Intelligent agents are gaining widespread applications in many trading markets. Although the US wholesale power market comprises a large commodity market, the mechanism of power trading is not clearly investigated. We explore the problem via an intelligent agent-based approach. We create an artificial wholesale market, where many different traders are equipped with learning capabilities. We validate the agent based model with the help of a data set from PJM electricity market. Using the new intelligence system, we investigate the bidding strategies of traders and examine how the price changes occur under different environments.