Computational modeling of the accident in the fourth power-generating unit of the chernobyl nuclear power plant

Abstract

During the accident in the fourth power-generating unit of the Chernobyl nuclear power plant complicated spatially distributed processes (neutron-physical, thermohydrodynamic, chemical, and thermomechanical) were focused and became intertwined. This has made it difficult to model the accident numerically and it has made international collaboration in this field urgent.Experience gained in the joint work performed by ENEA and the Scientific-Research and Design Institute of Electrotechnical Machinery [1] showed that organization of such investigations requires a parallel solution of a set of complicated scientific-methodological problems with the participation of different types of specialists. In Russia several groups of investigators used different programs (codes) to perform difficult calculations of the first phase of the accident. The main results of these investigations were summarized at the Paris conference in April 1991 [2]. The conclusions drawn there about the accident can be briefly formulated as follows:The main factors influencing the development of the accident were the high positive steam coefficient of reactivity and deficiencies in the construction of the safety and control rod system which came to light during the irregular state of the reactor prior to the accident; spatial dynamic processes played an important role in the development of the accident; to reproduce a real accident process, the physical characteristics must be reproduced in detail throughout the entire volume of the reactor in the period prior to the accident; and the codes used for comprehensive investigations of the first phase are incomplete and they must be further updated and verified.Specialists in different countries performed a series of methodological investigations of the effect of different factors on the positive reactivity arising as a result of the insertion of the safety and control rods [3][4][5][6][7]. These works confirmed that the positive reactivity is highly sensitive to the state of the core prior to the accident and they substantiated the need for reproducing in detail the preliminary initial conditions during computational modeling of the first phase of the accident.The first stage of a combined comprehensive computational analysis of the Chernobyl accident were quasistatic estimates of the positive reactivity according to the DINA [8] and CITATION [9] computational codes. The results of the reconstruction of the three-dimensional neutron fields on the basis of information recorded approximately 2 min prior to the accident by the SKALA system were used as the initial information for constructing the preaccident state of the reactor.Formulation of the Problem. The main problems addressed in the investigations were as follows: formation of the initial state of the reactor, so as to agree correctly with the data recorded; to make, using these initial data, quasistatic estimates of the positive reactivity according to different three-dimensional programs (DINA and CITATION) on the basis of a polycell library ...