Flow-induced vibration in nuclear reactors: A review

Abstract

1. PREFACE Flow-induced structural vibrations are driven by a highly energized flow which contacts a relatively flexible structure. In nuclear reactors, flow-induced vibrations are nearly always the result of the reactor coolant impinging on flexible reactor internals, fuel rods, shielding, or heat exchanger tubes. The resultant vibrations are undesirable and often unanticipated. They are frequently the unfortunate by-product of the tendency of structures becoming more flexible and flows increasing in velocity as reactors are scaled up. There are many mechanisms for flow-induced vibration, ranging from the obscure, such as jet switching, to the common, such as turbulenceinduced vibration. However, reactors are an illdefined environment for the analyst, and it is often difficult to relate a particular in-reactor problem to one of the well-known flow-induced vibration mechanisms. There are two reasons for this. First, parameters such as flow velocity, flow spectra, or structural damping are often not accurately known, and retrofitting instruments to make the desired measurements can be prohibitively expensive. Second, the mathematical models which are used to analyze flow-induced vibration are nearly always limited to specific geometries and certain idealizations of structure and flow that are almost never achieved, in a pure sense, in an operating reactor. What usually happens when a failure occurs in an operating reactor is that the operator observes anomalous behavior, such as oscillating temperatures, high radioactivity level in the coolant, or excessive temperatures, moisture, vibration, or noise, which compels him to reduce the power level. The operator then notifies the reactor owners, constructors, and the governing safety board. The analyst is called upon to diagnose and correct the problem under severe economic and safety restraints. Each day of down time of a large power reactor will generate secondary source electricity costs many times the annual salaries of most engineers, and even the slightest breach of reactor safety is intolerable. This is certainly not the atmosphere to encourage leisurely scientific work covering a spectrum of test conditions. Moreover, once a problem is resolved, it still may not be known in quantitative detail just how the solution worked, and the analyst may not be permitted to publish his experiences. As a result, relatively little information is available in the literature about specific flow-induced vibration problems in operating reactors. Nevertheless, there is a rapidly growing body of knowledge on flowinduced vibration based on model tests and analytical results which is proving very useful in analyzing certain flow-induced vibration problems that can occur in reactors. Analysis methods have evolved, mostly since 1970, which are capable of predicting a number of important flow-induced vibration phenomena. In this paper, some of the more widely used methods of analysis of flow-induced vibration in nuclear reactors will be reviewed, and some of the more important results will be presented. Considerable attention will be paid to where these methods work and, perhaps more important, where they fail. However, before beginning the analytical treatment, three cases of flow-induced vibration failure in nuclear reactors will be reviewed to give insight into the nature of an actual in-reactor failure and the human problems that result.