New_Market_Designs_for_the_Power_Sector_in_Armenia,_September,_2016.pdf

September 2016

This publication was produced for review by the United States Agency for International Development (USAID). It was prepared by IRG, an Engility Company.

NEW MARKET DESIGNS

FOR THE POWER SECTOR

IN ARMENIA

ENHANCING CAPACITY FOR LOW EMISSION DEVELOPMENT STRATEGIES

(EC-LEDS) PROJECT

CONTRACT NUMBER AID-OAA-M-15-00005

ARMENIAN NEW ELECTRICITY

NEW MARKET DESIGNS

FOR THE POWER SECTOR

IN ARMENIA

Enhancing Capacity for Low Emission Development Strategies (EC-LEDS) Project

Contract Number AID-OAA-M-15-00005

September 2016

DISCLAIMER

The author’s views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development or the United States Government.

This page is intentionally left blank.

NEW MARKET DESIGNS FOR THE POWER SECTOR IN ARMENIA I

TABLE OF CONTENTS

1. INTRODUCTION

2. EXECUTIVE SUMMARY

3. OBJECTIVES

4. NEW MARKET MODEL

4.1 Market Structure and Activities

4.2 Long-Term Planning

4.3 Capacity Market

4.4 Monthly Contracts

4.5 Day-Ahead Market

4.6 Balancing Market

4.7 Ancillary Services

5. ELECTRICITY TRADE MECHANISM

5.1 Capacity Market

5.2 Electricity Trade

5.2.1 Monthly Contracts

5.2.2 Day-Ahead Market

5.2.3 Balancing Market

5.2.4 Approaches to Settlement (Electricity Trade)

5.2.5 Ancillary Services

5.2.6 Electricity Transmission

6. REQUIRED SOFTWARE

6.1 Month- and Day-Ahead Planning

6.2 Market Simulation Software

6.3 Software for Parallel Process of Markets Operation

7. KEY MILESTONES FOR THE FIRST STAGE OF REFORMS OF

DOMESTIC MARKETS

8. ARMENIA-GEORGIA ELECTRICITY TRADE OPTIONS

8.1 Bilateral Contracts

8.2 Trade Opportunities on the DAM

8.2.1 PX Node Mechanism for Armenia-Georgia Trade

8.2.2 Market Coupling Mechanism for Armenia-Georgia Trade

9. INITIAL ACTION PLAN TO SUPPORT CROSS-BORDER TRADE

10. CONCLUSION

APPENDIX 1. REGULATORY CONCEPT OF IMPROVEMENT OF

THE ARMENIAN POWER MARKET AND

IMPLEMENTATION OF NEW MECHANISMS

II NEW MARKET DESIGNS FOR THE POWER SECTOR IN ARMENIA

LIST OF FIGURES

Figure 1. Energo-Pro Consumption Load Shapes for 2013 (in Kilowatts (kW)) Figure 2. Consumption Planning Load Shapes for May 2014 (in MW) Figure 3. Ladjanuri HPP Planning Load Shapes for February 2015 (First Week) Figure 4. Partial Pool Structure Figure 5. DAM Options for Each Hour Figure 6. Pricing on the DAM Figure 7. Hourly Deviations Traded on the Balancing Market Figure 8. MAP/DAP Software Figure 9. Market Simulation Software Figure 10. Software for Parallel Markets Implementation Figure 11. Bilateral Contracts Mechanism Figure 12. PX Node Mechanism Between Georgia and Turkey Figure 13. PX Node Mechanism Between Armenia and Georgia Figure 14. Market Coupling Figure 15. Export from Country 1 to Country 2 on the DAM Figure 16. Import to Country 1 from Country 2 on the DAM

LIST OF TABLES

Table 1. Pricing Options for Deviations Settlement Table 2. Settlements Options on the DAM and BM

NEW MARKET DESIGNS FOR THE POWER SECTOR IN ARMENIA III

ACRONYMS

ANPP Armenian Nuclear Power Plant

BM Balancing Market

DAM Day-Ahead Market

DISCO Distribution Company

EC-LEDS Enhancing Capacity for Low Emission Development Strategies

EMM Electricity Market Model

ENA Electric Networks of Armenia

EPSO Electric Power System Operator

ETM Electricity Trade Mechanism

EU European Union

GTMax Generation and Transmission Maximization Software

HPP Hydro Power Plant

HVEN High-Voltage Electric Networks

IRG IRG, an Engility Company kW Kilowatt(s)

MoENR Ministry of Energy and Natural Resources of Armenia

MW Megawatt(s)

MWh Megawatt Hour(s)

NPP Nuclear Power Plant

PSRC Public Services Regulatory Commission of Armenia

PSS/E Power System Simulation for Engineers

PX Power Exchange

SHPP Small Hydro Power Plant

TPP Thermal Power Plant

USAID United States Agency for International Development

WG Working Group

1. INTRODUCTION

The New Market Designs for The Power Sector in Armenia was prepared under the USAID Enhancing Capacity for Low Emission Development Strategies (EC-LEDS) project managed by IRG, an Engility Company and a team of local and international experts. The main goals of the EC-LEDS project in Armenia are: 1) to implement activities in support of the national energy strategy, 2) to propose a structure for market reforms, 3) to promote cross-border trade with Georgia and 4) working toward the harmonization of Armenian regulatory practices and electricity system standards with those of Georgia in compliance with European Union (EU) directives.

This report builds on the Gap Analysis with List of Major Market Challenges and Legal Barriers Report, which focused on necessary market design changes and opportunities for cross-border trade between Armenia and Georgia. This report also provides a new market model and several market opening scenarios, transitional phases, and an action plan for implementing mutually beneficial cross-border trade with Georgia.

Any reform process takes years. Implementation takes place in stages, and at each stage, detailed solutions flesh out concepts identified in earlier stages. This report presents the initial design for a proposed new market structure. As with other reform processes, further detailing will be needed in subsequent stages of implementation of this concept.

The issues relating to reforms of the Armenia domestic market in this report are grouped under the following four headings:

• Market Structure and Model

• Electricity Trade Mechanisms

• Implementation Stages Including Required Software Development

• Action Plan

2. EXECUTIVE SUMMARY

The Armenian power market’s current design was primarily developed in 2004, when ArmEnergo’s single-buyer responsibility was shifted to the Electric Networks of Armenia (ENA), the distribution company (DISCO).1 The removal of ArmEnergo was a necessary, but insufficient and incomplete, step for creating an effective and transparent power market. The DISCO was established to provide electricity at regulated prices, even if generation costs rose for any reason. Prior to 2012, there was no balancing mechanism to compensate for DISCO’s financial loses/profits due to mismatches of actual and forecasted cost of purchased electricity. Thus, end-user tariffs could increase or decrease on an annual basis to reflect the generation costs of the previous year in order to cover the full market cost of generation, but consumers did not receive better market reliability in return. The key disadvantages of the existing market and measures to increase its efficiency are presented in the Gap Analysis with List of Major Market Challenges and Legal Barriers Report.2 This report explains the new market model and an electricity trade mechanism (ETM), proposes a path to implement it, and discusses how it can improve the performance of market participants through liberalization of overall market operation.

Armenia’s transition to a fully competitive power market is based on the needs of the domestic market, but it also takes advantage of trade opportunities with Georgia. Cross-border trade would yield advantages while also improving the efficiency of the internal Armenian market. The key objectives of the new market structure are:

• Transition to an hourly market (at least) to enhance electricity sector operation

• Efficient allocation of responsibilities between market participants and end-users

• Protection of domestic consumers

• Realization of economic cross-border trade opportunities

• Creation of an enabling environment for investors

The new market model requires both structural and functional transformations, especially in the functions of market participants and operators. In particular, the following should be established:

• Market Operator – To carry out a wide range of functions in the new domestic market, including preparation of rules and cross-border trade processes

• Suppliers – To increase efficiency and competition in retail markets that enable unbundling of distribution and supply functions

• Traders – To conduct export/import transactions in wholesale markets and to aggregate individual generation units to increase their competitiveness in domestic markets and in regional trade

• Large Consumers Entering the Wholesale Market – To participate in wholesale and retail markets and provide more depth to those markets

1 ArmEnergo was a government owned company which was responsible for buying electricity from the generation sector and selling it to the Distribution Company at average prices. The ArmEnergo Company was liquidated after the Single Buyer market was established.

2 Gap Analysis with List of Major Market Challenges and Legal Barriers Report, IRG, an Engility Company, June 2016.

The suggested modifications for market participants and the System Operator in the new market require new regulatory approaches by the Public Services Regulatory Commission of Armenia (PSRC). These range from the planning of trading volumes and tariff calculation methodologies to monitoring overall market performance and participants’ market activities in particular.

The new market model, presented in Chapter 4, includes several important differences from the existing market model. The new market model organizes electricity trade through the 1) Capacity Market; (2) Monthly Contracts (3) Day-Ahead Market; (4) Balancing Market; and (5) Ancillary Services. Such a market model will increase the efficiency of the power sector, provide better protection for domestic consumers, provide clearer assignment of responsibilities, enhance competitiveness, provide more transparency of functions, and enhance conditions to attract investments. Ideally, competitive markets begin with a competitive surplus of generation. Where a surplus is lacking, as is the case in Armenia, a transitional market model needs to be implemented. Chapter 5 discusses each option, the appropriate transition phases, and the details of the proposed ETM.

This major mechanism and options for each market time segment are as follows:

• Capacity Market through capacity tariff implementation

• Monthly contracts based on 1) a Partial Pool (PP) concept and 2) direct negotiations between market participants, including export/import transactions

• DAM based on 1) cost-reflective prices with partial liberalization, 2) generators’ offers only, and

3) generators’ offers and off-takers’ bids. The marginal and weighted average pricing principle for settlement are used in this proposal

• BM –A deviations market without bids must be implemented at the initial stage. Deviations are distinguished based on source: 1) dispatcher-initiated orders ask generators to respond to external factors—generators may be rewarded with bonus payments; and 2) market participant-initiated deviations—penalties may be imposed for generator noncompliance. Marginal and weighted average pricing principles are again considered

• Ancillary Services through Capacity Market mechanism and System Operator’s special fee

During the transition period, numerous multi-optional calculations and analyses for co-optimization should be conducted. The required software for this stage is presented in Chapter 6, namely:

• Month-Ahead Planning and Day-Ahead Planning for market participants to increase planning accuracy and price optimization. Using this software will allow market participants to improve skills on the hourly market

• Simulation software incorporating the main principles of the new market functions to enable more informed policy choices

• Parallel market operation. This software carries out all calculations in accordance with the new model and ETM (without financial obligations during the first stage) and provides results that can be compared with the existing market (settlement and invoicing will be performed by existing models until the final decision is made to transition to the new market)

Key milestones for the first stage of domestic market reform (see Chapter 7) are:

• Establishment and approval of the detailed Armenia Electricity Market Model (AEMM) and

ETM

• Acceptance of the concept of the new electricity market in Armenia

• Establishment of the market development working group (WG) operating on a permanent basis with the inclusion of specialists from the PSRC, Ministry of Energy and Natural Resources of Armenia (MoENR), Electric Power System Operator (EPSO), and Settlement Center.

• Initial software development

• Implementation of multi-optional calculations and analysis in order to select the optimal principles of market functioning and ensure a smooth transition to the new market

• Creation of new and necessary amendments to the existing legal and regulatory framework, as well as their harmonization with EU legislation

• Development of new market rules and procedures

• Capacity building for market stakeholders (e.g., presentations, training, roundtables)

• Action Plan for transition to the new market

A timetable on the electricity market reformation process should be developed upon approval of the concept and formation of an appropriate WG.

Cross-border trade with Georgia is an important tool to enhance operation of the electricity sector.

Possible principles for organization of such trade between Armenia and Georgia are considered in Chapter 8 and include:

• The need to transition to hourly trading for DAMs and BMs in both countries

• Trade through bilateral contracts between market participants (Market Operator and DAM are not needed in both countries; only System Operators are necessary)

• Implementation of the Power Exchange (PX) Node mechanism for cross-border daily trade (Market Operator or responsible bodies must be developed in both countries; domestic DAMs are not mandatory)

• Market Coupling mechanism implementation as a recommended option (market Operators and domestic DAM establishment are needed in both countries)

• The need to develop balancing rules, including metering and settlement

The main steps required for organizing Armenia-Georgia cross-border trade are presented in the Initial Action Plan (see Chapter 9), including:

• Declaration of transition to hourly cross-border trading based on both bilateral contracts and daily agreements in both countries

• Authorization of trading through bilateral contracts between market participants

• Agreement between Armenia and Georgia on establishment of Market Operators (or responsible bodies at the initial stage) in both countries

• Agreement on the introduction of DAMs to the domestic markets of both countries allowing Day-Ahead Trade using the Market Coupling mechanism and implicit auctions

• Selection of an optimal mechanism for Day-Ahead Trade

• Choice of cross-border trade options and decisions at the time of their implementation (e.g., timetables for establishment of Market Operators, domestic DAMs, Market Operator-to-Market Operator agreements, coupling software)

• Extending the existing Joint Working Group of Armenian and Georgian experts Working on the details and development of the following regulatory framework to support cross-border trade:

– Bilateral contracts principles

– DAM mechanism implementation

– Balancing rules

– Metering rules

– Settlement rules

3. OBJECTIVES

The Armenian power market’s current design dates back to 2004, when the MoENR and PSRC ended ArmEnergo’s monopoly as a state-owned single-buyer and single-seller trading company with a primary charge to keep end-use prices stable when generation costs fluctuated. The disbanding of ArmEnergo puts the privately owned electricity distributor, ENA (the DISCO), and the generating companies in a new power market where ArmEnergo’s single-buyer responsibility, including financial responsibility for the sector, has been shifted de facto to the DISCO. The removal of ArmEnergo was a necessary, although insufficient and incomplete, step for creating an effective and transparent power market.

The DISCO was created to provide electricity at regulated prices even if generation costs rise. Before 2012, there were no balancing mechanisms to adjust the DISCO’s margin to make up for any loss or gain incurred by a mismatch between actual and forecasted cost of purchased electricity. Since 2012, PSRC has used a mechanism allowing the DISCO to compensate its losses/gains occurring when EPSO’s results did not achieve the forecasted generation mix. This mechanism was based on an annual ex-post assessment of the generation mix; it allows recoupment of the losses/gains over the next three-year period without recognizing any interest rate. In April 2016, PSRC improved this mechanism by allowing an annual recoupment with interest at 12 percent. When the generation mix is not achieved, for whatever reason, the entire financial burden, including interest, will be covered by the end-users’ tariff during next year. On one hand, this mechanism adds stability to the power market and avoids situations like the recent financial crisis. On the other hand, there is no longer a reason to keep the power market totally regulated if all risks resulting from the generation mix are shifted to the end-users on an annual basis, along with 12-percent interest. More simply, end-user tariffs fluctuate on an annual basis to reflect the generation costs of the previous year or cover the full market price without having sufficient market responsibility mechanisms. The value of not having responsibility mechanisms is included in the DISCO’s interest rate covering the fluctuation of the generation mix. At this stage, it is unclear what value is provided by this market model.

The key disadvantages of the existing market and the measures needed to increase its efficiency are presented in the Gap Analysis with List of Major Market Challenges and Legal Barriers Report.3 This report discusses the development of the new market model and ETM that will increase efficiency for market participants and avoid negative impact on end-users. In the course of developing a new market model, the following issues must be solved:

• Structure and management of the market

• Planning and approval of tariffs

• Optimization of the trade mechanism with sharing of responsibility

• Cross-border trading with Georgia as a tool to increase efficiency of domestic market

• Investment policy

The following issues are the first priority:

• Market structure optimization

• Market rules development

• Ensuring fair and impartial conduct of market participants

3. Gap Analysis with List of Major Market Challenges and Legal Barriers Report, IRG, an Engility Company, April 2016.

• Market operations transparency

• Transition to state-of-the-art hourly trading in DAM and BM time segments, improving on the monthly basis of trading currently used

• Sharing of responsibility between the participants and retail consumers

• Defining transition timing and mechanisms to prevent any possibility of a “shock scenario”

4. NEW MARKET MODEL

4.1 MARKET STRUCTURE AND ACTIVITIES

Development and implementation of a new market model is aimed at increasing the efficiency of market operations and improving the legal and regulatory basis of the market. The new model contemplates changes in market structure, new principles of operation, and reformed trade mechanisms. In particular, the following structural reforms are needed:

• Establishment of a Market Operator with the following functions:

– Participation in the development of regulatory documents governing wholesale market operation

– Organization of wholesale trading of electricity, capacity, and services

– Registration of bilateral contracts for sale/purchase of electricity and capacity

– Organization of metering and data acquisition of actual generation and consumption in the wholesale market

– Interaction with entities to forecast the volumes of generation and consumption

– Implementation and maintenance of software and information systems

– Monitoring actions of the System Operator and market participants

– Implementation of settlements in compliance with market rules

– Ensuring transparent market functioning

– Dispute resolution among market participants

• Establishment of suppliers and traders able to conduct the following activities:

– Export/Import Transactions – The analysis of regional trading opportunities shows the economic benefit that can be gained from cross-border trading; suppliers and traders need to be prepared to choose the best options (e.g., export, import, daily exchange)

– Sale of Power – With the gradual opening of the market, the traders may sell directly to customers competing with the ENA

– Aggregation of Individual Generation Units – The 15-year period provided for by the law for guaranteed sale of electricity at approved tariff rates has expired for many small hydro power plants (SHPPs) but is currently being extended. In the future, the buyer (presently the monopolized ENA) will be able to dictate decreasing sales prices of these plants (although the price of these hydro power plants (HPPs) is currently lower than the price proposed by thermal power plants (TPPs)). An aggregator for the group of small generators may be a more effective negotiator on behalf of these plants. Moreover, an aggregator may be able to deal with foreign markets more effectively than single-plant operators

• Entry of large consumers into the wholesale market. Rules and mechanisms need to be identified for large countries to enter the wholesale market. Particularly to be addressed are rules for electricity purchases, connection rules, and charges for use of the distribution network.

These consumers will participate only when it is economically advantageous for them

Transitioning the DAMs and BMs at least to hourly trading units is justified by the following:

• Technological characteristics

• Modern principles of power markets in unbundled markets

• Principles of regional trading and strong interconnection with the domestic market

• The values of economic dispatch

• Congestion management

• Necessity to share responsibility between market participants

• The mechanism of market balancing

For existing participants (i.e., generators, consumers), enlargement of functions is required in compliance with trade principles of the new ETM (see Chapter 5). The main difference with the current situation is the requirement to participate in short-term and long-term planning, to prepare and submit bids (load schedules and prices), and to take responsibility for market deviations.

The existing System Operator (i.e., Independent System Operator or Transmission System Operator in new model) will have greater responsibilities to solve market challenges, in particular:

• Administration of economic mechanisms to support the maintenance and development of generating capacities in the required volumes, with required technical parameters, in time to meet customer needs and considering the network infrastructure development forecast

• Selection of generation dispatch configuration

• Actualization of the designed power system model

• Economic dispatch with consideration of limitations on the generators’ regimes and network infrastructure (congestion management)

• Competitive selection of offers/bids of suppliers and buyers defining the prices and volumes of supply for each hour of the coming day, together with the Market Operator (DAM)

• Real-time management of balancing based on minimization of electricity supply cost (BM)

• Ensuring functioning of the system services market

4.2 LONG-TERM PLANNING

Accuracy of long-term planning is lower than short-term planning. Lacking short-term planning tools, the current market model uses annual planning volumes for calculation and approval of end-user and wholesale tariffs. It is recommended to keep this instrument mainly during the transitional period for tariff calculation.

4.3 CAPACITY MARKET

The main objective of the Capacity Market is to ensure sufficient amounts of operational generating capacity to cover electricity demand at any time with appropriate reliability and quality parameters (including generating capacity reserves).

A working Capacity Market prevents a capacity deficit in the middle- and long-term perspective, defines the responsibilities of owners of generating units to maintain equipment ready for operation, and com-pensates part of the fixed costs of operating power plants (the amount of fixed-cost coverage may vary depending on the characteristics of the power system). The investment attractiveness of construction is also enhanced. Availability of an efficient Capacity Market makes exports more competitive.

However, there is currently no mechanism to control capacity implementation, and the consumer actually pays for capacity as declared by generators. The specific nature of generation capacities in Armenia softens this requirement to a certain degree because several power plants consist of one unit only (e.g., Armenian Nuclear Power Plant (ANPP), Hrazdan-5, Yerevan TPP).

4.4 MONTHLY CONTRACTS

Monthly contracts are expected to be the most used competitive market in the first stage of reform.

This contract market brings several advantages: 1) it ensures price stability for buyers (contracts should not be at tariffs exceeding the PSRC-approved tariffs) and 2) they allow a buyer to plan its activities while recognizing the anticipated expenses of power purchase. Although the contracts might be signed for a one-year period, it would be reasonable to specify the contracted volumes in months.

Monthly contracts might be either regulated or nonregulated. While signing nonregulated contracts, the market participants themselves may choose contractors and set prices and volumes of power supply.

Monthly contracts, will start bringing responsibility to the power market. The key principle of the contractual relationship will be the principle of firm supply contracts. Specifically, if the generator produces less than the contract volume, then the generator must buy the missing shortage on the DAMs or BMs to fulfill its obligation. Correspondingly, unused power is at the risk of the buyer and is settled in the same way. The monitoring of the contracts is on an hourly basis.

The contracts should also take into account network losses. The most acceptable method is for a contract to set the delivery point as the point of generation, meaning the buyer will cover losses (mechanisms to identify losses are discussed further in Chapter 5).

4.5 DAY-AHEAD MARKET

In electricity planning, longer term plans require more adjustment as the delivery time approaches.

Volumes of electricity not anticipated by monthly contracts require fine tuning to adjust differences with day-ahead planning load shapes. The differences are traded on a DAM with volume and price determination for each hour. A DAM is conducted every day before the actual delivery day.

In general, the DAM price is defined as an equilibrium node price, which is characterized by a cross point of demand and supply curves, determined by prices and volumes set by bids and offers of consumers and suppliers. If this mechanism is applied and not all the day-ahead planning volume can be covered, the remaining part is traded on the BM.

If only the generators’ offers are used, the consumers will buy electricity by the established generation price, and the day-ahead planning shape for each market participant will be fully covered. The alternative to this approach would be to include no price bids of market participants. In this case, the prices are determined by a centrally accepted mechanism; for example, application of the approved tariffs with the respective defined limits of increasing coefficients (partial liberalization of price).

These mechanisms will be discussed in detail in Chapter 5.

4.6 BALANCING MARKET

Actual usage varies from day ahead forecasts just as day ahead forecasts vary from monthly or annual forecasts. Consumption level alterations and network restrictions cannot be forecast in advance with absolute accuracy. Real-time management of the power system requires a way to compensate for the deviations from the regime planned even a day before.

Real-time management is based on minimizing power delivery cost, which calls for measures that encourage market participants to honor dispatch orders and make it disadvantageous to deviate from planned values.

Unauthorized deviations are traded on the BM, and market participants are either “fined” (if the deviation is self-initiated) or “rewarded” if the deviations are made to comply with dispatch orders.

Possible options for the BM will be addressed in Chapter 5.

4.7 ANCILLARY SERVICES

The main task of the system services market is to ensure functioning of economic mechanisms that stimulate support and development of power system equipment. These services are specific and nonobligatory; they depend on individual participants’ characteristics, such as:

• Ability to participate in normalized primary frequency control

• Ability to participate in automatic secondary frequency control at the TPP

• Reactive power control

• Participation in emergency recovery

The System Operator defines the technical requirements, selects service suppliers, and makes long-term contracts to compensate providers for these services. The costs of these services are recovered across all energy usage.

5. ELECTRICITY TRADE MECHANISM

This chapter explains the principles and mechanisms of trading on the domestic market and with neighboring countries. These principles and mechanisms lead to a discussion of the modifications to Armenia’s legal and regulatory framework that should be planned in the next stage of the project. The suggested ETM will ensure reliable power supply with minimum growth of tariff rates for end-users a prerequisite for efficiency of the domestic market. Cross-border trade with Georgia is treated as a tool to improve the efficiency of the domestic market. As described in Chapter 4, the new power market will be divided into several time segments of trade, which are presented below.

5.1 CAPACITY MARKET

In the current single-buyer model, capacity charges are paid for by the DISCO only, which allows generators to reduce prices for export. Capacity payments may be implemented almost without a special mechanism to control operating capacity due to these facts:

• The ANPP, Yerevan TPP, and Hrazdan-5 are power plants with only one unit each

• The Vorotan Cascade of HPPs operates at maximum load at peak hours. Its available capacity is easily monitored through implementation of the dispatch schedule

• The Sevan-Hrazdan Cascade operates, in general, on an irrigation schedule; thus, the availability of its full functionality can also be easily controlled

If several wholesale consumers appear on the Armenian power market, the payments for capacity will be made proportionally based on their consumption, and payments to each generator will be proportional to its capacity contribution.

When the market is opened, it will be marked by the appearance of energy retail companies and the entry of large consumers in the wholesale market. Export volumes will increase and will include new generation capacity. At that point, it will be necessary to develop payment methods in the Capacity Market that will include:

• Payment for capacity proportional to participation in the market

• Allocation of generation between consumers

• Possible distribution of payments between domestic consumers and exporters

• Control over construction, connection, and cost allocation for newly contracted generating capacity

5.2 ELECTRICITY TRADE

5.2.1 MONTHLY CONTRACTS

The Capacity Market is intended to satisfy multiyear needs. The market also needs a contracting mechanism for participants to supplement their annual, seasonal, and monthly requirements. For this, bilateral long-term contracts are recommended, with terms of at least a month and up to one year, with adjustment of their parameters at least each month. This will allow demand-side parties to account for seasonal differences in consumption (the form of load shapes and volume changes) and for generators to reflect more accurate estimates of their capability. Although monthly contracts will differ from actual regimes, this will help consumers to predict and react to the expected level of prices. The first task is to establish monthly planning for each hour. The following format is recommended:

• Market participants prepare and submit offers (by producers) and bids (by users)

• The System Operator monitors offers and bids, making corrections if needed

• The System Operator (or Market Operator, if established) schedules optimal dispatch, recognizing congestion management needs

• The System Operator (and Market Operator, if established) approves the hourly balance for the planned month

Because market participants lack experience in trading on an hourly basis, the algorithm and software will be provided to them.

Preparing Bids The bidding process anticipates that the following mechanisms will be available:

1. Historical load shapes for each market participant will be taken from the metering system database with one to three years’ depth.

2. Load shapes are separated by type of day. For Armenia, breakdown by working days, Saturdays, and Sunday plus holidays is sufficient.

3. The initial load shapes are filtered to remove the anomalous days or hours where there may be problems with meter readings or data.

4. Curves are defined for typical days of the month.

5. “Daily coefficients” are defined, which account for the difference between anticipated daily volumes and typical days.

6. “Weekly coefficients” are defined, which account for the dynamics of changes of volumes by weeks.

7. The calendar is synchronized (to align the planned month with the corresponding historical month).

8. Weight coefficients are defined by each hour of the month.

9. Capacities are defined for every hour of the month as a product of the planned monthly volume and the corresponding coefficients from item #8 above.

This methodology has been tested in Georgia by Energo-Pro Company (it balances 15 power plants with over 400 megawatts (MW) total capacity) and has proven its functionality.4 As is shown on Figure 1, there is a significant difference between daily consumption load shapes depending on the day of the week and the dynamics of changes by months (using metering data for 2013). This methodology yields a breakdown of the monthly volumes into volumes for each hour of that month. Figure 2 shows the daily consumption load shapes for the third week of May 2014. Using this methodology produced accurate results for February 2015 (Figure 3), even for the largest regulated generator, Lajanauri HPP,

4. Hourly planning for market players for the Georgian electricity market based on hourly metering data from Energo-Pro, Deloitte, April 2014.

which is regulated in winter, but, in summer, due to the “high water” effect, it operates almost at maximum capacity.

Figure 1. Energo-Pro Consumption Load Shapes for 2013 (in Kilowatts (kW))

Figure 2. Consumption Planning Load Shapes for May 2014 (in MW)

Figure 3. Ladjanuri HPP Planning Load Shapes for February 2015 (First Week)

Monitoring of Offers and Bids The System Operator will initially review offers and bids it receives for apparent compliance with the following criteria:

• Incorrect monthly volume

• Incorrect minimal or maximal loads

• Large or unexplained difference between daily volumes

• Large or unexplained difference by weeks

If the System Operator detects violations, it should consult the market participant to make the appropriate correction(s).

Optimal Dispatch Scheduling and Congestion Management The System Operator calculates an economic dispatch schedule covering all consumption based on consumers’ bids using the price minimization criterion. This is called Optimal Dispatch Scheduling.

To solve the issue of generation dispatch in Armenia, the Generation and Transmission Maximization Software (GTMax)5 software installed at the System Operator is being used. The suggested method-ology of receiving offers and bids from market participants allows a significantly simplified use of that software because it already contains fixed load shapes for almost all participants; therefore, there is no need for detailed models of participants.

The initially calculated results are tested for feasibility (i.e., for possible network limitations). This is done by means of the load flow and stability Power System Simulation for Engineers (PSS/E) software also installed at the System Operator. If network limitations are exceeded, then a re-dispatching of generation is performed by changing the limitations in the GTMax software.

Final Planning Balance The System Operator determines the final hourly planning balance of the power system and gives each market participant its operating plan for the month. Based on that balance, the monthly bilateral contracts can be signed. Two options are proposed for execution of monthly contracts:

5. Economic Efficiency of Armenian Power System Integration and Analysis of Impacts of New Renewable in Armenia, Tetra Tech, November 2012.

• Pool principle. In some cases, less than a full pool is available; for example, where the entities offer only a part of their planned monthly volumes (from 0 to 99 percent). This is referred to as a partial pool

• Direct negotiations between market participants

Partial Pool Calculations in the partial pool suppose the availability of planned daily load shapes of market participants for the planned month. To protect the end-users, the prices may not exceed the PSRC-approved generator tariffs. All domestic consumers are partial pool participants, but the generators can be selected by criteria to be developed and approved. For example, generators may enter the partial pool when prices are cheaper than preliminary set values (this is a principle of economic generation dispatch that protects consumers). Another example is newly constructed power plants that are permitted to enter the partial pool even though the total volume of electricity to be purchased at the contract rate has already been determined in a signed power purchase agreement (supporting investors’ interests). In yet another example, generators might be included in the partial pool by an agreement to reduce their offer price. Export/import operations with provisions for long-term deliveries are determined through direct contracts with specified volumes and prices; therefore, these volumes are not included in a partial pool.

The partial pool structure may differ depending on the season. Practically, the partial pool is an hourly allocation of generation of selected power plants among consumers proportional to their hourly consumption. Average prices in the partial pool will be different for each consumer depending on daily consumption load shapes. Generator prices will also be differentiated by hour in order to preserve the average daily price due to the following:

• Favorable night tariffs are sometimes modified when the price of that generation exceeds the corresponding price during the hours of maximum load (current situation in Armenia). This occurs particularly when production is limited at the Vorotan Cascade

• Hourly rates may be modified to provide economic incentives to TPPs to minimize production when nighttime export flows are constrained

• Prices may be reduced to increase potential export of night power

The partial pool calculations produce volumes purchased by consumer “i” from generator “j” for hour “h” (Pijh) and the cost of each bilateral contract (Cijh) (Figure 4).

These contracts are assigned at the point of generation, meaning that losses of electricity are also considered while calculating the cost of electricity to be paid by a consumer.

Considering that generators might be connected both to the transmission and the distribution network, the term “market losses” is used without differentiation and represents the difference between generation plus import and consumption plus export at the sell/purchase nodes (for example, in Georgia they use the notion “transportation expenses”).

Based on the method of structural analysis6 it is possible to calculate each consumer’s portion of the occurring losses. However, this method has certain disadvantages for monthly planning:

• The need to plan for reactive load by hours

• The impact of the consumption level of one consumer on the portion of losses created by others

6. Structural Analysis of Flows and Losses in Electric Network, V. Safaryan, National Academy of the Republic of Armenia, 2001.

• Adjustment of monthly planned load shapes of market participants in the DAM

• The historically developed generation structure produces unequal conditions for consumers

Where, Lih = Planning capacity of consumer “i” for hour “h” loss%h = Percent of system losses for hour “h” Gjh = Generator “j” net generation for hour “h” gj = Portion of generation of “j” generator in partial pool pji = Coefficient of priority in the generation allocation of generator “j” for consumer “i” tjh = Price of generator “j” for hour “h”

Figure 4. Partial Pool Structure

The percent of losses will vary by hour. Differentiation of losses by hour is possible, for example, using the analysis of retrospective metering data. However, planned losses are not considered for DAM because final settlement calculation will include actual losses. Average percentages of market losses will be used in this stage.

In the abovementioned approach, the priority coefficient pji is also included. Ideally, all these coefficients will be equal to 1 (equal conditions for all consumers). However, depending on the target function and created conditions, certain adjustments will probably be needed (the decision will need to be justified). For example, if the target function is limitation of tariff growth for residential customers, then a higher coefficient might be accepted for DISCOs when allocating cheap generation.

As a result of these calculations, monthly bilateral contracts could be signed between all pairs of the partial pool participants. The structure of the partial pool may vary by seasons. The sum by months will be combined for annual bilateral contracts. In a general partial pool case, only a part of a consumer’s demand may be satisfied due to the following:

• Not all generators participate with 100 percent of their generation

• Differences between monthly planned load shapes and day-ahead planning load shapes for a particular day

Direct Monthly Bilateral Contracts Market participants should have the opportunity to sign direct contracts. Some generators may not be able to sell desired volumes in the partial pool; other generators may not be included in the partial pool at all due to competitive selection of offers. At the same time, consumers that did not cover their expected consumption demand should have the opportunity to make additional purchases. These contracts are executed on an hourly basis as a result of direct negotiations between two participants. The price in such contracts is negotiable.

This contracting opportunity is also available to export/import transactions, which should be equally accessible for each market participant subject to the overriding criterion—these transactions should not result in an increase of the generation price for domestic consumers. Relatively cheap power will presumptively be consumed inside the country. Exceptions are also possible. For example, if Armenia is disconnected from Iran7 in the spring and summer months, the system will be operating in an isolated regime. Therefore, the low nighttime demand in Armenia will make it necessary to significantly reduce the load of the ANPP. Given that Georgia itself has a surplus of generation in these seasons, it may only be interested in cheaply priced imports -purchase of low cost nighttime energy will allow it to accumulate additional water for later use at peak hours. While export from Armenia’s TPPs may not be possible, export from the nuclear power plant (NPP) may contribute in this indirect way to the reduction of the generation price in Armenia at peak hours.

Granting market participants the right to export power may create a situation where their aggregate export exceeds the transfer capability of the interconnection. In that situation, explicit or implicit auctions are usually conducted. In the absence of a unified regional market, explicit auctions are conducted. They can be annual, monthly, and day ahead.

As a result of such auctions, one or another market participant may reserve the transfer capability. If there is no restriction by transfer capability, all exporters get access free of charge. If there are restrictions, the priority of access to interconnection is be awarded based on the price proposed for the reservation.

It is possible to have both export and import contracts for the same hour even in the case of a single interconnection line. The physical flow will be equal to the difference of the contractual values.

Operators of both countries participate in realizing such contracts. The limitation is lifted by trading, when, for example, a long-term contract on export had become a limitation on possible imports, either long term or short term.

Monthly Contract Principles The order of signing monthly contracts may be random. However, after a market participant signs a bilateral contract, its participation in the partial pool is limited by the difference between the month-ahead planning load shapes and the aggregate load shape it has contracted.

All signed monthly contracts will follow the principle of firm supply contracts.

Consumption that was not covered on the contracts market must be traded on the DAM, making adjustment for the difference between day-ahead planning load shapes and load shapes for that particular day developed on a monthly basis.

5.2.2 DAY-AHEAD MARKET

Volumes of trading on the DAM for each market participant will be defined by the difference between the day-ahead planning load shape and the load shape formed by the sum of the hourly values of its monthly contracts for the planned day. Planned losses calculated by the System Operator for the planned day are presented in the calculations.

7. Economic Efficiency of Armenian Power System Integration and Analysis of Impacts of New Renewable in Armenia, Tetra Tech, November 2012.

A day before the given day, the market participants submit their load shapes to the Market Operator and System Operator, who calculates the optimal balance of the system taking into account the prices of generation and network limitations.

Prices on the DAM may be calculated based on:

• Generator tariffs approved by the PSRC (cost-based prices with possible partial liberalization)

• Generator offers (liberalized or limited prices)

• Both generator offers and consumer bids

Most developed competitive markets use liberalized generator prices for the DAM. Armenia may reach partial liberalization in the near future and allow generators to offer prices lower (any price) or higher (up to acceptable level); but, at the moment, there are few generators with similar price ranges. Permission to increase the price is often an incentive for investment into new generating facilities in markets with small volumes.

In all cases, generators would be permitted to differentiate prices by hours, which is also an element of competition. For example, run of river HPPs would prefer to generate at a lower price than release the water.

The first two options imply that by using contracts and DAM, the market participants will completely cover their day-ahead planning load shapes (Figure 5a).

It may happen that the volume of signed monthly contracts will exceed the day-ahead planning load shapes (Figure 5b); if so, the consumer becomes a seller and the generator becomes a buyer on the DAM. In other words, consumers sell back volumes which were committed to them for the month, but are not needed for the next day. Special mechanisms may be needed to limit the price for sale of surplus electricity to consumers. A market participant with this concern might reasonably adopt a strategy to limit the volumes of monthly contracts to the amount of planned consumption/generation and cover any remaining volume on the DAM. That strategy can also account for the difference between the day-ahead planning load shapes and load shapes planned for that day in the monthly plan.

Volumes being covered by monthly contracts may be assumed to be about 80 to 85 percent, becoming less as the competitive market becomes more developed. The DAM becomes a mechanism to cover the remaining required consumption volumes. Given that economic dispatch will be used to dispatch the power system, an increase of consumption will bring an increase of both marginal and weighted average prices (Figure 5c) compared to the prices used in the monthly contracts market. Prices will depend on the choice of a price formation mechanism.

Options 1 and 2 Option 3 (possible)

a. b. c.

Figure 5. DAM Options for Each Hour

Option 3 (Figure 5) presents a case where consumer bids are used as well. The consumer always wants to buy at the lowest price and the seller wants the highest price, and it is possible to satisfy only a part of the buyers’ bids at the equilibrium point of supply and demand (Figure 6b). If the total required day-ahead planning volume (VDAP) is not completely covered on the DAM, the clearing price on the DAM and the traded volume (VDAM) are being determined. The difference (VDAP minus VDAM) will be traded on the BM. Making a further assumption that the balancing segment of the market is often the most expensive market, it would be more profitable for the consumer to offer maximum realistic prices while submitting bids on the DAM. At the first stage of reforms, allowing several optional bids for each hour is unreasonable.

a. Options 1 and 2. VDAM = VDAP

b. Option 3. VDAM ≠ VDAP

Figure 6. Pricing on the DAM

Participation in regional markets may have a favorable impact on domestic markets. Figure 6a demonstrates how a relatively cheap import or benefit from swap reduces the clearing price on the DAM.

In particular, having efficient active traders helps to ensure availability of desirable import and export transactions. A unified regional market may ultimately provide this market depth.

Settlement…