The IEC 61850 series of standards for electrical substations harness the potential of Ethernet based wired networks as the backbone for intra-substation data communication. However, the flexibility of the standard, particularly the IEC 61850-90-5, enables the integration of emerging communication technologies such as wireless 5G cellular network for both intra and inter-substation data communication. In this paper, we evaluate and compare the reliability of a digital substation by analyzing different network architectures used based on wired Ethernet networks and wireless 5G networks. Reliability Block Diagram (RBD) approach have been implemented to conduct the comparison and the availability and Mean Time to Failure (MTTF) have been evaluated. The results show that the wireless 5G networks have a positive impact on the availability of the substation but the MTTF and Mean Time to Repair (MTTR) was found to decrease when compared to the wired Ethernet networks. However, such results indicate that the wireless 5G network is a sensible upgrade for the digital substation when considering the advantages and flexibility offered by the technology.

Digital substation technology adhering to the IEC 61850 standard has provided several opportunities and flexibility for the rapid growth and complexity of the present and future electrical grid. The communication infrastructure allows complete interoperability between legacy and modern devices. The emergence of 5G wireless communication and its utilization in substation operation presents significant advantages in terms of cost and scalability, while also introducing challenges. This paper identifies research gaps in the literature and offers valuable insights for future analysis by providing a simulation study using an empirical latency dataset of a 5G network to illustrate three aspects of substation operational challenges: coordination of protection schemes, sequential reception of packet data streams, and time synchronization processes. The findings show a mean latency of 8.5 ms for the 5G network, which is significantly higher than that of a wired Ethernet network. The results also indicate that the high latency and jitter compromise the selectivity of protection schemes. The variability in latency disrupts the sequence of arriving data packets such that the packet buffering and processing delay increases from around 1.5 ms to 11.0 ms and the buffer size would need to increase by 6 to 10 times to handle out-of-sequence packets. Additionally, a time synchronization success rate of 14.3% within a 0.1 ms accuracy range found in this study indicates that the IEEE 1588 protocol is severely affected by the latency fluctuations.

The implementation of renewable energy sources such as solar and wind for electricity production has picked up an enormous pace in recent years, which not only gives rise to a more sustainable process of electricity production but also minimizes carbon dioxide emissions. However, the intermittent behaviour of the sources coupled with the absence of kinetic energy contribution to the electrical grid introduces a negative impact on the frequency stability of the grid. Hence, the grid requires further reinforcements in its infrastructure, and one such promising solution lies in the usage of grid-connected Energy Storage Systems (ESS) as a source of Frequency Containment Reserve for Disturbance (FCR-D) ancillary service for frequency regulation in Europe. This paper presents an extension of our previous work [1] to realize ESS as a favourable candidate for FCR-D service. In our previous work, we showed the benefits of ESS in terms of performance and cost compared to hydro turbines. However, the extended research takes a deeper dive into the characteristics of ESS and shows how the deployment characteristics, such as delay and power level, affect the frequency trajectory of an electrical grid with variable renewable energy penetration (REP) levels after an N-1 disturbance. The research found that ESS adhering to FCR-D deployment characteristics for the same disturbance yields diminishing impact at higher peak power levels, and with an increase in delay of deployment, the peak level must be increased by around 0.8MW to 5.08MW for per second increase in delay as the REP level of the grid rises to 30%. The study also found that the deployment delay must be reduced by up to 33.33% to keep the same level of impact of the disturbance. Finally, the total cost of FCR-D support may increase by up to 148.8% depending on the FCR-D deployment and grid characteristics.