A heterogeneous network architecture for distributed energy management in the smart grid

A fully functional smart grid needs a large number of protection, control and monitoring devices located across a large geographic area to perform various distributed energy management (DEM) operations such as wide-area situational awareness, distribution automation and demand response. Hence, the corresponding communications system needs to operate in different propagation and deployment scenarios at different parts of the grid. Moreover, allocation of network resources (i.e. signalling and traffic) and guarantee of QoS (quality-of-service) among multiple DEM applications pose additional challenges. In such a paradigm, multi-domain resource allocation and QoS control functionalities such as hierarchical data aggregation and group resource allocation can significantly enhance the performance of the overall network and optimize its available resources.

With a view to meet the aforementioned challenges, this project is about a multi-tier heterogeneous network (Hetnet) architecture comprised of a primary network based on a long-range wireless technology such as UMTS/LTE/WiMAX and one or more secondary network based on short/medium range wireless technologies such as IEEE 802.15.4 based ZigBee and IEEE 802.11 based WLAN (Wireless Local Area Network) networks. A conceptual model of the proposed network architecture is shown in Fig. 1.


Figure 1

Source: Carnegie Mellon University

Figure 1

A resource management entity is the key enabler of our architecture that adaptively and opportunistically varies radio resources, i.e. signalling and traffic channels among the transmission links to meet the QoS requirements of each active connection. There are several advantages of this architecture, such as:

  • The architecture provides physical separation between the home area network (HAN) and the wide area network (WAN) of the smart grid communications system which improves security, resolves ownership issues, and encourages development of new independent HAN and WAN applications.
  • The access points/coordinators of the secondary network can act as Hetnet relays for the primary network which can extend the range, improve link quality, and eliminate dead spots of the overall wireless network.
  • The end-devices may considerably save the transmit power due to improved SNR (signal-to-noise ratio) values. The improved link margin will also allow the primary network links to operate on a higher modulation and coding scheme (MCS) which can significantly increase the overall data rate of the network.
  • The hybrid network can use spectrums from both licensed and licensed-exempt bands which could increase the spectrum efficiency of the system without raising the interference margin.
  • The end devices can use relatively less expensive WLAN, ZigBee chips which may reduce the overall cost of the network deployment.
  • The secondary networks can aggregate the traffic from multiple nodes into a single data burst which will reduce the amount of network access in the primary network which in turn will improve its data transmission efficiency by reducing the amount of signaling and protocol overheads.

To validate the proof of concept, the IEEE 802.16 based WiMAX and IEEE 802.15.4 based ZigBee will be used as the primary and the secondary network respectively for the proposed Hetnet architecture. Simulations might be conducted using MATLAB/OPNET to illustrate the performance of the proposed network architecture. Utilization of such a Hetnet will faciliate demand side applications such as EV charging, residential power generation with PVs and demand side response. These applications require communication between players at macro-scale, e.g. Network Operators, and players at micro-scale, e.g. any component in a smart house (thermostat of HVAC etc.). Hetnet concept allows for comprehensive managemenet by zooming in and out over the entire smartgrid.