Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Use of neural networks to evaluate the impact of proposed transactions on the security of a pool of electric utilities
Abstract
Power system networks are traditionally managed by utility companies that control the process of generation, transmission and distribution in their local region. These power systems are also known as pool systems, where all generators inject their power into the system (pool) and all loads satisfy their needs from the system. In traditional systems, each utility company determines a single price that changes with time, known as the spot price. Today, power systems are under pressure to open their “domains” of control and provide their services in a competitive market. Known as deregulated power systems, generation, transmission and distribution may be owned and managed separately. Another new entity, known as the Independent System Operator (ISO), coordinates the activities of these companies. In an auction-like process, generation and distribution companies propose power transactions (bids) and the ISO decides which ones to honor. Selecting a set of transactions from all bids is a challenging problem for a number of reasons. A summary of these reasons is: The selection process usually repeats every hour; the number of bids to select is not limited to one; the selection process should not be in favor of any entity; the selection process should be feasible; and it should contribute to the competitors' profits. Most studies on deregulation of power systems consider the feasibility problem and use security as the main criterion. These studies usually assume that transactions are given in a form that can be used by traditional power system measures. This adds another dimension to the difficulty in deregulation: the lack of a one-to-one relationship between the entities' bidding network (virtual) and the power system network (physical). This study provides a general framework that identifies details of a selection process in deregulated power systems and also provides a remedy for reducing the selection time by employing feasibility as a constraint. In the proposed framework, three issues are addressed separately: combining bids to increase the efficiency of the selection process; transforming bid information to power flow information to allow measuring the performance of each combination of bids in traditional measures; and the selection process. The selection process is further separated into two steps: elimination and final selection. In the elimination step, most of the transactions (bid combinations) are eliminated due to security constraints. The final selection can be tailored to satisfy other selection criteria. To show how the elimination process may be implemented and also to measure its contribution in terms of saving process time, a simulation study is conducted. Results of the simulation for three different test systems support the effectiveness of this two-stage selection process.
Subject Area
Electrical engineering|Energy
Recommended Citation
Jannati, Ali Seyed, "Use of neural networks to evaluate the impact of proposed transactions on the security of a pool of electric utilities" (2003). ETD collection for University of Nebraska-Lincoln. AAI3092559.
https://digitalcommons.unl.edu/dissertations/AAI3092559