The energy and environment problems have become increasingly prominent, the direct load control and emergency response requirements are combined. Considering the self elasticity and cross-elasticity price, the multiperiod dynamic elastic loads are modelled, the cost-emission demand response dispatch model is established. The system is decomposed into the optimization of interactive agents by multi?agent system and agents’are solved by auction algorithm, the adaptive coevolution of agents is reached by cooperative co evolution agent with the adaptive cooperative multipliers.And the optimal incentive compensation value at each time interval is obtained. The influence of the incentive compensation value on the optimization resultsis compared under the fixed value and the optimal value. The results show that the combination of MAS and auction algorithm can improve the computation efficiency. Considering the demand elasticity, the dispatch can achieve the reduction and transfer of the load curve. The maximum level of reducing load is obtained with the maximum values of elasticity by the optimization of incentive value. Comprehensive energy saving and emission reduction into the system increase the comprehensive scheduling system.