It has been recognised over many years that the ‘defence-in-depth’ principle is fundamental to the design of nuclear power reactors and other types of nuclear plant (Table 3.10). The important feature is the requirement that multiple barriers exist against the release of radioactivity to the environment. The defence-in-depth principle is generally assessed using either or both deterministic or probabilistic methods.

Various means of strengthening the defence-in-depth principle are being considered in current generation reactors and indeed implemented, with respect to accident prevention

(Hogberg, 1998). Additional levels of protection have been installed in many European and other reactors worldwide. In particular measures have been taken in a number of countries to improve the capability of existing components to withstand severe accident loads. The main objective is to mitigate the release of radioactive isotopes to the environment, particularly iodine and caesium. These measures have been complemented with the development of severe accident strategy improvements.

Clearly there are economic and technical constraints on back-fitting improvements in existing reactors. There are many types of design in operation and the feasibility of such improvements is design specific. Nevertheless significant improvements have been achieved at acceptable cost. Many of the desired measures have been identified in Periodic Safety Reviews, which are now a common-place regulatory requirement in most countries. They are being introduced within modernisation programmes, which may also be in place for other reasons, e. g. to replace out-of-date systems or instrumentation that has become too costly to maintain. There may be a requirement to improve the older operating plants to a standard commensurate with later models. If this is not achieved, it may be necessary to shut the older plants down.

It has been realised for many years that the defence-in-depth in many of the earlier Russian designed reactors only applies to a much more limited design basis than Western reactors. The safety of VVER and RBMK reactors has been extensively studied in a number of international projects over the last decade. Numerous safety recommendations have been made, including back-fitting of safety systems, etc. Some of these plants are operating in the EU Enlargement Countries, which will be joining the EU over the next few years. There is, therefore, a driver to accelerate the safety improvement process.

Table 3.11. Examples of back-fits on current plants

Availability of additional water-delivery sources Filtered venting

Hydrogen control with ignitors and catalytic recombiners

Improved safety valves

Reinforcement of containment penetrations

Sehgal.

A number of safety improvements have been recommended for the early VVER-440 designs in respect of control of the reactor pressure vessel embrittlement, improved emergency core cooling systems and, improved reliability of residual heat removal systems. Additionally there are recommendations for improved instrumentation and control systems, including the reactor protection and shutdown systems and improved capability of the confinement to limit radioactive releases. Safety improvement programmes are underway to address these concerns.

There were greater drives for immediate safety improvements of RBMK designed reactors in the wake of the Chernobyl accident. Some of the early plants have now been shutdown but a number of safety improvements have been implemented in the newer RBMK reactors still in operation.

For example the Ignalina power plant in Lithuania (an EU Candidate country) has recently undergone international peer review and various short-term safety improvements have been recommended. These relate to control and protection system reliability, the structural integrity of the major primary circuit components and the confinement function, improved emergency operating procedures, and the need to address fire hazards that could impact safety systems.

The large amount of severe accident phenomenological research carried out for Western water reactors has led to various mitigation measures being introduced and back-fits to be implemented (Table 3.11). The safety of current generation plants has been substantially improved by the development of this knowledge base. Containment research has been supported by experimental programmes on the removal of aerosols with sprays and on the modes of hydrogen combustion using igniters. The results of research for present day reactors are also benefiting the designs for future plants. Many of these plants include severe accident mitigation concepts in their design. Advanced designs include measures for improved in-vessel coolability of debris and ex-vessel debris coolability and retention. These are discussed in the subsequent chapters.