Tim Breikin

Disturbance Attenuation in Fault Detection of Gas Turbine Engines: A Discrete Robust Observer Design

This study is motivated by the onboard fault detection of gas turbine engines (GTEs), where the computation resources are limited and the disturbance is assumed to be band-limited. A fast Fourier transformation (FFT)-based disturbance frequency estimation approach is proposed and performance indexes are improved by integrating such frequency information. Furthermore, in the left eigenvector assignment, both eigenvalues and free parameters are optimized. As illustrated in the application to the actuator fault detection of a GTE, significant improvements are achieved compared to the existing methods. By combining the frequency estimation and eigenvalue optimization, the main contribution of the paper is the reduction of the computation complexity and the avoidance of the local optimal solution due to fixed eigenvalues.

Zero assignment for robust H2/H∞ fault detection filter design

In practical engineering, it is inevitable that a system is perturbed by noise signals. Unfortunately, H infin /H infin filtering may fail to detect some faults when the noise distribution matrix are the same as the fault distribution matrix. In this paper, it is shown that the dynamic feedback gain of a dynamic filter introduces additional zeros to the filter, and both the filter poles and the additional zeros can be assigned arbitrarily. In order to attenuate band-limited noises, the zero assignment technique is used, and an optimal dynamic fault detection filtering approach is proposed by locating the zeros to the noise frequencies and optimizing the poles. Compared to other dynamic filter design approaches, the zero assignment technique gives a better tradeoff between more design freedom and computation costs. As shown in the simulation, a better noise attenuation and fault detection performance have been obtained. The zero assignment in multivariable fault detection filter design would be the main contribution of this paper.

Zero Assignment for Robust Fault Detection Filter Design

In practical engineering, it is inevitable that a system is perturbed by noise signals. Unfortunately, H infin /H infin filtering may fail to detect some faults when the noise distribution matrix are the same as the fault distribution matrix. In this paper, it is shown that the dynamic feedback gain of a dynamic filter introduces additional zeros to the filter, and both the filter poles and the additional zeros can be assigned arbitrarily. In order to attenuate band-limited noises, the zero assignment technique is used, and an optimal dynamic fault detection filtering approach is proposed by locating the zeros to the noise frequencies and optimizing the poles. Compared to other dynamic filter design approaches, the zero assignment technique gives a better tradeoff between more design freedom and computation costs. As shown in the simulation, a better noise attenuation and fault detection performance have been obtained. The zero assignment in multivariable fault detection filter design would be the main contribution of this paper.

Dynamic modelling and Robust Fault Detection of a gas turbine engine

Dynamic modelling and fault detection play an important role in the condition monitoring of gas turbine engines (GTEs). Although system identification and robust fault detection observer (RFDO) have been studied intensively, on-board fault detection raises challenges. A fast identification and discrete observer design is required because of the limited computation ability. In this paper, an output error model is identified first and a discrete observer is designed to avoid the discrete-continuous conversion. With the aid of disturbance frequency estimation, an improved performance index and a fast left-eigenvector based robust observer design method are proposed. As illustrated in the application results, a better disturbance attenuation and fault detection performance have been achieved.

An Algorithm for Identification of Reduced-Order Dynamic Models of Gas Turbines

Model based approaches show a lot of advantages for fault detection and condition monitoring. Particularly, it is true in employing reduced order models for real-time parameter identification and output prediction of gas turbines. Many algorithms have been developed, but most of them focus on one-step-ahead prediction models and involve complex computation. These algorithms are not acceptable for long-term prediction and real-time condition monitoring. In this paper, an improved gradient method (dynamic gradient descent) is proposed. The idea is to take account of the dependency of prediction errors and calculate the gradient information recursively. Not only low computation expense is achieved, but the non-Gaussian errors can also be overcome when this approach is applied to estimate parameters of a reduced order gas turbine model and to improve long-term prediction

High-gain observer-based parameter identification with application in a gas turbine engine

Abstract In this paper, a novel identification technique, that is high-gain observer-based identification approach, is proposed for systems with bounded process and measurement noises. For system parameters with abnormal changes, an adaptive change detection and parameter identification algorithm is next presented. The presented technique and algorithm is finally applied to the parameter identification of the gas turbine engine by using the recorded input data from the engine test-bed. The identified parameters and the response curves are desired. The simulations have proved the effectiveness of the proposed procedure compared with the previous identification approach.

Tuning of digital PID controller parameters using local optimal control

This paper introduces a new tuning technique for digital PID controller parameters. This technique based on the modified local optimal controller parameters for certain reduced order model structures. Different methods are used to obtain the local optimal controller parameters for these reduced order structures. The new tuning technique for digital PID controller parameters is compared with the existing genetic algorithm technique for tuning digital PID controller parameters. Experimental results on a lab-based test rig confirm the effectiveness of this new technique over the genetic algorithm technique from robust performance point of view.

Zero assignment for robust H 2 /H ∞ fault detection filter design

In practical engineering, it is inevitable that a system is perturbed by noise signals. Unfortunately, H infin /H infin filtering may fail to detect some faults when the noise distribution matrix are the same as the fault distribution matrix. In this paper, it is shown that the dynamic feedback gain of a dynamic filter introduces additional zeros to the filter, and both the filter poles and the additional zeros can be assigned arbitrarily. In order to attenuate band-limited noises, the zero assignment technique is used, and an optimal dynamic fault detection filtering approach is proposed by locating the zeros to the noise frequencies and optimizing the poles. Compared to other dynamic filter design approaches, the zero assignment technique gives a better tradeoff between more design freedom and computation costs. As shown in the simulation, a better noise attenuation and fault detection performance have been obtained. The zero assignment in multivariable fault detection filter design would be the main contribution of this paper.

Disturbance Attenuation and Fault Detection Via Zero-Pole Assignment: A Dynamic Observer Approach

Abstract In this paper, a zero-pole assignment approach is proposed for disturbance attenuation in dynamic fault detection observer design. The properties of dynamic observers are analysed and it is shown that the poles of a dynamic observer can be shifted and the additional zeros can be assigned arbitrarily. Then, a novel pole-zero assignment approach is proposed and its application to a continuous system is presented. In the simulation, the disturbances are low-frequency signals (

Estimation delay compensation in high-gain observer-based parameter identification

In this paper, the properties of the disturbance estimation in a high gain observer are analyzed, and a time delay calculation method is proposed to improve the performance of parameter identification. This paper shows that the estimation delay depends on the observer gain, but has nothing to do with the parameter variation. Then, a novel algorithm is proposed to calculate the delay according to the phase response of disturbance estimation transfer function. Hence, in terms of system identification, this delay is compensated by lagging the state estimate by the same amount of delay. This technique, on one side, improves the performance of parameter identification accuracy, and, on the other side, gives a new insight into the high-gain observer design.