Building the African smart grid


By Jonathan Spencer Jones

This planning needs to be done both from an operational and a strategic, long-term perspective, as investment decisions today will certainly affect the grid opportunities of the next decades. There is a worldwide trend to modernise the grid. This involves the transition from the grid being fundamentally an electro-mechanical system to a future grid which is highly digitised, and where computers, technology and advanced materials are used to optimise the electric power infrastructure. This is very similar to the evolution of telephone systems, where some years ago there was the transition from electro-mechanical exchanges to fully digital systems. The nature of the power grid is far more complex than the telecommunications system, however, and the implications of moving from the existing system to the future systems require many more decisions and include many more variables.

How will African utilities plan their grid investments in such a way that the decisions they take in the short term can be beneficial in the long term? It is easy to invest in new systems to achieve specific short term objectives, but the utility may be faced with an investment that does not contribute to long term goals. The buzz around smart metering in the South African market presents an interesting example. There is high level interest in implementing smart metering as soon as possible, as a mechanism to help address the power crisis being faced in the country. However, the decision to implement smart metering has many potential implications beyond simply trying to introduce a demand response mechanism.

By making the correct smart metering decisions a framework for grid data collection at the metering end-point can be established, which can be used for a variety of smart grid applications such as outage management, predictive maintenance, asset management, revenue protection and much more. So, the focus should not only be on implementing smart metering, but rather implementing smart metering in the context of a smart grid strategy.

The smart grid of the future will interact with customers and markets; will be adaptive and self-healing; will be optimised to make best use of assets; will support a range of generation options, including distributed generation and renewables; will be predictive rather than reactive, thereby improving reliability; will merge monitoring, control, protection, maintenance, EMS, DMS, marketing and IT; and will be more secure from attack. Many investments and applications related to the grid are made in isolation, and then integrated at great cost at a later stage. As more and more systems are added, there is an ongoing process of trying to make them work together. However, the correct long term strategic planning can make this process far more efficient.

Utilities need to develop a long term vision of how the grid will evolve in terms of architecture and capability. Key issues that need to be considered include network management, data management and security. These should ideally be based on open standards, allowing easy integration between components and systems on the grid. Although this approach may cost slightly more in the short term, the long term benefits will far outweigh any short term costs. Once these decisions have been made, it becomes far easier to implement new capabilities on the grid and incrementally add new applications.

Let’s revisit the example of smart metering. A key requirement of smart meters is to communicate bi-directionally. The communications channels that are chosen could take many forms –the proprietary communications mechanism provided by the meter vendor, for example, or any number of open or semi-open channels. Although the vendor-supplied communication may be the easiest and quickest, it will probably not contribute to the overall grid vision. It would be far better to implement an open communication system that could provide not only meter communications, but also capacity for future grid services.

Separating meter communications from the meter is a key step in leveraging investment and providing future capacity. To achieve this, for example, the utility could overlay an IP network over the grid, and use this to communicate with IP-enabled meters. Any other IP-enabled device could similarly share this channel. The channel itself could be PLC based (such as BPL) or appropriate wireless infrastructure, or a mix of both. And this is just one element that needs to be considered – others include the type of data that can and should be collected through the meter, how this is managed, and how it can be used to deliver value to the utility.

So, how does one plan for the future? Realise that the grid of the future will be fundamentally different to the present-day grid. It will be intelligent, will be requirementsdriven, will be flexible, will be based on open standards, will be secure and will be expandable. It will also be extremely complex. The smart grid of the future will be built in an incremental manner, over many years. With the right foundation, the process can happen efficiently and effectively, with minimal wastage. Conversely, the cost of doing it wrong will be huge.