Correlation of emergency accident levels with probabilities of occurrence: implications for the safe operation of facilities

Any technological installation, as well as the disposal facility of RAW, is a source of potential danger because in the course of operation, they are forced to interact with the environment and, therefore, to provide some impact on it. Adverse effects associated with contamination of the sur­rounding area can be assessed via the radiological condition of the territory using a five-score system: (1) normal; (2) satisfactory; (3) accident; (4) emer­gency; (5) disaster (Busygin et al., 2009). The ratio r = A/RL can be used for quantitative assessment. Parameter A is an actual level of contamination, and parameter RL is the reference level, typical for this area. The value of r is continuous and varies widely, which complicates the classification and interpretation of the effects of radioactive contamination. Therefore, we propose a system assessment area, based on rankings of r in accordance with Table 9.1 .

Contamination of the environment and the site can clearly result from accidental events in radiation-hazardous facilities. Regulatory and legisla­tive documents allow a posteriori estimation and classification of emergency events, based on measurements of actual contamination levels and compar­ing these results with a certain threshold. However, in practice it is neces­sary to calculate and predict the consequence of events prior to their occurrence, i. e. to give an a priori assessment of the events. Since the events themselves, as well as their effects, are effectively random, then the evalu­ation must be made in terms of random variables, i. e. must have a proba­bilistic nature. An international scale is used to link the seven levels of technological accidents at NPP and their consequences on the environment (INES, 2008) as given in the first two columns of Table 9.2 . Based on this scale, we propose an additional relationship between the levels of incidents with their probabilities for all radiation-dangerous objects, which do not belong to the nuclear fuel cycle facilities. These relationships are given in columns 3 and 4 of Table 9.2 (Puzanov et al., 2004, p. 220).

In some regulations (GAN, 2000) for objects which do not involve the nuclear fuel cycle, a three-score grading system for the class of incidents and

Table 9.1 Score system for the assessment of the state of the site

Range of

r < 2

2 < r < 4

4 < r < 6

6 < r < 9

r > 9

value r

Ball

1

2

3

4

5

Site state

Normal

Satisfactory

Abnormal

Emergency

Disastrous

Table 9.2 Levels and probabilities of incident

Consequences

Trouble-free

operation

probability

Probability should not

exceed

Incident

level

Non-essential difficulties in operation

0.80

0.20

I

Essential difficulties in operation

0.90

0.10

II

Short-term stop of equipment

0.95

0.05

III

Stop of equipment at large material losses

0.99

0.01

IV

Complete destruction of construction

0.999

0.001

V

Destruction accompanied with danger for people’s health

0.9999

0.0001

VI

Disastrous destruction accompanied with a lot of victims

0.99999

0.00001

VII

their consequences is recommended, as well as liaison on the levels of con­tamination. Using Table 9.3 we can associate a class of incident with the probability of their realization and we can specify the requirements for safe operation of facilities. As a case study, we consider the emergency situation at the NPP ‘Fukushima-1’. According to many experts, the situation is con­sistent with a IV-V level of complexity. Initially the reactor coolant system failed under exposure to the earthquake measuring 9 on the Richter scale. According to Table 9.3, we can conclude that, due to this level of earthquake, the failure probability of the cooling system must not exceed 10-3-10-2. It is important to emphasize that the initiating event, i. e. earthquake itself, is not included in the script, since it is not an element of the event tree but is the external condition under which the event tree is realized.

Добавить комментарий

Ваш e-mail не будет опубликован. Обязательные поля помечены *