In order to meet the needs of heating in winter, most of the units in the northern region use thermal power units with a working mode of "using heat to determine electricity". At night, the electric power of the thermal power unit will be maintained at a high level. A large amount of wind energy cannot be effectively used. The traditional solution is to simply add a heat storage device or an electric boiler to the system to alleviate the phenomenon of wind abandonment. In some cases, this scheme cannot completely absorb wind power. In this paper, heat storage devices and electric boilers are added to improve the utilization rate of wind power. First of all, the causes of the phenomenon of wind abandonment are put forward. Secondly, the dispatching model of cogeneration system is constructed including electric boiler and heat storage device. Then a probability-based crossover mutation operator is designed and applied to the artificial bee colony neighborhood search process, and an improved artificial bee colony solution algorithm based on probability crossover operation is given. Based on the data of a certain area in northern my country, the simulation verification is carried out, which proves the feasibility of the proposed scheme. Finally, the economic analysis of wind power consumption is carried out, and the results show that the addition of electric boilers has lower coal consumption costs and better economic benefits.