A trail-blazer in battery-based energy management
"My goal is to enable batteries city-wide to be bundled together and run like a power plant"
Koji Kudo received a Minister of Economy, Trade and Industry Award at CEATEC JAPAN 2014, Asia's largest state-of-the-art IT and electronics comprehensive exhibition. He is the leader of a project that created the world's first technology for enabling a variety of batteries spread across a city to be bundled together using ICT, and charged and discharged in a timely and stable manner as a massive, virtual large-capacity battery. Here he speaks passionately about why he made the switch from being a semiconductor researcher, and talks about his reasons for tackling technology he had no knowledge or experience of. He also discusses the determination that helped keep this project moving forward.
Bundling batteries to create a virtual large-capacity battery in cities
--Can you give us a basic overview of the world's first Real-time Demand Response Technology Using Consumers' Distributed Batteries?
Kudo: Our technology enables the balancing of power supply and demand for society as a whole by bundling together batteries distributed across a city, and controlling them as a single massive battery via the cloud. An increasing range of batteries are being used in society, such as those installed in residential homes, office towers, and commercial facilities, as well as the batteries used in electric vehicles. This technology coordinates these batteries, and utilizes them effectively by treating them as a single large battery for an entire community. In principle, when electricity is used the supply and demand must always match. Consequently, it is necessary to ascertain the status of electrical supply and demand across all of society in real time. At the same time, commands must be issued according to the status of each battery. Our technology can help fine-tune the regulation of supply and demand for the power system networks of power companies by ascertaining the status of these elements and controlling them in real time.
--I guess you could say the concept is similar to grid computing, which combines small PCs that individually have low performance, making them function as a virtual large-scale computer.
Kudo: That's right. Recently, there has been a great deal of interest in renewable energy like solar power generation. Although this enables the effective utilization of natural energy, the amount of power generated fluctuates based on factors such as the weather. This is thought to make balancing the supply and demand of electricity more difficult when a large amount is introduced.
Power plants have two important roles. The first is to create electricity, and the second is to deliver that electricity to customers stably on an ongoing basis while maintaining quality. To fulfill this latter role, power plants constantly adjust the balance of electrical supply and demand. Our technology shows promise for reinforcing the ability of power plants to balance electrical supply and demand with regard to the introduction of unstable renewable energy. As a result, this technology also improves convenience for battery users.
Our Real-time Demand Response Technology Using Distributed Batteries creates value in two ways. In addition to bundling smaller batteries together so they can be used as a large-capacity battery, it also contributes to the high-quality, stable, and continual supply of power through use in balancing supply and demand.
--What were the goals behind the development of this new technology?
Kudo: Put simply, the goal was to help build a society that can co-exist with renewable energy. NEC has focused on finding new uses for energy through the development and supply of on-board batteries for cars, as well as batteries for residential homes and office buildings. But in the past, the main purpose of batteries was for users to consume power they have saved up themselves for their own convenience (personal consumption), such as peak shaving and using power saved up during the night during the day, or using batteries as a backup power supply in a blackout.
The main objective of our research was to make vast changes to how batteries are appraised, and raise their social value. Configuring a large battery virtually by bundling many smaller batteries together, enabling a large capacity, high-quality power supply, contributes to the stable supply of power.
It is particularly important to achieve the timely and stable charge and discharge of batteries in real time. This technology shows its strength as a new method for enabling the stable supply of power with regard to renewable energy sources that generate a dramatically fluctuating amount of power based on the weather, such as solar power and wind power. We felt a system that creates benefits for the world at large as well as battery users could be achieved by increasing the social value of batteries, leading to the expansion of renewable energy.
Real-time synchronous control of vast numbers of consumer batteries
--Can you tell us why technology like this was not implemented up until now?
Kudo: One reason is how difficult it is to process information without communication delays and synchronously control the charge and discharge timing of vast numbers of consumer batteries together in real time. It has been said that control of such a large number of units would be difficult even for the large batteries run by power companies.
The other challenge is overcoming the variations in status of a range of different batteries. There are differences in how batteries are used by each user, in their storage status, and in their charge and discharge output.
It is really difficult to carry out the centralized management of each battery, and completely synchronize and control the charge and discharge timing while allocating the optimum amount of power in a timely manner, balancing them as a whole and taking all these varying statuses into account. I think that's why nobody even considered a system like this would be possible before.
--What technology was the most key in making the impossible possible?
Kudo: Battery state adaptive control software, and a hierarchical hybrid control system. These are complex terms, but they were the core technologies in achieving Real-time Demand Response Using Distributed Batteries.
Battery state adaptive control software incorporates a number of control algorithms that NEC developed in-house. It is installed on a cloud-based system, and acts as a brain, gathering status information for each battery over the Internet, and controlling the charge and discharge of each battery optimally.
The hierarchal hybrid control system combines two levels of control: remote control of batteries from the cloud side, and local control of charge and discharge on the battery side.
On the cloud side, information such as variance in the storage status and output of each battery is identified, and then groups of batteries are coordinated to assign the overall optimization of charging and discharging to each battery. Meanwhile, on the battery side, the discharge or charge assigned is controlled in second increments using information such as the frequency of the power system network based on information from the cloud.
These two unique technologies were the first to enable real-time charging and discharging on an ongoing basis, based on virtual large-capacity batteries that have been optimized overall.
--Tell us about other technological features.
Kudo: The other major feature is the ability to extend communication intervals for the cloud side and battery side to ten or more minutes. Longer communication intervals enable more data processing to be carried out during that time, making it possible to control more batteries.
Additionally, the continuity of control can be maintained even when a temporary communication failure occurs. Even the most refined control system loses its value as power infrastructure if the whole system shuts down when there is a communication failure. This new technology enables highly reliable power demand response in real time, while also being resistant to network disturbance.
Taking on an unknown field without prior knowledge or experience
--What role did you play in the development of the Real-time Demand Response Technology Using Distributed Batteries?
Kudo: My role in this project was to manage all the technologies necessary for the Real-time Demand Response Technology Using Distributed Batteries, rather than focus on a single one. I was mainly in charge of overall research management, including negotiating with internal divisions and coordinating with external research institutions.
I originally joined NEC as a researcher of compound semiconductor optical devices, and subsequently carried out long-term research and development on optical integrated devices such as semiconductor lasers as light sources for optical communications. Then, I arrived at a major turning point in 2006.
--Why did an expert in semiconductor lasers like yourself decide to make this change?
Kudo: In 2006, a new project was set up at NEC by volunteer members. The goal of this project was to anticipate future global trends in society, while identifying technology that NEC should be pursuing thought to be required by society in the coming years.
So researchers from a range of specialist areas came together to hold discussions and propose ideas based on a myriad of themes. For this project our team selected the theme of energy consensus C&C, which could be considered the starting point of the technology we developed.
I felt that energy management would be essential in future society, so I stepped away from my position as a researcher of optical integrated devices, and made a bid to launch this project for real. In the beginning, there were only three project members, including me.
--Tell us about the sequence of events that led to your research becoming this Real-time Demand Response Technology Using Distributed Batteries.
Kudo: Initially we focused on the theme of computerizing energy, but because I had no knowledge or experience in power, we were literally starting from square one.
So we proceeded with the project through repeated trial and error, setting a course toward joint research by dispatching NEC research staff to the laboratory of a university professor who is an authority on power systems, and holding discussions with a variety of NEC's internal divisions.
Then, when we arrived at the concept of bundling large numbers of batteries together virtually to charge and discharge them in 2009, our proposal was adopted for the national project promoted by NEDO (New Energy and Industrial Technology Development Organization), which carries out testing in the U.S. state of New Mexico. Following achievements such as this, the number of people approving of our research and divisions willing to provide us with research funds gradually grew, enabling us to continue development.
The Great East Japan Earthquake also hit in 2011, and the subsequent blackouts drew attention to the area of demand response for power around the world. We hit complications during our research, but in 2012 we made contact with some important researchers. At a new organization there were experts on apparatus control and batteries for satellites, and by having discussions with them and picking their brains, the framework for our core hierarchal hybrid control system technology came together.