Jonathan Edward Cooper

Dynamic Multivariate Statistical Process Control using Subspace Identification

In this article, the monitoring of continuous processes using linear dynamic models is presented. It is outlined that dynamic extensions to conventional multivariate statistical process control (MSPC) models may lead to the inclusion of large numbers of variables in the condition monitor. To prevent this, a new dynamic monitoring scheme, based on subspace identification, is introduced, which can (1) determine a set of state variables for describing process dynamics, (2) produce a reduced set of variables to monitor process performance and (3) offer contribution charts to diagnose anomalous behaviour. This is demonstrated by an application study to a realistic simulation of a chemical process.

A Comparison of Blade Tip-Timing Data Analysis Methods

Abstract The experimental determination of the vibration characteristics of rotating engine blades is very important for fatigue failure considerations. One of the most promising techniques for measuring the frequency of blade vibrations is blade tip timing. In this paper, three vibration analysis methods were specifically formulated and applied to the tip timing problem for the first time, using data obtained from a simple mathematical blade tip timing simulation. The results from the methods were compared statistically in order to determine which of the techniques is more suitable. One of the methods, the global autoregressive instrumental variables approach, produced satisfactory results at realistic noise levels. However, all of the techniques produced biased results under certain circumstances.

Flutter Prediction from Flight Flutter Test Data

The most common approach to flight flutter testing is to track estimated modal damping ratios of an aircraft over a number of flight conditions. These damping trends are then extrapolated to predict whether it is safe to move to the next test point and also to determine the flutter speed. In the quest for more reliable and efficient flight flutter testing procedures, a number of alternative data analysis methods have been proposed. Five of these approaches are compared on two simulated aeroelastic models. The comparison is based on both the accuracy of prediction and the efficiency of each method. It is found that, for simple aeroelastic systems, the Nissim and Gilyard method (Nissim, E., and Gilyard, G. B., Method for Experimental Determination of Flutter Speed by Parameter Identification, AIAA Paper 89-1324, 1989) yields the best flutter predictions and is also the least computationally expensive approach. However, for larger systems, simpler approaches such as the damping fit and envelope function methods are found to be most reliable.

Development of moving spars for active aeroelastic structures

This paper describes a research program investigating the development of “moving spars” to enable active aeroelastic control of aerospace structures. A number of different concepts have been considered as part of the EU funded Active Aeroelastic Aircraft Structures (3AS) project that enable the control of the bending and torsional stiffness of aircraft wings through changes in the internal aircraft structure. The aeroelastic behaviour, in particular static deflections, can be controlled as desired through changes in the position, orientation and stiffness of the spars. The concept described in this paper is based upon translational movement of the spars. This will result in changes in the torsional stiffness and shear centre position whilst leaving the bending stiffness unaffected. An analytical study of the aeroelastic behaviour demonstrates the benefits of using such an approach. An experimental investigation involving construction and bench testing of the concepts was undertaken to demonstrate its feasibility. Finally, a wind tunnel test of simple wing models constructed using these concepts was performed. The simulated and experimental results show that it is possible to control the wind twist in practice.

Development of An Adaptive Stiffness All-Moving Vertical Tail

An investigation into the aeroelastic characteristics of a wind tunnel model all-movable vertical tail with variable attachment position and stiffness has been made. An initial analytical study investigated the effect of varying the torsional stiffness and position of the single root attachment. An attachment was designed and manufactured that enabled an existing wind tunnel model to behave as an all-moving fin in order to validate the analytical results. This initial study showed that in order to get the full benefits of such a design, whilst still meeting aeroelastic constraints, the torsional stiffness must be adaptive so that it can be adjusted at different regions of the flight envelope. A further variable stiffness attachment was then designed and manufactured that enabled fully adaptive torsional stiffness control. Bench-top tests were performed to validate the structural behaviour of the models, followed by wind tunnel tests to examine the aeroelastic characteristics of the vertical tail models, in particular the aeroelastic effectiveness. The experimental results compared well with theoretical predictions. It was shown that it is possible to control the torsional stiffness, and hence the aeroelastic characteristics, of the all-moving vertical tail using the adaptive stiffness devi ce.

Limit Cycle Oscillation Control and Suppression

The prediction and characterisation of the limit cycle oscillation (LCO) behaviour of non-linear aeroelastic systems has become of great interest recently. However, much of this work has concentrated on determining the existence of LCOs. We concentrate on LCO stability. By considering the energy present in different limit cycles, and also using the harmonic balance method, it is shown how the stability of limit cycles can be determined. The analysis is then extended to show that limit cycles can be controlled, or even suppressed, by the use of suitable excitation signals. A basic control scheme is developed to achieve this, and is demonstrated on a simple simulated non-linear aeroelastic system

A method for identification of non-linear multi-degree-of-freedom systems

System identification methods for non-linear aeroelastic systems could find uses in many aeroelastic applications such as validating finite element models and tracking the stability of aircraft during flight flutter testing. The effectiveness of existing non-linear system identification techniques is limited by various factors such as the complexity of the system under investigation and the type of non-linearities present. In this work, a new approach is introduced which can identify multi-degree-of-freedom systems featuring any type of non-linear function, including discontinuous functions. The method is shown to yield accurate identification of three mathematical models of aeroelastic systems containing a wide range of structural non-linearities.

On the use of control surface excitation in flutter testing

Abstract Flutter testing is aimed at demonstrating that the aircraft flight envelope is flutter free. Response measurements from deliberate excitation of the structure are used to identify and track frequency and damping values against velocity. In this paper, the common approach of using a flight control surface to provide the excitation is examined using a mathematical model of a wing and control surface whose rotation is restrained by a simple actuator. In particular, it is shown that it is essential to use the demand signal to the actuator as a reference signal for data processing. Use of the actuator force (or strain) or control angle (or actuator displacement) as a reference signal is bad practice because these signals contain response information. It may also be dangerous in that the onset of flutter may not be seen in the test results.

On the Use of Adaptive Internal Structures to Optimise Wing Aerodynamic Distribution

There is much interest at the moment in the use of Adaptive Stiffness Structures for aircraft due to the potential of improved drag performance as well as roll and loads control, making changes in the internal structure rather than using traditional control surfaces. Previous work has demonstrated the feasibility of implementing a number of different adaptive aeroelastic concepts and some of this research is discussed. Consideration of the stiffness distribution that is required to meet aerodynamic performance requirements whilst meeting structural and aeroelastic constraints is made. The use of a genetic and particle swarm optimisation algorithms is described in order to be able to deal with the large number of different possible design cases that arise from even from the most simple of design cases. Some sample results from a simple rectangular wing structure are shown. The relationship between the desired stiffness distribution and what could be achieved using more sophisticated adaptive aeroelastic structures in practice is considered.

A time–frequency technique for the stability analysis of impulse responses from nonlinear aeroelastic systems

Abstract A time–frequency method is proposed for the analysis of response time histories from nonlinear aeroelastic systems. The approach is based on a time-varying curve-fit of the short time Fourier transform of the impulse response. It is shown that the method can be used in order to obtain a clear picture of the sub-critical stability of a number of aeroelastic systems with a variety of structural and aerodynamic nonlinearities. Additionally, frequency and amplitude information can be obtained for both the linear and nonlinear signatures of the response signals in the sub- and post-critical regions. Finally, it is shown that, given certain types of nonlinear functions, sub-critical damping trends can be extrapolated to predict bifurcation airspeeds.