IFMIF accelerator: Database, FMEA, fault tree and RAM

Tapia, Carlos; Dies, J.; Pesudo, V.; Abal, Javier; Ibarra, Á.; Alguacil, Natalia

doi:10.1016/j.fusengdes.2011.02.099

Cited by 13 publications

(3 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Historically, reliability analysis for accelerators was usually performed in the design phase [14,15]. Some of the methods in the first case include: Reliability Block Diagrams (RBD) [4,16], Fault Tree Analysis (FTA) [17][18][19], Failure Mode and Effect Analysis (FMEA) [20] and other methods [21,22]. The results of these methods are later validated by analyzing machine operations data [9,23].…”

Section: Introductionmentioning

confidence: 99%

Predicting particle accelerator failures using binary classifiers

Reščič

Seviour

Blokland

2020

Nuclear Instruments and Methods in Physics Research Section A:

View full text Add to dashboard Cite

Particle accelerator failures lead to unscheduled downtime and lower reliability. Although simple to mitigate while they are actually happening such failures are difficult to predict or identify beforehand. In this work we propose using machine learning approaches to predict machine failures via beam current measurements before they actual occur. To demonstrate this technique in this paper we examine beam pulses from the Oakridge Spallation Neutron Source (SNS). By evaluating a pulse against a set of common classification techniques we show that accelerator failure can be identified prior to actually failing with almost 80% accuracy. We also show that tuning classifier parameters and using pulse properties for refining datasets can further lead to almost 92% accuracy in classification of bad pulses. Most importantly, in the paper we establish there is information about the failure encoded in the pulses prior to it, so we also present a list of feasible next steps for increasing pulse classification accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

Predicting particle accelerator failures using binary classifiers

Reščič

Seviour

Blokland

2020

Nuclear Instruments and Methods in Physics Research Section A:

View full text Add to dashboard Cite

show abstract

“…Previous analysis performed in 2011 in the Fusion Energy Engineering Laboratory [31] concluded that further RF amplifiers analyses should be performed. RAMI analyses done in this chapter contain individual reliability models for the RF modules and availability calculations for the RF system as a whole.…”

Section: Probabilistic Analysis and Resultsmentioning

confidence: 99%

“…• CDA (1996) [18], KEP (2003) [22] and CDR (2004) [15] • Adjoint sensitivity analyses procedure of Markov Chains with application on reliability of IFMIF Accelerator-System Facilities (2004) [26] • Developing the IFMIF RAM planning (2009) [27] • RAMI Guidelines (2009) [28] • Data management and data capture methodology from IFMIF prototypes (2010) [29] • Exploration of reliability databases and comparison of former IFMIF's results (2010) [30] • IFMIF accelerator: Database, FMEA, Fault Tree and RAM (2010) [31] All these documents served as a basis to develop the analyses performed in this thesis. Guidelines, databases, experiences in other facilities and first availability analyses of the accelerator were used to perform the RAMI analyses of the accelerator facility during the Engineering Design Phase.…”

Section: Previous Analysesmentioning

confidence: 99%

IFMIF accelerator facility RAMI analyses in the engineering design phase

Bargalló Font

View full text Add to dashboard Cite

The planned International Fusion Materials Irradiation Facility (IFMIF) has the mission to test and qualify materials for future fusion reactors. IFMIF will employ the deuteron-lithium stripping reaction to irradiate the test samples with a high-energy neutron flux. IFMIF will consist mainly of two linear deuteron accelerators, a liquid lithium loop and a test cell. Accelerated deuterons will collide with the lithium producing a high-energy neutron flux that will irradiate the material samples in the test cell. A timely and relevant fusion neutron source is essential in the path towards DEMO and future fusion power plants. For this reason, IFMIF is required to have high availability to obtain a fusion materials database to find suitable materials for DEMO design within the anticipated timeline. RAMI (Reliability Availability Maintainability Inspectability) analyses are being performed in the very early stages of design to meet such requirements. The IFMIF accelerator facility is composed of two independent linear accelerators, each of which produces a 40 MeV, 125 mA deuteron beam in a continuous wave mode at 175 MHz. These beam characteristics pose several unprecedented challenges: the highest beam intensity, the highest space charge, the highest beam power and the longest RFQ (Radio Frequency Quadrupole). As a result of these challenges, many design characteristics are counter to high-availability performance: the design is reluctant to accept failures, machine protection systems are likely to stop the beam undesirably, cryogenic components require long periods for maintenance, and activation of components complicates maintenance activities. These design difficulties, together with the high availability requirements and the demanding scheduled operational periods, make RAMI analysis an essential tool in the engineering design phase. These studies were performed in collaboration with system designers, enabling the creation of RAMI models that reflect current accelerator design. This feedback has been of the utmost importance to propose plausible design modifications in order to improve the availability performance of the machine. Parallel activity on the design and construction of the Linear IFMIF Prototype Accelerator (LIPAc) provided the detailed design information needed to conduct these studies properly. An iterative process was followed to match IFMIF design and availability studies. These iterations made it possible to include recommendations and design change proposals coming from the RAMI analyses into the accelerator reference design. Iterations consist of gathering information from the design, creating or updating the RAMI models, obtaining and analyzing results, and proposing ways to improve the design. Three different approaches were carried out in the iterative process. First, a comparison with other similar facilities was performed. Second, an individual fault tree analysis was developed for each system of the accelerator. Finally, a Monte Carlo simulation was performed for the whole accelerator facility considering synergies between systems. These approaches make it possible to go from detailed hardware availability analyses to global accelerator performance, to identify weak design points, and to propose design alternatives as well as foresee IFMIF performance, maintenance and operation characteristics. The IFMIF accelerator facility design was analyzed from the RAMI point of view, estimating its future availability and guiding the design towards a high reliability and availability performance. In order to achieve the high-availability requirements several design changes have already been included in the accelerator reference design whereas other important design modifications have been proposed and will be further analyzed in future design phases.

show abstract