A-Z of Smart Grid Analytics
Smart grids fuse energy development with technological advancements. Using sensors, IoT, and other computing devices, there is a provision for two-way communication between consumers and utility providers in a smart grid. As an artificially intelligent system, a huge amount of data comes from various sources, e.g. smart meters. All the unstructured data gathered from these sources can only be valuable with smart grid analytics.Smart grid analytics are systematic computational analyses of the data produced in the grids. With these analytics, one can get a more precise interpretation, communication, and identification of data trends or meaningful patterns from the data that comes in. Thus, it is essential to improve grid operations and predict the next course of action.
A Brief History
From the 1990s, attempts at electronic metering, control and monitoring evolved into smart grids. From automated meter readings in the 1980s to Advanced Metering Infrastructures in the 1990s, attempts have been made to go beyond measuring power usage to maximizing the information.[caption id="attachment_8591" align="aligncenter" width="940"]
Image by Gerd Altmann from Pixabay[/caption]The concepts of analytics can be traced back to the 19th century with Frederick Winslow Taylor's time management exercises and Henry Ford's measurements of assembly lines' speeds. It would interest you to know that predictive analytics (which is now of high importance in smart grids) started in the 1940s. However, it did not attract any attention until the 1960s, when decision support systems became popular. By 2005, businesses applied analytics to make iterative explorations on past activities and make decisions to plan the future.Applying analytics to smart grid data is what birthed smart grid analytics. The problem of big data (as Roger Magoulas called it in 2005) has always existed as long as the internet. Around early 2012, big data in smart grid systems initiated collaboration between smart grid integration companies and data analytics start-ups.As grids became smarter, grid data analytics also developed, using available technologies such as machine learning techniques. Computing techniques like statistics, machine learning (under artificial intelligence), and data analytics are now being applied in various facets, and the power sector is not left out. As we will see, smart grid analytics gives relevant information that helps set the course of upcoming activities for the effective distribution of power.
Three Things You Should Know About The Current Trends In Smart Grid Analytics
For one, the smart grids analytics market in Europe was projected in 2019 to grow at the compounded rate of 11-12.7% by 2025. This growth is based on how advanced grid technologies are embraced on a broader scale. However, I have observed these trends;
1. There is currently rapid growth in investment in smart grids projects and, subsequently, smart grids analytics.
Many countries in the European Union have invested in smart grids projects and are recording successes. Of the projects, up to 59% are demonstration projects, 32% are for deployment, while 9% are research and development projects. A significant highlight is the smart meter roll-out in Italy that takes up to 71% of these projects aforementioned.
Smart meters are installed in all of these projects, and to get relevant information from the data, smart grids analytics have to be employed. These projects result from increased interest and initiatives channelled to the ongoing energy transformation and sustainability goals.
2. Smart grid analytics work with real-time data even with the increased speed and variety of requirements.
This easy adaptation is because they are entirely computerized and are built on the blocks of advanced technologies. Smart grid analytics can now generate information from high-speed data of various forms needed for the grids' operation and prior knowledge of what to put in as resources.
3. Digital technologies and cloud computing would continue to improve and allow for more data computation.
Digital data, which highest storage used to be terabytes, is now accessible on larger scales like exabytes and zettabytes. Manual methods and previous ways of analyzing this data are becoming redundant. Also, with the inclusion of renewable energy in the conventional grids, the adaptation of intelligent systems is increasing, and the need for grid data analytics will follow this trend.
Challenges Of Smart Grid Analytics
Despite the enormous advantages and improved technologies, there are still a few challenges. Some include:
- Cost implications - the initial costs of setting up smart grids make many grid operators sceptical about using smart grid analytics. For the grid as well, it usually includes the costs of sensors and other components in making it effective. The analytics themselves are part of what makes the smart grid a modern electric system. However, it is worth the investment to foster a low-carbon economy and a greener world.
- Security concerns - the fact that smart grids allow for two-way communication is a concern as the data is prone to cyber-attacks. Despite this, cybersecurity has continued to improve and is developing better solutions using codes and encrypted data.
- Customer demand - the demand needed for effective use of smart grid analytics is higher than what exists now. Not enough grid operators have adopted analytics, and the low-scale usage is not optimal. More large-scale energy supplying and distributing firms need to embrace the new technologies at this time.
Smart Grid Analytics Scope
As smart grids involve a more frequent measurement of rates and power usage and newer energy generation technologies like renewable energy is integrated, data is gathered from time to time. This measurement is not done manually but via smart meters and other data collection sources, thereby helping utilities and their customers manage their bills and rates of power usage. To interpret all these forms of data that come into the grid, smart grid analytics works across a broad scope, as illustrated in the table below.Smart Grid Analytics ScopeOverviewAspectsOperations analyticsThis involves the functions that manage the operational aspects of the smart grid, that is, how the entire system is run.
- Managing renewables
- Grid management
- System performance
- Adaptive analytics
- Operational effectiveness
- Resiliency analytics
Signal analyticsSignal analytics makes the most of the state of the signals that are gotten from the sensors on the advanced metering infrastructure.
- Meter data
- Sensor signals
- Substation waveforms
- Line sensor waveforms
State analyticsThis involves various analyses and interpretations of the state of the grid, even geographically.
- Electrical states
- Parametric identification
- Topology identification
- Asset utilization
Enterprise analyticsThis analyzes the business expectations and economic values of the entire grid management system.
Customer use analyticsThis analytics makes use of the operational data of customers, their demand, and their reactions.
- Customer segmentation
- Sentiment analysis
- Home energy management
- Enhanced billing
- Utility communication
- Non-linear load parameters
- Demand profile
- Increasing customer value
Event analyticsIn the case of unprecedented events, as well as planned ones, analytics also makes room for organizing energy schedules.
- Events correlation
Opportunities for Smart Grid Analytics
According to a report on the market analysis, trends, and forecasts on smart grid analytics, the dynamics affecting how much smart grid analytics can do, shift from time to time. These factors, asides from macroeconomic factors and internal market forces, include:
1. Customer acceptance and engagement
More electric power consumers have come to accept the smart grids deployed in projects by private firms and public institutions, though not so many have engaged the setups. More of those who engaged in trials have been typical volunteers, according to a report after a survey.
In the same study, 55% of the funding from smart grid projects in Europe was from the EU and other government agencies, while 45% came from private investments. 87% of smart grid projects in Europe have received funding despite the business case uncertainty that has restricted many private firms from investing in the ongoing projects. Meanwhile, only 65 out of 112 customers responded to the survey.
2. Regulatory policies
Regulations change from time to time, and recently, policies supporting the growth of renewable energy and smart grids have been put in place. Notably, the European Union has played a significant role in making smart grids effective by policies like the strategic research agenda road mapped for 2007-2035
3. Innovative structures
The structures available are key factors to how much smart grid analytics would thrive. The more innovative systems are implemented, the more innovations like this collaboration between technological advancements and the power grid would improve. Thus, there are opportunities for smart grid analytics to thrive when these factors are in favour and all work towards achieving the set goals.
Other contributing factors.
Investments have been made in many projects involving smart grids all around the world. These investments have happened over time because grids enabled with smart technologies ensure reliable, secure, and efficient electricity management, meeting the objectives of any power grid.
While these attractive features are of great importance, I cannot stop at them without mentioning that the only way to sustain that power quality management and proper communication, planning, and management of data derived intelligently by smart meters is by the use of smart grid analytics.
The constant development of smart grids has provided a platform for smart grid analytics. Coupled with the fact that smart grids utilize wired and wireless communication infrastructure, information systems, demand response management system, SCADA (Supervisory Control and Data Acquisition), GIS (Geographical Information System), CIS (Customer Information System), and advanced metering infrastructure. Each platform in a smart grid provides an opportunity for smart grid analytics. This is because of the complexity that comes with data from decentralized energy systems.
More so, these issues of data management are the ones that give smart grid analytics a chance. For example, various standards (like models) have been put in place because of differences between networks and devices in terms of bandwidth constraints, energy constraints, continuous and non-continuous data, and the likes. Also, the need to manage massive data and ensure data privacy has led to more and more opportunities for smart grid analytics.
Future of Smart Grid Analytics
I strongly believe that we can achieve a lot with smart grid analytics. Experts expect the market to grow at a CAGR of 25% between 2019 and 2024, given the increase in IoT and big data analytics. In addition, new information and communication technologies keep penetrating the grid, which is enough to make smart grid analytics grow stronger globally.Speaking of global locations, it is projected that the Asia-Pacific region is likely to witness the fastest growth, particularly in the two countries that dominate the region - India and China. This growth is predicted because the energy consumption in India since 2000 has doubled, and there is potential for more. Interestingly, the current market for smart grid analytics is also really competitive in Europe and worldwide, and this is a driver for growth.
IoT is one technology that is likely to take the grids to another level, despite the predicted challenges of cyberattacks, impersonation, the need for more IT infrastructure, data tampering, security, and privacy issues. While these challenges are already being combated using cybersecurity techniques and blockchain technology as a secured and distributed database, more researchers and developers are making better ways to handle them. As a result, the future of smart grid analytics would include IoT, more than it is now, in how smart grids operate. However, the existing IT infrastructure has led to the discovery and application of certain technologies used for smart grid analytics.
Best Technologies for Smart Grid Analytics
Smart grid analytics has evolved over the years and consists of various techniques involving the integration of data from electric power sources, analysis, processing, and visualization. There are generally four kinds of analytics. In order of complexity, they are
- predictive, and
There is also cognitive analytics though it is a recent advancement that combines many functions. Of the four, the smart grid operators prioritize predictive analytics to find out what could happen at any point in time.Nevertheless, the best technologies that have also evolved for smart grid analytics include Business Intelligence and Data Analysis (BI&DA), or big data analytics. The two terms were introduced to companies around 2009-2010 and have been unified as they work together to form the most relevant technology for smart grid analytics.
1. Business Intelligence (BI) and Data Analysis
Business intelligence is a broad term that comprises several activities aimed at helping companies make better use of the data at their disposal. Also, data analytics is used to make conclusions from raw data by computational analysis. Business intelligence and data analytics fuse together to maximize the benefits of the smart grid by applying business-centric methodologies to get useful information from the smart grid. BI makes use of tools and software to mine data, process it, and make meaning out of it using tools like spreadsheets, OLAP (Online Analytical Processing), data mining tools, and data reporting and visualization software. In the end, the data that has been collected is evaluated, optimized, and re-evaluated.
BI and DA also perform some processes to manage and create centralized data from various sources continuously. Some of these processes that occur before any form of analysis include pattern mining (that is, identifying patterns and similar arrangements), classification, the association of rule mining, clustering, making of regressions, and detection of outliers.
2. Other technologies that form the framework - Databases like Apache Hadoop, MapReduce, SQL
Asides from the techniques and technologies used to analyze data, a proper database is necessary for smart grid analytics. The currently available databases form software frameworks that are open source and they spread offline data across clusters and nodes for easy processing. They have improved with time, so new ones have more specific functions than the previous ones. For example, though Apache Hadoop is a popular and basic database, there is a need to use MapReduce because of background indexing. More so, for the sake of ad-hoc querying issues, SQL has been introduced.The NoSQL database allows the newer data management technologies that are easier to scale and perform optimally than its counterparts. Some of the NoSQL open source database types include Cassandra, Elastic Search, MongoDB, and Hbase. Hbase and Cassandra are each column stores based on the concept of BigTable. However, Hbase bases also on Apache Hadoop, while Cassandra bases on DynamoDB.
The Important Role of Big Data in Smart Grid Analytics
Big data refers to the massive amount of data that an institution, unit, or system has to manage to utilize efficiently. The amount of data chunked into the smart grid through the smart meters, weather updates, social media, programmable thermostat, traffic updates, remote terminal units, and so on can be overwhelming and difficult to evaluate using the existing traditional methods. Of course, big data goes beyond the size, but the data format is usually diversified in the power grid.[caption id="attachment_8614" align="aligncenter" width="940"]
In a day, there can be 30GB of PMU data and 120GB of smart meter measurements, not to mention 16GB of weather data from satellites, radars, and weather forecast models, and up to 2.7GB of vegetation and topography data.
They all come in at a high velocity and give updates in intervals of about 1-15 minutes. With the variety and complexities involved, big data plays a significant role.Some ways in which big data analytics play a role in smart grid analytics include:
- power generation analysis
- load management with demand
- performance analysis of energy consumption
- forecasting and scheduling of loads
- evaluation of economic effects and constraints.
Big data analytics plays its role in the smart grid and motivates all that has to do with smart grid analytics.
Market Analysis of Smart Grid Analytics technology in Europe
Without a doubt, there’s so much we can achieve with smart grid analytics. Experts project the global smart grid analytics market to grow at a CAGR of 25% between 2019 and 2024, given the development of the Internet of Things (IoT) and big data analytics. While for Europe, the market growth projected is a CAGR of 10.38% by 2028.
Despite North America and Europe leading the market significantly, experts predict that the Asia-Pacific region is likely to witness the fastest growth, particularly in the two countries that dominate the region - India and China. With India’s energy consumption increase (over 100% in the last 21 years), these predictions are on track.From observing Europe’s smart grid analytics market, these things are worthy of mention;
- There are key factors propelling the market, like increased investment in research and pilots in the UK, growth of renewable energy in France, Austria, the Netherlands and the likes, and the predominance of the use of the IoT in Italy.
- There have been constant technological advancements in IoT, consequently improving its application to smart grids.
- There is a growing interest in smart grid systems that have led to more funding and investments by the European Union Commission and other key bodies.
- More renewable energy sources are being integrated to meet energy demands from Industrial growth.
As a result of the factors highlighted above, more advanced grid analytics companies in Europe like Hive Power, ERIGrid, GridCure and the likes are on the path to making the grids effective with their solutions.
Smart Grid Analytics Use-Cases in Europe
Below are three use cases and projects on smart grid analytics in Europe.
To make customers active participants in their energy savings, Hive Power, in partnership with AEM (Azienda Elettrica di Masagno), created an application that updates customers with information regarding their electricity usage called Drainspotter.
In a preliminary survey of 9,000 homes in Lugano, we analyzed the data obtained from 15-minute sampled load profiles. The meters used to get this data were L+G E450. DrainSpotter allows users to monitor their pattern of energy usage, as well as summaries of customer behaviour. An exciting feature of the DrainSpotter is its ability to notify customers of anomalies in their energy consumption patterns. For example, Hive Power figured out that if AEM's residential users eliminated excessive standby power of more than 200W beyond 14 days, 5% of them would reduce their energy consumption by at least 20%. Also, for one and half years, 4.2% of the customers would save up to 500 CHF($546.49) on energy. In all, the DrainSpotter gives a system that supports the inclusion of users in their energy cost management and supports DSOs in providing expert advice on end-users.
Enexis Netbeheer, a grid operator in the Netherlands, started to make an internet of things inspired smart grid in 2016. They had 900,000 smart meters already installed for a start and sought to increase them with time to 2.8 million by 2020. One motivation for this project was that more stakeholders were interested in making grids more flexible, hence the installation of smart meters in their numbers. To get to the desired number of meters, 7,700 meters were installed weekly and gradually increased to 10,000 meters per week. With such a large number of smart meters, it calls for proper data management.
Also, a futuristic step they put in place was the plan to install sensors in all their 50,000 substations to know the details of their activities. In the same way, the meters that were already installed and only had 2G were upgraded by creating sim cards that helped them function with 4G enabled devices. Using an in-home display to show the energy use by customers in real-time, Enexis and the partner companies were able to achieve the awareness customers needed to understand the cost implications of their use of one electronic device or the other.
This is an ongoing horizon 2020 project by BRIDGE, and this is the project's final year. According to EU-SysFlex, their name stands for "Pan-European system with efficient, coordinated use of flexibilities to integrate a large share of RES." The aim is to provide new services to support systems with more than 50% renewable energy sources and find problems associated with renewable energy integration and solutions.With 34 partners, including TSOs, DSOs, researchers, and €20 million from the European Union horizon 2020 funds, this project is globally recognized as a very innovative one. Demonstrations for this project exist in Germany, Finland, Italy, Estonia, France, and Portugal. EU-SysFlex applies Smart Grid Analytics in the aspect of grid data management, flexibility management, and forecasting.The demonstration in Brandenburg, Saxony, and Saxony-Anhalt in Germany resumed live operation earlier this year (2021). Currently, up to 100 solar photovoltaic systems and wind power plants are being managed there. In addition, this project has innovatively provided flexible solutions to real-time data management for similar energy systems.
A lot of research goes into developing analytical models fit for smart power grids, especially when renewable energy is involved, which hints that the future holds more.From all I have discussed in this series, it is evident that smart grid analytics is of great advantage to the power sector in Europe and worldwide. The best way to keep it growing is for more grid operators and energy suppliers to explore it more and apply it to various aspects of their operations.