The invention relates to a load control method and system based on a game theory. The method includes: acquiring Historical data and real-time meteorological data to serve as prediction sample data; Formulating electricity prices of multiple power generation subjects according to the prediction sample data; Inputting the electricity prices of the plurality of power generation subjects and the loaddemand of the load side as dynamic game particles into a pre-constructed game model, and solving Nash equilibrium points meeting a profit maximization target of the plurality of power generation subjects by using a particle swarm optimization algorithm, Wherein the historical data comprises load demand historical data, photovoltaic power generation historical data and historical meteorological data. According to the invention, the economic optimization model of multi-benefit subject transaction is established, and based on the dynamic game theory, the Nash equilibrium problem about electricity price formulation and load requirements is solved by using the dynamic particle swarm optimization algorithm, so that benefit maximization of multiple power generation subjects is realized.