The traditional power system optimization model usually use the inherent technical output of the thermal power unitas the output constraint of the unit, in actual operation, the unit’sdis patching output range is also affected by the net load and theunit’s output during the previous period, there is a large optimization space for output constraints. In order to effectively reduce the feasible range of unit output variables, a net load incremental indexis proposed, and the index and the output state of the unit in the previous period are used to determine the dispatch able space of eachunit. With the scope of schedulable space as a constraint, an improved dispatching model for improved power systems is established. A hybrid particle swarm optimization algorithm combining standard particle swarm optimization and simulated annealing algorithm is used. The example results show that the hybrid particles warm optimization algorithm can effectively improve the shortcomings of the standard particle swarm optimization algorithm to fall into the local optimum, and improve the accuracy of the model solution. In addition, compared with the traditional optimization scheduling model, the improved optimization scheduling model of power system that introduces schedulable space constraints, while ensuring the accuracy of the solution, the calculation amount is greatly reduced, and it is not easy to fall into a local optimum. the improved ideas and methods can also be applied to the optimization scheduling model of other energy systems.