Putting our heads together

In the long run, access to knowledge, material data and practices for a broad spectrum of components and technologies will be needed to support future decommissioning of nuclear installations.

The Institute for Energy (IE) has developed a method for consolidation of nuclear knowledge. The method relies on the mobilisation of identified experts in the European Union to re-evaluate old knowledge and consolidate the necessary information to produce training materials for future generations of nuclear engineers.

This method was applied to a pilot study for consolidating and preserving VVER reactor pressure vessel (RPV) safety related knowledge, scattered in many countries and in different languages, is facing serious risks of being lost. The experience gained and lessons learned from the first exercise will be presented in this article. This initiative could be the start of a wider nuclear knowledge preservation and consolidation activity.

As the first generation of senior nuclear experts are now retiring, there is an increased risk that the nuclear community will lose valuable knowledge and capability. To avoid such a loss in the EU, action should be taken now to preserve and disseminate nuclear knowledge to the new generation of nuclear workers.

This conclusion is supported by various national and international organisations including the European Commission [1-3], the International Atomic Energy Agency [4,5], the Working Party on Nuclear Safety [6], the Council of the European Union [7,8], the European Atomic Energy Community [9], the OECD Nuclear Energy Agency [10], the European Nuclear Energy Forum [11] and IBM’s Nuclear Power Advisory Council [12].

One of today’s activities at the Institute for Energy concerns data management and dissemination in the area of nuclear safety. An online data & information network (ODIN) has been set up that maintains one document database (DoMa) and four engineering databases [Fig 2]. These systems aim to deploy networks for energy-related research & development, specifically for nuclear energy and to provide the public experimental data of European projects on mechanical and thermo-physical material properties relative to international standards and recommendations (see also [18]). They enable co-operation and technology transfer with member states with small nuclear programmes.

ODIN manages six nuclear databases, which have restricted access.

They cover:

• data on high temperature reactor (HTR) fuel elements

• HTR graphite element data

• data on the safety of Eastern European type nuclear facilities

• information on current research reactor safety assessment approaches

• data on hydrogen incidents and accidents;

• information on long-term radioactive waste management.

The web-enabled document management system DoMa is designed to enhance the dissemination of information amongst R&D community sectors. Any type of electronic record (document, spreadsheet, graphic, etc.) can be stored in the system, which combines open access to general information (title, author, abstract, etc.) about a particular record with controlled access to the actual files.

New Method

The Institute for Energy has developed a methodology for consolidation of nuclear knowledge [Fig 1]. The method relies on leading experts re-evaluating old knowledge and consolidating it to produce training and educational material for new generations of nuclear engineers.

Fig. 1: The Institute for Energy's general methodology for consolidation of nuclear knowledge. The process would need to be carried out for each important component, material property or process.

These experts were identified by JRC staff, based on their involvement in European nuclear safety networks since the early 1970s. The experts are asked to provide the papers in their possession related to a specific nuclear expert field. Furthermore, they are asked to identify more key experts in that area.

All papers are collected centrally and stored in the DoMa protected database within ODIN. The papers are stored in PDF format and additionally have information about the title, authors, keywords and abstracts stored separately in Microsoft Word for easy searching.

After the identification of some possible key reviewers amongst the expert group, the subject is subdivided, in order to reduce the heavy workload of review, summary and preliminary consolidation. After reading their subfield’s papers, the reviewers prepare a report containing short summaries and references of the individual papers, preliminary conclusions and preliminary ‘open issues’ – subjects that need further study in order to become commonly accepted. The report thus contains not only information from the papers, but also tacit knowledge from the experts’ own experience.

Then, the report is sent to all experts participating in an upcoming consolidation workshop. At the workshop the key reviewers present their review reports. They emphasise proposing commonly-accepted conclusions (referred to as ‘consolidated’ conclusions) and open issues in the subfield. In parallel they present the most relevant experimental data plots of research results as part of a data-validation exercise (however a complete experimental data validation process is out of the scope of this project). The experts discuss the material. A chairman leads the discussion toward a distillation of the key elements of the knowledge in each particular subfield with as much consensus as possible. Disagreements are logged as open issues. At the end of the workshop, recommendations are made to lead to further consolidation efforts in certain subfields or to a final consolidation document in others.

An additional important item in this process is the identification of consolidated open issues, which are areas where further research may be necessary in order to arrive to a commonly agreed result. They would complement the goal of a final report of the state of the art in the specific expert area.

The pilot study

Much of the information and knowledge about VVER reactor pressure vessel (RPV) embrittlement is at risk of being lost, because of:

• retirement of senior experts present when most VVER nuclear power plants were designed and put into operation

• generation gap between existing staff and new entrants today, caused by decline in new nuclear construction in the intervening years

• previous print-based publishing

• limited dissemination possibilities

• language of publication (non-English publications from Eastern Europe)

Therefore, the Institute for Energy and several key experts performed a pilot study using the previously described methodology for consolidation of VVER RPV embrittlement knowledge. To make the review process easier, several experts in RPV embrittlement sub-fields were proposed and the papers were allocated accordingly.

Eight reviewers received between seven and 21 papers in their field of expertise. A dedicated workshop was held 11-13 December 2007. In a brainstorming session at the workshop, the field was redefined as in Table 1. Although more than 500 papers had been collected already, it was decided to start with a limited number as a voluntary pilot study. The reports and presentations were requested to follow a standard structure for each paper that included: paper title, author(s), reference; reviewer’s summary/abstract; comments on the material’s current relevance.

In addition, they were requested to write a conclusion to include:

• mention of missing reference papers in the area known to the reviewer

• consolidated conclusion proposal

• open issues in the area known to the reviewer.

There are three scopes of work in this exercise, depending on the time-span available. In the short term, the goal of the exercise is to reach consolidated conclusions in a workshop setting following discussion of individual reviewers’ presentations. This short-term perspective is based mainly on the limited amount of papers reviewed and the tacit knowledge of the experts. It can be managed within one year. In the medium term, the goal of the exercise is a consolidated review of the individual expert fields. This would contain all available papers allocated to the sub-field. It has a considerable impact on resources, as many papers have to be reviewed, summarised,

referenced and compared before reaching consolidated conclusions in the field. In the present pilot study this is estimated to take around three years. In the long term, the goal of the exercise is to prepare a report of the state of the art for the entire field of VVER RPV embrittlement, including its history, and explanation of choices made in the past (of RPV material, RPV composition, and so on). The last such general document was produced in 1981 by Alekseenko, Amaev, Gorynin and Nikolaev [13]; it is in Russian and needs updating. This complete state of the art report could be available as a book in about five years’ time.

Lessons learned

It is evident that we cannot solve the problems caused by a structural shortage of nuclear experts just by starting a project. Also, initiating such a project cannot prevent the experts retiring with their specialist knowledge. The key problem is the effect of these developments: a shortage of human resources qualified to do the work that needs to be done. This shortage causes difficulties everywhere in the field, and it will also make it even more difficult to collect the knowledge of the retiring experts in a complete and systematic way.

However, the methodology described above has proven to be a step in the right direction. The experts themselves, most of whom have been working in the field since the beginning of the nuclear era, are proud of their work. They contributed in a very idealistic and positive way to this first circle of knowledge consolidation. Some even did the reviewing work in their spare time at home. The atmosphere during the workshops was relaxed and constructive, as were the discussions on open issues. The outcome was preserved in two summary records [14,15], which were the basis for the third and final workshop held in September 2009. The two consolidation workshops have reviewed nearly the full set of papers available. It was interesting to notice that the experts did agree on their consolidated conclusions and on open issues on the basis of a limited number of papers. It was clear that the experts drew on their entire range of knowledge and experience, and not just the information contained in the limited range of papers under discussion. This may be a very powerful tool in order to save time in the consolidation process. It will be interesting to see whether the consolidated conclusions and/or open issues varied between the workshops.

A further advantage of this consolidation methodology is that its summary reports can be published openly, with reference to all sourced papers, without violating intellectual property rights (IPR). Therefore, wide, free-of-charge dissemination to the interested public is guaranteed. It seems logical to continue applying this consolidation methodology to other fields of possible nuclear knowledge loss –  not only for materials, but also for technologies, components, systems and so on.

Some key aspects learnt from the pilot study were:

• The consolidation methodology proves to be efficient

• The participation of the experts to the consolidation is excellent

• Unified keywords are essential to trace the information needed

• The consolidated sub-field summaries give a good general overview of results, open issues and key references without violating IPR.

Ulrik von Estorff, Institute for Energy, Joint Research Centre of the European Commission, Postbus 2, NL 1755 ZG Petten, The Netherlands. Milan Brumovsky, Nuclear Research Institute Rez plc, Husinec - Rez 130, 25068 REZ, Czech Republic.