To improve the running efficiency and reduce the phenomenon of abandoning wind turbine output in north China, a rolling optimal dispatch for co-generation and wind turbine units is proposed. Due to the strong volatility of renewable energy like wind, rolling strategy is put forward and compared with single forecast results. Meanwhile, intelligent single particle optimizer (ISPO) algorithm, which divides a particle into sub-vectors in accordance to with unit type, is used in optimal dispatch model. Its ability of searching optimal state is stronger than traditional particle swarm optimization (PSO) algorithm. Distributed power driven heat pump is used to improve heat and electricity load structure and reduce the desert wind rate. The effectiveness and validity of model and strategy proposed are verified by the IEEE30 bus system case.