China
Africa
Explore chapters and articles related to this topic
Distributed Control Systems
Published in Richard L. Shell, Ernest L. Hall, Handbook of Industrial Automation, 2000
In modern control theory, the term self-tuning control [30] has been coined as alternative to adaptive control. In a self-tuning system control parameters are, based on measurements of system input and output, automatically tuned to result into a sustained optimal control. The tuning itself can be affected by the use of measurement results to: Estimate actual values of system parameters and, in the sequence, to calculate the corresponding optimal values of control parameters, or toDirectly calculate the optimal values of control parameters.
Parameter Estimation
Published in Alex Martynenko, Andreas Bück, Intelligent Control in Drying, 2018
So far, the unknown parameters have been considered time invariant either fixed or distributed. However, in general parameters may also vary during the process as result of unmodeled dynamics or changing process conditions, for example, fouling and deterioration processes. In such cases, it is advantageous to estimate parameters online. Well-known approaches include classical adaptive techniques, like recursive least squares techniques in which the parameters are updated with each measurement (Ljung, 1987), but also state observer techniques. The first are important elements of adaptive control algorithms, so-called self-tuning controllers. In each recursive step, the model parameters (and thereby the model) are updated with the current measurements, and the controller parameters are computed for this current model. More details on the adaptive control and self-tuning controllers are found in Chapter 7 of this book. The main idea of observers is to use a mathematical model of a process to infer information on nonmeasurable states of the system (Luenberger, 1971; Walter & Pronzato, 1997). To augment observers for simultaneous estimation of model parameters, the latter are considered as nonmeasurable (static) states of the corresponding processes. Some approaches, like Kalman filtering techniques, also allow estimation based on corrupted measurements. However, these methods are rather suited for processes described by models of medium complexity and medium-sized sets of unknown parameters. For complex systems with a high number of unknown model parameters, application may result in bad performance. Observer concepts have found broad application in many process engineering systems, for example, (bio)chemical processes (Ali et al., 2015) disperse systems (Bück et al., 2011), and drying processes (Velardi et al., 2009). In the latter, an extended Kalman filter–based soft sensor was developed for inline monitoring of the primary drying phase of the lyophilization of pharmaceuticals in vial using a simplified mono-dimensional model. Only the temperature at the bottom of the vial was measurable and the estimator was used to reconstruct the temperature and position of the drying front as well as mass effective diffusivity in the dried layer and the heat transfer coefficient between the shelf and the bottom of the product. The estimator shows a good performance for simulation scenarios and also for an experimental setup. However, the presented estimator is limited for state/parameter estimation of a single vial and it is mentioned that the estimator needs to be tuned.
An overview of self-engineering systems
Published in Journal of Engineering Design, 2021
Many electronic devices have built-in self-tuning or self-optimisation achieved using sensors and a feedback control circuits. Examples found in patents include: a medical lancing device (Alden and Freeman 2008) which prevents skin puncture from being too deep,an amplifier which adapts based on the signal strength and quality (Bayar and Quintall 2019; Sengupta et al. 2012),a phone which registers it has been dropped and tests the antenna, re-tuning it if needed (Asrani, Katragadda, and Ananthanarayanan 2016).