تجزیه و تحلیل حساسیت دیفرانسیلی مرتبه اول از یک سیستم ایمنی هسته ای توسط شبیه سازی مونت کارلو

In this paper we employ a Monte Carlo method to compute the first-order, differential sensitivity indexes of the basic events characterizing the reliability behavior of the containment spray injection system of a nuclear power plant. An exemplification is provided as to how the obtained sensitivity indexes can be used to drive improvements in the system design and operation.

In the nuclear field one of the principal activities for applications of risk-informed regulatory processes is the ranking of structures, systems and components with respect to their safety-significance [1] and [2]. With reference to safety systems in particular, this requires an analysis of how the safety performance is affected by the stochastic behavior of the components constituting the system: importance measures [1], [2] and [3] and sensitivity indexes [4] are often used for this scope [3]. When realistic issues of system operation are included, such as components' ageing and maintenance, load-sharing, etc. and when uncertainty in the components parameters' values exist, the computation of these measures and indexes is not straightforward [5]. In this paper we illustrate a Monte Carlo simulation method which allows to compute first-order, differential sensitivity indexes [6], [7] and [8]. The details of the method are illustrated and a realistic case of a nuclear safety system, the Containment Spray Injection System (CSIS) [7] is presented. The sensitivity indexes obtained are used to suggest changes in the system design and operation. In the application presented, the Monte Carlo calculations are standard analog ones (no biasing). The introduction of biasing techniques can be easily accommodated in the method to reduce the variance of the Monte Carlo estimates. The calculations have been performed by means of the MARA (Monte Carlo Availability Reliability Analysis) code, developed by the authors at the Department of Nuclear Engineering of the Polytechnic of Milan. In Section 2, we present the Monte Carlo method for the computation of first-order, differential sensitivity indexes. Section 3 contains the results of the application of the method to the Containment Spray Injection System. In Section 4, the findings are summarized and some remarks are given with respect to the method proposed.