Thermostatically controlled loads, which represent the major portion of consumptions, have considerable shifting potential for Demand Response (DR) applications. In this study an event-driven intelligent control system for rescheduling thermostatic loads is worked on. In addition to the discussion of areas of use for the proposed system, a number of terms that can be used in decision making processes are defined. The performance of the proposed system for working with refrigerators according to the power output of a local photovoltaic system is simulated by the use of field data. The results are discussed from both grid-side and customer-side perspectives.