Luis C. Trevino

A framework for integration of IVHM technologies for intelligent integration for vehicle management

As a part of the overall goal of developing Integrated Vehicle Health Management (IVHM) systems for aerospace vehicles, the NASA Faculty Fellowship Program (NFFP) at Marshall Space Flight Center has performed a pilot study on IVHM principles which integrates researched IVHM technologies in support of integrated intelligent vehicle management (IIVM). IVHM is the process of assessing, preserving, and restoring system functionality across flight and ground systems. The framework presented in this paper integrates advanced computational techniques with sensor and communication technologies for spacecraft that can generate responses through detection, diagnosis, reasoning, and adapt to system faults in support of IIVM. These real-time responses allow the IIVM to modify the affected vehicle subsystem(s) prior to a catastrophic event. Furthermore, the objective of this pilot program is to develop and integrate technologies which can provide a continuous, intelligent, and adaptive health state of a vehicle and use this information to improve safety and reduce costs of operations. Recent investments in avionics, health management, and controls have been directed towards IIVM. As this concept has matured, it has become clear that IIVM requires the same sensors and processing capabilities as the real-time avionics functions to support diagnosis of subsystem problems. New sensors have been proposed, in addition to augment the avionics sensors to support better system monitoring and diagnostics. As the designs have been considered, a synergy has been realized where the real-time avionics can utilize sensors proposed for diagnostics and prognostics to make better real-time decisions in response to detected failures. IIVM provides for a single system allowing modularity of functions and hardware across the vehicle. The framework that supports IIVM consists of 11 major on-board functions necessary to fully manage a space vehicle maintaining crew safety and mission objectives. These systems include the following: guidance and navigation; communications and tracking; vehicle monitoring; information transport and integration; vehicle diagnostics; vehicle prognostics; vehicle mission planning; automated repair and replacement; vehicle control; human computer interface; and onboard verification and validation. Furthermore, the presented framework provides complete vehicle management which not only allows for increased crew safety and mission success through new intelligence capabilities, but also yields a mechanism for more efficient vehicle operations. The representative IVHM technologies for IIVM includes: 1) enhanced communications and telemetry, 2) sensors for radiation materials, 3) vehicle controls and dynamics, 4) flight mechanics and control, 4) embedded sensors for structural integrity of tanking systems, 5) evolutionary concepts for embedded sensor placement in tank systems, 6) real time operating systems, and 7) computer architectures for distributed processing for IVHM. This paper presents the IIVM framework and the IVHM technologies developed under NASAs NFFP pilot project

Soft computing applications on a SR-30 turbojet engine

The use of soft computing algorithms in hardware-in-the-loop applications has been investigated. A fuzzy logic controller (FLC) was designed and successfully tested on the Turbine Technologies SR-30 turbojet engine for the main-stage operation of the engine. A transfer function model of the plant was obtained using frequency-response methods. To ensure safe operation of the engine, a PID controller was first tested on the engine, and both the PID and FLC controllers were tested in a simulated environment before being used with the engine.

Use of soft computing technologies for rocket engine control

NASA issues and remedial applications for rocket engine control are presented. A testbed for researching these applications is presented and further detailed. Automation and control of the operation of the testbed, a Turbine Technologies SR-30 (single radial) small turbojet engine, is discussed. A fast data acquisition board, fuel-flow rate meter, a valve controller are implemented to the existing system, which was already equipped with pressure, temperature, RPM and load sensors. Start, operation and stop sequences are automated using National Instruments Labview software. Classic PID control algorithm is applied in the initial phase of the automation. Data collected during the start, transient/steady operation, and stop sequences of the engine has been analyzed to design a Bayesian Belief control algorithm. PID algorithm implementation and fuzzy algorithms are currently being conducted.

Autonomous Component Health Management with Failed Component Detection, Identification, and Avoidance:

Abstract This article describes a novel scheme for autonomous component health management (ACHM) with failed actuator detection and identification and failed sensor detection, identification, and avoidance. This new scheme has features that are very superior to those with triple redundant sensing and voting, yet requires fewer sensors; it can be applied to any system with redundant sensing. Relevant background to the ACHM scheme is provided in this article. Simulation results for its application to a single-axis spacecraft attitude control system with a third-order plant and dual-redundant measurements of the system states are presented. The ACHM scheme fulfills key functions needed by an integrated vehicle health monitoring (IVHM) system. It is autonomous; is adaptive; works in real time; provides optimal state estimation; identifies failed components; avoids failed components; reconfigures for multiple failures, reconfigures for intermittent failures; works for hard-over, soft, and zero-output failures; and works for both open- and closed-loop systems. The ACHM scheme combines a prefilter that generates preliminary estimates of the system states, detects and identifies failed sensors and actuators, and avoids failed sensors in generating preliminary estimates of the system states with a fixed-gain Kalman filter that provides model-based state estimates, which are utilized in the failure detection logic, and generates optimal estimates for the system states. The simulation results show that ACHM can successfully detect, identify, and avoid sensor failures that are single or multiple; persistent and intermittent; and hard-over, soft, and zero-output types. It is now ready to be tested on a computer model of an actual system.

SR-30 turbojet engine real-time sensor health monitoring using neural networks, and Bayesian belief networks

This paper describes the use of artificial intelligence-based techniques for detecting and isolating sensor failures in a turbojet engine. Specifically, three artificial intelligence (AI) techniques are employed: artificial neural networks (NNs), statistical expectations, and Bayesian belief networks (BBNs). These techniques are combined into an overall system that is capable of distinguishing between sensor failure and engine failure—a critical capability in the operation of turbojet engines.The turbojet engine used in this study is an SR-30 developed by Turbine Technologies. Initially, NNs were designed and trained to recognize sensor failure in the engine. The increased random noise output from failing sensors was used as the key indicator. Next, a Bayesian statistical method was used to recognize sensor failure based on the bias error occurring in the sensors. Finally, a BBN was developed to interpret the results of the NN and statistical evaluations. The BBN determines whether single or multiple sensor failures signify engine failure, or whether sensor failures represent separate, unrelated incidences. The BBN algorithm is also used to distinguish between bias and noise errors on sensors used to monitor turbojet performance. The overall system is demonstrated to work equally well during start-up and main-stage operation of the engine. Results show that the method can efficiently detect and isolate single or multiple sensor failures within this dynamic environment.

Model Diagnostics of Jet Engines

Time-dependent model of a small-scale turbojet engine is developed to be used for engine health diagnostics. A thermodynamic model obtained by solving the one-dimensional and time-dependent governing equations of the jet engine, and a neural network based model of the engine are used together to predict the state variables in the engine in response to change in the fuel flow valve angle. The thermodynamic model is more efficient in tracking the temperature state variables exhibiting transitional responses varying visibly from the fuel- valve angle trend. The neural network based model on the other hand is a much less erroneous and a more competent model for diagnosing other state variables (pressures and RPM). These two models offer an accurate model of the engine.

Jet-Engine Health Monitoring and Operation Using Fuzzy- Logic Controllers and Bayesian Belief Networks

Development of a test-bed that allows hardware-in-the-loop testing of soft-computing tools for rocket engine control is described. A “Turbine Technologies” SR-30 turbojet engine has been automated to run via computer. A fuzzy-logic controller has been implemented to control engine start-up sequence, and several discrete health and safety monitoring controls have been implemented to safely start, run and shut down the engine. Along with the discrete health monitoring system, a statistical analysis is done concerning the functionality of the engine’s sensors during the start up phase. A Bayesian belief network is designed to not only prove the capability of soft computing technologies for engine control but also to alter the fuzzy control system to adapt to sensor malfunction without having to shut the engine down.

An Intelligent Integration Framework for In-Space Propulsion Technologies for Integrated Vehicle Health Management

The success of NASA’s Exploration activities hinges on the ability to make space systems safer, more affordable, and more self-sufficient. As these missions expand to ever increasing distances from earth, the systems that support the missions will be required to become more self-sufficient. Intelligent Integrated Vehicle Management (IIVM) is an approach that supports vehicle self-sufficiency. The framework presented in this paper integrates advanced computational techniques with in-space propulsion technologies for spacecraft that can generate responses through detection, diagnosis, reasoning, and adapt to system faults. IIVM supports all forms of propulsion systems including chemical and nuclear. Technologies exist today to develop IIVM, but there are many challenges in integrating these technical solutions.

A Testbed for Evaluating Soft Computing Technologies for Rocket Engine Control

NASA remedial applications for rocket engine controls are presented. A testbed for researching these applications is presented and further detailed. Automation and control of the operation of the testbed, a Turbine Technologies SR-30 small turbojet engine, are discussed. The automation of the engine required implementing a fast data acquisition board and a fuel flow valve controller to the existing system. Start, main stage, and stop sequences are automated using National Instruments LabVIEW software. Classic closed-loop control algorithm is implemented in a LabVIEW program. The data collected during the start, main stage, and stop sequences of the engine has been analyzed to design the closed-loop controller. The methods to design Bayesian Belief and Fuzzy Logic control algorithms are discussed.

Soft computing for propulsion control

The objective is to explore how soft computing technologies could be employed to improve overall vehicle system safety, reliability, and rocket engine performance by development of a qualitative and reliable engine control system (QRECS). This is addressed by enhancing engine control using soft computing technologies, data mining tools, and sound software engineering practices. Goals for addressing quality are improving software management, development time, maintenance, processor execution, fault tolerance and mitigation, and nonlinear control in power level transitions. The intent is not to discuss any shortcomings of existing engine control methodologies, but to provide alternative design choices for control, implementation, performance, and sustaining engineering, all relative to addressing reliability. The approaches presented require knowledge in rocket engine propulsion, software engineering for embedded flight software systems, and soft computing technologies (e.g., neural networks, fuzzy logic, data mining, and Bayesian belief networks), some of which are briefed. For this effort, the demonstration engine testbed is the MC-1 engine which is simulated with hardware and software. A brief plan of action for design, development, implementation, and testing a Phase One effort for QRECS is given, along with expected results. Phase One focuses on development of a Smart Start Engine Module for proper engine start operations. The final product that this paper proposes is an approach to development of an alternative low cost engine controller that would be capable of performing in unique vision spacecraft vehicles requiring low cost advanced avionics architectures for autonomous operations from engine pre-start to engine shutdown.