For China’s nuclear industry, 2016 has been a frustrating year. So far, construction has started on only one new plant, and its target of bringing 58 gigawatts of nuclear capacity in service by 2020 seems impossible to meet, writes Steve Thomas, professor of Energy Policy at the University of Greenwich in London. Article courtesy of China Dialogue.
At present, China has 19.3 gigawatts of nuclear supply under construction and a further 31.4 gigawatts already in service. Given that new plants take five years or more to build, the country faces a shortfall of more than seven gigawatts on its target.
In 2015, nuclear power accounted for only 3% of China’s electricity and at any plausible rate of building nuclear plants, it is unlikely that nuclear would achieve more than 10% of China’s electricity supply
All the plants started between 2008 and 2010 are online except for six imported reactors. These include four AP1000 reactors designed by Westinghouse, based in the USA but owned by Toshiba of Japan; and two European Pressurised Reactors (EPR), developed by Areva, a French multinational group specialising in nuclear power.
The plants are not expected to be completed before 2017 and all will be at least three years late, an unprecedented delay in China’s nuclear history. It would be surprising if China was not disillusioned with its suppliers and their technologies.
Technology problems
The EPR and AP1000 reactors have been problematic to build. The two EPRs are 3-4 years late although there is little available information detailing why. Meanwhile, EPR plants in Finland and France, which should have been completed in 2009 and 2012, respectively, will not be online before 2018.
There are no obvious problems that account for the majority of the delays at any of the sites, just a series of quality and planning issues that suggest the complexity of the design makes it difficult to build.
The four AP1000s are also running 3-4 years late. They are being built by China’s State Nuclear Power Technology Company (SNPTC), which has not built reactors before. There is some publicly available information about the problems suffered in China with the AP1000s, including continual design changes by Westinghouse. The reactor coolant pumps and the squib valves, which are essential to prevent accidents, have been particularly problematic, for example.
Still, China is expected to be the first country to complete construction of AP1000 and EPR designs, a scenario it did not expect or want. The government is required to develop and demonstrate test procedures for bringing the plants into service, which could take up to a year. These test procedures are developed by vendors and generally standardised although national safety regulators must approve them and can add specific requirements.
China is expected to be the first country to complete construction of AP1000 and EPR designs, a scenario it did not expect or want
In 2014, a senior official at China’s nuclear safety regulator, the National Nuclear Safety Administration (NNSA) complained that only a small number of test procedures had been developed for the AP1000, and no acceptance criteria had been submitted for review. He said the same issues affect the EPR.
China will likely be reluctant to commit to further AP1000s (and the CAP1400, a Chinese design modified from the AP1000) until the first of the Westinghouse designs is in service, passes its acceptance tests, and demonstrates safe, reliable operation. There are no plans to build additional EPR reactors.
In fact, state-owned China General Nuclear (CGN) and China National Nuclear Corporation (CNNC) opted instead to develop medium-sized reactors (1000 megawatts), the ACP1000 and the ACPR1000, respectively, based on Areva’s much older M310 design rather than the EPR.
Challenging circumstances
The slowdown in electricity demand growth at home has left China with surplus power-generating capacity. Nuclear is now competing against coal plants supplied with cheap fuel. Furthermore, nuclear has a lower priority for dispatch in winter than combined heat and power plants, which warm homes and factories and typically burn coal and gas.
In 2015, nuclear power accounted for only 3% of China’s electricity and at any plausible rate of building nuclear plants, it is unlikely that nuclear would achieve more than 10% of China’s electricity supply.
This year, one reactor (Hongyanhe 3) in Liaoning, operated for only 987 hours in the first quarter of 2016, just 45% of its availability, while reactors in Fujian (Fuqing) and Hainan (Changjiang) were shut down temporarily.
The challenge for the Chinese nuclear industry is to do what no other nuclear industry worldwide has been able to do; to bring the cost of nuclear generation down to levels at which it can compete with other forms of generation, particularly renewables
Another challenge is the strain placed on China’s nuclear regulators in the face of such an ambitious target. The NNSA is under particular pressure to oversee the operation of 36 plants and the construction of 20 plants, as well as being the first regulatory authority to review six new designs. Not even the US Nuclear Regulatory Commission, which monitored standards during the huge build out of the industry in the 1960s and 1970s, has faced such a workload.
Safety authorities are usually reluctant to appear critical of their international peers but in 2014, a senior French safety regulator described NNSA as “overwhelmed”, and claimed that the storage of components was “not at an adequate level”. A senior official from SNPTC said in 2015: “Our fatal weakness is our management standards are not high enough.” To build up the capabilities to support such a large construction programme a pause in ordering new plants and equipment may be necessary.
Uncertain future
The 58GW target of nuclear capacity in service by 2020 is not achievable and, like nuclear capacity targets in the past in China and elsewhere, it will be quietly revised down. The challenge for the Chinese nuclear industry is to do what no other nuclear industry worldwide has been able to do; to bring the cost of nuclear generation down to levels at which it can compete with other forms of generation, particularly renewables.
If the world nuclear market does not pick up soon, the Chinese government may decide to put its formidable resources behind other technologies that would develop influence with less economic risk
If it is unable to do this, China cannot afford to carry on ordering nuclear plants and nuclear will retain a small proportion of the electricity mix.
This leaves China’s nuclear export drive in a precarious position. Since 2013, China has turned its attention to nuclear export markets, offering apparently strong advantages over its competitors. The Chinese government can call on all the resources of China to offer a package of equipment, construction expertise, finance and training that none of its rivals, even Russia, can match.
Unlike its competitors, it also has a huge amount of recent construction experience allowing it to supply cheap, good quality equipment. Its attempt to build reactors in the UK is an important element to this strategy; convincing an experienced user of nuclear power that a Chinese plant is worth investing in is a strong endorsement of their technology.
Despite these advantages China has had little export success so far. In part, this is because there are fewer markets open to new nuclear. Such markets are typically found in developing countries where the financial risks are greater, and where governments have tried and failed to launch nuclear power programmes themselves.
It seems clear there is a political element to the Chinese nuclear export strategy, which is to gain influence and leverage in the importing countries. However, if the world nuclear market does not pick up soon, the Chinese government may decide to put its formidable resources behind other technologies that would develop influence with less economic risk. If China’s nuclear home market is not flourishing, this decision will be much easier.
Editor’s Note
This article was first published on China Dialogue and is republished here under a Creative Commons licence and with permission from the author.
[adrotate group=”9″]
Dennis Sweitzer says
Here’s an interesting paper (though technical):
The authors analyze the historical costs of 53 technologies to generalize Moore law for the rate at which costs decline with deployment.
One interesting thing: Nuclear power is the only technology that they examined that was increasing in cost with deployment.
One can only speculate about what is fundamentally different about nuclear power that such a trend exists.
http://www.sciencedirect.com/…/pii/S0048733315001699
Bas says
Why nuclear power costs increase while costs of other technology decrease when deployed more and longer?
I estimat that the main reasons are:
1. The ignorance regarding safety in the first generation of nuclear power plants. Nuclear scientists & engineers assurances that they were safe, were based on childish like considerations. They simply ignored the most obvious possibilities, such as a terror attack etc.
So when accidents showed how dangerous they were, safety had to be improved again and again. Increasing the costs again and again.
2. Also due to the problems stated in previous paragraph, nuclear never reached good economy-of-scale.
Construction of new Power Plants implied also new designs in order to implement the learned lessons….
Dennis Sweitzer says
I think you’re essentially correct.
The mathematics behind learning curves is that basically: (1) One starts making a product meeting a set of needs, within a set of parameters; (2) As one makes the product, people figure out ways of making it better/cheaper/faster while meeting the needs.
One can argue that most nuclear plants are custom built, so each is a different, hence economies of mass production can never occur. Or that we simply don’t know how to make a nuclear power plant that meets all of the desired specifications — so we’ve never reached step (1).
Some of the new designs (such as small modular liquid salt nuclear reactors) might over come these issues, but there are just so many complications with nuclear reactors (i.e., spent fuel, gird dispatchability, etc) that they may never get there.