Scroll

Insights and News /

28 June 2018

Wholesale power price projections for Turkey

Turkish wholesale power market background

Wholesale electricity trading in Turkey was transformed with the creation of wholesale electricity exchange Energy Exchange Istanbul (EPIAS) which was established in March 2015. For Independent Power Producers (IPPs) there are currently four main commercial routes to market:
  • Bilateral trading
  • Spot exchange trading
  • Balancing market
  • Forward / derivatives market for hedging purposes

Bilateral contracting: Long term bilateral contracts are generally referred to as power purchase agreements (PPAs). Contractual terms may last anywhere between 1 and 25 years, during which time the power purchaser buys energy, and sometimes also ancillary services, from the electricity generator. Bilateral contracts may exist between a generation or wholesale company, and a distribution company or an end-customer and the Energy Market Regulatory Authority (EMRA) has no supervisory power over these contracts. Bilateral contracts are not standardised and the terms are not usually made public. In 2017, bilateral contracts between private entities reached 129.5 TWh (2.5% decrease from 2016), representing approximately 28.8% of total annual market transaction volumes respectively [1].
 
Spot exchange trading: Spot exchange trading – whereby generators, traders and electricity supply companies place bids (for buying) and offers (for selling) on the spot electricity exchanges, thus determining the demand and supply curves which are used as a basis for determining the prices and the supply volumes – is rapidly growing in importance for near-term trade. EPIAS currently operates the Day Ahead and Intraday spot electricity markets in Turkey. In 2017, 123.32 TWh was traded in the Day Ahead market (8.1% increase from 2016), and 1.72 TWh in the Intraday market (105% increase from 2016). This represents approximately 27.5% and 0.4% of total annual market transaction volumes respectively [1]. Transactions on the Day Ahead and Intraday markets are executed on an hourly basis every day, with the Intraday market acting as a bridge between the Day Ahead and balancing market to give participants the opportunity to balance their portfolios closer to delivery. An important recent development in the Day Ahead market is the potential for participation from demand side traders, allowing them to adjust consumption based on price levels. Over time, increasing penetration from renewable energy technologies is expected to increase traded volumes in the Day Ahead and Intraday markets.
 
Balancing market: the balancing market is also operated by EPIAS to provide the System Operator (TEIAS National Load Dispatch Centre) with reserve capacity that can be utilised within 15 minutes for real-time balancing. Transactions on the balancing market are also executed on an hourly basis every day. In 2017, 20.0 TWh was traded in the balancing market (28.2% decrease from 2016), representing approximately 4.5% of total annual market transaction volumes (1). Market participants who are able to submit bids/offers on the balancing market are those who are regarded as balancing entities, meaning that they can flexibly change their output within specified limits with 15 minutes’ notice. Such entities are mainly gas-fired power stations and reservoir hydro power stations in Turkey. Note that in addition to the balancing market there is also a range of ancillary services that flexible generators can provide to the Transmission System Operator (TEIAS) such as primary and secondary frequency control.
 
Forward / derivatives market: Electricity can also be traded financially with the aim of hedging against price volatility. Derivatives traded via the organised market in Turkey are referred to as future power contracts or “base load electricity futures”. The forward/derivatives market is based on OTC (Over The Counter) trading which is a form of trading in which the contracts are standardised and with tenors of multiple time horizons. Borsa Istanbul operates the OTC forward/derivatives market in Turkey, with standardised products covering quarterly, yearly and 16-month contract periods. The reference price for base load electricity futures is the average of the Day Ahead Market Clearing Prices for each hour of the maturity period. The forward/derivatives market currently represents a very small proportion of total market volume.
 
 
Baringa analysis of the Turkish power market – modelling methodology
 
For detailed simulation of the Turkish power market, Baringa has developed a bottom-up model of the Turkish power system in which power stations are modelled individually and market dispatch is determined on a least-cost basis, namely to minimise the costs of power generation subject to a range of technical and commercial constraints.  In addition to the Day Ahead wholesale power market, which is where Independent Power Producers derive the majority of their revenues, the optimisation framework also takes the Secondary Frequency Control (SFC) market into account. This is because the secondary reserve requirement in the Turkish power system is significant (typically around 750-1,200 MW per hour) and as a result it can have a strong influence on generation dispatch and bidding behaviour in the Day Ahead market, given that eligible market participants (mainly gas and hydro reservoir plant) will attempt to co-optimise their operation across both markets [2].
 
At the heart of our modelling methodology lies a dispatch ‘engine’ based on a detailed representation of market supply and demand fundamentals at an hourly level of granularity. The supply mix is represented with the operating and commercial parameters of generating plant including costs, operational constraints (e.g. ramp rates, forced and unforced outage rates) and commercial constraints (e.g. existing commercial PPAs). For wind and solar power stations, we use regional hourly output profiles based on historical data for seven regions across the country (Marmara, Aegean, Mediterranean, Central Anatolia, East Anatolia, Southeast Anatolia, Black Sea) and sourced from the same year.  This allows correlations in output across the different regions to be taken into account in the market model.  For hydro reservoir plant, the model calculates the value of energy stored within an associated reservoir, the ‘water value’.  This value is then used to optimally dispatch the hydro generators for a given hydro inflow profile (normalised to represent an average year), subject to storage and operational constraints. The demand side is represented as a projected hourly profile, derived from historical calibration to represent average weather conditions.  Market dispatch is then simulated with system-level constraints optimised to deliver the least cost solution for the Day Ahead and SFC markets.  The marginal cost for each plant is calculated from heat rate curves, fuel costs, transportation costs, non-fuel variable operating costs and carbon costs (if any).  Start-up and no-load costs are used in the ‘unit commitment’ decisions that are taken within the market model. This allows calculation of revenues and gross margins for all simulated power stations in the model.
 
The market modelling methodology employed is illustrated schematically in the Figure below:



Baringa analysis of the Turkish power market – scenario and sensitivity modelling
 
Using the methodology described above, Baringa conducts detailed, forward-looking modelling of the Turkish power market. We model a range of scenarios designed to explore different market, policy and regulation conditions in Turkey, and investigate key issues such as the prospects for gas-fired power generation, the outlook for renewable energy technologies, future changes to the capacity margin in the system, and the potential introduction of a carbon price.
 
Baringa’s fundamental scenario framework is based on the following scenarios. There are no probabilities associated with these scenarios, rather each scenario is discrete and internally consistent.



In addition to the fundamental scenarios that underlie Baringa’s modelling of the Turkish power market, we also model a range of sensitivities that help to characterise some of the key uncertainties in the market such as:

  • Potential introduction of a carbon price in the power generation sector
  • Changes to the Regulation on Electricity Market Capacity Mechanism
  • Changes to commodity prices (e.g. due to potential delays in liberalising the wholesale gas market)
  • Changes to the maximum build rate assumptions for various technologies (e.g. lignite, coal or renewables) due to environmental, economic or policy considerations
  • Changes to the penetration of flexibility technologies such as batteries, demand side response (e.g. from industrial customers or electric vehicles), and future electricity interconnection levels
  • Changes to assumptions on the bidding behaviour of gas and hydro reservoir plant. 
________________________
 
 
[1] EPIAS – 2017 Electricity Market Report (in Turkish). Available at: https://www.epias.com.tr/wp-content/uploads/2018/03/EPIAS_2017_Yillik_Bulten_V2.pdf
 
[2] In addition to the SFC market, a Primary Frequency Control (PFC) market also exists in Turkey to allow the system to automatically respond to frequency deviations.  The primary reserve has to be spinning since the required response time is 2-4 seconds, ramping up to full output within 30 seconds.  The purpose of the secondary reserve is then to relieve the primary reserve back to its normal condition.  It is controlled on-line by the system operator and participating plant must be capable of responding in 30 seconds, ramping up to reach full output within 5 minutes.  The primary reserve requirement of the Turkish power system is considerably smaller relative to the secondary reserve requirement (typically around 200-300 MW per hour) having decreased considerably following ENTSO-E (European Network of Transmission System Operators for Electricity) interconnection with Bulgaria and Greece.  As a result, the SFC market has a stronger influence on generation dispatch and bidding behaviour in the Day Ahead market relative to the PFC market, and this is reflected in the modelling methodology developed by Baringa.