A virtual power plant is proposed to aggregate various distributed renewable resources with controllable resources to overcome the uncertainty and volatility of the renewables so as to improve market involvement. As the virtual power plant capacity becomes remarkable, it behaves as a strategic price maker rather than price taker in the market for higher profit. In this work, a two-stage bi-level bidding and scheduling model is proposed to study the virtual power plant strategic behaviors as a price maker. A mathematical problem with an equilibrium constraints-based method is applied to solve the problem by transforming the two level problem into a single level multi-integer linear problem. Considering the deficiency of computational burden and implausible assumptions of conventional stochastic optimization, we introduce interval numbers to represent the predicted output of uncertainty resources in a real-time stage. The pessimism degree-based method is utilized to order the preferences of profit intervals and tradeoff between expected profit and uncertainty. An imbalance cost mitigation mechanism is proposed in this pessimism degree-based interval optimization manner. Results show that the bidding price directly affects the cleared day ahead of the locational marginal price for higher profit. Interior conventional generators, energy storage and interruptible loads are comprehensively optimized to cover potential power shortage or profit from market. Moreover, controllable resources can decrease or even wipe out the uncertainty through the imbalance cost mitigation mechanism when the negative deviation charge is high. Finally, a sensitivity analysis reveals the effect of interval parameter setting upon optimization results. Moreover, a virtual power plant operator with a higher pessimism degree pursues higher profit with higher uncertainty.