Simplified Tuning of Long-Range Corrected Density Functionals for Use in Symmetry-Adapted Perturbation Theory

Long considered a failure, second-order symmetry-adapted perturbation theory (SAPT) based on Kohn-Sham orbitals, or SAPT(KS), can been resurrected for semiquantitative purposes using long-range corrected (LRC) density functionals whose asymptotic behavior is adjusted separately for each monomer. As in other contexts, correct asymptotic behavior can be enforced via "optimal tuning" of LRC functionals, based on the ionization energy theorem, but the tuning procedure is tedious, expensive for large systems, and comes with a troubling dependence on system size. Here, we show that essentially identical results are obtained using an automated tuning procedure based on the size of the exchange hole, making tuned "SAPT(wKS)" fast and convenient. In conjunction with SAPT-based methods that sidestep second-order dispersion, this procedure achieves benchmark-quality interaction energies, along with the usual SAPT energy decomposition, without the hassle of system-specific tuning.

Simplified Tuning of Long-Range Corrected Density Functionals for Use in Symmetry-Adapted Perturbation Theory

Long considered a failure, second-order symmetry-adapted perturbation theory (SAPT) based on Kohn-Sham orbitals, or SAPT(KS), can been resurrected for semiquantitative purposes using long-range corrected (LRC) density functionals whose asymptotic behavior is adjusted separately for each monomer. As in other contexts, correct asymptotic behavior can be enforced via "optimal tuning" of LRC functionals, based on the ionization energy theorem, but the tuning procedure is tedious, expensive for large systems, and comes with a troubling dependence on system size. Here, we show that essentially identical results are obtained using an automated tuning procedure based on the size of the exchange hole, making tuned "SAPT(wKS)" fast and convenient. In conjunction with SAPT-based methods that sidestep second-order dispersion, this procedure achieves benchmark-quality interaction energies, along with the usual SAPT energy decomposition, without the hassle of system-specific tuning.

Automated tuning of a piezoelectric power harvesting cantilever beam via the application of an axial load

This thesis deals with automatic tuning of piezoelectric power harvesters. A prototype was constructed to verify the effect of the application of an axial load on a cantilever beam and the effectiveness of increasing the power harvested from a piezoelectric beam via axial loading. It was shown, experimentally, that the natural frequency of a piezoelectric beam harvester can be changed over a range of 25Hz. From the experimental results it was shown that the power can increase up to seven times when tuned in comparison to unturned. Computer simulations were used to demonstrate a closed loop tuning system’s ability to apply an axial load effectively onto a piezoelectric cantilever beam in response to ambient vibrations. The closed loop tuning system is a viable option for tuning and can lead to increased power when applying the axial load suggested by the tuning system.

Automated tuning of a piezoelectric power harvesting cantilever beam via the application of an axial load

This thesis deals with automatic tuning of piezoelectric power harvesters. A prototype was constructed to verify the effect of the application of an axial load on a cantilever beam and the effectiveness of increasing the power harvested from a piezoelectric beam via axial loading. It was shown, experimentally, that the natural frequency of a piezoelectric beam harvester can be changed over a range of 25Hz. From the experimental results it was shown that the power can increase up to seven times when tuned in comparison to unturned. Computer simulations were used to demonstrate a closed loop tuning system’s ability to apply an axial load effectively onto a piezoelectric cantilever beam in response to ambient vibrations. The closed loop tuning system is a viable option for tuning and can lead to increased power when applying the axial load suggested by the tuning system.

The case for NLP-enhanced database tuning

A large body of knowledge on database tuning is available in the form of natural language text. We propose to leverage natural language processing (NLP) to make that knowledge accessible to automated tuning tools. We describe multiple avenues to exploit NLP for database tuning, and outline associated challenges and opportunities. As a proof of concept, we describe a simple prototype system that exploits recent NLP advances to mine tuning hints from Web documents. We show that mined tuning hints improve performance of MySQL and Postgres on TPC-H, compared to the default configuration.

An inquiry into machine learning-based automatic configuration tuning services on real-world database management systems

Modern database management systems (DBMS) expose dozens of configurable knobs that control their runtime behavior. Setting these knobs correctly for an application's workload can improve the performance and efficiency of the DBMS. But because of their complexity, tuning a DBMS often requires considerable effort from experienced database administrators (DBAs). Recent work on automated tuning methods using machine learning (ML) have shown to achieve better performance compared with expert DBAs. These ML-based methods, however, were evaluated on synthetic workloads with limited tuning opportunities, and thus it is unknown whether they provide the same benefit in a production environment. To better understand ML-based tuning, we conducted a thorough evaluation of ML-based DBMS knob tuning methods on an enterprise database application. We use the OtterTune tuning service to compare three state-of-the-art ML algorithms on an Oracle installation with a real workload trace. Our results with OtterTune show that these algorithms generate knob configurations that improve performance by 45% over enterprise-grade configurations. We also identify deployment and measurement issues that were overlooked by previous research in automated DBMS tuning services.

AUTOMATIC SYNTHESIS OF PETRI NETS AT TUNING UP OF THE COORDINATING AUTOMATIC CONTROL SYSTEMS

The process of automated tuning for the coordinating automatic control system is considered in this paper. This process of tuning for the coordinating control system is linked to the automatic synthesis of Petri nets based on functioning of the artificial neural network. Thereby, we can automate the process of tuning and synthesis of system models and also solve the urgent task linked to the minimization of tuning time for the multilevel control systems. The purposes of the scientific work are time reduction of the tuning and automatization of the tuning for the multilevel coordinating systems of the automatic control. In order to achieve this purpose in the MATLAB \ Simulink software environment it is necessary to devel- op the system for automated tuning of the regulators of various levels for the coordinating automatic control system. The application of artificial neural network with automatic synthesis of Petri nets allows to introduce intelligent technology in the automated tuning system. In this work we have presented the corresponding block diagrams of considered automated tuning system and the principles of its functioning. The certain principle of the formation of Petri nets is proposed. These Petri nets represent the algorithms of tuning in the systems for analysis the corresponding processes. The formation of the composition in the scheme from Petri net during the functioning of the artificial neural network is presented in the paper. The results of experiment are presented in the final part of this work. This time characteristics of the pro- cess of setting up for the coordinating automatic control system of foodstuffs cooling in tunnel chamber. The experiments were conducted in the Matlab 2012a environment. Based on the results of the experiment we have depicted the process of synthesis of the Petri net representing the system tuning algorithm. The performed experiments have showed the principal suitability of the automated search system for the settings of the regulators of various levels of the coordinating control system. The technique of automatic synthesis of Petri nets based on the functioning of artificial neural networks has obtained the further devel- opment while performing the approved task in the scientific paper.