Nuclear safety in question answer through project concept of nuclear power station safety.

• Author: Macak, Tomas

1 INTRODUCTION--MOTIVATION FOR THE PAPER AND OBJECTIVE

Two scares in quick succession occurred in the nuclear power plants of Forsmark in Sweden and Temelin in the Czech Republic approximately two years ago. Both incidents reminded Europeans of the Chernobyl disaster and the risks inherent in nuclear technology, one of Europe's chief energy sources. West European countries used to worry about the antiquity of East Europe's nuclear reactors, but since the fall of the iron curtain, power stations in the former Eastern bloc countries have been modernized and upgraded to EU safety standards--thanks to the know-how of Western contractors. In the 1990s the American company, Westinghouse, undertook the renovation of the Soviet-designed WER 100 pressurized water reactors at the Temelin plant in South Bohemia (Czech Republic), adding, for example, complete physical containment. That has not, however, prevented the Temelin plant from showing a "serious failing", and in the wake of the incident at the end of July of the same year at Sweden's Forsmark plant, the European nuclear safety debate is once again an issue. Probably it's no coincidence that two serious incidents at European nuclear power stations occurred on the same day. At Sweden's Forsmark power station, central safety systems failed. One expert referred to this as the most serious incident since Chernobyl. This effectively silences claims that a serious accident couldn't occur at a Western nuclear power station. At almost the same time several thousand liters of radioactive water leaked from the Czech Republic's Temelin power station, which is just a few dozen kilometers away from the Austrian border. Only shortly beforehand a reactor block had been shut down because of a leaking oil pipe. Both incidents are a signal and will reignite the discussion about the safety using nuclear power in Europe. This is a good thing, because it was just looking very much like the use of this source of energy would increase on the continent.

The objective of this paper is to introduce the principle of a new methodology for static reliability measurement of a large system which would have a serious environmental impact in the case of a defect.

2 LITERATURE REVIEW & METHODS

In contemporary management, in whatever system is employed, we must consider the possibility that the required output may not always be reliably obtained (Hron, 2007). In general, we can formalize the uncertainty of the output system to be like the probability of a failure of the system element during its activity time. If we know the probability of any component's failure-free working during its lifetime (p), then we can determine a value of the component's unreliability (h), by way of simple subtraction from the expected reliability: h=1- p. For example the management of a vehicle servicing organization can statistically calculate that a modern, best selling car does not need to be repaired during the duration of the guarantee in 96 of 100 cases. The reliability of the car during the guarantee period is therefore p =0.96, and its measure of unreliability is then h=0,04. This data about the unreliability of a system (here about manufactured product) is very valuable for a manager. The date about unreliability allows the manager to identify what additional costs must be added to production costs for the purposes of calculating profit.

Shown diagrammatically, it is possible to represent a methodology for reducing unreliability in the following way:

1. Couplers' safety optimalization,

2. Adding duplicate or standby components.

[FIGURE 3.1 OMITTED]

Now, based on the schematic shown in figure 3.1 we will show how to determine the reliability of the resulting behavior of two parallel connected elements. Reliability theory is based on the application of probability theory to systems theory.

Let us make an experiment. We are playing with the same two dice cubes, which are identical in every way, it can therefore be assumed that the probability of a 1 showing is the same as the value of any other number, 2 to 6. Imagine that each element of the game cube is a system. If the failure of an element is a random variable, then it cannot it removed, and it is only possible to attempt to predict the likelihood of failure. The malfunctioning component will represent the value of number "six" on the dice. The question is, what is the probability that both dice in one throw show sixes? This reasoning is similar to considering the possibility of both main and standby channels failing simultaneously.

The probability that any one dice falls as a six is equal to [p.sub.1] = 1/6, however the probability that both fall six, is equal to the resulting probability:

p([p.sub.1] a [p.sub.2]) = [p.sub.1] from [p.sub.2] = 1/6 from 1/6 = 1/6 x 1/6 = 1/36

In a similar way we can consider the unreliability of two elements connected in parallel, where one appears as a redundant element. Let us consider the general case, where the elements do not have the same uncertainty:

(3.1) h = [h.sub.1] x [h.sub.2]

Where [h.sub.1] is the unreliability of first element and [h.sub.2] is the unreliability of second element.

If we have n-1 redundant elements, (where n is the total number of elements connected in parallel), the resulting uncertainty would be obtained by multiplying together the unreliability associated with each interconnected parallel element:

(3.2) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

For the system where the unreliability of each element is the same value, i.e. [h.sub.1] = [h.sub.2] = ... = [h.sub.n] = const, the formula (5.2) becomes:

(3.3) h = [h.sub.i.sup.n];

Where n is the number of parallel elements and i is any element: i [member of] {1,2, ..., n}

If we use formula (3.2) needed to determine the reliability of a system composed entirely of parallel elements, we can identify it as a...