Coordinated optimal operation of a joint virtual energy storage system based on an improved non-dominated sorting genetic and multi-objective particle swarm optimization algorithm

High renewable penetration intensifies the source–load temporal mismatch in building-integrated microgrids. This study introduces a joint virtual energy storage concept—combining building thermal inertia with electric-vehicle (EV) batteries—and formulates a multi-objective dispatch model to minimize operating cost and CO₂ emissions. An improved NSGA-MOPSO solver is employed to obtain well-distributed Pareto fronts with fast convergence under operational constraints. Case studies on a summer day in Beijing show that, relative to no virtual storage, EV-only storage reduces daily operating cost by 5.21 % and CO₂ emissions by 3.00 %. In comparison, the joint virtual storage achieves 13.33 % cost reduction and 17.08 % emission reduction. Coordinated scheduling of EVs lowers the peak-to-valley spread by 8.65 %, 20.1 %, and 24.9 % under progressively responsive user modes, while indoor temperature remains within 22–26 °C. The enhanced solver yields orders-of-magnitude improvements in convergence accuracy and stability on benchmark functions. These results demonstrate quantifiable economic and environmental gains and practical feasibility for low-carbon microgrids. • Joint virtual storage combining building thermal inertia and electric vehicle batteries. • Improved NSGA-MOPSO for multi-objective microgrid optimization. • Raises efficiency and economy while reducing emissions. • Coordinated scheduling mitigates source–load temporal mismatch.