106 research outputs found
Unsupervised Physics-Informed Health Indicator Discovery for Complex Systems
No full textDiscovering health indicators (HI) is essential for prognostics and health management of complex systems, as an HI enables timely interventions and effective maintenance strategies. However, most of the existing methodologies for HI discovery rely on labeled data which is expensive and complicated to obtain in the real world. In this paper, we propose a novel, unsupervised physics-informed model structured after expert knowledge in the form of a graphical representation of the expected relationships between sensor readings, operating conditions, and degradation. In addition, a soft constraint is used to guide the representation of the HI according to generally available expert knowledge about degradation. We evaluated the model on a turbofan engine dataset and conducted four experiments by manipulating the original data to create realistic real-world scenarios. The proposed method discovers an HI that exhibits better intrinsic qualities than the current state-of-the-art methodologies, leading to enhanced prognostic performance. Notably, in situations where the initial health state of each system varies, the proposed method achieves an average prognostic performance improvement of approximately 20% compared to existing state-of-the-art methods.Air Transport & Operation
Generic Hybrid Models for Prognostics of Complex Systems
No full textHybrid models combining physical knowledge and machine learning show promise for obtaining accurate and robust prognostic models. However, despite the increased interest in hybrid models in recent years, the proposed solutions tend to be domain-specific. As a result, there is no compelling strategy of what, where, and how physics-derived knowledge can be integrated into deep learning models depending on the available representation of physical knowledge and the quality of data for the development of prognostic models for complex systems. This Ph.D. project aims to develop a general strategy for hybridizing prognostic models by exploring multiple methods to incorporate physical knowledge at various stages of the learning algorithm. The project will prioritize general expert knowledge as the primary source of information, while domain-specific knowledge will serve as an additional feature when applicable.Air Transport & Operation
SL<SUP>~</SUP>(2,R)-homogeneous vector bundles
No full textWe describe all the holomorphic Hermitian vector bundles (E, h) over the upper half plane in C with the property that (f* E, f* h) is holomorphically isometric to (E, h) for any holomorphic automorphism f of the upper half plane. We give an explicit construction of all such holomorphic Hermitian vector bundles using some linear algebraic data
Adaptive Prognostics: A reliable RUL approach
No full textIn the past decade, data-driven methodologies have gained increasing popularity, offering a foundation for predicting the remaining useful life (RUL) of engineering systems and structures using condition monitoring (CM) data. A particularly intriguing challenge lies in accurately predicting the RUL of systems that exhibit exceptional performance, whether underperforming or overperforming, owing to unforeseen phenomena occurring during their operational life. These unique systems, often referred to as outliers, pose a formidable challenge for RUL prediction. This research addresses this challenge by introducing a novel data-driven model, which is known as the Similarity Learning Hidden Semi-Markov Model (SLHSMM) and extends the capabilities of the Non-Homogeneous Hidden Semi-Markov Model (NHHSMM). The training dataset comprises strain data obtained from open-hole carbon-epoxy specimens exposed solely to fatigue loading. In contrast, the validation-testing dataset includes strain data from two specimens subjected to both fatigue and in-situ impact loading, representing an unexpected and previously unseen event in the training data. The study compares RUL estimations generated by the SLHSMM and NHHSMM. The results indicate that the SLHSMM outperforms the NHHSMM, offering superior accuracy in predicting outliers' RUL. This underscores its capability to adapt to unexpected phenomena and seamlessly incorporate unforeseen data into prognostics.Structural Integrity & Composite
Mélange: Multi-tenant scheduling with adaptive eviction for graph processing clusters
No full textMulti-tenancy is an important approach to resource consolidation in cluster management. In this thesis we design and evaluate Mélange, an efficient multi-tenant scheduler targeted towards graph processing jobs. Mélange supports job priorities and eviction, while attempting to avoid starvation. We propose novel ways of exploiting domain-specific knowledge to achieve better scheduling decisions for graph processing jobs. We evaluate static eviction policies and design Mélange to adapt to the cluster and job state at run time to reduce overhead costs during eviction. We have developed Mélange as a cross-layer scheduler built over Apache Giraph and YARN, and show experimental results with synthetic as well as production workloads.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2020-05-01The student, Jayasi Mehar, accepted the attached license on 2018-04-23 at 16:28.The student, Jayasi Mehar, submitted this Thesis for approval on 2018-04-23 at 17:20.This Thesis was approved for publication on 2018-04-24 at 15:17.DSpace SAF Submission Ingestion Package generated from Vireo submission #12434 on 2018-08-31 at 17:21:18Made available in DSpace on 2018-09-04T20:36:52Z (GMT). No. of bitstreams: 2
MEHAR-THESIS-2018.pdf: 2131888 bytes, checksum: 3d6639d87f3c3efbb1bee31e21a20dda (MD5)
LICENSE.txt: 4209 bytes, checksum: 53175c3bd8e036182ec8cdbe3a27034e (MD5)
Previous issue date: 2018-04-24Embargo set by: Seth Robbins for item 107296
Lift date: 2020-09-04T20:37:00Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 107296
Lift date: 2020-09-04T20:42:08Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 107296 on 2020-09-05T09:15:09Z
Adaptive control for availability and consistency in distributed key-values stores
Full text linkThe CAP theorem says that distributed key-value stores can only provide bounded consistency (C) and availability (A) under the presence of partition (P). Recent work has proposed the ability for applications of such stores to specify either an availability SLA or a consistency SLA. In this paper, we propose an adaptive algorithm that automatically controls the underlying storage system in real-time to meet such an SLA while optimizing the other C/A metric. We also present an implementation of the algorithm based on the popular key-value store Riak. Our experiments with the modified system, under realistic workloads, show that the control technique is able to change the system’s configurations to quickly and stably satisfy the SLAs.Item withdrawn by Laura Spradlin ([email protected]) on 2014-12-09T16:37:00Z
Item was in collections:
University of Illinois Theses & Dissertations (ID: 1)
No. of bitstreams: 1
CanhSon_NguyenBa.pdf: 950628 bytes, checksum: 6fa85ced91c072b5a6e905a8e6ee3594 (MD5)Made available in DSpace on 2015-01-21T19:59:27Z (GMT). No. of bitstreams: 1
Canh Son_Nguyen Ba.pdf: 950424 bytes, checksum: 2f2180bc253219320a79cc8806fa6dfd (MD5)Embargo set by: Seth Robbins for item 73282
Lift date: 2017-01-21T19:59:39Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction Lifted for Item 73282 on 2017-01-22T10:15:43Z
Topology-aware distributed graph processing for tightly-coupled clusters
Full text linkCloud applications have burgeoned over the last few years, but they are typically written for loosely-coupled clusters such as datacenters. In this thesis we investigate how one can run cloud applications in tightly-coupled clusters and network topologies, namely super-computers. Specifically, we look at a class of distributed machine learning systems called distributed graph processing systems, and run them on NCSA Blue Waters. Partitioning the graph is key to achieving performance in distributed graph processing systems. We present new topology-aware partitioning techniques that better exploit the structure of the network topologies in supercomputers. Compared to existing work, our new Restricted Oblivious and Grid Centroid partitioning approaches produce 25-33% improvement in makespan, along with a sizable reduction in network traffic. We also discuss optimizations such as smart network buffers that further amplify the improvement. To help operators select the best graph partitioning technique, we culminate our experimental results into a decision tree.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2020-05-01The student, Mayank Bhatt, accepted the attached license on 2018-04-23 at 17:13.The student, Mayank Bhatt, submitted this Thesis for approval on 2018-04-23 at 17:20.This Thesis was approved for publication on 2018-04-24 at 15:21.DSpace SAF Submission Ingestion Package generated from Vireo submission #12435 on 2018-08-31 at 17:21:19Made available in DSpace on 2018-09-04T20:36:52Z (GMT). No. of bitstreams: 2
BHATT-THESIS-2018.pdf: 1415794 bytes, checksum: e08311d8168967b2e47baf1ef67f7fdc (MD5)
LICENSE.txt: 4209 bytes, checksum: b810a770b0873fc45062dd7e9ce83fde (MD5)
Previous issue date: 2018-04-24Embargo set by: Seth Robbins for item 107297
Lift date: 2020-09-04T20:37:00Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 107297
Lift date: 2020-09-04T20:42:08Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 107297 on 2020-09-05T09:15:32Z
An experimental comparison of partitioning strategies in distributed graph processing
Full text linkIn this thesis, we study the problem of choosing among partitioning strategies in distributed graph processing systems.
To this end, we evaluate and characterize both the performance and resource usage of different partitioning strategies under various popular distributed graph processing systems, applications, input graphs, and execution environments.
Through our experiments, we found that no single partitioning strategy is the best fit for all situations, and that the choice of partitioning strategy has a significant effect on resource usage and application run-time.
Our experiments demonstrate that the choice of partitioning strategy depends on (1) the degree distribution of input graph, (2) the type and duration of the application, and (3) the cluster size.
Based on our results, we present rules of thumb to help users pick the best partitioning strategy for their particular use cases. We present results from each system, as well as from all partitioning strategies implemented in two common systems (PowerLyra and GraphX).Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2019-05-01The student, Shiv Verma, accepted the attached license on 2017-04-17 at 19:28.The student, Shiv Verma, submitted this Thesis for approval on 2017-04-17 at 19:40.This Thesis was approved for publication on 2017-04-24 at 09:06.DSpace SAF Submission Ingestion Package generated from Vireo submission #10830 on 2017-08-10 at 15:05:59Made available in DSpace on 2017-08-10T20:33:03Z (GMT). No. of bitstreams: 2
VERMA-THESIS-2017.pdf: 1176883 bytes, checksum: e49f2de22c65fd67d96121626f710849 (MD5)
LICENSE.txt: 4207 bytes, checksum: eb422c7d45cb7c49bb2212e387d9fcaf (MD5)
Previous issue date: 2017-04-24Embargo set by: Colleen Fallaw for item 102777
Lift date: 2019-08-10T21:27:21Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 102777 on 2019-08-11T09:15:39Z
Henge: An intent-driven scheduler for multi-tenant stream processing
Full text linkThis thesis presents Henge, a system that supports intent-based multi-tenancy in modern stream processing applications. Henge supports multi-tenancy as a first-class citizen: everyone inside an organization can now submit their stream processing jobs to a single, shared, consolidated cluster. Additionally, Henge allows each tenant (job) to specify its own intents (i.e., requirements) as a Service Level Objective (SLO) that captures latency and/or throughput. In a multi-tenant cluster, the Henge scheduler adapts continually to meet jobs’ SLOs in spite of limited cluster resources, and under dynamic input workloads. SLOs are soft and are based on utility functions. Henge continually tracks SLO satisfaction, and when jobs miss their SLOs, it wisely navigates the state space to perform resource allocations in real time, maximizing total system utility achieved by all jobs in the system. Henge is integrated in Apache Storm and the thesis presents experimental results, using both production topologies and real datasets.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2019-12-01The student, Faria Kalim, accepted the attached license on 2017-12-04 at 13:41.The student, Faria Kalim, submitted this Thesis for approval on 2017-12-04 at 13:51.This Thesis was approved for publication on 2017-12-04 at 16:54.DSpace SAF Submission Ingestion Package generated from Vireo submission #11818 on 2018-03-13 at 10:37:19Made available in DSpace on 2018-03-13T17:35:43Z (GMT). No. of bitstreams: 2
KALIM-THESIS-2017.pdf: 3149824 bytes, checksum: 1a1d9956c624f4aa3d379806d5627319 (MD5)
LICENSE.txt: 4208 bytes, checksum: 12a2f1944760a6e0d6e4e380714f485b (MD5)
Previous issue date: 2017-12-04Embargo set by: Seth Robbins for item 105472
Lift date: 2020-03-13T17:36:05Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction Lifted for Item 105472 on 2020-03-14T09:15:12Z
- …
