Dependability analysis of a very large volume neutrino telescope

https://doi.org/10.1016/j.ress.2010.05.003Get rights and content

Abstract

This work considers a first order approximation to the dependability analysis of complex large scale installations. The dependability criterion used here is quantitative unavailability, and an appropriate unavailability model is presented. The model assumes that the system is symmetrical, has various levels of hierarchy, and components found in the same level are similar and function independently.

The application example comes from very large volume neutrino telescopes installed under water or ice, consisting of several thousands of optical modules. The readout architecture of the detector has several levels of multiplexing including optical detection towers, branches and tower sectors. The paper presents results for various alternative detector layouts and distances of the detector from the onshore facilities. It also develops dependability requirements for major components and/or subsystems consistent with an overall system performance target. The results depict the dependence of the system unavailability on the number of optical modules and the alternative deep sea infrastructure configurations for transferring the measured signals.

Introduction

Astronomical objects can be observed using Cherenkov telescopes to detect the high-energy astrophysical neutrinos they emit. In Cherenkov telescopes, water or ice has the dual role of absorbing daylight and the vast majority of charged cosmic rays, while allowing the detection of atmospheric or astrophysical neutrinos through the Cherenkov light emitted by muonic reaction products in aqueous transparent media [1]. These telescopes employ a grid of several thousands of optical modules (OM) to measure this light and enable optical neutrino detection. According to solid theoretical arguments, the total volume of the detector grid should be at least of the scale of a cubic kilometer, as in the cases of IceCube [2] which is expected to be complete in 2011 and KM3NeT [3] which is in preparatory phase [4]. The OMs consist of the photon sensors and the associated electronics housed in a glass sphere. To form a spatial grid, the OMs are placed on mechanical structures, the detection units (DUs), supporting and connecting them in vertical assemblies. The OMs are interconnected via watertight and pressure-resistant connectors, and their measured data travel through an underwater network of multiplexed passive optical or active electronic equipment and optical fibers to the onshore base [5].

Dependability is the ability of a system to perform its required function, under determined operating conditions, during a given period of operation. The dependability measure used here is the overall telescope unavailability, namely the percentage of time that the telescope is expected to be unavailable given the desired period of operation. The telescope is treated as a complex system; therefore, its unavailability can be expressed as a function of the unavailabilities of the telescope components. The following sections discuss the development of an appropriate model to determine: (a) the system unavailability given the unavailabilities of its components and (b) the component unavailabilities that satisfy a given performance criterion.

The paper is organized as follows. Section 2 describes the main relevant characteristics of the neutrino telescope. Section 3 presents the method of analysis and the model linking the component unavailabilities with the system unavailability. Section 4 presents and discusses some numerical results, and Section 5 concludes this work.

Section snippets

System description

The large volume neutrino telescope consists of various components and subsystems that can be distinguished into three major subsystems: the detector network, the deep sea infrastructure and the onshore infrastructure [5] (see Fig. 1). All three subsystems must be successful for the telescope to be successful. This work considers the dependability analysis of the OM network and the deep-sea infrastructure. Depending on the specific design, the components belonging to each of the above

Unavailability model

The development of the DN unavailability model is based on (a) the network component failure assumptions discussed in the previous section, (b) a success criterion based on the minimum number of available OMs and (c) a set of design assumptions described below. It is also assumed that all the system components can be in one of the two states: available or unavailable to perform their required operations.

The functional relationship of the various parts of the DN is developed as a hierarchical

Component availability allocation

The objective of the dependability analysis presented in this paper is to obtain a first order approximation of the required component availability characteristics in order to achieve an overall system availability requirement.

The base case design considered here is a cuboid grid of N4=17×18×20=6120 OMs. The OMs are arranged in N3=306 DUs each containing 20 OMs. The DUs are grouped into N2=102 branches each containing 3 DUs. The branches are grouped into N1=17 sectors each containing 6

Conclusions

This work presents a conceptual dependability analysis for a very large volume submarine neutrino telescope. The dependability criterion used here is quantitative unavailability, and an appropriate unavailability model is developed and presented here. The telescope unavailability is estimated based on the unavailabilities of the telescope components, namely the optical modules, the detection units where the optical modules are hosted, and the cable network transferring the measured neutrino

Acknowledgements

This work was partially supported by the EU in FP6, KM3NeT Design Study, Contract no. 011937. The authors would like to thank Dr. Gregory Hallewell for his comments concerning references to the KM3NeT project.

References (14)

There are more references available in the full text version of this article.

Cited by (0)

View full text