Integrated Systems Health Management includes as key elements fault detection, fault diagnostics, and failure prognostics. Whereas fault detection and diagnostics have been the subject of considerable emphasis in the Artificial Intelligence (AI) community in the past, prognostics has not enjoyed the same attention. The reason for this lack of attention is in part because prognostics as a discipline has only recently been recognized as a game-changing technology that can push the boundary of systems health management. This paper provides a survey of AI techniques applied to prognostics. The paper is an update to our previously published survey of data-driven prognostics.

In this paper we propose ν-Anomica, a novel anomaly detection technique that can be trained on huge data sets with much reduced running time compared to the benchmark one-class Support Vector Machines algorithm. In ν-Anomica, the idea is to train the machine such that it can provide a close approximation to the exact decision plane using fewer training points and without losing much of the generalization performance of the classical approach. We have tested the proposed algorithm on a variety of continuous data sets under different conditions. We show that under all test conditions the developed procedure closely preserves the accuracy of standard oneclass Support Vector Machines while reducing both the training time and the test time by 5 − 20 times.

Complex engineering systems require efficient on-line fault diagnosis methodologies to improve safety and reduce maintenance costs. Traditionally, diagnosis approaches are centralized, but these solutions do not scale well. Also, centralized diagnosis solutions are difficult to implement on increasingly prevalent distributed, networked embedded systems. This paper presents a distributed diagnosis framework for physical systems with continuous behavior. Using Possible Conflicts, a structural model decomposition method from the Artificial Intelligence model-based diagnosis (DX) community, we develop a distributed diagnoser design algorithm to build local event-based diagnosers. These diagnosers are constructed based on global diagnosability analysis of the system, enabling them to generate local diagnosis results that are globally correct without the use of a centralized coordinator. We also use Possible Conflicts to design local parameter estimators that are integrated with the local diagnosers to form a comprehensive distributed diagnosis framework. Hence, this is a fully distributed approach to fault detection, isolation, and identification. We evaluate the developed scheme on a four-wheeled rover for different design scenarios to show the advantages of using Possible Conflicts, and generate on-line diagnosis results in simulation to demonstrate the approach.

Sample data, including nominal and faulty scenarios, for Tier 1 and Tier 2 of the First International Diagnostic Competition. Three file formats are provided, tab-delimited .txt files, Matlab .mat files, and tab-delimited .scn files. The scenario (.scn) files are read by the DXC framework. See the Support/Documentation section below and the First International Diagnostic Competition project page for more information.

Failure prognosis - as a natural extension to the fault detection and isolation (FDI) problem - has become a key issue in a world where the economic impact of system reliability and cost-effective operation of critical assets is steadily increasing. Failure prognostic algorithms aim to characterize the evolution of incipient fault conditions in complex dynamic processes, thus allowing to estimate of the remaining useful life (RUL) of subsystems and components. Several examples can be used here to illustrate the range of possible applications for these algorithms: electro-mechanical systems, continuous-time manufacturing processes, structural damage analysis, and even fault tolerant software architectures. Most of them have in common the fact that they are highly complex, nonlinear, and affected by large-grain uncertainty. We introduce in this chapter an integrated failure prognosis architecture that is applicable to a variety of aircraft systems and industrial processes. We are targeting a specific rotorcraft system as a prototypical testbed for proof-of-concept. The overall architecture consists of an on-board and an off-board module for eventual on-platformimplementation purposes.

This paper describes the results of a significant research and development effort conducted at NASA Ames Research Center to develop new text mining algorithms to discover anomalies in free-text reports regarding system health and safety of two aerospace systems. We discuss two problems of significant import in the aviation industry. The first problem is that of automatic anomaly discovery concerning an aerospace system through the analysis of tens of thousands of free-text problem reports that are written about the system. The second problem that we address is that of automatic discovery of recurring anomalies, i.e., anomalies that may be described in different ways by different authors, at varying times and under varying conditions, but that are truly about the same part of the system. The intent of recurring anomaly identification is to determine project or system weakness or high-risk issues. The discovery of recurring anomalies is a key goal in building safe, reliable, and cost-effective aerospace systems.

One-class nu-Support Vector machine (SVMs) learning technique maps the input data into a much higher dimensional space and then uses a small portion of the training data (support vectors) to parametrize the decision surface that can linearly separate nu fraction of training points (labeled as anomalies) from the rest. The exact solution of standard one-class nu SVMs assigns (at least) nu fraction of training points as support vectors. However some of these support vectors may be unnecessary or redundant. Hence the computational issue turns alarming especially when SVMs based novelty detectors with nonlinear kernels are trained on data sets of huge size. The proposed nu-Anomica algorithm can solve this problem. The idea is to train the machine such that it can provide a close approximation to the exact decision plane using far less number of training points and without loosing much of the generalization performance of the classical approach. The developed procedure closely preserves the accuracy of standard One-class nu-SVMs while reducing both training time and test time by several factors.

Round 4 Test DatasetThe data being generated and disseminated is the test data used to construct trojan detection software solutions. This data, generated at NIST, consists of human level AIs trained to perform image classification. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers. This dataset consists of 288 adversarially trained, human level, image classification AI models using a variety of model architectures. The models were trained on synthetically created image data of non-real traffic signs superimposed on road background scenes. Half (50%) of the models have been poisoned with an embedded trigger which causes misclassification of the images when the trigger is present.

인공지능기법을 이용한 고압 LNG Pump에 대한 결함 진단 기술 개발

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of instruction fine tuned LLMs. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting that trigger behavior in the trained AI models.