to their own devices will devise a custom setup to manage all this information.   The new process would likely leave out something or be organized slightly differently.   Sometimes you get quite an inexperienced modeler and they leave out some crucial issues in the power model.

Power modeling implementation is not as simple as rolling out a new version of office and should not be treated that way.   Even with “out of the box” software implementations from People Soft, Oracle, SAP, CRM, you still need to bring a whole team of consultants to make sure the process gets implemented effectively.   Why would you not take the care to make sure your power modeling is implemented effectively?   It does not take a whole team of consultants as People Soft, Oracle, SAP, CRM software would, but you should take advantage of experienced professionals.  I know some of the software vendors offer implementation help – but just like People Soft, Oracle, SAP, CRM software, it is best to get people who actually do it as a living.   The software people are spending their main resources on their software rightly so.  Therefore, talent is not necessarily on par to achieve your business needs.

My power modeling offering is here for you to maximize your investment in both software and staffing.   To read more about my offering please visit – http://allenergyconsulting.com/2011/11/modeling/

Your Power Modeling Consultant,

David K. Bellman

614-356-0484

Leave a Reply

Your email address will not be published. Required fields are marked *