The Lessons of the Great Tohoku Earthquake and Its Effects on Japan’s Economy (Part 6)

Thoughts on Rolling Blackouts: Redesigning the Power System

January 10, 2012 (Tuesday)

1. Japan’s Power Crisis in 2011

After the Great Tohoku Earthquake of March 11, Japan faces its greatest power system crisis since WWII. The radiation leakage incident resulting from the damage to Fukushima’s nuclear power plants is said to be the second most serious in history, the most serious being Chernobyl. For the first time since it began operations, TEPCO is being forced to implement rolling blackouts in Kanto, and trains and other city facilities are currently paralyzed.

Members of electric utilities, the government, and municipalities have been working tirelessly to avoid the worst possible scenario, for which they have my deepest respect. This is an ongoing problem, and in reality there is much information which remains unclear. However, as a researcher at a think tank, I would like to lay out the current points of debate related to the power problems caused by the recent tragic disaster.

The current power crisis can be divided into two general issues: the nuclear power plants’ radiation problem and the issue of rolling blackouts due to insufficient power supply. Both are important issues that can shake Japan’s power system to its core, but I will refrain from discussing the former here.

As a small aside, Japan’s government had planned to rely on nuclear power for 41.5% of its power by 2020 and 48.7% by 2030.(*1) As Chief Cabinet Secretary Edano is reported to have said on March 18, if by some chance this plan is revised, it will be necessary to massively spread renewable energies over the mid-to-long term.

As I pointed out in a Current Topics article on February 4, in order to increase its national security, Japan should move to a distributed power system with smart grids and renewable energy at its core. However, before Japan could do so, it was confronted by this crisis.

2. Problem of Rolling Blackouts due to Power Supply Shortage

In this article, I would like to discuss the latter of the above issues, i.e., power supply shortage. The rolling blackouts being conducted recently are likely unfamiliar to Japanese people, who should be careful not to confuse them with sudden blackouts, which have occurred frequently in the past during earthquakes and are qualitatively different. In Japan, “blackout” generally means an unexpected stoppage in power distribution due to external forces such as an earthquake or accident. Conversely, in developing countries where power supply and demand is constantly tight, it is not unusual for electric utilities to give notice to users beforehand and stop distribution to a given region at a planned time. (*2)

The fact of the matter is that the recent Great Tohoku Earthquake also caused generalized damage to power facilities resulting in widespread, sudden blackouts, but TEPCO’s efforts have already brought about a full recovery. Figure 1 shows the number of users who suffered blackouts after the earthquake. Initially, 4.05 million users all across Kanto experienced blackouts, but transformer stations, transmission and distribution lines were quickly repaired and all blackouts were resolved by 1:00 AM on March 19. In other words, TEPCO recovered from the blackouts in 178 hours, which compares favorably with the 153 hours taken by Kansai Electric Power Co. (KEPCO) to recover from the Great Hanshin Earthquake. Kanto, which suffered very limited damage compared to Tohoku, has many power supply facilities.

Figure 1: Number of blackouts

Source: created by author based on TEPCO press releases and KEPCO’s website “Great Hanshin Earthquake.”

However, Figure 1 does not include the effects of rolling blackouts. The reason that more than 10 million households have to undergo rolling blackouts is because supply capacity is severely lacking (Figure 2). During the Great Tohoku Earthquake, Fukushima nuclear power plants 1 and 2 (7 reactors, 6.4 GW) and also thermal power plants in Hirono (2 reactors, 1.6 GW), Hitachinaka (1 reactor, 1 GW), Kashima (4 reactors, 3.2 GW) all stopped operating, and the following day saw a total reduction in supply capacity of 14 GW. TEPCO is the largest electric utility in Japan, boasting facilities capable of generating 64.4 GW, but its supply capacity dropped to between 31-37 GW (*3), thus rendering it unable to supply the 41 GW necessary to meet weekday peak demand.

Figure 2: TEPCO’s supply capacity and expected demand

Source: created by author based on TEPCO press releases. Blackout numbers were extracted from various news reports.

The Great Hanshin Earthquake also stopped the operations of 10 thermal power facilities containing 12 units owned by KEPCO, resulting in a drop in supply of 1.76 GW. However, four units, e.g., Himeji 2nd, restarted operations on the same day, and within ten days of the earthquake all twelve units were up and running, thus obviating the need for rolling blackouts. While capacity did decline temporarily, it was not enough to affect the system as a whole, and these facilities recovered in parallel with other power plants. Why then have we seen such a large supply deficit this time, and why has it not been corrected even after more than half a month? The reason lies in the locations of the power plants owned by TEPCO.

3. Uneven Distribution and Central Management of Large-scale Power Plants

The single greatest reason that rolling blackouts are necessary is because large-scale power plants are unevenly distributed along Japan’s pacific coast from Fukushima to Ibaraki. The lower bar in Figure 3 shows the energy source mixture of TEPCO. Thermal accounts for 59.3% and nuclear for 26.9% of the 64.40 GW of power output—very standard ratios for Japan. The problem is the top bar, which shows the cumulative lost power output for facilities damaged in the earthquake and still out of commission as of March 27. All these facilities are situated along the pacific coast of Fukushima and Ibaraki Prefectures. (*4) This suggests that the recent earthquake, and especially tsunamis, were extremely destructive. The problem, therefore, is that 28.4% of power plants in Japan are located along this coast. (*5) Among these, Fukushima 1 and 2 and Hirono in particular are aligned north-to-south over a range of 40 km.

Figure 3: TEPCO power source mix (power output base)

Source: created by author based on TEPCO website

TEPCO is not the only one with unevenly distributed large-scale power plants. KEPCO relies on three nuclear power plants, all of which are situated along Wakasa Bay in Fukui prefecture, for 9.77 GW (almost 30%) of its 34.87 GW power output. Japan’s nuclear power plants rely on sea water for cooling and so must be situated along the coast, but concentrating them into specific sparsely populated areas only causes greater risk. (*6)

Because Japanese electric utilities depend on overseas sources for most of their energy sources including uranium, they have attempted to disperse risk by diversifying their supply sources under the term “power source mix.” However, it is very difficult to say that the same risk dispersion was carried out from the perspective of “location mix.” Those in the power business may make excuses saying the size of the recent tsunami was “unpredictable,” but with that many plants gathered in one location, this large-scale power shortage was entirely predictable.

Why, then, are we so attached to large-scale power plants? It is because they are more efficient in terms of operation. That large-scale plants have a lower electric power unit price should be easily intuited. Furthermore, it is easier to create a power transmission network and manage the power system as a whole if one’s power plants are limited in number.

Nuclear power plants face particularly strong-rooted opposition from residents of possible locations, and therefore in accordance with Japan’s policy large-scale plants were built in sparsely populated areas and their power sent through the power grid—this was made out to be an ideal power system. It was also explained that if efficiency could be attained through such a centrally managed system, then ultimately this would be reflected in the price of electricity and consumers would reap the benefits.

Figure 4: Power by provider

Source: METI “Total Power Demand Initial Report”

The electric utilities’ explanation for opposing liberalization is consistently that stable power supply is paramount. They assert that power and liberalization don’t fit together; even if Japan’s power is a bit more expensive than that of other countries, it is essential that one utility be responsible from generation to transmission to distribution in order to continuously supply power without interruption. Indeed, stable supply is an obligation of electric utilities imposed by the Electricity Business Act—a task no less important than efficiency or global warming countermeasures. However, this time TEPCO was unable to fulfill this obligation, which, no matter how favorably one tries to spin it, is not something that can be waved away with an “it was unpredictable.” Were protective industry establishments such as regional monopolies and vertical integration really indispensible to fulfilling the mandate of stable supply, or were they not in fact preventing flexible supply and demand throughout the market? An objective verification is required.

5. Smart Grids as Autonomous Distributed Open Systems

To summarize the above discussion, we have arrived at the understanding that the recent power outages were caused by the risk of unevenly distributing large-scale power plants and the lack of flexibility in the regional monopolistic system. Japan’s power system is characterized by these problems and as such I have called it a centrally-managed closed system. When the myth of nuclear power’s safety is added on top of these problems, the biggest assumption of Japan’s post-war power policy groans and teeters precariously. For those like me who have criticized this policy for years, it was a harsh reality to be told “it was unpredictable.”

However, simply thinking along the same tracks to find countermeasures now that the existing power system has been completely refuted will do us no good. If, for example, we build even larger-scale power plants because several large-scale plants were destroyed, or we build even taller sea walls because the tsunamis were higher than predicted, we will certainly regret it when a new “unpredictable” disaster strikes. We must question what has until now been common sense and devise an entirely new system. The resulting conclusion is an autonomous distributed open system.

First of all, Japan must switch from large-scale to small-scale power plants distributed evenly across all regions, in which case, hydro and thermal will become viable power sources. The greatest hope for the future, however, lies in renewable energies such as solar, wind, and geothermal, which are being actively pursued by various countries in Europe but have yet to exceed 1% of Japan’s power production (Figure 5). Looking at TEPCO alone, “New Energy” does not make up even 0.01%. In general, the high cost of renewable energy hinders its spread, but if we take into account the damages due to the recent radiation incident and power outages and recalculate nuclear power’s cost, the superiority of the centrally-managed system decreases greatly.

Furthermore, evenly distributing power plants would not only distribute locational risk but the burden on citizens as well. It is well-known that residents of Fukushima prefecture have suffered more damage from the radiation incident than from the earthquake itself. Wind power, a renewable energy, faces locational opposition due to its detrimental effects on landscape aesthetics and the noise it produces, but it is important to distribute that burden (with appropriate compensation) for the sake of creating the power essential to all citizens’ daily lives.

Figure 5: Power source mixture of various countries (2009; power generation base)

Source: created by author based on IEA, Montly Electricity Statistics, 2010. China data is from the IEA website, dated 2007.

Secondly, Japan must do away with regional monopolies, connect the entire country’s power grids with each other, and integrate their operation; using this as a neutral infrastructure, power generation and retail should be unbundled from transmission. Instead of chopping Japan’s small land mass up into ten sections with no real give and take between them, Japan should instead create a flexible and accommodating common power infrastructure and reform its regulations so that power generators and power sellers can use it freely.

These two points provide the framework of an autonomous distributed open power system, but in reality controlling a distributed power grid is easier said than done. Just as we see TEPCO hard pressed to predict power demand in order to enact its rolling blackouts, so too have major countries struggled with their power systems: without expecting anything from the wanton, power-using masses, electric utilities have prepared a surplus of supply capacity with their large-scale plants and unilaterally judged the amount of power to supply in order to meet demand. Additionally, renewable energy is affected by natural phenomena and thus predicting its supply is difficult.

The key to realizing an autonomous distributed open power system lies in the talk of the town, i.e., smart grids. These borrow the power of both market mechanisms and IT to optimize supply-demand adjustment, including unstable power sources such as solar and wind power generators, which continue to increase in number. Some ideas under consideration are: introducing real-time billing to give incentive to users to cooperate with supply-demand adjustment, or equipping smart meters so that families can easily save electricity. (*8) Further, the rapidly spreading electric car is expected to take on the role of storing electricity in it storage battery, which will help exponentially increase the efficiency of supply-demand adjustment.

It is true that at present smart grids do not yet exist and Japan’s existing system is relatively complete, and therefore electric utilities are reluctant. However, some European companies are making advances in practical business and policy towards such systems. In Europe, power has been liberalized and the transmission network has been unbundled and set up as a public monopoly, thus allowing linkage lines to be laid down not only between regions but even across national borders. Power is traded internationally, not only in emergencies but every day. Meanwhile, Europe’s longstanding care for the environment has led to advanced introduction of renewable energy (Figure 5), but flexible market-based trading of power helps absorb instability of supply and therefore large-scale rolling blackouts are unheard of. In order to use IT to effect further advances, substantiative experimentation on smart grids is being actively pursued.

6. Redesigning the Power System

Even now, the rolling blackouts continue. Mid-April brought warmer weather and a temporary reprieve, but demand will reach its annual peak during the summer months and large-scale outages will again become unavoidable. The entire country is employing countermeasures to minimize the extent of the power outages: TEPCO plans to restart operations at stopped plants and build new LNG thermal power plants; the government is debating the possibility of adopting daylight saving time and distributed summer vacation; and companies are reworking their production plans. These are all important but they are short-term measures. At the same time, we have to look at the long term and redesign Japan’s power system.

The key is an autonomous distributed open power system. Now that the myth of stable supply has crumbled away, proposing liberalization and distribution may seem like a backwards step to those power experts who have opposed such movements in the past. However, the real danger lies in a lack of discontinuous thinking. The story of how the internet, a spider web-like network for packet transmission, was a research project by the US Department of Defense with the aim of developing a network more robust than the tree-structured phone network is a very famous one. We tend to think that a centrally-managed closed system is efficient and highly stable, but there are times when this is not the case.

Even if the entire country worked together, a smart grid is not a power system that could be completed in a year or two. It would take 10 to 20 years for renewable energies to make inroads and grow to tens of times their current level. Electric cars, which can be used to store electricity, have as yet only sold around 10,000 units. Looking at the current rolling blackouts, the existing nuclear power plants can hardly be stopped tomorrow, and even if Japan were to choose to denuclearize, it would take many decades to complete. That is exactly why we must now look at the big picture and begin discussing redesigning the power infrastructure for the sake of our children and grandchildren.

Obviously, to do this we must gather the wisdom and knowledge of experts. However, the recent power crisis has caused electric utilities’ engineers, professors on nuclear power, and heads of regulatory bodies to forfeit their authority. When computer scientists brought about the dawning of the internet, large communications companies and communications engineers showed absolutely no interest. I hope, however, that even if people coming from different fields and backgrounds lead the necessary structural reforms, they can hold constructive discussions with the current power experts to address the power crisis.

Notes

1) Agency for Natural Resources and Energy, “Long-term Energy Supply-Demand Forecast (recalculation).”

2) In Japan, too, notices of temporary power stoppages due to construction sometimes arrive by post; these are one type of rolling blackout.

3) Supplying power requires power reserves, and not all power sources included in the 64.4 GW were necessarily operating at the time, e.g., the Kashiwazakikariwa plant was stopped for inspection.

4) At first, the power plants at Chiba and Goi were stopped due to damage, but they were brought back online within 24 hours. Additionally, the Ōi power plant was back up and running on March 17, and the Higashiōgishima plant on March 24.

5) In a press conference on March 19, TEPCO Vice President Takashi Fujimoto said that by restarting operation of the Kashima thermal power plant damaged in the disaster before the end of April, TEPCO would be able to increase its power generation capacity by 20% of current levels, up to 42 GW, but that even so it would not be able to meet demand at the peak of summer.

6) Japan currently has a total of 54 nuclear reactors at 10 locations, of which Fukushima 1 and 2 hold 10 reactors. If we add plants under construction or about to begin construction (Namie-Odaka nuclear power plant), the total climbs to 13 reactors in and around Fukushima 1 and 2 which generate as much as 12.7 GW.

7) TEPCO receives support power from the Shin Shinano (600 MW), the Sakuma (300 MW), and the Higashi-Shimizu (100 MW) frequency converter stations, and the Kitahon HVDC Link (600 MW).

8) The target areas of the recent rolling blackouts were decided comprehensively, and many criticized the inclusion of refugee centers and hospitals in those areas. If smart meters had been equipped, it would have been possible to constrain power outages on a building-by-building basis.