The dynamic load and emission dispatch in daily cycles is an important problem in power supply-demand management. In this problem, the goal is to meet energy demand at the lowest possible cost and with the lowest possible environmental impact due to power generation. With the rising prices of fossil fuels and the advancement of power generation technologies, especially those concerning renewable energies, it is now impossible to ignore the potential effects of renewable sources on the dispatch problem. To address this issue, this study formulated the problem of dynamic load-emission dispatch for a single day period. Given the inherent uncertainty in the outputs of renewable sources, probabilistic models were used to reach a more accurate mathematical model for wind and solar generators. After formulating the problem with all operational constraints of power plants, a multi-objective particle swarm optimization algorithm was developed for solving this problem. In the proposed method, local and global searches of the algorithm are improved by the modeling of particle behavior. The resulting performance improvement is demonstrated through comparison with alternative solution methods.