Dynamic control and information processing in chemical reaction systems by tuning self-organization behavior

Signal design using nonlinear oscillators and evolutionary algorithms: application to phase-locked loop disruption

This work describes an approach for efficiently shaping the response characteristics of a fixed dynamical system by forcing with a designed input. We obtain improved inputs by using an evolutionary algorithm to search a space of possible waveforms generated by a set of nonlinear, ordinary differential equations (ODEs). Good solutions are those that result in a desired system response subject to some input efficiency constraint, such as signal power. In particular, we seek to find inputs that best disrupt a phase-locked loop (PLL). Three sets of nonlinear ODEs are investigated and found to have different disruption capabilities against a model PLL. These differences are explored and implications for their use as input signal models are discussed. The PLL was chosen here as an archetypal example but the approach has broad applicability to any input∕output system for which a desired input cannot be obtained analytically.

Targeting characteristic wave properties in reaction-diffusion systems by optimization of external forcing

We consider the targeted manipulation of reaction-diffusion waves by optimization of an external forcing parameter. As an example, we present numerical results for the FitzHugh-Nagumo system exploiting model-based optimization capable of targeting characteristic wave properties such as wavelength, shape, and propagation speed by spatiotemporally controlling electric current. The conceptual basis of our approach is optimal control of periodic orbits in a wave-variable coordinate system. The results are transferred back to the partial differential equation context and validated in numerical simulations. The whole procedure is applicable to any reaction-diffusion model.

Annihilation of limit-cycle oscillations by identification of critical perturbing stimuli via mixed-integer optimal control

We present a novel model-based mixed-integer optimal control method to automatically identify the strength and timing of critical external stimuli leading to the transient annihilation of limit-cycle oscillators. Biochemical oscillators of this type play a central role in regulating cellular rhythms. Their specific manipulation is a promising perspective to control biological functions by drugs and tailored treatment strategies. We demonstrate our new optimal control approach in an application to a biochemical model for oscillatory calcium signal transduction.

Real-time nonlinear feedback control of pattern formation in (bio)chemical reaction-diffusion processes: a model study

Theoretical and experimental studies related to manipulation of pattern formation in self-organizing reaction-diffusion processes by appropriate control stimuli become increasingly important both in chemical engineering and cellular biochemistry. In a model study, we demonstrate here exemplarily the application of an efficient nonlinear model predictive control (NMPC) algorithm to real-time optimal feedback control of pattern formation in a bacterial chemotaxis system modeled by nonlinear partial differential equations. The corresponding drift-diffusion model type is representative for many (bio)chemical systems involving nonlinear reaction dynamics and nonlinear diffusion. We show how the computed optimal feedback control strategy exploits the system inherent physical property of wave propagation to achieve desired control aims. We discuss various applications of our approach to optimal control of spatiotemporal dynamics.

Specific external forcing of spatiotemporal dynamics in reaction-diffusion systems

Self-organization behavior and in particular pattern forming spatiotemporal dynamics play an important role in far from equilibrium chemical and biochemical systems. Specific external forcing and control of self-organizing processes might be of great benefit in various applications ranging from technical systems to modern biomedical research. We demonstrate that in a cellular chemotaxis system modeled by one-dimensional reaction-diffusion equations particular forms of spatiotemporal dynamics can be induced and stabilized by controlling spatially distributed influx patterns of a chemical species as a function of time. In our model study we show that a propagating wave with certain shape and velocity and static symmetrical and asymmetrical patterns can be forced and manipulated by numerically computing open-loop optimal influx controls.

Manipulation of surface reaction dynamics by global pressure and local temperature control: a model study

Specific catalyst design and external manipulation of surface reactions by controlling accessible physical or chemical parameters may be of great benefit for improving catalytic efficiencies and energetics, product yield, and selectivities in the field of heterogeneous catalysis. Studying a realistic spatiotemporal one-dimensional model for CO oxidation on Pt(110) we demonstrate the value and necessity of mathematical modeling and advanced numerical methods for directed external multiparameter control of surface reaction dynamics. At the model stage we show by means of optimal control techniques that species coverages can be adjusted to desired values, aperiodic oscillatory behavior for distinct coupled reaction sites can be synchronized, and overall reaction rates can be optimized by varying the surface temperature in space and time and the CO and O2 gas phase partial pressure with time. The control aims are formulated as objective functionals to be minimized which contain a suitable mathematical formulation for the deviation from the desired system behavior. The control functions pCO(t) (CO partial pressure), pO2(t) (O2 partial pressure), and T(x,t) (surface temperature distribution) are numerically computed by a specially tailored optimal control method based on a direct multiple shooting approach which is suitable to cope with the highly nonlinear unstable mode character of the CO oxidation model.