Leveraging Smart Grid Information
Written by: Keith Houghton (CTO) & Gary Michor (CEO), Screaming Power Inc.
Even if you don’t fully understand what the “smart grid” is, no doubt you have heard the term. It could be argued that the smart grid is the result of a technological evolution within the electricity supply chain, where reliability and quality have been improved by the adoption of modern technology, particularly communications technology. But there are a few game-changing concepts that take this evolution idea to another level. One such concept is that of the microgrid.
The microgrid is a small part of the distribution network with an arrangement of small, embedded generation units, usually fuelled by renewable resources. It can be isolated from the main network at times of disturbance. While isolated, the generators can supply the local loads, thereby maintaining service until the main network is available again. This ability to “island” generation and loads together has the potential to provide a higher local reliability than that provided by the traditional supply model. It provides:
- Better use of the local resources, creating a more positive image;
- Improvements to the local economy by reducing “downtime” and minimizing costs related to power outages;
- Lower emissions and the potential to have lower cost negating traditional economies of scale;
- Reduced risk of cyber security attacks that can affect all areas of the network.
Penetration of distributed generation in many areas of the world has not yet reached significant levels. That situation is changing rapidly, partly due to government policies driven by environmental change. To arrange generation into microgrids requires attention to technical issues related to renewable generation and new technologies such as electricity storage.
The objective should be to achieve the desired functionality without extensive custom-engineering and still having high system reliability combined with flexibility of generation placement. To achieve this we must rely heavily on standards and adopt a peer-to-peer and plug-and-play model for each component of the microgrid. The peer-to-peer concept ensures that there are no components – such as a master controller or central storage unit – that are critical for operation of the microgrid as a whole.
Distributed generation encompasses a wide range of prime mover technologies, such as internal combustion engines, micro-turbines, photovoltaic, fuel cells and wind power. Most emerging technologies such as micro-turbines, photovoltaic and fuel cells require an inverter to provide the AC supply that most customers need, so the harmonics and DC current it injects to the grid must be managed.
When a microgrid is main grid-connected, it can be treated as a controllable load or source. In this mode of operation, the primary function of the microgrid could be to satisfy all of its load requirements and contractual obligations with the grid.
The microgrid’s most important feature is its ability to disconnect from the main grid when an abnormal condition occurs. This is known as an “islanded” mode of operation. The microgrid is faced with the following issues:
- If the microgrid is exporting or importing power to the grid at the time of disconnection, then immediate control actions are needed to balance generation and consumption in islanded mode.
- If the connected load within the microgrid exceeds the available generation, demand-side management should be implemented.
- A microgrid usually has renewable generating resources that cannot be relied on to produce the maximum output all the time and there are usually no units that have the fast-response reserves that are inherently present in the conventional grid. Therefore, fast-response intermediate storage units are required.
- The microgrid should maintain acceptable power quality while in islanded operation. The primary energy storage device should be capable of reacting quickly to frequency and voltage deviations and injecting or absorbing large amounts of real or reactive power.
- When connected to the main grid, the voltage and frequency are established by the grid. When the microgrid “islands,” one of the primary generating sources should become “master” by establishing its voltage and frequency as the reference.
There are other issues, as well, such as the ownership of the distributed resources, so one must consider a microgrid very carefully.
Virtual Power Plants
Since energy generation near the consumer offers economic and ecological benefits, over time decentralized generation will become common. Distributed generation in a non-islanded mode or distributed generation not part of a microgrid can now be considered a so-called virtual power plant. A virtual power plant is a collection of small and very decentralized generation units that are monitored and controlled by the utility’s distribution management system as a single entity.
In effect, the distribution utility needs dispatch algorithms similar to the transmission system operator with complex mathematical models for optimization. The functions of distributed energy resource can be subdivided into planning functions and control functions:
- Generation and load management
- Generation forecast
- Load forecast
- Unit commitment
- Weather forecast
Generation management functions allow for the control and supervision of generation and storage units of the virtual power plant. Load management functions allow the control and supervision of controllable loads in the virtual power plant.
The electric grid operates with power generated at the same time it is consumed, and traditionally with little storage of electrical energy. Fluctua-tions in demand are handled by different types of reserves, such as spinning reserve. This means that the transmission and distribution system must be built to accommodate maximum demand rather than average demand, resulting in it being underutilized most of the time. It’s known that energy storage can enhance network reliability, enable efficient use of base load generation, and support a higher penetration of renewable energy resources, but capital expenditure tends to be high.
In some countries, energy storage already exists. At the grid level, pumped storage hydro power plants represent most of the facilities today. At the microgrid level, ultra-fast storage will be needed to address short-term power-quality disturbances and frequency regulation needs. Medium-term storage will be needed to shave peak demand.
Microgrid battery storage technology is most frequently used because it is simple technology to apply. Lead-acid batteries are used for backup power in small power plants. In larger-scale applications, sodium sulphur and vanadium redox flow batteries can be used. Large-scale, longer-term energy storage can be applied to peak shaving, and there are moves afoot to utilize the batteries in plug-in electric vehicles.
Electrical storage can also be connected directly into the distribution grid. It can be used to support microgrids and can even be integrated into building automation systems. Electri-cal storage should fulfil a number of functions on the grid, including:
- Load leveling
- Power system stabilization
- Reactive power support
- Acting as spinning reserve
Energy storage will be a major element of smart grid and microgrids and is particularly important in the case where there is a lot of non-reliable energy (e.g., solar or wind). Communication is Vital.
The communication infrastructure between microgrid components and the grid is of vital importance and must be considered when choosing the right islanded microgrid ap-proach. The microgrid should have a plug-and-play architecture for flexibility and general removal of technical barriers. Where communication between components is required, the delay within the communication network cannot present a problem for the microgrid itself.
Security and privacy of data are important. Because future networks will be IP-based, every consumer becomes a potential entry point into the utility’s networks. Failure to achieve security of the infrastructure may have severe outcomes. Interest-ingly, multiple microgrids improve security because multiple microgrids need be attacked to have a global effect on the grid.
Monitoring and Control
Distribution companies take measurements for load forecasts, real-time monitoring and control. With this information the network state can be estimated, predicted and controlled fairly accurately, but with microgrids the problem is more pressing. To have a flexible and adaptable system, it is necessary to have a great deal of high-speed digital monitoring; even when islanded the utility needs to know what is going on.
In recent years, utilities have been undertaking extensive advanced meter projects with the intention that the communications system installed to support the advanced meters will also be used as the communications platform for multiple solutions. With the deployment of advanced metering infrastructure (AMI), additional measurements are often available and could be added with minimum cost.
Generation not included as part of a microgrid is usually operated at the owner’s discretion. As customers decide to install more renewable generation, it changes the topology of the distribution network as well as the load patterns perceived by the utility’s systems.
Electric vehicle (EV) support may have a devastating effect. The future grid is significantly more complex and unpredictable. A large amount of EVs could require new distribution management applications to manage the issues. In addition, the power generated (D.C. instead of A.C.) means power quality measurements are often needed at or near power production.
Most types of renewable power production, especially wind turbines, have large fluctuations in their output. Distributed energy resources increase the needs for voltage quality monitoring and control. Monitoring of the generation is needed even for very small energy resources. This effectively requires a SCADA (Supervisory Control And Data Acquisition) system at the microgrid level.
Demand dispatch is a relatively new operating concept for utilities and represents a different approach to balancing generation and load. Unlike traditional demand response, demand dispatch is active and deployed all the time, not just during peak times – it aggregates and precisely controls individual loads on command. Resources like EVs, which can actually adjust the load they represent to the grid while charging, could vary their charging rates to follow the variations in intermittent resources (e.g., wind, solar); similarly, air conditioner set-points can be adjusted. Large numbers of EVs operating in this mode could be a valuable asset for optimizing the integration of renewables. Other consumer-owned devices could also be used to support demand dispatch.
The choice of dispatch strategy should take into account the types of installed distributed resources. If non-reliable or combined heat and power distributed resources are available, there must be at least one resource that would act as a load-following unit. The load-following resource simply adjusts its output power to balance the instantaneous generation and consumption.
The microgrid or storage device owner should have a means of recovering the installation and operating costs of that device. In the future, the distributed devices may be able to participate in the market operation of the microgrid.
The ownership of the microgrid components affects the choice of dispatch approach. If the distributed resources have several owners, they tend to be operated in a way that will optimize the individual needs of the owner.
The easiest control approach to be used is the microgrid with one owner. Having one owner implies that the distributed resources do not have to compete and the objective of the microgrid controller is to maximize the value of the entire microgrid. Some advocate that the utility should be responsible for the microgrid. There are also arguments that larger consumers could take responsibility.
Additional research is needed in the area of microgrid deployment and operation that can address the key value areas of the smart grid vision.
A smart grid will generate huge volumes of data; however, how this data will be handled is not well defined. For example, advanced asset management depends on data and access to it. An objective evaluation is needed to fully understand all the pros, cons and value propositions.
Applications must ensure that all users, processes and technologies can access required information. What data is needed and how will it be retrieved and managed remains in question. Further work is needed to identify a data management process. Standards are needed to ensure smart grid data is effectively utilized by the utility/market processes and by consumers to make smart decisions.
New smart grid sensing and measuring technologies will have the capability to generate vast amounts of data. This data will include both operational parameters and economic information. Data may be collected on a near-real-time basis.
Further work is needed to explore how the use of this data can be optimized. A number of questions remain unanswered:
- At what frequency do the various data feeds need to be collected?
- How will data be validated?
- How will the data be protected?
- What are storage requirements?
- What tools are needed for data mining and filtering?
Final Points of Interest
The speed at which microgrids are deployed will to some extent be determined by the technologies that will “glue” the distributed resources together. A lot of work is being done on the individual components, but what is needed quickly is a standardized platform that enables microgrids to be configured reliably, economically and easily so that more advanced control strategies can be developed later.
Some debate the value of microgrids, suggesting that they may introduce unintended consequences and perhaps may not be a cost-effective solution for optimizing around reliability, economics and environmental opportunities; and this may be true until plug-and-play is a reality. Further objective research is needed to evaluate and validate these claims and concerns for both utility-owned and consumer/community-owned microgrids. We don’t have all the answers yet, but we do know that microgrids are needed for many reasons and in one way or another they will be coming soon.
September 23rd, 2013