Showing papers presented at "AUTOTESTCON in 2001"

Fault prognosis using dynamic wavelet neural networks

Abstract: Prognostic algorithms for condition based maintenance of critical machine components are presenting major challenges to software designers and control engineers. Predicting time-to-failure accurately and reliably is absolutely essential if such maintenance practices are to find their way into the industrial floor. Moreover, means are required to assess the performance and effectiveness of these algorithms. This paper introduces a prognostic framework based upon concepts from dynamic wavelet neural networks and virtual sensors and demonstrates its feasibility via a bearing failure example. Statistical methods to assess the performance of prognostic routines are suggested that are intended to assist the user in comparing candidate algorithms. The prognostic and assessment methodology proposed here may be combined with diagnostic and maintenance scheduling methods and implemented on a conventional computing platform to serve the needs of industrial and other critical processes.

An open standard for Web-based condition-based maintenance systems

Abstract: An effort is underway to develop an open system architecture for condition-based maintenance (CBM). The architecture development has focused on the definition of a distributed software architecture. The general requirements for a CBM software architecture are defined and an outline of the developing architecture is given. This paper will discuss in detail a web-based architecture which uses Extensible Markup Language (XML) to define the syntax of messages between software components. The core of the architecture is a CBM domain data model which is defined using Unified Modeling Language (UML). A procedure for mapping from the UML data model to a general XML schema, as well as to a schema for system messages, is described.

Optical fiber-based corrosion sensor systems for health monitoring of aging aircraft

Abstract: Increased service requirements for current military aircraft fleet are extending lifetimes and environmental exposure past original design criteria. It is therefore critical that corrosion be detected before structural integrity is compromised and repair becomes economically prohibitive. Robust, cost effective methods of corrosion detection and monitoring are needed. One approach to corrosion monitoring is to measure the environmental species that typically preclude the initiation of corrosion or reveal its propagation. Novel optical fiber-based sensors for real-time in situ detection of corrosion precursors and by-products have been developed for health monitoring and maintenance of aging aircraft. The developed system consists of optical fiber long-period grating (LPG) sensors multiplexed using an optical switch and a low-cost miniature spectrum analyzer. The LPG sensors operate based on specially designed hydrogel coatings which cause a measurable change in the refractive index measured by the LPG in the presence of specific target species. Application of optical fiber-based chemical sensors for the detection of water ingress in lap joints, and metal-ion detection are presented and discussed.

Improved diagnostic and prognostic assessments using health management information fusion

Abstract: Various data, feature and knowledge fusion strategies and architectures have been developed over the last several years for improving upon the accuracy, robustness and overall effectiveness of anomaly, diagnostic and prognostic technologies. Fusion of relevant sensor data, maintenance database information, and outputs from various diagnostic and prognostic technologies has proven effective in reducing false alarm rates, increasing confidence levels in early fault detection, and predicting time to failure or degraded condition requiring maintenance action. The data fusion strategies discussed in this paper are principally probabilistic or AI-based in nature and are used to aid in directly identifying confidence bounds associated with specific component fault identifications and predictions. Dempster-Shafer fusion, Bayesian inference, fuzzy-logic inference, neural network fusion and simple weighting/voting are the algorithmic approaches that are discussed in this paper. Data fusion architectures such as centralized fusion, autonomous fusion, and hybrid fusion are described in terms of their applicability to fault diagnosis and prognosis. The final goal is to find the optimal combination of measured system data, data fusion algorithms, and associated architectures for obtaining the highest overall prediction/detection confidence levels associated with a specific application. A gas turbine engine test cell sensor validation example is provided in the paper that is specifically related data fusion approaches for test cell sensor validation using Dempster-Shafer fusion.

An architecture to implement Integrated Vehicle Health Management systems

Abstract: Modern aircraft benefit from having health management functions integrated with other onboard and offboard systems. Integrated Vehicle Health Management (IVHM) systems provide a reduction in the amount of data traffic required between systems, more accurate diagnostic capability, addition of a prognostic capability, and lower support costs. A flexible and extensible architecture that directly supports the implementation of IVHM systems is presented. This architecture embodies industry standardization concepts, and has been used as a flexible demonstration platform for several IVHM applications.

A software architecture for building interchangeable test systems

Abstract: Test system developers can benefit greatly from a software architecture that allows for easy interchangeability of instruments in those systems. Using open industry standard software architectures such as Virtual Instrument Software Architecture (VISA), and Interchangeable Virtual Instruments (IVI), developers are able to create systems with interchangeable test instrumentation. This paper describes the VISA and IVI software standards and demonstrates how their use within a broader software architecture, which includes standard development environments and flexible test executive software, facilitates the creation of interchangeable test systems.

Remote diagnosis server architecture

Abstract: Modern systems such as the Space Shuttle, the International Space Station, or nuclear power plants are examples of mission critical systems that need to be monitored around the clock. Such systems typically consist of embedded sensors in networked subsystems that can transmit data to central (or remote) monitoring stations. At Qualtech Systems, we are developing a Remote Diagnosis Server (RDS/sup TM/) to implement a remote health monitoring system based on received data from such systems. RDS/sup TM/ can also be used to provide online monitoring of sensor-rich, network capable systems such as jet engines, building heating-ventilation-air-conditioning systems, and automobiles. This paper presents the various components and architecture of the RDS/sup TM/. The RDS/sup TM/ is built on a three-tier architecture with a "Broker" application in the middle layer, and multiple TEAMS-RT/sup TM/ and TEAMATE/sup TM/ based reasoners at the backend. The client layer consists of sensor agents. that collect test results and transmit them over a message-passing network, or technicians with web browsers being guided through intelligent troubleshooting sessions. A database in the backend, TEAMS-KB/sup TM/, is used to manage models and content, and collect diagnosis logs for data mining. the solution scales easily to hundreds of sessions in any modern workstation or server.

Formal specification of testability metrics in IEEE P1522

Abstract: The objective of the IEEE P1522 Testability and Diagnosability Metrics standard is to provide notionally correct and mathematically precise definitions of testability measures that may be used to either measure the testability characteristics of a system, or predict the testability of a system. Notionally correct means that the measures are not in conflict with intuitive and historical representations. Predictive testability analysis may be used in an iterative fashion to improve the factory, field-testing, and maintainability of complex systems. The end purpose is to provide an unambiguous source for definitions of common and uncommon testability and diagnosability terms such that each individual encountering it can know precisely what that term means.

Fiber optic sensors for predictive health monitoring

Abstract: The ability to predict the onset of failure in mechanical systems is key to the reduction in maintenance costs, downtime, and health hazards in industrial environments. Vibration monitoring can be used to spot mechanical problems, trigger preventative maintenance and diagnose the health of rotating machinery. A variety of MEMS accelerometers and piezoelectric acoustic emission sensors are used in health monitoring that are limited by their size, cost, and performance. Optical fiber sensors are rapidly emerging as viable alternatives to these devices as effective means of health monitoring in harsh environments. These sensors are tolerant to extreme temperature, EMI, shock and vibration, and offer reduced weight and increased accuracy over conventional instrumentation. As a result, these sensors have begun to replace conventional sensors in harsh environment applications. Optical fiber sensors offer much smaller size, reduced weight, ability to operate at temperatures up to 2000/spl deg/C, immunity to electromagnetic interference, resistance to corrosive environments, inherent safety within flammable environments, and the ability to multiplex multiple sensors on a single optical fiber. This paper presents the research and development of both low-profile fiber optic-based acoustic emissions sensors for use in nondestructive evaluation systems and fiber optic accelerometers for harsh environment health monitoring applications.

A process and tool for determining the cost/benefit of prognostic applications

Abstract: Efficient application of prognostics requires the capability to estimate its benefit in specific applications. A cost/benefit analysis process and tool-set to generate such estimates is described. The application of this process and tool to an example application is provided to further illustrate the approach. Inputs to the process are a system description, failure modes and distributions. and implementation cost estimates. The output is a life cycle payoff or cost as well as an assessment of the technical risk of the application. The process and tool was developed for use on the Joint Strike Fighter program. It is currently being integrated into a more complete Boeing cost/benefit analysis model that addresses the complete spectrum of diagnostic and prognostic methods.

IEEE-1149.x standards: achievements vs. expectations

Abstract: This paper examines the IEEE-1149 standard and its .x sub standards. Since the JTAG/IEEE-1149.1 boundary-scan was first standardized in 1990, it has undergone some close scrutiny. The paper discusses how this pioneering standard met expectations, how it achieved some unplanned utilities, and what other surprises came to light. The paper also looks at the progress of some of the other standards that were published later. The IEEE-1149.5 system level testability standard was published in 1995, but has not gained as much popularity. The reasons are examined and information is provided on how the 1149.5 could be used more effectively. The IEEE-1149.4, released in November 1999, incorporates the features of the digital boundary-scan while adding analog capabilities. The other two assigned numbers, 1149.2 and 1149.3 never became standards. This paper reports on the present status and utility of the 1149.x standards with an eye to their future applications. It concludes that the standards meet the challenges brought about by greater complexities in electronics and indicates confidence that they will remain an integral part of the board and system level testability needs well into the foreseeable future.

Determining how to do prognostics, and then determining what to do with it

Abstract: Understanding how to do prognostics is rapidly evolving, but an area of study that is in its infancy is the development of doctrine regarding what to do with prognostic information. The US Army's Logistics Integration Agency has initiated a program to accelerate development and fielding of the hardware, information systems and business processes needed to seamlessly generate, transmit and use real-time, platform-based readiness data. This program is investigating how future logistics support concepts need to be adapted to fully exploit such data. This paper describes an approach that has been adopted by this program to facilitate the development of proactive business practices and logistics system architectures.

Semantic requirements on information integration for diagnostic maturation

Abstract: The diagnostic maturation process requires those who support complex systems to have ready access to design, maintenance, and other logistic support information sources. The heterogeneous nature of these sources poses unique challenges to those who would extract meaningful knowledge from them. There have been numerous Information Integration efforts over the last ten or so years, with varying degrees of success. There have been data exchange standards created for product data that concern primarily the design and manufacturing aspects of the product life cycle. Finally, there are information-based standards in the diagnostic design and implementation phases that utilize formalisms for representing the semantics of the data and knowledge over which they operate. This paper will look at each of these aspects of the Information Integration challenge and examine them for their relevance to the issues inherent in the collection and integration of information that supports diagnostic maturation.

ATLAS2K and the IVI Signal Interface-the framework for an open, modular and distributed ATS architecture

Abstract: This paper investigates the combined use of ATLAS2K and the IVI Signal Interface in ATS architectures. The analysis of these initiatives identifies a set of common features and demonstrates that they address complementary areas of ATS architectures. This situation makes them an excellent fit for establishing the foundation of a modular, open, distributed and coherent COM-based ATS architecture.

Active aircraft fault detection and isolation

Abstract: In this paper, we describe development and analysis of an active fault detection and isolation system for the commuter and business aircraft. To accommodate faults wherever possible, we develop an algorithm that can reliably detect and isolate system faults with minimal disruption of normal aircraft operations. We define several candidate fault scenarios (e.g., aircraft icing, faults of control surface actuators, stuck or floating control surfaces, etc.) that are common occurrences, and construct a jet aircraft model, suitable for simulation studies. Aircraft faults are detected and isolated using a hierarchy of techniques. Successive layers in the hierarchy are increasingly invasive, higher layers being invoked only when lower layers indicate a potential problem.

Mitigating COTS obsolescence in military test

Abstract: This paper seeks to identify the causes of obsolescence in military test equipment, specifically in automated test equipment, while offering potential alternatives to reduce the occurrence and impact of the changes required. The analysis and alternatives is at a system level to advocate a proactive approach rather than the reactive approach forced by individual occurrences. The use of automated tools for predictive analysis is emphasized. An approach to reduce the impact of obsolescence is detailed.

Whither AI in test and diagnosis

Abstract: In an increasingly competitive marketplace system complexity continues to grow, but time-to-market and lifecycle are reducing. This has driven the need for automated test and diagnostic tools. As test and diagnosis is a high-level human activity, AI-based solutions have been pursued. This has been an active research area for some decades, but the industrial acceptance of AI approaches, particularly in cost-sensitive areas, has not been high. This paper reviews the history and current state of AI in test and diagnosis, discusses future challenges, and introduces the authors' work in addressing some of these challenges in a diagnostic context.

Implementation strategy for AI-ESTATE compliance in the Remote Diagnosis Server (RDS) framework

Abstract: The main objective of the paper is to outline, in detail, the implementation strategy and process of IEEE Standard 1232, Artificial Intelligence Exchange and Service Tie to All Test Environments (AI-ESTATE) compliance of the product suite from Qualtech Systems, Inc., while maintaining the product suite's existing mechanism of information interchange and service specifications. The product suite includes Testability Engineering and Maintenance Systems (TEAMS)-the diagnostic model development system, Remote Diagnostic Server (RDS)-a framework for distributed remote diagnosis using QSI's runtime diagnostic reasoners, and TEAMS-Knowledge Base (TEAMS-KB)-database tool for model management and data collection. As a key factor in enabling AI-ESTATE compliance in our runtime environment, namely RDS, we envisage encapsulating the AI-ESTATE service format in the RDS communication protocol packet. Since all AI-ESTATE services are represented as part of an information model defined in EXPRESS language, and does follow a textual services-and-attributes format without specifying an underlying implementation technology, we envision a relatively straight-forward encapsulation of both types of AI-ESTATE services, namely Model Management and Reasoner Manipulation, within the RDS protocol. This solution allows both AI-ESTATE compliant and non AI-ESTATE based diagnostic reasoners to provide their services using the same protocol and hence service their respective clients within the RDS framework.

The IVI foundation signal interface; a new industry standard

Abstract: The Interchangeable Virtual Instruments (IVI) Foundation is leading major changes in virtual instrumentation technology. The main focus of standardization has moved from IVI Instrument Driver Class Specifications to IVI Components that allow users to support more robust semantic and behavioral compatibility. There is a new emphasis on interoperability and interchangeability at a deeper level among ATE Systems Integrator and Test System Users. On a grander scale within the environment of synthetic and virtual instruments, users are requesting absolute interchangeability. The potential of full interoperability and absolute instrument interchangeability is examined along with its likely impact. Some of the gaps in current standards are examined, with a focus on the information and IVI Components required for filling those gaps.

Augmenting legacy military ate for functional test using NxTest technology

Abstract: Typical military Automatic Test Equipment (ATE) usually consists of a number of single channel stimulus and measurement devices connected to the UUT with a switch matrix, providing traditional serial, parametric test with its lengthy test times. Functional Test methodology tests a unit by simulating its environment and verifying that the unit operates correctly in that environment. This requires simultaneous stimulus and measurement capability that is not usually found in traditional military ATE. Boeing Support Systems is currently in the process of augmenting an existing military test station to provide functional test capability through the use of NxTest Technology. This paper will discuss our approach to adding Functional Test capability to an existing piece of equipment.

An algorithm for diagnostic reasoning using TFPG models in embedded real-time applications

Abstract: Embedded diagnostic reasoners require compact modeling representations and efficient reasoning algorithms given limited available computational resources. Timed failure propagation graphs (TFPG) are compact representations used to model failure causes and progressions of conditions that are symptoms of failure occurrence, together with the temporality and likelihood of these symptom progressions and the observation of some of the aberrant conditions. Algorithms for design-time diagnosability analysis using TFPG models have previously been reported, but these algorithms have different design objectives leading to different computational strategies and optimization criteria. This paper presents and discusses an algorithm specifically designed for efficient failure isolation based on reported observations of abnormal conditions that is suitable for use in embedded real-time applications.

Using thermal analysis to enhance fault isolation techniques

Abstract: The present methods of diagnostic testing for printed circuit boards (PCB) using automatic test equipment (ATE) presents the Navy with technical and cost of ownership concerns. These concerns are manifested in several areas: (1) the escalating complexity of PCBs requires an ever-increasing amount of maintenance processing time, (2) the increased processing time results in escalating repair costs, and (3) protracted component turnaround-times are usually managed by procuring additional spares to meet operational requirements. When coupled with current PCB design constraints such as limited test points, in many cases ATE software that cannot isolate a fault to an acceptable ambiguity level, and a maintenance philosophy that places high reliance on ATE and less on the technical skills of maintenance personnel, the situation becomes bleaker. Naval Aviation cannot continue to absorb the increasing costs of ownership associated with current maintenance practices. Neither can it afford to make major changes in its current maintenance philosophy. The solution to this dilemma rests in the ability to successfully challenge the current ATE diagnostic testing methodology and develop a means of enhancing existing ATE capabilities.

The cost of test

Abstract: The cost of test is an important issue and one which continues to attract a lot of attention even though it is often not fully understood. Most industry designers do not have enough time to devote to an economic analysis of testing options, but it is quite clear that the benefits of doing so are enormous-ultimately very large sums of money can be saved through the use of 'appropriate' test. This paper will explore the issues of testing costs and touch on all aspects of design through to field service.

Re-configurable hardware for "synthetic" wideband instruments

Abstract: Using current commercially-available programmable devices, ranging from ASICs to PCs, it is possible to configure a single set of hardware that can serve as the measurement kernel for software-defined instruments suitable for the creation and/or analysis of modulated signals up to 80 MHz - bandwidth. Furnished with an appropriate software interface, such a device can replace individual instruments in many cases, and in other cases can be used to respond rapidly to changing measurement needs.

Visual programming versus textual programming in automatic testing and diagnosis

Abstract: The present paper investigates the categories of functionality implemented by software in Automatic Test Systems and the programming paradigms that are commonly used for implementation. Building on the conclusions of this analysis, the paper proposes a combined development paradigm, where visual programming is used at diagnosis level, while both textual and visual languages are supported for implementing test and instrument control functionality. This approach offers important benefits in reducing development and maintenance costs for ATS software, while maximizing its potential for reuse. The proposed development paradigm is illustrated by an actual ATS architecture, implemented in software product.

Residual stress measurement of thermal barrier coatings using laser fluorescence technique and their life prediction

Abstract: Non-destructive determination of the remaining life of coatings of gas turbine parts is highly desirable. The present paper describes early attempts to prove the feasibility of doing this based on the optical measurement of the stress in the oxide that attaches the coating to the metal component. Both standard regression methods and neural network methods are compared and it was found that the neural network approach was superior for the case where multiple signal features were present. All methods provide useful predictions for the idealized case considered.

PIP-enhanced "at system" diagnosis on the flightline

Abstract: Flightline component false removals are costly and add to the logistics tail of every weapon system. Integrating portable maintenance aids (PMAs) and maintenance information systems into the test repertoire can help significantly; however, these solutions do little to augment the information available for failure diagnosis. Adding conventional test instrumentation under PMA control would provide the additional information we seek; however, these hardware intensive solutions are typically large, heavy and expensive. VXI offers an opportunity to consolidate instrumentation into a convenient package, however, the majority of VXI products provide a single function. Therefore many VXI modules are required to provide the needed functionality, resulting in an instrumentation package that is large and awkward to use at best, and quite often just not useable on the flightline. This paper addresses false removals, the challenges involved in addressing this issue, and solutions that are being implemented on the RAH-66 Comanche program. The PMA Instrument Pack (PIP) approach to flightline instrumentation facilitates powerful test techniques that augment diagnostic capability at the weapon system.

Full-scale testing of fuselage panels

Abstract: A unique state-of-the-art facility to assess the structural integrity of aircraft fuselage structure was established at the Federal Aviation Administration (FAA) William J. Hughes Technical Center. The Full-Scale Aircraft Structural Test Evaluation and Research (FASTER) facility is capable of testing full-scale fuselage panel specimens under conditions representative of those seen by an aircraft in actual operation. The test fixture features a novel adaptation of mechanical, fluid, and electronic components and is capable of applying pressurization, longitudinal, hoop, frame, and shear loads to a fuselage panel. A high-precision, Remote Controlled Crack Monitoring (RCCM) system was developed to inspect and record crack initiation and progression over the entire fuselage panel test surface. A detailed description of the FASTER facility along with representative results from a variety of experimental test programs will be presented.

AI-ESTATE model editing in a component based ATS

Abstract: "All Test Environments" that are based on the IEEE 1232 Standard, that is, are AI-ESTATE compliant, utilize diagnostic information models as specified by the 1232 Standard. Thus the creation and maintenance of these models are critical to the operation of the Test Environment. This paper details a Model Editing Component (MEC) that operates as a plug-n-play component in a component based ATS architecture. The diagnostic information model is represented within the Standard in the EXPRESS language. Due to empirical assessments of test engineering human factors the editor hides this representation from the test engineer. The model editor instead presents the information model in a combination of well-understood graphical forms and data dialogs while providing functional operations through well-understood drag-n-drop and click operations. The important information model relationships are handled automatically behind the graphical representation layer relieving the test engineer from learning and maintaining an understanding of the EXPRESS language. Also discussed are the morphological and topological display algorithms as well as operational usage, model storage/retrieval management, import/export, exchange and maturation.

Passive fiber optic position sensor with digital encoding

Abstract: Position sensors are ubiquitous in many industrial environments. Currently, resolvers, LVDTs, and electro-optic encoders are common. We report on fiber optic rotary and linear absolute position sensors developed for the aerospace industry that have broad potential. The use of fiber optics in sensors has many well known advantages including immunity to electromagnetic interference, remote positioning of the sensors on very long leads, reduced size, and increased tolerance to harsh environments. In addition, the sensors reported here encode the position information on a serial digital bit stream that interfaces easily with digital electronics and decreases sensitivity to lead fiber losses. This paper reports on the development of a class of passive fiber optic rotary and linear encoders for use in aerospace and industrial applications.