Ph.D. Positions

 

Suitably qualified people will be considered as candidates for PhDs in the Theory and Applications of Nonlinear Dynamics and Control Theory. If you have a  degree in the areas of Mathematics, Physics or Engineering and are interested in working in this field, please contact  Mario di Bernardo.
Funding could be available from grant awarding bodies, such as the EPSRC and the European Union.  Departmental bursaries may also be available.  Possible projects include the study of nonlinear phenomena in systems of relevance in Engineering, the analysis of bifurcations and chaos in piecewise smooth dynamical systems and more generally the modelling, analysis and control of complex dynamical systems, i.e.:
 

Statistical Properties of Power Converters: Analysis and Control

Recently, it has been proposed that power electronic converters can exhibit several types of bifurcations and chaos [1]. From a mathematical viewpoint, power converters can be modelled by piece-wise linear models, whose configuration changes according to a given control law. In applications these circuits are often used to generate voltage and current waveforms with prescribed statistical characteristics (probability density function and power spectral density). This is usually achieved by rather inefficient and expensive filtering actions. An interesting open problem is the characterisation of the stochastic properties of these converters when operating under chaotic regimes [2]. This, in facts, could be used to solve the inverse problem, i.e. to design chaotic circuits with the desired statistical properties.

This project is concerned with the analysis and control of the statistical properties of a widely used class of power converters, the so-called DC/DC converters [1]. In particular, we shall seek to analyse the stochastic properties of a set of  discrete maps derived from these systems [4]. The analysis will be developed by focussing on the average behaviour of voltage and current waveforms associated with power converters operating in chaotic regime.  In so doing we shall seek to characterise the probability density functions associated to different converter configurations and the related power spectral densities. The aim will be that of relating  these stochastic properties with the parameters of the circuits [3]. This in turn could lead to the interesting possibility of designing the spectral properties of the system under investigation by appropriately controlling its chaotic dynamics.  To this end, we shall seek to extend and apply the control approach based on the idea of stabilising an arbitrary invariant measure recently presented in [5]. We anticipate that this will result in the careful analysis of the inverse Frobenius-Perron Problem as outlined in [5].

References

[1]  J.B. Deane & D.C. Hamill, "Instability, subharmonics and chaos in power electronics systems",  IEEE Trans. on Power Electronics, 5:3, pp. 260-268, July 1990

[2] A. Baranovski, W. Schwarz, A. Moegel, “Statistical Analysis and Design of Chaotic Switched Dynamical Systems”, Proceeding IEEE ISCAS99, vol. 5,
pp. 467-470, May 1999

[3] A. Lasota, M.C. Mackey, “Probabilistic properties of determinstic systems”, Cambridge University Press, 1985

[4] M. di Bernardo, F. Vasca, “On Discrete Time Maps for the analysis of bifurcations and Chaos in DC/DC converters”, to appear IEEE Trans. on Circuits and Systems I, July 1999

[5] E. M. Bollt, “On the Inverse Frobenius-Perron Problem: Global Stabilisation of Arbitrary Invariant Measures”, to appear on International Journal of Bifurcations and Chaos, 1999
 
 

Controlling Nonlinear Systems using Chaos

 
Over the past few year, the problem of controlling nonlinear systems has become increasingly relevant in many areas of engineering and applied science. Nonlinear dynamical systems are used, in facts, to model electronic circuits, motors, machines and many other devices in applications [1].

At the department of Engineering Mathematics, we have recently devised new method to control this class of dynamical systems joining the wider research-effort to understand and exploit the occurrence of nonlinear phenomena such as bifurcations and deterministic chaos [2]-[4].

This project is concerned with the analysis and investigation of novel control schemes aimed at exploiting the chaotic nature of the systems involved to achieve given control objectives. Preliminary results show that by using chaos, optimal control strategies for example can be used effectively to  ame the dynamics of a fully nonlinear system. Surprisingly this seems to indicate the possibility of improved minimization algorithms, based on  Pontryagin's Maximum Principle, to solve effectively the control problems [5].

The research student will be involved with both the analytical and numerical investigation of these schemes and encouraged to develop new theoretical control strategies that make explicit use of  bifurcations and chaos exhibited by the systems involved.
Applications to communications, flight dynamics and mechanical engineering will be considered with the possibility of carrying out experimental investigations in the Automatic Control Laboratory of the University of Bristol. The work will be carried out in strict collaboration with coworkers at
the University of Texas at Houston, USA and the University of California at Berkeley and visits to both these institution will be carried

References
----------

[1] M.di Bernardo and G. Chen in  "Controlling Chaos and Bifurcations in Engineering Systems", CRC Press, 1999

[2] D.P. Stoten, M. di Bernardo, "An Application of the Minimal Control Synthesis Algorithm to the Control and Synchronization of Chaos", International Journal of Control, Vol. 6, pp. 925-938, December 1996.

[3] M. di Bernardo, "A purely adaptive controller to synchronize and control chaotic systems",Physics Letters A, May 13, vol. 214, 3/4,  p. 139, 1996.

[4] M. di Bernardo, "An Adaptive Approach to the Control and Synchronization of Continuous- Time Chaotic Systems",  International Journal of Bifurcation and Chaos, Vol.6, No. 3,  pp. 557-568, 1996.

[5] S. Perry, "Analysis and Control of a double inverted pendulum", M.Eng. Project, University of Bristol, 1999 (preliminary report)
 
 

Numerical Analysis of discontinuous dynamical systems


Most of the theory of nonlinear dynamics has focussed on the analysis of nonlinear phenomena for smooth dynamical systems. In so doing, efficient algorithms have been proposed to continue their solutions, analyse their Lyapunov exponents and in general characterise their nonlinear behaviour.

In many applications, though, the occurrence of discontinuous events (such as switchings in electronic circuits or impacts in mechanical devices) takes it necessary to consider nonsmooth or discontinous dynamical systems. Specifically, systems characterised by switching between different configurations whenever certain conditions are satisfied [2]-[4].

Understanding the complex behaviour often exhibited by this class of dynamical systems is often a cumbersome task.  It has been shown, for  example, that nonsmooth system can exhibit novel class of bifurcations termed border-collisions or grazing which have been observed in real-world systems [1]. Nevertheless, few algorithms have been proposed to investigate numerically the behaviour of this type of systems and even fewer have
been actually implemented.

The aim of this project is to study and implement appropriate methods to continue the solutions of a given piecewise smooth dynamical system,
compute its Lyapunov exponent, classify its bifurcations  and characterise its behaviour. The platform chosen is the powerful software package MATLAB and the associated toolbox SIMULINK which offer an ideal tool for numerical investigations  of these systems. As part of the project, specific routines in C or FORTRAN will also haveto be written.

This project is very well suited for students with a keen interest in Mathematics and Computer Science. It involves both aspects of theory and
computer programming. By the end of the research period, the student will be confident in an increasingly important area of modern Mathematics. At the same time he will be an expert in the  analysis and characterisation of  discontinuous dynamical systems which are extremely relevant in applications.
 

References

[1] M. di Bernardo, M.I. Feigin, S.J. Hogan, M.E. Homer, "Local analysis of C-bifurcations in n-dimensional piecewise-smooth dynamical systems",  Chaos, Solitons and Fractals, vol. 10, no. 11, pp. 1881-1908, 1999.

[2]  G.M. Maggio, M. di Bernardo, M.P. Kennedy,  "Nonsmooth bifurcations in a piecewise linear model of the Colpitts Oscillator",  IEEE Transactions on Circuits and Systems Part I (submitted) 1999.

[3]     M. di Bernardo, F. Vasca, "On Discrete Time Maps for the analysis of bifurcations and Chaos in DC/DC converters", IEEE Transactions on Circuits and Systems Part I (to appear).

[4]     M. di Bernardo, C.J. Budd, A.R. Champneys, "Grazing, Skipping and Sliding: analysis of the non-smooth dynamics of the DC/DC buck converter", Nonlinearity, vol. 11, no. 4, pp. 858-890, July 1998.

(see  here for some other examples)