With the large-scale integration of distributed renewable energy sources, virtual power plants have emerged as important carriers for aggregating distributed resources, and multi-agent coordinated operation through peer-to-peer trading has become a significant development trend. However, peer-to-peer trading among virtual power plants requires simultaneous consideration of coordinated optimization between electricity markets and green certificate markets, as well as compatibility with the secure operation of distribution networks. To address this challenge, a bi-level optimization model for virtual power plant electricity-green certificate collaborative peer-to-peer trading considering distribution network operational constraints is proposed. The upper-level model establishes a framework for virtual power plant operational optimization and electricity-green certificate collaborative peer-to-peer trading decision-making, while the lower-level model develops a distribution network market clearing mechanism, with nodal electricity prices serving as key variables connecting the upper and lower levels to achieve coordination between peer-to-peer trading decisions and distribution network security constraints. An electricity and green certificate collaborative peer-to-peer trading model is developed to achieve dual-commodity collaborative optimization, reducing virtual power plant green certificate acquisition costs and incentivizing renewable energy development. An adaptive alternating direction method of multipliers based on consensus is employed to handle distributed coordination problems among virtual power plants, combined with the bisection method to achieve efficient and stable solution of the bi-level model. Case studies demonstrate that the proposed method can effectively enhance peer-to-peer trading economic benefits while satisfying distribution network security constraints, providing technical support for virtual power plant participation in diversified electricity markets.