| Planning for contingencies has become increasingly important as power systems are being operated closer to their limits. In the past, system operators placed great emphasis on the security of the system, often at the expense of operating cost. In today's competitive and deregulated environment, operators need tools that will enable them to take prudent risks in the interest of reducing the economic premium of secure operation.; Contingency planning is a multi-criteria optimization problem with cost, security, and probability being the most important criteria. The problem is compounded by the existence of multiple probable contingencies. In the current deregulated power industry, the “flows” in the transmission lines are transaction based (comprising of many varied transactions between various generating and distribution companies). Only those transactions can be accepted which will keep the system in a “contingency-safe” state. The greater the number of transactions accepted, the greater is the profit, but if a transaction is curtailed once it is accepted, there is a penalty which is to be paid for that transaction. Thus there is a trade-off between the current gain and a future loss. The operator has to intelligently decide the trade-off. In this job it is aided by the scheduling system. This problem can be cast as a real-time scheduling problem with producers and consumers. We have proposed a market-based multi-agent approach for this kind of problems. The scheduling approach consists of two kinds of agents: producers that produce or own resources and consumers that consume or use these resources. The scheduling process consists of a market-based approach in which the consumers bid for the resources and the producers and the producers choose the consumers that give them the maximum their profit. The same framework is modified for solving the power-scheduling problem. |
