Content for  TR 22.867  Word version:  18.2.0

Utilities (DSO) make use of both Private and Public Wireless networks as access technologies for Smart Grid assets providing Smart Grid services, some of which can have a high level of criticality. Public cellular networks are sometimes the only option to provide remote access to Distribution Automation (DA) assets. Remote access to DA assets in a fault situation causing a power outage is a necessary condition to isolate the fault and restore the service promptly. In order to guarantee high availability, the communication system providing remote access to the Smart Grid assets needs an uninterruptable power supply. When DA assets are connected by means of Public 3GPP networks, they are typically served by a Node located in the proximity of the fault and hence affected by it because it happens to be fed by the faulty power line. Power autonomy of cellular nodes is a key variable of a fault scenario in order to guarantee the correct and prompt resolution of the issue and the service recovery. Power autonomy of the mobile telecommunication system nodes is a variable that escapes the control of the DSO and relies exclusively on MNO's policies and 'implementation'. A standard communication flow between parties would enable mutually beneficial coordination towards energy recovery. Both MNO and DSO could adapt their processes to the existing constraints,

Utility "U" has a significant percentage of Distribution Automation assets connected by Public cellular services. U relies on MNOs A and B for 5G services. Distribution Automation is considered a critical service requiring high availability. Cellular routers that enable remote access to the assets have dual cellular configuration in order to provide an extra level of redundancy and availability. This configuration enables a backup cellular connection with MNO B in case of failure of MNO A. Access to Distribution Automation assets such as reclosers or SCADA switches must be available in a fault situation in order to restore the service. If the 5G node providing cellular coverage to the connected Smart Grid asset is fed by a line affected by the fault and the Node does not have an adequate UPS that enables access to remote operations to restore the fault, the loss of connectivity and the corresponding loss of remote control of the asset is inevitable. DSO and MNO have an available channel for communication of energy outage incidents. This channel is bidirectional. The MNO can be made aware real time of an incident affecting DSO's distribution network with a result of power failure affecting MNO's Node(s) serving the Smart Grid asset. The communication channel can also inform the DSO of the MNO's autonomous power status in areas affected by energy outage.

Remote access and operation of Distribution Automation assets is performed from the Energy Control Center (ECC). Charles is an operator working at the ECC.

The outcome of Service Flow 1, alternative 1 is unfortunate because it causes a lengthy service interruption for both energy and telecommunications service. The outcome of Service Flow 1, alternative 2, allows this situation to be avoided. As the DSO and MNO have a standard channel for communication of energy outage incidents, they are both able to plan and execute recovery effectively. The DSO can provide locality, communication requirements and estimated time of repair of an issue in DSO's Distribution network. The MNO can use this information, especially the recovery schedule, the location and the communication requirements, to enable remote access to the Smart Grid asset that will solve the issue. For example, in the event of a service interruption or outage affecting a 5G node with a limited power autonomy, the Node has access to information coming from the Network so that it is able to detect the root cause of the problem and react accordingly. The Node is able to efficiently manage the resources and prioritize granting access to U's traffic. All the rest of processes that are not strictly necessary to the final aim of serving U's need of cellular resources become dormant. MNO will inform DSO of the Power autonomy and location of the Node serving the Smart Grid asset that needs to be accessed for remote operations. The DSO will react accordingly, adjusting the operations taking into account the MNO communication constraints. Thus, both parties will react accordingly, adjusting their processes to facilitate the resolution of the issue. U is able to successfully restore the issue affecting the line remotely so that the Node recovers the power supply after a short period of time. Once Power supply is back to normal, 5G node can resume its normal processes. Energy system comes back online, taking into consideration the timing and UPS resource constraints of both MNO and DSO, as well as the locality of the incident. This is a marked improvement of the current system in which for availability considerations DSOs cannot rely upon mobile telecommunications for recovery with building a redundant infrastructure. This reduces the applicability of mobile telecommunications infrastructure to support energy utility communication. Another alternative, shown in Service Flow 2, is that the MNO provisions their UPS resources sufficiently to suffer loss of service in the event of an energy system outage for a sufficiently long period of time to enable the DSO to restore service. The provisioning of the UPS can be informed by the outcome of past energy outage incidents, as well as historical information exchanged as per Service Flow 1, alternative 1 and 2 - as lessons learned.

Network slicing and QoS implementation in 5G networks enable the identification of DSO's service criticality.

[PR.5.18-001] [PR.5.18-002]

As electric grids are expanding, there is an increasing need for fault detection and location as well as maintaining system stability in real time which requires precision timing. There is also a raising interest to have other clock synchronization sources in addition to GNSS. IEC 61850-9-2 standard [49] specifies the protocol for transmitting measurement information in a power system using Ethernet. Standard also recommends the use of PTP (Precision Time Protocol) for clock synchronization and IEC 61850-9-3 standard [7] shall be followed whenever PTP is used. PTP accuracy is affected by network structure and devices in the network. Since different types of clock devices produce different time errors and jitters, overall time inaccuracy must be verified to make sure that application time accuracy requirements are fulfilled. Requirements in IEC 61850-9-3 [7] aim at achieving a network time inaccuracy better than 1μs after crossing 15 transparent clocks or 3 boundary clocks. PTP profile for power utility automation defines the inaccuracy as follows:

Grandmaster clocks in substations today use a GNSS receiver to achieve accurate clock synchronization with holdover time even up to 24h (corresponding to <1μs accuracy). However, since the time source (GNSS) may become unavailable due to failure (e.g. broken antenna, satellite availability or interference), 5G is a candidate resiliency solution (see clause 5.2 of TR 22.878). In order to reach the same level of measurement accuracy, 5G capable grandmaster clock should fulfil the requirement of maximum 250ns inaccuracy when its timing reference is used in the same location as of the Grandmaster clock that is used to synchronize the Smart Grid devices over Ethernet. Synchronization requirements are related to various aspects. In particular, IEC/IEEE 60255-118-1:2018 [50] defines a Phasor Measurement Unit (PMU) that must maintain less than a 1% total vector error (TVE). Such error includes time, phasor angle, and phasor magnitude estimation errors. Standard also recommends that a time source that reliably provides time, should be at least 10 times better than values corresponding to 1 % TVE. Therefore IEC 61850-9-3 [7] should be followed whenever substation has PMUs either as standalone devices or PMU capable protection relays. By using PTP within a substation the system can have multiple grandmaster capable clocks which, in case of failure of the current grandmaster, can be used to maintain accurate time between devices and processing sampled values. However, when sampled values are processed outside the substation, time inaccuracy limits defined by IEC 61850-9-3 [7] must be followed since the devices are not connected to the same grandmaster clock and therefore measurement timestamps may have too much inaccuracy. Application outside substation may be but not limited to Wide Area Protection.

A Smart Grid operator utilizes 5G wireless communications across primary and secondary substations. These communications report events and error detection recordings to a centralized monitoring station. Accurate time stamping across the sources is essential to provide a clear picture of the system. As outages in one area may have impact throughout the system, the 5G system may be used as a supplement to (e.g., integrated as alternative radio in the grandmaster clock) or as an alternative grandmaster clock. As a capable grandmaster clock, the 5G system should fulfil the accuracy requirements defined by IEC 61850-9-3 [7].

Sensors across the secondary distribution centres provide timestamped event reporting to a centralized monitoring system. Each event is timestamped to aid in analysis. The 5G system and the Smart Grid operator have 5G wireless communications configured across the Smart Grid network. Each source uses the same master clock and communicates over the 5G system, timestamps on the recordings are aligned for accurate analysis to be done at a centralized point. Additionally, the 5G system will need to be compatible with requirements from IEC 61850-9-3 [3] when used in conjunction with GNSS for a timing service in a Smart Grid environment.

• Power system time accuracy from time source (ex. GNSS or UTC) to 5G end device when used as part of IEEE 1588 PTP sync device is 1μs (absolute time-synchronization in substation local area networks IEC 61850 Sample Values).
• For a 5G system entity acting as a PTP grandmaster in a subsystem, 250 ns accuracy applies between time source (ex. GNSS or UTC) and 5G end device.

The Smart Grid can use 5G system as a supplement or backup to their GNSS receiver based (or wired) clock synchronization systems to improve accuracy and availability of the clock synchronization across the Smart Grid.

Rel-16 and Rel-17 requirements cover a good portion of the accuracy requirements for a 5GS timing resiliency system, including:

• Capability to meet 5GS synchronicity budget of ≤900 ns (factory) to <1 μs (wide area) as defined for Industry 4.0 applications in TS 22.104
• Propagation delay compensation, to ensure that clock synchronization can be conducted accurately across wide area deployments needed for power sub-station support
• Support for IEEE 1588 PTP [23]
• Clock synchronization to UTC leveraging 5G system C-Plane and/or U-Plane