Automated Operating Procedures for Transfer Limits

A Modern Energy Management System (EMS) provides sophisticated online security analysis applications to assist operators in ensuring that the power system can survive credible contingencies. Still in current practice, system operators generally refer to written operating procedures to establish system constraints, particularly in regards to transfer limits across major interties. The limits are based on numerous power system studies that represent the stressed system and satisfy specific performance criteria following select contingencies. The relations between these critical paths and operating conditions are tabulated and often plotted as nomograms. With such a simplified view of system conditions, the operator is unable to have a complete understanding of operational limits. Thus, transfer ratings are typically conservative, as studies are based on highly stressed system conditions, and incomplete, as studies cannot analyze all combinations of equipment out-of-service. This study investigates some approaches to improving such operator procedures. Ensuring system security usually means operating so as to maintain a specified margin, for example, real power reserve within a particular area. The required margins are generally mandated by the regional reliability organizations. Unfortunately, it is time-consuming to compute such limits and thus, the margins are primarily determined off-line. The operators then use the conservatively tabulated values to operate the system within limits. Ideally, as the system operating conditions change, the margins would be recomputed to precisely verify security. Since this is computationally infeasible, an alternative approach is to employ pattern matching methods. That is, by using the detailed off-line studies to interpolate between unstudied operating conditions, one can estimate the margins from the present operating point without employing detailed calculations. This project looked at general ways to improve the operator procedures with particular emphasis on estimating the margin based on such pattern matching schemes. As voltage security is an increasingly important issue as systems are operating under greater stress, this study focused on voltage issues. Specific contributions from this study were: • A typical set of operating procedures was tabulated into an on-line database. Conceptually, these margins can be easily modified on-line to reflect changing system conditions. • Several pattern matching type approaches were investigated using a modified New England 39 bus system. Based on these results, a system based on feedforward artificial neural networks (ANN) was designed. • A modified ANN system, employing multiple networks and a voting system, was applied to the Western System Coordinating Council (WSCC) system. The analysis was based on …