1,721,111 research outputs found
Identification of gene interaction networks based on evolutionary computation
No full textThis paper investigates applying a genetic algorithm and an evolutionary programming for identification of gene interaction networks from gene expression data. To this end, we employ recurrent neural networks to model gene interaction networks and make use of an artificial gene expression data set from literature to validate the proposed approach. We find that the proposed approach using the genetic algorithm and evolutionary programming can result in better parameter estimates compared with the other previous approach. We also find that any a priori knowledge such as zero relations between genes can further help the identification process whenever it is available
Event-based intelligent control of saturated chemical plant using an endomorphic neural network model
No full textAn event-based control system with an endomorphic neural network model is designed and realized to control a saturated non-linear plant. The scheme employed in this system is based on an event-based control paradigm previously proposed to control monotonic plants. However, this scheme is different from the previous one in that it can be used to control plants with saturation property. This new scheme may be viewed as a combined method of a time-based diagnosis mechanism in an event-based control system and a state-based control mechanism in a neural network control system. A chemical plant having strong non-linearity and complicated dynamics is controlled using this realized event-based control system. This paper discusses the structure of an event-based controller, the neural network modelling methodology, some related problems, and the simulation results.Dr J.J. Song at MI
EVENT-BASED INTELLIGENT CONTROL USING ENDOMORPHIC NEURAL-NETWORK MODEL
No full textIn event-based control, a controller checks the responses of sensors about commands with time constraints. To do this, the event-based controller should have some information about the dynamics of the plant at discrete levels, its desired state transitions, and inputs to move the state transitions. In an existing modelling method, the information is represented by a tabular form, which is not adaptable to the variation of set positions. An artificial neural network was taken as a new modelling method to solve this problem. Experiments show that this neural network model works well in the dynamic variation of set positions. This endomorphic neural network modelling helps us to construct a more autonomous event-based controller
Investigations into the design principles in the chemotactic behavior of Escherichia coli
No full textInspired by the recent studies on the analysis of biased random walk behavior of Escherichia coli[Passino, K.M., 2002. Biomimicry of bacterial foraging for distributed optimization and control. IEEE Control Syst. Mag. 22 (3), 52-67; Passino, K.M., 2005. Biomimicry for Optimization, Control and Automation. Springer-Verlag, pp. 768-798; Liu, Y., Passino, K.M., 2002. Biomimicry of social foraging bacteria for distributed optimization: models, principles, and emergent behaviors. J.Optim. Theory Appl. 115 (3), 603-628], we have developed a model describing the motile behavior of E. coli by specifying some simple rules on the chemotaxis. Based on this model, we have analyzed the role of some key parameters involved in the chemotactic behavior to unravel the underlying design principles. By investigating the target tracking capability of E. coli in a maze through computer simulations, we found that E. coli clusters can be controlled as target trackers in a complex micro-scale-environment. In addition, we have explored the dynamical characteristics of this target tracking mechanism through perturbation of parameters under noisy environments. It turns out that the E coli chemotaxis mechanism might be designed such that it is sensitive enough to efficiently track the target and also robust enough to overcome environmental noises. (C) 2007 Elsevier Ireland Ltd. All rights reserved
Use of an Adaptive Window in PID-plus Bang-Bang Control : A Motor Control Experiment
No full textPID-type controllers have been widely used in many industrial applications. Regulation properties of those can be improved through the addition of the Bang-Bang action. In spite of the potentials of this PID-plus Bang-Bang controller, their regulation properties are still limited if a fixed window limit is used in selection of a control action between PID and Bang-Bang action. Thus, this paper proposes an approach to improving regulation properties. Our approach changes window limits adaptable to plant dynamics by use of a Gradient Based Prediction Model. We experimented our control scheme with a DC servo-motor system. It has been shown through experiment that our control scheme outperformed than existing one in terms of overshoot, rise time and settling time.The authors would like to thank reviewers. Their helpful comments enalble us to improve the quality of this paper
Interstitial oxygen incorporation into silicon substrate during plasma enhanced atomic layer deposition of Al2O3
No full textThe incorporation of oxygen into a silicon substrate was investigated during the growth of Al2O3 gate oxide. Al2O3 films were grown on p-type (100) silicon wafers, at 100 degrees C, using plasma enhanced atomic layer deposition. Methylpyrrolidine alane (C5H14NAl) and capacitively coupled O-2 plasma were used as the sources of Al and O, respectively. The interstitial oxygen in the silicon substrate was found by Fourier transform infrared spectroscopy, and the amount of interstitial oxygen found increased when the thickness of Al2O3 was increased. This phenomenon was attributed to the diffusion of oxygen atoms throughout the Al2O3 layer. (c) 2005 The Electrochemical Society
Ripple Feedback Filter Suitable Suitable for Analog/Digital Mixed-Mode Audio Amplifier for Improved Efficiency and stability
No full text- …
