Looking at past nuclear construction projects, eg, in Finland, what are the factors that contributed to success?

Developer/owner/operator’s strong ownership of the project

From the outset, the utility/developer/ owner/operator has a strong management organisation that closely monitors and assesses design, manufacturing and construction works, even in the case of turnkey projects. This inspires good performance from the vendor.

In addition there are direct and open lines of communication between developer/ owner/operator and the regulatory body. Formal communication is supported by informal contacts to ensure correct mutual understanding.

Participants share the common goal of finding acceptable technical solutions to problems encountered during construction.

Vendor takes full responsibility for its work

The vendor and developer/owner/operator cooperate constructively, emphasising their joint interest in building a safe and reliable plant to the planned schedule and meeting quality targets.

The vendor has a strong management system for assuring quality of its own work and of the work of all subcontractors.

The vendor demonstrates its sense of responsibility by correcting swiftly and without hesitation identified non-conformances. Systematic information transfer is established within and between organisations involved in the project implementation.
Adequacy of information transfer is monitored and confirmed by regular audits.

Experience and professional competence of all parties

Many of those responsible for design and manufacturing have a solid background in research and development work, and have contributed to production of technical codes and standards and are familiar with the background to the requirements.

Experience of project managers often extends to construction of more than one nuclear plant.

Projects are managed by teams with adequate knowledge of nuclear technologies and good internal communication, permitting information exchange between different technical discipline.

Positive business environment

Looking back to the 1970s, when most of the currently operating fleet was constructed, the nuclear industry was relatively strong. The power plant vendors had plenty of orders, the future looked bright. Each vendor country had domestic manufacturing capacity that was able to supply most of the components. The vendors themselves had extensive and experienced engineering departments and comprehensive in-house capabilities in design and manufacturing.

Examples of success

Loviisa 1 and 2

Loviisa was the first nuclear power plant in Finland, and prior to this project a national nuclear infrastructure was almost non-existent.

At start of the project the utility had only a handful of experts with nuclear experience, acquired via a Triga Mark II reactor and/or the carrying out of reactor and thermal physics calculations.

On the other hand the utility had a strong engineering department with previous experience from design and construction of hydro and fossil-fired power plants.

The nuclear regulatory body was established inside the existing radiation protection institute, STUK, starting from one person in 1968 and achieving a level of about 50 professionals in 1975.

The Finnish technical research centre, VTT, had, from the late 1960s onwards, know-how in materials research and testing, safety analysis, reactor physics, reliability calculations, and nuclear plant systems simulation. VTT developed sufficient competence to verify independently the safety relevant information received from the vendor. This was a necessary condition set by STUK for an operating licence.

The utility (licensee) was responsible for project management and took ownership of safety and quality assurance.

In addition to general project management, the utility was also responsible for parts of the design, equipment purchase, and site works.

Utility staff designed electrical systems and civil structures, notably the ice condenser containment, under the licence and guidance of Westinghouse.

Electrical, I&C, and ventilation equipment, plus certain auxiliaries, eg, the seawater cooling system, were purchased from separate contractors.

Inspections of nuclear fuel and mechanical equipment were conducted at manufacturers before shipping to site.

Civil construction and the installation of electrical, I&C, and ventilation systems were conducted under direct supervision of utility experts.

Prior to start-up, the utility developed the skills needed to operate and maintain the plant (independent of vendor support), including reactor core calculations, in-service inspections, and water chemistry.

Atomenergoexport (AEE) of Russia, as main supplier, listened to requests from the customer and was committed to meeting Finnish nuclear regulatory and quality requirements.

AEE provided most process systems and mechanical equipment: reactor and fuel; primary circuit; turbine plant; and most auxiliary systems.

The plant was designed to be very different from the original reference plant, with the objective of meeting US requirements set out in ten CFR 50, Appendixes A and B.

Significantly, departures from the reference plant were implemented in lay-out and all safety, electrical and I&C systems.

The reference design was reproduced in reactor and fuel, reactor vessel and its internals, steam generators, turbine plant (although with modified lay-out), and some auxiliary systems.

The successful outcome of the project and rapid implementation of design changes was much facilitated by the experience, competence, and cooperative attitude of the Soviet experts.

Loviisa 1 started power production about five and a half years after the first concrete casting.

Since start-up, the plant has operated almost without any forced outages for more than 38 years, with an annual load factor of around 90%.

The construction time and the high levels of safety and reliability in operation demonstrate that the project has been able to cope with potential sources of trouble, such as: a major redesign of the layout and safety systems during the early stages of construction; very limited experience and knowledge of nuclear technology initially on the Finnish side; strict regulatory oversight, which was a new experience for the suppliers; and different native languages of key people (Russian, Finnish, German), necessitating interpreters for communication.

Barakah

A more recent example of nuclear construction success is provided by the Barakah project, the first nuclear power plant in an Arabic country and the first export project of the South Korean nuclear industry, under a contract signed in 2009.

Four units are under construction in parallel and are planned to start up at one year intervals, starting in 2017.

The reference plant is Shin Kori 3, which was supposed to start up in 2013 but was delayed by two years, principally by cable quality issues (see below). It was connected to the grid and supplied first electricity in January 2016, with commercial operation expected in May 2016.

The Barakah plant has been built to the planned schedule and it seems that targeted start-up dates will be achieved although several significant changes to enhance safety are being implemented, including upgrades based on lessons learned from Fukushima Daiichi and 911.

Among the factors contributing to success to date are:

  • strong political support from government and determined building of the national nuclear and technology infrastructure;
  • experience collected from the South Korean nuclear new build programme, which has been underway since 1972;
  • continuous development of the nuclear plant design, building on the Combustion Engineering 80+ model;
  • very strong and experienced international project management team working for the utility and cooperating well with the vendor team;
  • experienced international management team and foreign experts working for the regulatory body;
  • successful recruitment of talented local people, coupled with an extensive training programme in the UAE and abroad.

Delays and cost increases

At the other end of the spectrum are nuclear new build projects beset by delays and cost increases. What are the factors contributing to these problems?

First of a kind plant – absence of a detailed design

Example: The contract for Olkiluoto 3 in Finland was signed and a promise of first power in 2009 was given although the design was only at a conceptual stage, and the supplier had never really faced a rigorous regulatory process.

The main contract was signed in December 2003 and the construction licence, accompanied by a large number of conditions, was issued in February 2005. The licensing decision was made in good faith, with the understanding that open issues would be resolved in the early stage of construction.

First concrete for the reactor building bottom plate was poured in late summer 2005 but after that the vendor had no more work drawings.

Readiness to start erection of reactor building walls was not achieved until late spring 2007, when the detailed design had been completed – at that time the construction was already delayed by two years and also after that each concrete casting took more time than scheduled.

First of a kind plant – increase in component sizes

Example: AREVA’s EPR, at 1600MWe net, is larger than any other previous PWR.

Manufacturing of the primary circuit components, main primary coolant pipes, the turbine and the generator has been a challenge.

Manufacturing of some of the large components and large subassemblies failed at the first attempt and re-manufacturing was required.

On a positive note, AREVA demonstrated its safety responsibilities by rejecting failed pieces on its own initiative.

Due to delays for other reasons, failures in manufacturing have not impacted on the main schedule in the case of Olkiluoto 3.

First of a kind plant – new I&C technology

Example: Olkiluoto 3 civil construction works and installation of mechanical components and pipes were completed a couple of years ago but I&C system re-design, testing and installation are delaying the plant start-up, which is currently scheduled for 2018.

Adequate reliability of programmable I&C systems has been difficult to prove in this and many other new build and refurbishment projects.

It seems that I&C system design and qualification are more advanced in some other fields of technology where high levels of safety and reliability are required, eg, car and aviation industries.

Standards have been developed by IEC and IEEE for new nuclear equipment that require design and qualification of new I&C platforms to be conducted in a traceable manner from the very first stages of design because qualification by comprehensive testing is not feasible.

First of a kind plant – new manufacturing technology

Application of new manufacturing technology can enhance the reliability of mechanical components but demonstration of safety generally needs testing and experience.

Example 1: Nozzles for connecting main coolant lines to the EPR reactor vessel are fitted and welded in a new way that improves inspectability of welds, but the first welds required time consuming repairs. However, after the new techniques had been practiced and some experience gained, the quality was much improved.

Example 2: The cylindrical part of 1200MW VVER pressure vessel, ie the section around the reactor core, can now be manufactured in one piece without welds. However, there is a need to demonstrate that material properties over the whole height of the cylinder are at least as good as in the proven former design, in which two shorter cylindrical pieces are joined with a weld at the reactor core height.

First of a kind plant – new types of components

New types of component have been designed especially for passive safety systems. Such components require thorough testing that prove their intended function in all design conditions.

Example: The main coolant pumps of the Westinghouse AP-1000 PWR enhance safety because, unlike currently operating Westinghouse plants, they have no seals that could cause loss of reactor coolant in the event of a total loss of AC power. Demonstration of adequate pump reliability has required extensive testing, resulting in construction delays for the first AP-1000 plants in China (although the problems are now resolved).

New business models in the industry

The nuclear business environment for vendors has changed significantly since the 1970s, especially in Western Europe and North America: some vendors have very small or non-existent domestic markets; export markets are reduced and highly competitive; design and manufacturing are subcontracted globally in long supply chains; management and oversight functions are subcontracted; customer countries request localisation of manufacturing and construction, often to companies that have no nuclear experience and have limited capabilities for achieving quality.

It is not unusual for lower level sub- contractors to make bids without previous nuclear business experience, and without knowledge of inspection practices and nuclear safety culture.

Vendor failing to take onboard customer and regulator expectations

To better manage regulatory risks is necessary that vendors become well acquainted with the relevant national nuclear regulations and safety requirements.

It is also important that the power company planning new construction has itself a clear understanding of the regulators’ expectations and can provide a comprehensive set of national requirements to the vendor.

Furthermore, to avoid regulatory risks the vendors need an opportunity to get direct explanations from the regulator as to the intent of requirements that may not be expressed clearly enough.

It is also important that the vendor takes seriously the regulatory requirements.

Example: It seems that in the Olkiluoto 3 case the vendor did not believe that the regulations were seriously enforced. This caused problems especially in the area of I&C design.

Inexperienced nuclear plant customer or regulator

In some projects an inexperienced nuclear plant customer or regulator has enforced strict formal requirements that actually do not influence the safety or reliability of the plant but only cause much work and delay.

Especially in countries that are planning to build their first nuclear power plant it is not unusual for the developer/power company and the regulatory body to have limited knowledge, with no capability for proper safety assessment.

In such situations it is necessary to allow time for training.

Example: In Vietnam the start of nuclear power plant construction has been postponed by several years to establish first a proper legislative and regulatory framework and to train the national experts in nuclear technology and safety.

False economy of cheap components

There are examples of misjudgments where an initially cheaper component has turned out to be very expensive due to failure to meet requirements and consequent negative impact to the project schedule.

Accurate assessment of real capabilities of low cost manufacturers is necessary before making a decision on purchase.

Deterioration of manufacturing know-how

Starting new nuclear build after what amounts to a moratorium of about ten years has turned out to be difficult because many of the manufacturers of nuclear components have moved out of the business and those who remain have lost their most experienced staff to retirement.

Maintaining the quality certificates, quality management systems and manufacturing specifications on paper has been found not to guarantee maintaining of the skills. Examples of elementary manufacturing errors have also been observed at the long established manufacturers.

In considering the capabilities of manufacturers it is necessary to assess the situation at the shop floor level: staffing; shop organisation; available equipment; and recently supplied products.

Counterfeit components

Fraud by some suppliers and manufacturing of counterfeit components is a new concern that requires new approaches to quality assurance.

Example: It was noticed in the autumn of 2013 during construction of the Shin Kori 3 plant that counterfeit cables had been installed in key control systems. The fraudulent supply by a South Korean manufacturer was revealed when the installed control cabling failed to pass flame tests, as well as tests to confirm the cable’s performance in the event of a loss of coolant accident. All cables had to be replaced and the incident (including all necessary investigations) delayed the planned fuel loading by almost two years. Furthermore, investigations conducted at operating plants prompted by the Shin Kori 3 incident showed that safety related control cabling with falsified documentation had been installed at four of KHNP’s reactors: Shin Kori units 1 and 2; and Shin Wolsung units 1 and 2. The regulator ordered KHNP to shut down all four reactors and not to operate them until the cabling had been replaced.

Strategies and solutions

So, based on the foregoing, how can nuclear new build construction risks be minimised? Priorities should include the following:

National safety requirements

National safety requirements or proper references to recognised international or other countries’ requirements need to be clearly specified as part of the licensing and regulatory framework.

Understanding of regulatory practices

Understanding of regulatory practices is essential for successful project implementation.

All parties (vendor, developer, regulator) should be familiar with licensing, regulatory oversight, and inspection practices both
in the vendor country and in the customer country. The vendor should take regulatory approach into account in the project planning.

Early contacts between vendor, customer and regulator

Early contacts between vendor, customer and regulator after signing the contract help to avoid licensing uncertainties.

Safety issues that could create uncertainties in the licensing process need to be identified and thoroughly discussed before applying for a construction licence.

Regular project management meetings and other direct contacts among the three parties at management level are most useful.

Moving the vendor’s project management to the site, and maintaining a continuous presence there, significantly improves shared understanding of current issues.

Adequate decision making power

Adequate decision making power needs to be vested in the vendor’s project management team, and there needs to be clear separation of contractual issues from project management.

Decisions coming from the vendor’s head office that go against a consensus on technical issues reached in project meetings between all parties are detrimental to smooth progress.

Developer/owner/operator’s responsibility for safety

The nuclear licensee (developer/owner/ operator) is responsible for the safety of the plant, and needs to verify during design and construction that it is getting a plant that is reliable and safe to operate.

This requires detailed oversight of the construction and manufacturing, even in for a turnkey project.

The licensee management needs to build and implement a strong quality management system during the construction phase, with the competence and resources to verify that safety and quality requirements are being met, the ability to resolve non-conformances, the authority to require use of proven state- of-the-art technology in manufacturing and construction (not just to accept final products that meet minimum agreed quality requirements) and to ensure the vendor uses only suitably qualified sub-contractors.

Adequacy of resources for construction stage

Resources needed in the vendor and developer/owner/operator/licensee organisations for the construction phase, and the recruitment schedule, must be planned during the licensing preparation and licensing stage.

The developer/owner/operator/licensee and the vendor must have: adequate project management and quality management skills; experience from management of a large construction project; knowledge and experience in all technical areas relevant for nuclear power plant construction and operation, including: civil, mechanical, electrical, I&C, and nuclear technologies (water chemistry, nuclear fuel, reactor physics, thermo-hydraulics, safety analysis).

In addition, the vendor needs to have at its disposal experienced generalists who have broad technical knowledge across the areas relevant to nuclear power plant design and safety ("chief engineers") and experienced designers who have a realistic view of the actual challenges involved in implementation and can set out the requirements clearly to constructors and manufacturers.

Timely completion of design

The importance of timely completion of design and engineering work cannot be overemphasised.

It may not be realistic to require that all details of design, such as choice of all components, shop drawings for manufacturing, and specifications for each construction step are completed before construction start.

However before starting construction it is necessary to have approved design documentation and a construction plan and schedule for at least six months ahead at all times.

Selecting sub-contractors

In selecting sub-contractors for construction, one should not underestimate the importance of proven experience in management of
large projects.

For contracting suppliers and sub-suppliers with no previous experience in the nuclear field, the vendor needs to ensure that all relevant nuclear-specific work practices are clearly brought out in each call for tender. These may include: requirements for design documentation to be provided for approval before manufacturing; oversight (audits, inspections, etc) to be conducted by several different organisations during manufacturing; and expectations regarding safety culture.

If the nuclear specific work practices are not recognised and understood by the sub-contractors at the time of signing the contract, difficulties are to be expected later.

Promotion of safety culture

From the outset and throughout the project strong messages and transparent actions and decisions are needed from the vendor and developer/licensee management to promote safety culture. Safety culture cannot be turned on overnight at the plant start-up.

Managers need to demonstrate their attitude towards safety and quality through: choice of qualified sub-contractors; use of state-of-the-art tools and methods; uncompromising compliance with the agreed requirements; carrying our regular walk downs; and taking into account safety concerns expressed by workers and answering their questions.

Above all it must be emphasised that safety and quality have higher priority than costs and schedule.