Analyzing and Designing Price-based Power System Operation for Active Power with the Feedback Control Mechanism

Abstract

In a decentralized and deregulated environment, price-based operation (PBO) is suggested as an alternative to the centralized scheme. In the PBO, time-varying prices coordinate independent suppliers and consumers who may try to maximize their own profit. The most important thing in the PBO is to determine the appropriate prices, but it is not an easy task. Thus, analyzing and designing methods for the PBO is necessary to guide the determination of the price signals and to provide insights for the dynamic results. In order to perform the analysis and the design considering the dynamic performance of the coordination in view of the frequency stability, various methods and structures for the PBO have been presented in this dissertation. Firstly, the continuous-time model of the PBO is described under the assumptions that there are a large number of the participants and that the superposition of the asynchronous discrete responses of them can be approximately represented as the continuous function. Then, the power market dynamics is formalized as the feedback control structure, and the PBO is interpreted as the controller within it. Then, the approximation method is composed to express the target system with a simple typical form. After determining the approximate target system, the basic tuning rule is derived by applying a selected controller tuning rule. In addition, the modified designing rule is composed by supplementing an asymmetric damping control to the basic tuning rule in order to enhance the dynamic characteristics for the frequency control. As a method for further improving the dynamic performance in view of the frequency stability, a new structure for the PBO using price information as a kind of price offset is constructed. The effects of the price information on the maximum deviation of the energy imbalance are quantitatively analyzed. The analysis shows that the maximum deviation is likely to be reduced linearly to the difference between the price information and the system marginal price in the steady state. It is also found that the actual maximum deviation is the smallest when the price information is a little greater or less than the steady-state price. Finally, a general framework for the optimal design of the PBO is constructed. Not only that the power/energy balancing and the congestion management functions are successfully performed within the framework, but it is also guaranteed that the converged values in the steady state should be equal to the optimal solutions of the OPF method. The framework is suitable in a decentralized environment in the sense that each congestion management controller can be separately designed and operated. The effectiveness of the presented methods is verified by two case studies using the IEEE 39 bus network. The results show that the PBO designed by the designing rules is better than that without the application of them in view of the frequency stability. Moreover, the modified tuning rule performs even better than the basic tuning rule. The use of price information gives the considerable reduction of the energy imbalance regardless of whether the PBO is appropriately designed or not. In the case study for the framework, the fundamental property of it is verified that the converged values are all equal to the optimal solution irrespective of the specific design of the PBO, even though the time variation differs by the designs. A few kinds of the trade-off in the PBO are identified such as

between the recovery speed to the normal state and the smooth change of the variables such as the nodal prices

between the power/energy balancing and the congestion management functions.