Computational intelligence: The key to making smart grids really smart
By Shawkat Ali
Around the world, in developed as well as developing countries, utilities are embracing smart grid technologies to improve the grid's efficiency and incorporate renewable energy resources into the grid. The strategy is both practical and laudable because it will create the means to meet society's future electricity demands in a way that also minimizes carbon emissions and reduces electricity's environmental impacts. The recognition of smart grid's pivotal roles, along with many regional and national policies acknowledging its importance and often mandating its adoption, are inspiring others to move into the new smart grid era. The smart grid is inevitable now. We should all join in to help it advance and succeed.
Computational intelligence (CI) is one of the techniques utilities will use to make sure their smart grids are as intelligent as possible and delivering tangible, worthwhile benefits to utilities, users and the environment. A comparatively new era in computing, CI is already at the heart of many types of "smart" technologies and services in a variety of industries. In particular, CI enables us to interpret big data and create knowledge from it to make decisions in real-time for critical business functions. The purpose of this article is to discuss some of the exciting ways in which CI is being applied in smart grids.
Computational intelligence for smart grid optimization and planning
CI can be used throughout a utility to optimize operations and planning. Broadly speaking, it can be used in power generation, transmission, distribution and consumption applications to meet the Smart Grid's safety, security, reliability, resilience and efficiency needs. Many examples are available to illustrate very practical applications.
For example, CI can be used to forecast the amount of renewable energy that might be injected into a grid during a particular time period. We can then use this information to determine how much power production is needed from traditional sources, such as coal.
We can use CI for both short-term and long-term load forecasting. This is particularly important, given that population growth is one of the main factors inspiring the industry to produce more power. Similarly, we can use CI to inform our demand management programs.
CI can be used to help utilities respond to outages caused by natural disasters, such as storms or downed trees, which impact high-voltage transmission lines. When incorporated in a smart grid, CI can detect asymmetric single-phase-to-ground or two-phase-to-ground faults or symmetric three-phase to-ground faults and determine where the faults have occurred in the transmission and distribution system. This is a significant attribute in any power system.
CI can also be applied in conjunction with sensor networks to monitor power quality. If a power quality issue occurs, the system will alert the administrator, who can then take the steps necessary to address the problem.
CI can be used to monitor power losses from the grid. These losses usually average about 7% of a country's power production but in some countries, like South Africa, the losses represent about 17 percent. CI can not only monitor the losses but figure out the cause.
Finally, utilities can use CI to optimize the real-time process of purchasing energy from multiple suppliers based on price, variable rate or other data for recent or forecasted time periods. They can identify the best suppliers through CI as well.
Security applications for computational intelligence
We need to keep the energy supply chain secure. We must protect the grid's IT technologies, for instance, from hackers who might want to stop renewable energy production or access a utility's billing information.
In computer networks, many attacks are known to us. For example, denial-of-service (DoS) attacks are known attacks. We understand these attacks and can apply technologies and systems to help protect our networks against them. But how can we stop unknown attacks?
CI is an algorithmic approach that responds to the unknown in the way human beings respond to problems they haven't encountered before: the system uses the knowledge it has already acquired from previous experiences to develop a new mechanism to help stop the new attack. This is the beauty of CI and explains why CI has value in smart grid networks. It also explains how CI can create robust and dynamic security systems.
Moreover, because CI acquires and learns from traditional attack information, it will study a new attack and compare it to previous attacks. It will alert the network manager or administrator that the attack has occurred and that the system has stopped it. Even if a very strong unknown attack occurs in the smart grid and CI fails to stop it, the system will instantly alert the network manager that further action is required. So CI can determine how to address unknown attacks and it also automatically provides warnings when needed.
Cloud computing and data mining in smart grid systems
Cloud computing also has many useful applications in smart grid and data mining is one of its most significant attributes. A smart grid will continuously generate huge volumes of data about the weather, solar or wind characteristics, network security, people's electricity consumption and how much electrical power people add from their own roof-top systems to the grid. Utilities will need to store that data and cloud computing is the solution for that.
|The Internet of Things will play a significant role in smart grid. It will be used in network safety management, network operations and maintenance.|
Once smart grid data is stored in the cloud, utilities will need data mining techniques to develop knowledge from the raw data. They will use the data to determine the relationship between demand and supply or to explain customer impressions or opinions about their smart grid services. So cloud computing, data mining, and smart grid are all very closely related.
I want to mention, however, that the quality of the intelligence we gain from data mining will be influenced by the quality and quantity of data we have available to us. Smart grid is creating a wonderful database that is a resource for monitoring the system and even solving operational problems. But the database needs to be comprehensive. We need a very good database for every section of the smart grid.
The "Internet of Things" will help implement smart grids
The Internet of Things brings connectivity to any object, whether the object is an element in the smart grid, a person or any other physical entity. It has the capability to automatically transfer data over a network without any human or human-to-computer interaction. It uses a combination of wireless technologies, micro-electro-mechanical systems (MEMS) and the Internet.
The Internet of Things will play a significant role in smart grid. It will be used in network safety management, network operations and maintenance. It will be used to monitor the security of the smart grid, to manage end user interactions, and many other things. The Internet of Things is not just a big world of connected devices, it's also a tool that we can use to implement smart grid.
In order to meet the growing demand for electricity, we cannot be afraid of these new techniques and systems. The technologies and applications needed for smart grid are a 50-50 combination of modern information and communication technologies (ICT) and traditional grid technologies. We have these technologies. We just need to integrate existing systems.
While smart grid technologies are available, people from both ICT and utility engineering communities do need to work together to make the smart grid dream come true. ICT experts, statisticians, and mathematicians need to collaborate with people from the engineering side of the business to create successful systems and programs.
About the Author
Shawkat Ali is an IEEE Smart Grid Technical Expert, a senior member of IEEE and a member of the IEEE Computational Intelligence Society. He is an authority in computational intelligence and its role in Smart Grid and speaks frequently on these and related topics at IEEE conferences. He has also chaired IEEE international workshops on computational intelligence topics. He is a senior lecturer in the school of engineering and technology at Central Queensland University in Australia.