H. Benitez-Perez

A CANbus-based safety-critical distributed aeroengine control systems architecture demonstrator

Abstract Recent advances in microelectronics coupled with ever decreasing costs mean that it is now possible to produce very compact and cheap intelligent modules. For instance, it is now quite common for cars to use a number of intelligent units with intercommunication to implement complex functions such as traction control. There has also been a move towards embedding processing in sensors and actuators directly with application to the process control, automotive and aerospace sectors. When considering aerospace applications there are major benefits to be gained by adopting a distributed controller. However, this has to be carried out within the strict design constraints for safety-critical systems. This paper discusses design tools and a distributed system demonstrator that has been developed to explore future distributed control systems.

Implementation of a Smart Sensor Using Analytical Redundancy Techniques

Abstract Increasingly control systems are becoming more distributed in nature. This affords opportunities to enhance fault diagnosis capabilities through the utilisation of distributed fault diagnosis. Smart elements in the system can be used to perform a variety of different diagnostic functions. In this paper we consider integration of analytical redundancy strategies employing model based and fuzzy concepts. A smart element has been developed which incorporates these different techniques and the performance of the element subjected to a variety of faults is presented.

Simulation of Distributed Fault Tolerant Heterogeneous Architectures for Real-Time Control

Synopsis Increasingly, there is a move towards in-built intelligence for sensors and actuators to produce smart components leading to distributed heterogeneous systems. Of particular interest to safety-critical systems is utilisation of the local intelligence to provide health monitoring, fault detection and fault tolerance. However, crucial to system safety is the ability to predict time delays in the system and analyse their effect on control system perfonnance. For real-time systems the delays associated with communicating diagnostic messages may result in safety deadlines being missed. At present, smart components are integrated in an ad-hoc way. This research is exploring how distributed processing capability can be exploited to improve controller performance and increase fault tolerant capability to create high availability systems for gas turbine engine control.

An approximation for reconfigurable Fuzzy Takagi-Sugeno networked control

The paper propose a control strategy from the definition of a Takagi-Sugeno approach, considering computer network reconfiguration. The focus is on computer networks and how control techniques are modified using Fuzzy Takagi-Sugeno control. Control reconfiguration is presented as an available approach for fault coverage in order to keep system performance. Reconfiguration is pursued as a response of time delay modification rather than fault appearance, although this is the basis for control reconfiguration. The focus is on reconfigurable control law due to the presence of local faults and its resulting time delays.

Application of Parameter Estimation Techniques to a Smart Fuel Metering System

Abstract The application of parameter estimation techniques to monitor the health of a smart fuel metering unit for a gas turbine engine is reported. The process dynamics of this fuel metering unit are multivariable and non-linear. The parameter estimates are used to form fault residual signals that are processed by a fuzzy logic-based evaluation procedure to form health monitoring information. The condition monitoring information resulting from this evaluation is presented in a modified SEVA standard.

Integration and Comparison of FDI and Fault Masking Features in Embedded Systems

Abstract This paper addresses the benefits of integrating fault masking (FM) and fault detection and identification (FDI) methods to an embedded control system. It is shown that using a “smart” sensor in a Triple Modular Redundancy (TMR) configuration improves the system performance. Within the proposed hybrid system, “smart” sensors are used as redundant modules of the TMR configuration. A novel quality measure, which indicates the confidence level of a “smart” sensor output is used by a purpose-built voting algorithm to produce outputs. The proposed voting algorithm is a modification of weighted average voter, which produces results based on both the sensor output and the associated quality measure. This algorithm performs better than the conventional weighted average voter within a simple TMR configuration for low error ratios (