This paper studies the customer behavior in the electricity market through price elasticity matrices（PEMs）

so we can determine the power reduction by customers. Here we develop a market optimization model for a microgrid with demand response（DR） by residential consumers. The model mainly considers two types of load: the transferable load and the Interruptible load. The participation of them brings great benefits for the whole microgrid. We use the linearizing techniques to convert this market optimization model into a mixed-integer linear programming model. Under the Real Time Pricing program

we get the solution with the professional mathematical programming solver

CPLEX. The testing results indicate that with the participation of transferable load

the total cost reduces by 7.85%

with the participation of both transferable and the Interruptible load

the total cost reduces by 17.56%. The performance of the optimization is remarkable.