Operating Reserve Procurement In Deregulated Power Systems

The success of electric power industry restructuring depends on whether reliability can be maintained. Reliability problems have occurred in some deregulated power systems. Although the degree to which deregulation contributes to these problems is not known, it is clear that pre-restructuring policies, institutions, and reliability models have not kept pace with the new industry structure.Under the new deregulated environment there are not many publications in the area of reliability of bulk power systems. Reliability is not only an engineering problem, but also an economic problem. Addressing this multifaceted problem requires a comprehensive analytical framework.As the industry moves towards deregulation, the generator's responsibility to supply reserve capacity is no longer well defined. The past quantity-constrained methods do not explicitly address economic criteria and, when employed in competitive markets, they do not capture the value of added reliability provided by the provision of operating reserve. The development of mechanisms to procure, allocate and price operating reserve, which is the key to ensuring the success of operation of an electricity market, is an urgent research need. The objectives of this dissertation are to design an efficient approach for evaluating reserve requirements in deregulated power systems by taking into account their economic value and capacity-reliability correlation. Both price and reliability will be jointly optimized to achieve maximum net social welfare.Despite facilities' outage and load fluctuation, the deregulation presents more challenges to reserve procurement, including the handling of network congestion and uncertainties about the bidding outcomes. Whichever mechanism is adopted, the constraints must be accommodated and the procurement should be based on market methods in order to fulfill the objectives of deregulation. How reserve should be procured, priced and allocated is studied thoroughly in this dissertation.First, a new approach is proposed to analyze the users' reserve procurement problem and a novel reserve trade mechanism is developed between electricity users and the retailer of the market. The differences between the procurement of operating reserve in decentralized and centralized ways are analyzed. The comparison of the equilibrium solutions reveals that the centralized procurement that results in a systemic optimal solution is better than the decentralized procurement that results in a Nash equilibrium solution. Furthermore, an incentive contract based on a Principal-agent model, that is able to induce a systemic optimality as well as a Pareto equilibrium and manage risks at the same time is designed. The proposed model is equitable and beneficial to all participants.Secondly, the chance-constrained programming is applied to determine spinning reserve requirement and a novel procurement model is developed. Under the premise of well balancing the system reliability requirement and spinning reserve procuring cost, the determination of the optimal spinning reserve capacity to account for the uncertainties of forced outage of generators as well as load forecasting errors is an important issue to be dealt with. In a chance-constrained programming framework, a mathematical model with the minimization of spinning reserve procurement cost as the objective and the system security requirement as a chance constraint is formulated. The problem is solved by a Monte-Carlo simulation based genetic algorithm. Comparing with other methods for the determination of spinning reserve requirement, the proposed probability based model can more flexibly deal with different uncertainties and different types of constraints.Thirdly, transmission capacity constraints are considered in reserve procurement. In the framework of chance-constrained programming, a mathematical model is first developed with the minimization of spinning reserve procurement cost as the objective function, the system security requirement and transmission limits as the chance constraints. A Monte-Carlo simulation based genetic algorithm is next proposed for solving the problem. Since spinning reserve is called in for generation randomly in real-time dispatching, it's appropriate to formulate the problem as a stochastic optimization problem.Finally, a reserve bidding strategy is presented by using average probability of wins. Similar to the energy market, development of optimal bidding strategies in the reserve market is also important to be considered for generation companies. In building the optimal strategy for a hydro unit, it is necessary to consider the reservoir capacity, hereby the water resource utilization problem. Based on the widely-used stepwise bidding protocols, a probability model is first presented for describing the electricity price in the market environment, using the "average probability of wins" method. Taking account of water limit and the randomness that reserve capacity is called in for generation, the optimal bidding strategy for a price-taker hydro-unit is developed under the given number of bidding blocks, in the framework of the chance-constrained programming method. The minimal possible profit is taken as the measurement of bidding risk, which is compatible with the generation companies' risk attitude.