*3.1. Aggregator Architecture*

This part describes the architecture of the presented aggregator model for DR programs. In fact, the responsibility of the aggregator in this paper is to gather all small and medium scale DR participants located in the same geographical area and present this flexibility to the electricity market negotiation as a unique resource. To do this, the aggregator has to make bidirectional contracts with the electricity consumers who intend to participate in one or more DR programs. This enables the aggregator accordingly to control and monitor the consumption of the end-users. Figure 1 shows the role of aggregator as a network player in electricity systems and smart grid technology.

As Figure 1 illustrates, the aggregator has a transactional role between the downstream (medium and small-scale DR participants) and upstream (electricity markets and network operators) sides of the network. Indeed, the aggregator has two layers, namely communication-based controlling and monitoring sides. In the upper layer, the aggregator is in touch with network operators or electricity market operators. While some technical or economic instabilities occur in the network, and it is required to reduce network consumption, the aggregator will be notified by the upstream level players to apply DR programs. Subsequently, in the lower layer, the aggregator has a multi-round communication with the downstream level of the network (i.e., consumers), as it may have some voluntary DR programs. In other words, the aggregator cannot forecast the response of consumers to each DR event. This leads to having several iterations of DR requests from aggregator to the consumers, and in response, the consumers reply with their preferences, demand bids, reduction capacity, price signal, etc. This procedure will be continued until the aggregator reaches the reduction baseline, which

is, in fact, one of the responsibilities of the aggregator during the ramp period. In the last step, the aggregator presents DR bids in the electricity markets with a certain rate based on the real-time price of the market. Generally, the aggregator has day-ahead information of DR bids, such as forecasted reduction rate. Therefore, it should consider the reduction rate a bit higher than the forecasted baseline. This is due to preventing possible failures in the case of some consumers opting out during the event.

**Figure 1.** Participation of aggregator in the electricity system as a third-party entity.

In the practical phase, communication infrastructures are very important as they play as a base for the network management strategies, as all network players must exchange data continuously in real-time. This is more visible in critical moments, such as the ramp period, as the aggregator should have real-time information of the DR participants. Furthermore, there are instances during the DR event or ramp period whereby aggregator needs to verify that the consumers have followed the DR programs and contractual reduction correctly, according to the request of the operator (in the case of incentive-based DR). Therefore, all consumers should be equipped with a local energy management system, smart metering, and advanced metering infrastructure (AMI) to monitor real-time data, especially power consumption.
