China Southern Power Grid, together with China Mobile and ZTE, has deployed the industry's first slicing-enabled electric power private network for commercial use in Guangzhou, covering the whole Nansha New Zone with over 100 sites, and has successfully verified over 50 categories of power communication services. The 5G slicing-enabled private network not only provides high-security resource isolation for services in the power grid production area but also deterministic, low latency and low jitter communication.

The grid operator's services involve five links: power generation, transmission, transformation, distribution, and consumption. Different services have different network requirements covering security isolation, bandwidth, latency, jitter, and timing accuracy. According to their functions and SLA requirements, grid services are divided into two types: production area and management area. Different types of smart grid services impose different 5G network performance requirements. For services in the production area (e.g. differential protection, PMU and power distribution automation), high reliability, low latency, and high-precision time synchronization are required. For services in the management area (e.g. substation services and integrated services), high bandwidth is required.

 

Preservation of PRB Resources 
The 5G radio air interface resources can be divided into different resource blocks from such dimensions as time, frequency and air interface. The frequency domain can be partitioned into physical resource blocks (PRBs) with each PRB covering 12 subcarriers. To meet the high security and low latency requirements of services in the production area, the wireless side of the grid uses PRB reservation to guarantee fast forwarding, that is, a proper PRB resource reservation proportion is set for services based on their bandwidth and latency requirements. 
The reserved PRB resources can be dedicated to a slice group, which is used for all the services within the slice group. The data radio bearers (DRBs) for different slices are mapped to different PRBs to achieve physical security isolation. Resources can be obtained without competition, and high reliability can be guaranteed through reservation. Compared with QoS-based slicing with shared PRB resources, PRB reservation-based slicing provide hard isolation similar to dedicated radio spectrum resources while supporting the flexible policies of assigning PRB resources in a dedicated or shared manner. By reserving a certain proportion of PRB resources, ZTE's solution guarantees the bandwidth resources of the private power network and provides isolation and deterministic guarantee.

QoS Priority-Based Scheduling
The QoS priority-based scheduling is used for services with low requirements for isolation in the management area. For services with high isolation requirements in the production area, differentiated QoS guarantee may be required. For example, in the slice of the production control zone I, second-level latency is required for remote monitoring, control and measurement in distribution automation, and 50 ms latency required for PMU application. These services can be classified, allocated with different 5QIs, and matched with different radio policy parameters based on a specific S-NSSAI. In this way, different power services in the same slice have differentiated QoS guarantees.

Dedicated UPF
To meet the high isolation and low latency needs of grid services, UPF is directly deployed in the grid equipment rooms to provide local service access and ensure the security of grid service with no data leaving the zone. Thus, low latency and deterministic communication can be enabled.
In accordance with the security characteristics of grid services, ZTE also proposes the multi-network instance and multi-slice isolation technology for UPF sharing to solve the resource isolation problem when the services in the production or management areas share UPF. A physically-independent dedicated UPF is deployed in the production control area with high security requirements. On this basis, multiple logically independent slices can be created, providing isolation between the services in the security zone I and security zone II and SLA guarantee.
—Multi-instance: To meet the resource isolation and SLA requirements of services in different areas, different network instances are configured in the UPF, and different resources and VPN isolation strategies are configured in each network instance.
—Multi-slice: Multiple network instances can be mapped to power grid slices, and resource isolation, network plane isolation (VLAN) and bandwidth isolation between power grid slices are implemented through the utilization of network instances.

TSN Technology 
URLLC + TSN is the best solution for the grid in the future as it can solve low latency, high reliability and low jitter problems at the same time. At present, URLLC or TSN is not mature in terms of industry chain. In line with the progress of the industry chain, both URLLC and TSN can be gradually introduced into grid communications.
 
In the demonstration project between ZTE and China Southern Power Grid, ZTE provides wireless, bearer, and core networks, and deploys end-to-end 5G deterministic-network slicing with low latency and jitter to provide differentiated SLA guarantees for power grid services in different security areas. At the same time, ZTE is committed to providing a variety of key technologies, such as 5G LAN, TSN, URLLC, precise timing, and small-granularity FlexE, to facilitate the rapid development and deployment of new 5G+ smart grid services.