scholarly journals Adaptive PID Control of a Nonlinear Servomechanism Using Recurrent Neural Networks

10.5772/13020 ◽  
2011 ◽  
Author(s):  
Reza Jafari ◽  
Rached Dhaouadi
Keyword(s):  
Neural Networks ◽  
Pid Control ◽  
Recurrent Neural Networks ◽  
Adaptive Pid Control

Motion Control of a Piezoelectric Microprecision Positioner Using Diagonal Recurrent Neural Networks

2010 ◽  
Author(s):  
Jua-Liang Chen ◽  
Zong-Lin Chen
Keyword(s):  
Neural Networks ◽  
Motion Control ◽  
Pid Control ◽  
Recurrent Neural Networks ◽  
Tracking Error ◽  
System Model ◽  
Phase Lag ◽  
Tracking Errors ◽  
Positioning Errors ◽  
20 Nm

The objective of this paper is to design a suitable controller for a piezoelectric microprecision positioner. The dimensions of this positioner are 150 mm × 150 mm × 10 mm and with X-Y-θZ three degrees of freedoms. Piezo-electric actuators are used to drive the positioner, which is constructed by flexure structures. In order to improve the short stroke of PZT, simple-levers are added to the structures. In this research, a diagonal recurrent neural networks (DRNN) controller is added to the system with aim to reduce the effect causes by the hysteresis, inaccurate system model and phase lag, and to save time for adjusting control gains for PID control. From the experimental results, it shows that the positioning errors for the X-axis, Y-axis, and θ-axis of continuous stepping test are less than 20 nm and 0.15 μrad. For the ramp tracking test, the tracking errors are less than 30 nm and 0.3 μrad. For the circular tracking test, the tracking error is less than 55 nm for both X- and Y-axis.

Identification and adaptive PID Control of a hexacopter UAV based on neural networks

2018 ◽  
Vol 33 (1) ◽  
pp. 74-91 ◽  
Author(s):  
Claudio Rosales ◽  
Carlos Miguel Soria ◽  
Francisco G. Rossomando
Keyword(s):  
Neural Networks ◽  
Pid Control ◽  
Adaptive Pid Control

Research on Single Neuron Adaptive PID Control

2012 ◽  
Vol 150 ◽  
pp. 174-177 ◽  
Author(s):  
Yan Hong Zhang ◽  
De An Zhao ◽  
Jian Sheng Zhang
Keyword(s):  
Neural Networks ◽  
Intelligent Control ◽  
Pid Control ◽  
Control Algorithm ◽  
Single Neuron ◽  
Control Parameters ◽  
Control Effect ◽  
The Common ◽  
Single Neuron Pid ◽  
Adaptive Pid Control

As a branch of the intelligent control, neural networks is applied in control more and more widely, the single neuron adaptive PID control algorithm is studied in this paper, and the program is written by MATLAB, the common object of single neuron adaptive PID is simulated, and the effect of single neuron adaptive PID control parameters on control effect is analyzed, experimental results show that the single neuron PID control has more obvious advantages than general PID control.

Adaptive PID control based on orthogonal endocrine neural networks

2016 ◽  
Vol 84 ◽  
pp. 80-90 ◽  
Author(s):  
Miroslav B. Milovanović ◽  
Dragan S. Antić ◽  
Marko T. Milojković ◽  
Saša S. Nikolić ◽  
Staniša Lj. Perić ◽  
...  
Keyword(s):  
Neural Networks ◽  
Pid Control ◽  
Adaptive Pid Control

Levenshtein Augmentation Improves Performance of SMILES Based Deep-Learning Synthesis Prediction

2020 ◽  
Author(s):  
Dean Sumner ◽  
Jiazhen He ◽  
Amol Thakkar ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Deep Learning ◽  
Recurrent Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Sequence Similarity ◽  
Learning Models ◽  
Underlying Network

<p>SMILES randomization, a form of data augmentation, has previously been shown to increase the performance of deep learning models compared to non-augmented baselines. Here, we propose a novel data augmentation method we call “Levenshtein augmentation” which considers local SMILES sub-sequence similarity between reactants and their respective products when creating training pairs. The performance of Levenshtein augmentation was tested using two state of the art models - transformer and sequence-to-sequence based recurrent neural networks with attention. Levenshtein augmentation demonstrated an increase performance over non-augmented, and conventionally SMILES randomization augmented data when used for training of baseline models. Furthermore, Levenshtein augmentation seemingly results in what we define as <i>attentional gain </i>– an enhancement in the pattern recognition capabilities of the underlying network to molecular motifs.</p>

Sign in / Sign up

Export Citation Format

Share Document