• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Generation planning and market operation strategies in the Southwest Power Pool energy imbalance service market

Dissertation
Author: Kittipong Methaprayoon
Abstract:
The transition of the Southwest Power Pool (SPP) from its traditional operation under tariff towards the new market environment has significant impact on the utilities in the SPP footprint, both in terms of planning and operations. The new market regulation creates the need for the utilities to reassess their planning strategy, primarily to ensure that operational decisions conform to the new market rules while remain profitable under the new operating regime. The utilities are referred to as Market Participants (MP) under the new market scheme. The new market environment will expose the MPs to a large amount of information with regard to market operation and settlement. The management of this information in a limited timeframe presents a great challenge to the MPs and necessitates the development of evolutionary tools to perform effective data analysis in a timely manner in order to ensure that the MP stays competitive in the new market environment. This dissertation presents the results of a comprehensive study of the structure and rules of the developing SPP Energy Imbalance Service (EIS) market and discusses aspects of planning functions with regard to the utility participation in the EIS market. The functions presented include an artificial neural network short-term load forecast, energy price forecast, economic dispatch, cost-based and price-based unit commitment scheduling, energy imbalance settlement analysis, as well as the validation of the RTO imbalance statement. The evolutionary EIS market planner (EISMP) software has been developed as a decision support tool for the utility's participation in the EIS market. The EISMP helps the utility to perform effective generation planning and analysis of real-time market data through an easy-to-use graphic user interface. The integration and implementation of this market planning tool have been verified in a real market environment through the assistance of the sponsor utility.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS....................................................................................... ii

ABSTRACT.............................................................................................................. iii

LIST OF ILLUSTRATIONS..................................................................................... x

LIST OF TABLES..................................................................................................... xiii

Chapter

1. INTRODUCTION………............................................................................. 1

1.1 U.S. Electric Supply Industry and the Shift towards Deregulation......... 1

1.2 SPP Regional Transmission Organization............................................... 7

1.3 Motivation, Objective, and Development Framework............................ 11

1.4 Contents of the Dissertation..................................................................... 14

2. ENERGY IMBALANCE SERVICE MARKET OF SPP............................. 16

2.1 Introduction to Energy Imbalance Service Market.................................. 16

2.1.1 Energy Imbalance Service Market Structure............................ 17

2.1.2 Key Features of SPP-EIS Market............................................. 19

2.1.3 Market Activity Timeline......................................................... 23

2.2 Functions Required by MPs under EIS Market Scheme......................... 23

2.2.1 Market Operation: Data Communication Link between MP and SPP................... 24

vi 2.2.2 Market Operation: Real-time Resource Allocation from Market Dispatch............................................ 25

2.2.3 Market Operation: Market Settlement Calculation................... 25

2.2.4 Generation Planning: Short-term Load Forecasting................. 26

2.2.5 Generation Planning: Hourly Price Estimation........................ 26

2.2.6 Generation Planning: Economic Dispatch and Unit Commitment............................. 27

2.3 Energy Imbalance Settlement.................................................................. 27

2.3.1 Energy Imbalance Service........................................................ 28

2.3.2 Under-Scheduling Charge........................................................ 28

2.3.3 Over-Scheduling Charge.......................................................... 31

2.3.4 Uninstructed Deviation Charge................................................ 32

2.3.5 Revenue Neutrality Uplift......................................................... 32

2.3.6 Self-provided Loss and Financial Settle Loss Amount............ 33

2.4 Operating Cost Components under EIS Market Scheme........................ 33

2.4.1 Cost Components without Under/Over-Scheduling Charge............................................... 38

2.4.2 Cost Components with Under/Over-Scheduling Charge............................................... 39

2.5 Optimal Strategy for Real-time Market Dispatch.................................... 41

2.5.1 Optimal Real-time Dispatch without Under/Over-Scheduling Charge.................................. 44

2.5.2 Optimal Real-time Dispatch with Under/Over-Scheduling Charge....................................... 46

2.6 Chapter Summary…………………………………………………........ 50

vii 3. REAL-TIME ARTIFICIAL NEURAL NETWORKS SHORT-TERM LOAD FORECAST MODULE.......................................... 51

3.1 Introduction to Load Forecasting............................................................. 51

3.2 Methods for Short-Term Load Forecasting............................................. 53

3.3 Load Series Analysis ............................................................................... 55

3.3.1 Peak Demand and Energy Use by Month................................. 55

3.3.2 Pattern of Demand.................................................................... 57

3.4 Neural Networks Structure and Training................................................. 59

3.4.1 Load Pattern Clustering............................................................ 59

3.4.2 Selection of Input Variables..................................................... 60

3.4.3 Data Preprocessing................................................................... 64

3.4.4 Activation Function and Training Algorithm .......................... 65

3.4.5 Number of Hidden Units Selection........................................... 67

3.5 Simulation and Testing of the ANNSTLF Model................................... 69

3.6 Chapter Summary…................................................................................ 77

4. HOURLY PRICE FORECAST MODULE................................................... 79

4.1 Introduction to Price Forecast.................................................................. 79

4.2 Price Forecast Timeline in the SPP EIS Market...................................... 81

4.3 Analysis of Price Characteristics............................................................. 82

4.4 Hourly Price Forecast by Similar Load Profile Approach....................... 84

4.5 Price Forecast Performance Index........................................................... 87

4.6 Simulation and Results............................................................................ 88

4.7 Chapter Summary………........................................................................ 100

viii 5. GENERATION PLANNING IN SPP-EIS MARKET.................................. 101

5.1 Cost-based UC for Self-scheduled Resources......................................... 102

5.1.1 Aggregation of Multiple Resources from the Same Settlement Location......................................... 103

5.1.2 Economic Dispatch Methodologies.......................................... 106

5.1.3 Unit Commitment Scheduling Methodologies......................... 109

5.1.4 Cost-based UC – Problem Formulation.................................... 113

5.1.5 Cost-based UC – Simulation Results........................................ 115

5.2 Price-based UC for Market Dispatch Resources..................................... 117

5.2.1 Price-based UC – Problem Formulation................................... 118

5.2.2 Price-based UC – Simulation Results....................................... 121

5.3 Energy Imbalance Settlement Calculation............................................... 123

5.4 Energy Schedule Sensitivity Analysis..................................................... 124

5.5 Chapter Summary……............................................................................ 128

6. SPP-EIS MARKET PLANNER SOFTWARE............................................. 130

6.1 Introduction………….............................................................................. 130

6.2 EISMP Program Structure....................................................................... 131

6.2.1 SPP Portal Connection Module................................................ 133

6.2.2 XML-to-database Transfer Module.......................................... 134

6.2.3 Short-term Load Forecast Module............................................ 135

6.2.4 Hourly Price Forecast Module.................................................. 136

6.2.5 Generation Planning Module.................................................... 136

6.2.6 Settlement Analysis Module..................................................... 137

ix 6.3 Database Design…….............................................................................. 138

6.4 Software Interface and Implementation Examples.................................. 142

6.4.1 Example 1: Generation Planning.............................................. 144

6.4.2 Example 2: Imbalance Settlement Calculation......................... 148

6.4.3 Example 3: Resource Allocation from Dispatch Instruction.... 152

6.5 Chapter Summary……............................................................................ 155

7. CONCLUSION…………............................................................................. 156

7.1 Conclusion…………............................................................................... 156

7.2 Contributions……………....................................................................... 159

7.3 Recommendation for Future Studies....................................................... 160

Appendix

A. 10-UNIT TEST SYSTEM…......................................................................... 162

B. GENERATION SCHEDULING COMPARISON FOR DIFFERENT DYNAMIC PROGRAMMING APPROACHES.......... 165

C. GENERATION SCHEDULING FROM PRICE-BASED UNIT COMMITMENT......................................... 169

D. SPP-EIS MARKET PLANNER DATABASE TABLE STRUCTURE...... 172

E. SPP-EIS MARKET PLANNER SOFTWARE USER INTERFACE CONTROL.................................................................. 176

REFERENCES.......................................................................................................... 190

BIOGRAPHICAL INFORMATION......................................................................... 196

x

LIST OF ILLUSTRATIONS Figure Page

1.1 Regional Transmission Organizations as of September 2006......................... 9

1.2 Proposed architecture of SPP market decision support tool........................... 14

2.1 SPP-EIS market system................................................................................... 18

2.2 SPP-EIS activity timeline................................................................................ 22

2.3 Locational imbalance price on congestion path and counterflow path........... 42

2.4 Real-time dispatch strategies........................................................................... 49

3.1 Daily maximum and minimum energy usages from 2003 to 2005................. 58

3.2 Mean hourly loads for each day of the week in 2004 and 2005...................... 58

3.3 Three-layer feed forward neural network structure......................................... 60

3.4 Training and testing MSE comparison for hidden unit selection.................... 68

3.5 Structure of ANNSTLF model........................................................................ 70

3.6 Hourly load forecast and actual values from February 14-27, 2005 (winter)............................................................... 75

3.7 Hourly load forecast and actual values from May 9-22, 2005 (spring)......................................................................... 75

3.8 Hourly load forecast and actual values from August 8-21, 2005 (summer).................................................................. 76

3.9 Hourly load forecast and actual values from November 14-27, 2005 (fall).................................................................. 76

3.10 MAPE gap between DA and HA forecasts using ANNSTLF........................ 77

xi 4.1 Time framework to forecast the hourly energy clearing price........................ 82

4.2 MCP series from MISO interconnection (05/01/06 to 07/31/06)................... 83

4.3 One week hourly MCP from MISO interconnection (06/19/06 to 06/25/06)........................................................... 83

4.4 One week hourly load from MISO interconnection (06/19/06 to 06/25/06)........................................................... 84

4.5 Hourly demand series from MISO interconnection (05/01/06 to 07/31/06)........................................................... 85

4.6 Logarithm of MCP series from MISO interconnection (05/01/06 to 07/31/06)........................................................... 85

4.7 Price forecast of MISO interconnection from 06/01/06 to 06/30/06.............. 90

4.8 Price forecast of MISO interconnection from 07/01/06 to 07/31/06.............. 90

4.9 Price forecast results from 4 selected weeks of PJM interconnection data............................................................................ 94

5.1 Incremental cost curve of three units in the same settlement location............105

5.2 Aggregate offer curve of settlement location with three units........................106

5.3 Economic dispatch by the lambda iteration method.......................................107

5.4 Flowchart of cost-based unit commitment scheduling by DP-STC................112

5.5 Flowchart of price-based unit commitment scheduling..................................120

5.6 Forecasted LIP and load obligation for price-based UC simulation...............122

5.7 Energy schedule sensitivity analysis...............................................................128

6.1 SPP-EIS market planner components.............................................................132

6.2 SPP data communication process....................................................................134

6.3 EIS market planner interface...........................................................................143

6.4 Short-term load forecast by load area for example 1......................................145

xii 6.5 Hourly price estimation for example 1............................................................146

6.6 Ancillary service obligation for example 1.....................................................147

6.7 Cost-based UC results for example 1..............................................................147

6.8 Resource plan submission for example 1........................................................148

6.9 Energy imbalance settlement for example 2...................................................151

6.10 SPP dispatch instruction for example 3...........................................................152

6.11 Real-time resource dispatches for example 3..................................................154

6.12 Real-time resource dispatches for example 3 with alternate combination of online resources.......................................................154

xiii

LIST OF TABLES

Table Page

1.1 Comparison of SPP Market Structure to Other RTO Markets........................ 11

2.1 Under-Scheduling Charge Example................................................................ 30

2.2 Over-Scheduling Charge Example.................................................................. 30

2.3 Cost-benefit Changes due to Different Participation Options in EIS Market............................................................... 35

3.1 Monthly Peak Energy Demand from 2002 to 2006........................................ 56

3.2 Monthly Energy Usages in MWh from 2002 to 2006..................................... 56

3.3 Correlation Coefficient and Partial Derivative Index in Summer................... 62

3.4 Correlation Coefficient and Partial Derivative Index in Winter..................... 62

3.5 Performance Index of ANNSTLF Model for Network D0 (Monday)............ 72

3.6 Performance Index of ANNSTLF Model for Network D1 (Tue to Thru)...... 72

3.7 Performance Index of ANNSTLF Model for Network D2 (Friday)............... 73

3.8 Performance Index of ANNSTLF Model for Network D3 (Saturday)........... 73

3.9 Performance Index of ANNSTLF Model for Network D4 (Sunday)............. 74

3.10 Performance Index of ANNSTLF Model for All Networks........................... 74

4.1 Performance Index Calculation of Price Forecast on 5/22/2002..................... 93

4.2 Price Forecast Daily-MAPE [in percent] of PJM Interconnection data.......... 95

4.3 Price Forecast RMSE [in $] of PJM Interconnection data.............................. 96

xiv 4.4 Price Forecast Error Variance of PJM Interconnection data........................... 97

4.5 Mean Square Error from Last Week Price Forecast of All Months in 2002..................................................................................... 98

5.1 Incremental Cost Aggregation for Plant with Three Units.............................105

5.2 Twenty-four Hourly Demand for Unit Commitment Simulation...................115

5.3 Total Production Cost Comparison for Different DP Algorithms..................115

5.4 Unit Combination Results from Different DP Algorithms.............................116

5.5 Hourly Forecast Price and UC Solution from Priced-base UC.......................121

5.6 Energy Imbalance Settlement Results from Different Energy Schedules............................................................................125

6.1 Summary of Input/Output of EISMP Modules...............................................138

6.2 Summary of Data Tables in the EISMP Database..........................................140

6.3 Generation Planning and Market Operation Functions in EISMP..................143

6.4 Load Forecast Results for Example 1.............................................................145

6.5 Price Forecast Results for Example 1.............................................................146

6.6 Energy Schedules Submission for Example 2.................................................149

6.7 Actual Load and Actual LIP for Example 2....................................................150

6.8 Charge Type Summary by Hour Ending for Example 2.................................151

1

CHAPTER 1 INTRODUCTION

The restructuring of the United States electricity supply industry has been underway for more than a decade. The restructuring is resulting in dramatic changes in the way the U.S. electricity supply industry operates. This chapter presents a history of electricity supply restructuring in the U.S., the background of the Southwest Power Pool (SPP) Regional Transmission Organization, and the motivation and structure of this dissertation. 1.1 U.S. Electric Supply Industry and the Shift towards Deregulation

The electric power industry in the United States is in the midst of evolution. For more than a decade, the wholesale and retail U.S. electricity supply industry has been forced to transform its structure to a competitive market-based operation. In the past, electric utilities were vertically integrated and operated as regulated monopolies. The local utility was the sole supplier for the energy service to customers within its service area. The economies of scale had encouraged electricity supply to operate as a natural monopoly, i.e. the electricity can be generated from a large power plant under single utility control in a much more efficient manner than any combination of small generators from multiple providers. Since electricity operation requires a high level of security, reliability, and a large investment to secure adequate transmission

2 infrastructures to support the delivery of electricity service, the conventional operation of utilities was regulated by the state public utility commission (PUC). The regulatory commission ensured that utilities had sufficient funding to invest in generation and transmission infrastructure while protecting consumers from unreasonably high prices. The vertically integrated utility performed all generation, transmission, and distribution functions to serve all industrial, commercial, and residential customers in its territory. Retail customers were charged an average system cost of production plus a reasonable rate of return on the utility's investment as approved by the state regulators. The revolution in the electricity supply industry towards competition was initiated by the question of how efficiently the utilities operated under this regulated monopoly structure. With the exclusive control over the electricity supply in a geographical region, the customers in one region might observe a better quality of service and lower rate compared to customers in a neighboring region. The oil embargo in 1973-1974 triggered an energy crisis, raising the electricity price drastically throughout the 1970s and early 1980s period. The average retail electricity price increased 53 percent from 5.7 cents/kWh in 1972 to 8.7 cents/kWh in 1982 [1]. Embracing the hypothesis that competition could increase the efficiency of energy production, economists began to debate over restructuring of electricity supply. Technological advancements in small-scale power generation lessened the economy of scale constraints, while growing concern over the environmental impact of traditional power generation technology provoked interest in incentives to encourage the development of alternative, green energy producing technologies. Other industries,

3 including the gas, airline, and communication industries, also revealed the benefits of privatization, enabling consumers to purchase their services at lower cost. All these factors encouraged a restructuring of the U.S. electric supply towards a market structure to promote competition. To response to the energy crisis and critiques of utility regulation, Congress enacted the Public Utility Regulatory Policy Act (PURPA) in 1978. The goal of PURPA was to promote energy conservation by encouraging development of alternative energy technologies. In addition, the PURPA created a new opportunity for non-utility power producers, who were qualified as qualifying facilities (QF), to generate and sell power to local utilities. The utilities were required to buy power from the QF at a so-called avoided cost, which is the production cost that utilities would have to pay if they generated the same power at their owned plant. A high electricity selling price attracted many Independent Power Producers (IPP) and Non-Utility Generator (NUG) to establish and participate in the QF program. Technological changes during the 1980s enabled the development of small generation resources at production costs competitive with utilities’ large power plants. The PURPA allowed these non-utilities to sell power to the utility. However, the electric supply on the consumer side remained a monopoly as consumers could only buy energy from their local utility. The transmission network also remained under control of the utility, which discouraged competition and inhibited the development of a wholesale market. To overcome this, Congress took action to promote competition in wholesale electricity supply by passing the Energy Policy Act (EPA) in 1992. The EPA

4 granted the Federal Energy Regulatory Commission (FERC) the authority to order interstate power transmission owners to open transmission access to other non-utility producers on a case-by-case basis. This change created an opportunity for large industrial customers to purchase wholesale electricity from other resources at lower rates. The FERC envisioned that undue transmission discrimination was a main barrier to a competitive electricity market. Therefore, utilizing its authority under the EPA, FERC issued Order No. 888 and Order No. 889 in 1996. Order No. 888 required all public utilities to file an Open Access Transmission Tariff (OATT) that outlined the terms and conditions for open access to their transmission services. Order No. 889 mandated public utilities to develop an Open Access Same-time Information System (OASIS) to publish information about the rates and transmission capacity available to all potential transmission customers on the same basis. The OASIS ensured equal access to all transmission facilities by all potential customers on a non-discriminatory basis. Even though Order No. 888 and Order No. 889 encouraged the establishment of an Independent System Operator (ISO) to operate the transmission grid on behalf of the transmission owners, the rule did not mandate the transfer of transmission control from the utilities to the ISO. Most transmission facilities remained operating under control of the owning utilities, which resulted in pricing inefficiencies in transmission services. The transfer of power across different regions might incur charges from multiple utilities operated under different OATTs, causing a problem referred to as pancaking

5 transmission charges. To correct this, the FERC issued Order No. 2000 in December 1999. Order No. 2000 mandated all public utilities with transmission assets to develop a plan to transfer transmission control to the Regional Transmission Organization (RTO). The Order No. 2000 was the landmark that forced utilities to separate their transmission services from other power market functions. As a result, various entities, namely the Generation companies (GENCOs), Transmission companies (TRANCOs), Distribution companies (DISCOs), and Load Serving entities (LSE) were created in the electricity market. Order No. 2000 set forth the development of wholesale electricity trading and regional transmission management under the RTO control. The generation competition under the open-access transmission services has resulted in a large volume of power being transferred between regions. However, regulated by the state commission, the transmission service charge did not provide enough incentive for investors to invest in new transmission facilities. As a result, the transmission grid has become increasingly congested as a result of competition. Several proposals have been put forth to address the shortage of transmission capacity. The noteworthy one was the Notice of Proposed Rulemaking on Standard Market Design (SMD) published by the FERC in 2002. The SMD incorporated market design structures that FERC believed could sustain system reliability and promote better efficiency of energy delivery in a common market framework. Under the SMD, the RTO manages transmission congestion by charging a congestion fee, in addition to the base rate, as an economic incentive. The Locational Marginal Pricing (LMP) was suggested for transmission congestion management. The LMP reflects the spot-market

6 energy clearing price, which can vary between the location of delivery and receipt of power under a period when there is transmission congestion. FERC believed that the higher congestion charge would encourage the construction of new generation or additional transmission capacity at the right locations. The SMD also introduced the Congestion Revenue Right (CRR) to enable holders to manage the risk associated with the congestion charge. Even though the SMD laid out comprehensive rules for virtually all aspects of the wholesale electricity market that would benefit electricity trading through market competition, the SMD proposal faced strong opposition from many consumer groups, public utilities, and other investor-owned utilities. Twenty-two states had asked FERC to abandon SMD. Many opposing states are located in the Southern and Western regions of the U.S., which currently have cheaper power than the national average, and consumers were skeptical that SMD would raise their price and threaten reliable service [2]. Many public utilities that are against deregulation do not wish to be forced to participate in the deregulated environment under the SMD. Concern for the environmental was also raised by federal agencies, fearing the impact from diesel- generators whose owners may produce power under the provision of SMD demand response, ignoring state air pollution policies. The most controversial discussion was regarding the mandatory regulation of the RTO proposal, which it was felt did not sufficiently accommodate regional differences. Many preferred that markets should be developed on a voluntary basis. As a result of the strong opposition, FERC announced the termination of SMD proceedings in July 2005 [3].

7 Since the termination of the SMD, the wholesale electricity market has been developing on a regional basis. Each region establishes a market structure based on the best judgment as to how to accommodate needs within the region. For those regions where the ISO/RTO has been established, the wholesale competition is operating under different rules as specified by the regional protocols. Nevertheless, the improvement in the transmission grid is still the most critical issue at the present time. The cascading blackout in the eastern states and large parts of Ontario on August 14, 2003 demonstrates the fact that nation’s transmission grid is vulnerable and immediate action is needed to improve the reliability of the grid. The latest enactment of the Energy Policy Act in 2005 by Congress emphasized the need of reliability standards for the bulk power system and authorized FERC to establish incentive-based rate treatment programs to promote capital investment in the transmission system [4]. Whether deregulation has achieved the goal for energy cost savings remains debatable. However, competition has definitely pushed the system to operate more aggressively and assures that long term transmission reliability stands as the greatest challenge to today’s U.S. electricity supply infrastructure. 1.2 SPP Regional Transmission Organization

After FERC issued Order No. 2000 in 1999, the restructuring of the electricity supply industry spread rapidly across the states. Given a strong and complex interaction of the electricity network, the transmission system plays the most important role in supporting the delivery of energy product. The undue discrimination of access to transmission services is a fundamental barrier to competition in the wholesale electricity

8 market. The Regional Transmission Organization (RTO) is the independent entity introduced by Order No. 2000 to remedy the transmission related impediments to a competitive market. The goal of the RTO, as outlined by the FERC, is to coordinate the transmission grid on a regional basis. RTO is responsible for ensuring non- discriminatory access to the transmission services, eliminating transmission rate pancaking, maintaining adequate transmission structure, and promoting efficiency and reliability in the planning and operation of region-wide transmission systems. The benefit of the RTOs quoted from the Order No.2000 includes [6] 1. The increased efficiency through regional transmission pricing and the elimination of rate pancaking of inter-state transmission service 2. A more effective management of parallel path flows 3. An improved transmission congestion management 4. More accurate estimates of Available Transfer Capability (ATC) 5. A more efficient planning of transmission and generation investments. 6. An increased coordination among state regulatory agencies 7. Reduced transaction costs 8. Facilitation of retail access programs and state deregulation 9. Improved grid reliability 10. Fewer discriminatory transmission practices Order No. 2000 strongly encouraged the transmission owning entities, including non-public utilities, to place their transmission facilities under the common control of appropriate RTO. As a result, several RTOs have been developed in many regions of

9 the U.S.. The RTOs coordinate dispatch of generation in their system and offer transmission services under the RTO specific Open Access Transmission Tariff (OATT). As of September 2006, nine functioning ISO/RTOs form the ISO/RTO council (IRC) in North America. This RTO configuration is shown in Figure 1.1.

Figure 1.1 Regional Transmission Organizations as of September 2006 Source: http://www.ferc.gov/industries/electric/indus-act/rto/rto-map.asp

The Southwest Power Pool (SPP) is a RTO operating transmission services in the central states. SPP has planned to develop market operation and submitted a request seeking authorization of RTO status since October 2000. However, its first filing was rejected by the Commission because the scope of the proposal did not satisfy Order No. 2000 [7]. In August 2001, the market development of the SPP was put on hold when SPP shifted its effort towards consolidation with Midwest ISO to operate transmission

10 facilities under a jointed RTO basis. However, after almost two years of discussion and collecting comments from the public and participants, the two organizations mutually agreed to terminate their merger plan in March 2003. After that, SPP resumed its plan to develop its own market in 2003 and re-filed the request for formal recognition as an RTO in October 2003. SPP’s effort to be an authorized RTO succeeded on October 1 st , 2004. However, the Commission required that SPP develop market mechanisms for its transmission service to fulfill the RTO status [8]. As a result, SPP is currently planning new markets for imbalance energy, ancillary services, and congestion management. SPP chooses the phase-in implementation for its markets. At present time, SPP is pushing for the first phase to its market development to settle real-time energy imbalance. This market is named the Energy Imbalance Service (EIS) market. The day-ahead energy markets will be constructed as a second phase where SPP plans to add the financial transmission right for market based congestion management. The final phase will be the implementation of Ancillary Service market, which completes the obligation at full market structure as an RTO. The implementation of the EIS market has encountered several delays due to a glitch in market system stability and incomplete readiness of the market participants. As of December 2006, SPP scheduled the launch date for the EIS market to be February 1 st , 2007. After implementation of the SPP-EIS market, the characteristics of SPP RTO as compared to the wholesale market in other RTO regions can be summarized as in Table 1.1 [9]. The transition of SPP towards the new market operating environment has

11 potentially significant impacts for SPP customers, especially for generation companies or utilities under the SPP footprint. These impacts are reflected both in terms of planning and operation, which are the major focus of this research. Table 1.1 Comparison of SPP Market Structure to Other RTO Markets Services provided SPP MISO PJM ERCOT ISO-NE NYISO CALISO Regional transmission scheduling √ √ √ √ √ √ √ Regional economic dispatch √ √ √ √ √ √ √ Regional transmission planning √ √ √ √ √ √ √ Regional interconnection process √ √ √ √ √ √ √ Bilateral Transactions √ √ √ √ √ √ √ Real-time energy market √ √ √ √ √ √ √ - Locational energy price √ √ √ √ √ √ √ - Hourly energy price √ √ √ √ √ √ √ - Congestion price √ √ √ √ √ √ √ - Losses price √ √ √ √ √ √ √ Day-ahead energy market √ √ √ √ - Locational energy price √ √ √ √ - Hourly energy price √ √ √ √ - Congestion price √ √ √ √ √ - Losses price √ √ √ √ Virtual Bidding √ √ √ √ Financial transmission rights √ √ √ √ √ √ Ancillary services market √ √ √ √ √ Regulation service market √ √ √ √ Capacity markets √ √ √ Independent market monitor √ √ √ √ √ √ √ Market mitigation √ √ √ √ √ √ √

1.3 Motivation, Objective, and Development Framework

The transformation of the SPP transmission operation towards the implementation of a real-time energy imbalance market represents a revolution in traditional operations both in terms of the technical and economic perspectives. In the past, the main objective for utilities was to minimize the cost of energy production

12 required to serve the demand obligation. In a competitive market environment, the rate of return is no longer guaranteed. The participation of sellers and buyers in the competitive market is profit driven. As a result, the utilities will have to change their strategy to achieve the maximum profit and ensure cost recovery of energy supply. Tools to support market planning and decision are indispensable in the competitive world of market operation. However, as discussed earlier, no common market structure currently exists. Regulation and rules governing energy markets may vary significantly from region to region. Any potential tool which can be practically applied to support the market decision-making must conform to the structure and rules associated with the market of interest. The SPP-EIS market is a new developing market to be implemented soon in the SPP footprint. Once the market starts, all market participants (MP) must follow the regulations specified in the market protocols. There are many new functions that MPs need to prepare to support operation in the EIS market. To mention a few, these functions include resource plan and offer submission, energy imbalance settlement, price estimation, and statement verification from the RTO. A new cost component associated with energy imbalance service is also introduced as a part of the overall operating cost. All of these requirements necessitate a comprehensive study and the development of a decision support tool for generation planning, which is an intended contribution of this dissertation. The ultimate goal of the development is to provide a complete environment to assist the MPs in participating actively and effectively in the EIS market. The objective of the research can be summarized as follows:

13 a) To study and understand the regulatory requirements as stated in the EIS market protocols. Identify the need and utility practice to participate in the real-time energy market as suggested by the utility engineer. b) According to the unique restrictions and characteristics of the EIS market, develop generation planning strategies taking into account the rules specified in the latest available market protocols at the time of development. c) Develop a data communication module to facilitate the data communication between the MP and the SPP market operation systems. d) Develop and integrate essential components required for generation planning. The components include the regional short term load forecast to forecast the demand at many different settlement locations, the hourly price estimation tool to forecast price for settlement calculation and generation planning, and the imbalance settlement calculation to verify the RTO settlement statement for accounting purposes. The proposed architecture to achieve aforementioned goals is illustrated in Figure 1.2. The architecture has been designed taking into account the practical recommendations from the utility engineer that will participate into the EIS market activities when the market starts. With a huge amount of information involved in generation planning and business process, a proper structure of database must be prepared to ensure the storability and transferability of information between many planning functions developed in this research. The integration and performance testing of the developed software is performed to ensure the practical implementation and reliability of developed work in the real market environment.

Full document contains 211 pages
Abstract: The transition of the Southwest Power Pool (SPP) from its traditional operation under tariff towards the new market environment has significant impact on the utilities in the SPP footprint, both in terms of planning and operations. The new market regulation creates the need for the utilities to reassess their planning strategy, primarily to ensure that operational decisions conform to the new market rules while remain profitable under the new operating regime. The utilities are referred to as Market Participants (MP) under the new market scheme. The new market environment will expose the MPs to a large amount of information with regard to market operation and settlement. The management of this information in a limited timeframe presents a great challenge to the MPs and necessitates the development of evolutionary tools to perform effective data analysis in a timely manner in order to ensure that the MP stays competitive in the new market environment. This dissertation presents the results of a comprehensive study of the structure and rules of the developing SPP Energy Imbalance Service (EIS) market and discusses aspects of planning functions with regard to the utility participation in the EIS market. The functions presented include an artificial neural network short-term load forecast, energy price forecast, economic dispatch, cost-based and price-based unit commitment scheduling, energy imbalance settlement analysis, as well as the validation of the RTO imbalance statement. The evolutionary EIS market planner (EISMP) software has been developed as a decision support tool for the utility's participation in the EIS market. The EISMP helps the utility to perform effective generation planning and analysis of real-time market data through an easy-to-use graphic user interface. The integration and implementation of this market planning tool have been verified in a real market environment through the assistance of the sponsor utility.