Information Continuity and Advanced Reasoning for Improved System

Information Continuity and Advanced Reasoning for Improved System Diagnostics and Prognostics Carl S. Byington Patrick W. Kalgren Impact Technologies, LLC 220 Regent Court State College PA 16801 carl. byington@impact-tek. com 814 -861 -6273

Example High Cost of BIT False Alarms BIT false alarm $ costs in the F/A-18 program are very high u False alarms also negatively impact fleet readiness and safety u 75% of all cannot duplicate (CND) maintenance on the F/A-18 C airplanes was deemed the result of BIT false alarms u Table details F/A-18 A/B/C/D organizational and intermediate level wasted maintenance labor that resulted from BIT false alarms during 1999 u Based on these numbers, the annual wasted maintenance due to BIT false alarms causes a yearly loss of $1. 7 million (F/A-18 alone!) u Addressing these CND would save these $$’s and provide improvements in readiness, manpower, web: http: //www. dtic. mil/ndia/systems/Bainpaper SOURCE: F/A-18 E/F Built-in-Test (BIT) Maturation Process; logistics, and safety u www. impact-tek. com Impact Technologies, LLC

Information Continuity Motivation www. impact-tek. com Impact Technologies, LLC

Growing the Embedded Diagnostics Pie Integrated Diagnostics Verification and Repair BIT Results www. impact-tek. com Box 1 2 3 3 Impact Technologies, LLC Information Continuity

Systems Perspective and Evidence Legacy Federated System composed of many LRUs from different manufacturers, independent BIT u Future Integrated System with specified interfaces and encapsulated interdependencies u Treated as a system, the individual components have relationships and dependencies that can be exploited to gain evidence. • BIT • Power Monitor • Environmental • Operational • Historical Usage www. impact-tek. com Evidence Impact Technologies, LLC

Example OSA XML Documents XML Implementation with guidance from Open System Architecture for Condition-Based Maintenance schema u Document structure specified by Schema at multiple Functional layers u n Data Acquisition n Data Manipulation n Condition Monitor n Health Assessment n Prognostics n Decision Support n Presentation u www. impact-tek. com Documents are created and validated by schema on local or remote site (ATML) Impact Technologies, LLC

XML Insertion into Database OSA-CBM Functional Layers Impact DA Module Impact DM Module Impact CM Module www. impact-tek. com event_ID x. Axis. Start x. Axis. Delta values. . . event_ID LRU_ID frequency Onboard DB Global Voltage Table Auto. Pilot Table . . event_ID mission_id trigger_type GMSStart. Time. Year Auto. Piliot_Fault. Code_PBIT Auto. Piliot_Fault. Code_CBIT Auto. Piliot_Fault. Code_IBIT Impact Technologies, LLC Event Detection Table

AFCP 1553 Interface and XML Conversion u Interfaced with legacy hardware n Honeywell Aircraft Flight Control Processor u Communicated through 1553 data bus n Laptop and Ballard Technology CM 1553 -3 PCMCIA Card u Extracted raw, proprietary hexadecimal data from AFCP Remote Terminal memory n Created C executable using Ballard Technology Application Programmer’s Interface Converted raw data to Meaningful Fault Codes u Wrapped Technologies, LLC OSAFault Codes Impact u www. impact-tek. com

Problem Classification Binary Fault Intermittent but Repeatable Intermittent but Pseudo-Random Graceful Degradation www. impact-tek. com Simple fault/no fault. Can be detected by low level reasoners and BIT. Intermittent fault occurs with high correlation to input parameter set (can be repeated). Can be isolated by a combination of low level reasoning and high level time and feature set correlation. Pseudo-Random intermittent faults are the most difficult to isolate. Require multiple levels of reasoning, adaptability of reasoners and continuous learning. Graceful component degradation can be detected and predicted using system models and time correlated tracking parameters. Refinements to predictions are made when usage profile diverts from norm or tracking parameters indicate. Impact Technologies, LLC

Reasoning Techniques www. impact-tek. com Impact Technologies, LLC

Bayesian Network u High Level Reasoner u Describe Entities u Describe Relationships n Process n Physical n Proximal u Encapsulates a priori knowledge u Permits robust diagnostics with incomplete knowledge or modeling capability www. impact-tek. com Impact Technologies, LLC

Top Level Reasoning 1. Initial State a priori relationships 2. BIT & Sensor Knowledge 3. Failure and Inference www. impact-tek. com Impact Technologies, LLC

Evidence Fusion and Bayesian Networ www. impact-tek. com Impact Technologies, LLC

At-Wing Evidence Analysis and Fusion Techniques u Data or Knowledge fusion - the process of using collaborative or competitive information to arrive at a more confident decision both in diagnostics and prognostics n Should play a key role in terms of producing useful features, combining features, and incorporating new evidence u Several different architectures and implementation choices for fusion n Bayesian and Dempster-Shafer Combination, Voting, and Fuzzy Logic Inference Ex. Bayesian Fusion Where: = probability of fault (f) given a diagnostic output (O) = probability that a diagnostic output (O) is associated with a fault (f) = probability of the fault (f) occurring. www. impact-tek. com Impact Technologies, LLC

Positive + Negative Evidence Reasoner www. impact-tek. com Impact Technologies, LLC

Integrated Diagnostics Results u Prioritized list of actions to be performed by maintainer u Rankings by confidence u Rankings by greatest benefit for ambiguity reduction u Opportunity for maintainer feedback to reconfigurable TPS u Executed repair history u Linked to Maintenance Action Form www. impact-tek. com Impact Technologies, LLC

Potential Technology Transition Ø Integration of multiple OSA Health Indications Ø Bus Monitoring and Data Fusion Ø Neural Network and low level reasoners Ø Wrapping proprietary data streams in OSA Ø Storing and Brokering in OSA database Ø System Level Diagnostics Ø Prognostics and Prediction Ø On-board and At-wing Reasoning Ø Bayesian Belief Network Ø Case-based Reasoning Ø Novel Evidence-based BIT Metadata and ATML ARGCS and At-Wing Verification and Link to Logistics www. impact-tek. com Impact Technologies, LLC

Backup Slides www. impact-tek. com Impact Technologies, LLC

Summary of Progress u Demonstrated Multiple Component Avionics Health Management with Bayesian Belief Network u Demonstrated OSA Data Representation and Transport n Automated 1553 Data Interface and Code Extraction n Proprietary Fault Code to XML n XML to Database n Database to Reasoners u Developed Innovative BIT Reasoner for Ambiguity Reduction u Proposed Architecture to Support Information Continuity www. impact-tek. com Impact Technologies, LLC u Coordinated Prototype Development with

Evidence Sources OSA Data Transformation DATA Bus Middleware System Power OSA Wrapper OSA Data Broker OSA Database Environmental Impact Proprietary Low-Level Reasoning 3 rd Party Evidence Source Dempster. Shafer Condition Monitor OSA-XML 3 rd Party Reasoning Module OSA Knowledge Broker System-Level Knowledge Fusion OSA-XML Middleware Bayesian Case-Base Neural-Fuzzy Genetic AHM Design Concept Model-Based High-Level Reasoning Temporal Overlay www. impact-tek. com Impact Technologies, LLC Continuous Learning

H-1 Upgrade Program AH-1 Z Super Cobra H-1 Program to remanufacture/upgrade U. S. Marine Corps fleet of AH-1 W Super Cobra attack helicopters and UH-1 N Huey combat utility helicopters u Strong emphasis on commonality between the vehicles in order to reduce logistics support costs – onboard and offboard u Current plan for integrated avionics suite upgrades u 180 Super Cobras will be upgraded to AH-1 Z u 100 UH-1 N helicopters upgraded to UH-1 Y u Low-rate initial production (LRIP) to begin in 2004 and initial operating capability in 2007. u Bell Helicopter Textron forecasts increasing demand for the AH-1 Z, as other nations, such as Turkey and Israel, are considering upgrading their http: //www. awgnet. com/shownews/01 paris 4/intell 04. htm fleet of AH-1’s. u UH-1 Y utility helicopter SOURCES: http: //www. flug-revue. rotor. com/frtypen/FRErstfl/FR 01 Erst/PRUH-1 Y. htm http: //www. chinfo. navy. mil/navpalib/policy/vision/vis 02/vpp 02 -ch 3 c. html www. impact-tek. com Impact Technologies, LLC http: //www. helis. com/news/2001/uh 1 yff. htm

Intelligent Embedded Diagnostics and Open Architecture for Avionics Health Management (AHM)- SBIR Phase II DESCRIPTION / OBJECTIVES / METHODS • Capable of avionics subsystem and component identification, performance monitoring, prognostic prediction, and severity classification. • Implement specific evidence-based and neural network reasoners for on-board or at-wing diagnostic assessment. Operational Concept MILITARY IMPACT / SPONSORSHIP • AHM technology development targeted for upgradeable and future weapons systems • UH-1 Y & AH-1 Z at-wing and test equipment • F/A-18 Smart TPS Analysis modules • Honeywell D&SS is partner on project and working towards additional transition: • RAH-66 Commanche and C 130/141 • Technology adaptable for on-board use in newer integrated modular avionics www. impact-tek. com • V-22, F 22 & JSF • Demonstrate applicability, adaptability in open architecture, and effectiveness of the advanced diagnostic/prognostic reasoners applied to legacy avionics systems. Reduce 'I' level turnaround time and repair costs. BUDGET & SCHEDULE TASK FY 03 FY 04 FY 05 Design & Develop AHM Architecture Develop AHM Software Modules Customize & Apply to application arenas Impact Budget: 0. 75 M through 3 Q 05 (initiated 3 Q 03) Technologies, LLC