This paper proposes a peer-to-peer energy trading system based on multi-objective game-theoretic optimisation for a clustered microgrid to find suitable sizes of distributed generations, including energy storage systems. The chosen architecture includes three different microgrids, and each microgrid can be a combination of solar panels, wind turbines and batteries, to meet the requirements of their residential loads. In the clustered microgrid, all microgrids are connected with one another, and also with the main grid, and they have the ability to perform both peer-to-peer and peer-to-grid energy trading based on load requirements and to reach the optimum value of the outlined objective function. To formulate a multi-objective function, two different criteria’ the annual profit and the loss of power supply probability are considered, and the Nash equilibrium-based game theory technique is modelled for clustered microgrids using a particle swarm optimisation algorithm to obtain suitable sizes of the players and payoff values. In the end, for each microgrid, suitable sizes of solar panels, wind turbines and batteries are proposed to meet the residential load requirements in the grid-connected mode. The outcomes of the multi-objective function are analysed and compared for two energy trading schemes, including peer-to-grid and peer-to-peer. The trends of objective functions and sizes of the players are analysed to verify the design of the proposed architecture.