scholarly journals Closed-loop separation control using machine learning

2015 ◽  
Vol 770 ◽  
pp. 442-457 ◽  
Author(s):  
N. Gautier ◽  
J.-L. Aider ◽  
T. Duriez ◽  
B. R. Noack ◽  
M. Segond ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feedback Control ◽  
Genetic Programming ◽  
Closed Loop ◽  
Recirculation Zone ◽  
Low Frequency ◽  
Learning Control ◽  
Separation Control ◽  
Periodic Forcing ◽  
Model Free

We present the first closed-loop separation control experiment using a novel, model-free strategy based on genetic programming, which we call ‘machine learning control’. The goal is to reduce the recirculation zone of backward-facing step flow at $\mathit{Re}_{h}=1350$ manipulated by a slotted jet and optically sensed by online particle image velocimetry. The feedback control law is optimized with respect to a cost functional based on the recirculation area and a penalization of the actuation. This optimization is performed employing genetic programming. After 12 generations comprised of 500 individuals, the algorithm converges to a feedback law which reduces the recirculation zone by 80 %. This machine learning control is benchmarked against the best periodic forcing which excites Kelvin–Helmholtz vortices. The machine learning control yields a new actuation mechanism resonating with the low-frequency flapping mode instability. This feedback control performs similarly to periodic forcing at the design condition but outperforms periodic forcing when the Reynolds number is varied by a factor two. The current study indicates that machine learning control can effectively explore and optimize new feedback actuation mechanisms in numerous experimental applications.

scholarly journals Machine Learning Control for Experimental Turbulent Flow Targeting the Reduction of a Recirculation Bubble

2017 ◽  
Author(s):  
Camila Chovet ◽  
Marc Lippert ◽  
Laurent Keirsbulck ◽  
Bernd R. Noack ◽  
Jean-Marc Foucaut
Keyword(s):  
Machine Learning ◽  
Turbulent Flow ◽  
Recirculation Zone ◽  
Learning Control ◽  
Control Logic ◽  
Periodic Forcing ◽  
Recirculation Bubble ◽  
Turbulence Control ◽  
Model Free ◽  
Free Machine

We experimentally control the turbulent flow over backward-facing step (ReH = 31500). The goal is to modify the internal (Xr) and external (Lr) recirculation points and consequently the recirculation zone (Ar). A model-free machine learning control (MLC) is used as control logic. As benchmark, an optimized periodic forcing is employed. MLC generalizes periodic forcing by a multi-frequency actuation. In addition, a sensor-based control and a non-autonomous feedback, open- and closed-loop laws, were use to optimize the control. The MLC multi-frequency forcing outperforms, as expected, periodic forcing. The non-autonomous feedback brings a further improvement. The unforced and actuated flows have been investigated in real-time with a TSI particle image velocimetry (PIV) system. The current study shows that a generalization of multi-frequency forcing and sensor feedback significantly reduces the turbulent recirculation zone, far beyond optimized periodic forcing. The study suggests that MLC can effectively explore and optimize new feedback actuation mechanisms and we anticipate MLC to be a game changer in turbulence control.

scholarly journals Control Synthesis as Machine Learning Control by Symbolic Regression Methods

2021 ◽  
Vol 11 (12) ◽  
pp. 5468
Author(s):  
Elizaveta Shmalko ◽  
Askhat Diveev
Keyword(s):  
Machine Learning ◽  
Genetic Programming ◽  
Mathematical Formulation ◽  
Small Variation ◽  
Learning Control ◽  
Symbolic Regression ◽  
Control Technique ◽  
Control Synthesis ◽  
Basic Solution ◽  
Regression Methods

The problem of control synthesis is considered as machine learning control. The paper proposes a mathematical formulation of machine learning control, discusses approaches of supervised and unsupervised learning by symbolic regression methods. The principle of small variation of the basic solution is presented to set up the neighbourhood of the search and to increase search efficiency of symbolic regression methods. Different symbolic regression methods such as genetic programming, network operator, Cartesian and binary genetic programming are presented in details. It is shown on the computational example the possibilities of symbolic regression methods as unsupervised machine learning control technique to the solution of MLC problem of control synthesis for obtaining the stabilization system for a mobile robot.

scholarly journals Three Closed-Loop Feedback Control System with Dual Disturbance Observers of an Optoelectronic Stable Control Platform

Electronics ◽  
2020 ◽  
Vol 9 (2) ◽  
pp. 359 ◽  
Author(s):  
Hanwen Zhang ◽  
Yao Mao ◽  
Jiuqiang Deng ◽  
Huabo Liu
Keyword(s):  
Feedback Control ◽  
Disturbance Rejection ◽  
Closed Loop ◽  
Low Frequency ◽  
Tracking Accuracy ◽  
Visual Axis ◽  
Frequency Bands ◽  
Loop Feedback ◽  
Closed Loop Feedback ◽  
Stable Control

Disturbances presented in aeronautical imaging equipment can cause visual axis jitter, which directly leads to a reduction in closed-loop bandwidth and a decrease in tracking accuracy. The disturbance frequency affecting the stable control platform is mainly concentrated in the low- and middle-frequency bands, but the commonly used three closed-loop feedback control methods do not perform well in the disturbance rejection of those frequency bands. Moreover, the only disturbance observer in the acceleration loop cannot improve the low-band disturbance rejection capability due to the drift of the micro-electro-mechanical-system (MEMS) accelerometers in the low-frequency range. To solve these problems, this paper proposed dual disturbance observers (dual DOB) based on the disturbance information in the acceleration loop and the position loop. This design used two compensators to observe and compensate for the disturbances, which did not require additional sensors, and therefore did not increase system cost. Theoretical demonstrations and physical experiments showed that the designed method of the dual DOB not only improved the disturbance rejection capability of the low- and middle-frequency band of the optoelectronic stable control platform, but also improved the robustness of the system.

Active Damping by Acceleration Feedback Control and Redistribution of Stresses in the Structure

1999 ◽  
Author(s):  
Maxime P. Bayon de Noyer ◽  
Patrick J. Roberts ◽  
Sathya V. Hanagud
Keyword(s):  
Control System ◽  
Feedback Control ◽  
Vibration Control ◽  
Closed Loop ◽  
Driving Forces ◽  
Active Vibration Control ◽  
High Stress ◽  
Low Frequency ◽  
Acceleration Feedback ◽  
Active Vibration

Abstract In most structures, fatigue critical areas are associated with regions of high stresses. Passive stiffening of structures usually displaces these high stress regions. Thus, for most applications, active vibration control is preferred. However, the question of whether an active vibration control scheme involving a set of actuators will reduce stresses in the whole structure or create high stress areas in the vicinity of the actuators arises. In this paper, the stresses induced by an active vibration control system based on the use of an offset piezoceramic stack actuator with acceleration feedback control are investigated. Using a modal analysis of the actuator acting on a cantilever beam, a low frequency approximation of the actuator is developed in the form of a spring and two driving forces. Based on this approximation, a 3-D finite element simulation of the closed loop active vibration control system is developed and the closed loop stresses are studied.

Visual trajectory tracking of industrial manipulator with iterative learning control

2015 ◽  
Vol 42 (1) ◽  
pp. 54-63 ◽  
Author(s):  
Bingxi Jia ◽  
Shan Liu ◽  
Yi Liu
Keyword(s):  
Feedback Control ◽  
Trajectory Tracking ◽  
Control Strategy ◽  
Iterative Learning Control ◽  
Learning Control ◽  
Image Noise ◽  
Iterative Learning ◽  
Control Signal ◽  
Content Type ◽  
Model Free

Purpose – The purpose of this paper is to propose a more efficient strategy, which is easier to implement, i.e. the engineer can directly operate the target object without the robot to do a demonstration, and the manipulator is regulated to track the trajectory using vision feedback repetitively. Generally, the applications of industrial robotic manipulators are based on teaching playback strategy, i.e. the engineer should directly operate the manipulator to perform a demonstration and then the manipulator uses the recorded driving signals to perform repetitive tasks. Design/methodology/approach – In the teaching process, the engineer grasps the object with a camera on it to do a demonstration, during which a series of images are recorded. The desired trajectory is defined by the homography between the images captured at current and final poses. Tracking error is directly defined by the homography matrix, without 3D reconstruction. Model-free feedback-assisted iterative learning control strategy is used for repetitive tracking, where feed-forward control signal is generated by iterative learning control strategy and feedback control signal is generated by direct feedback control. Findings – The proposed framework is able to perform precise trajectory tracking by iterative learning, and is model-free so that the singularity problem is avoided which often occurs in conventional Jacobean-based visual servo systems. Besides, the framework is robust to image noise, which is shown in simulations and experiments. Originality/value – The proposed framework is model-free, so that it is more flexible for industrial use and easier to implement. Satisfactory tracking performance can be achieved in the presence of image noise. System convergence is analyzed and experiments are provided for evaluation.

scholarly journals Closed-Loop and Robust Control of Quantum Systems

2013 ◽  
Vol 2013 ◽  
pp. 1-11 ◽  
Author(s):  
Chunlin Chen ◽  
Lin-Cheng Wang ◽  
Yuanlong Wang
Keyword(s):  
Robust Control ◽  
Feedback Control ◽  
Quantum Control ◽  
Closed Loop ◽  
Learning Control ◽  
Quantum Systems ◽  
General Idea ◽  
Control Measurement ◽  
Control Of Quantum Systems ◽  
Control Feedback

For most practical quantum control systems, it is important and difficult to attain robustness and reliability due to unavoidable uncertainties in the system dynamics or models. Three kinds of typical approaches (e.g., closed-loop learning control, feedback control, and robust control) have been proved to be effective to solve these problems. This work presents a self-contained survey on the closed-loop and robust control of quantum systems, as well as a brief introduction to a selection of basic theories and methods in this research area, to provide interested readers with a general idea for further studies. In the area of closed-loop learning control of quantum systems, we survey and introduce such learning control methods as gradient-based methods, genetic algorithms (GA), and reinforcement learning (RL) methods from a unified point of view of exploring the quantum control landscapes. For the feedback control approach, the paper surveys three control strategies including Lyapunov control, measurement-based control, and coherent-feedback control. Then such topics in the field of quantum robust control asH∞control, sliding mode control, quantum risk-sensitive control, and quantum ensemble control are reviewed. The paper concludes with a perspective of future research directions that are likely to attract more attention.

scholarly journals Closed Loop Control of a Series Class-E Voltage-Clamped Resonant Converter for LED Supply with Dimming Capability

Electronics ◽  
2019 ◽  
Vol 8 (12) ◽  
pp. 1380 ◽  
Author(s):  
Javier Ribas ◽  
Pablo J. Quintana ◽  
Jesus Cardesin ◽  
Antonio J. Calleja ◽  
Juan M. Lopera
Keyword(s):  
Feedback Control ◽  
Closed Loop ◽  
Light Emitting Diode ◽  
Low Frequency ◽  
Resonant Converters ◽  
Closed Loop Control ◽  
Control Loop ◽  
Loop Control ◽  
Dimming Control ◽  
Class E

In this work, a new closed-loop control system is applied to a class-E resonant DC–DC converter with voltage clamp used for light-emitting diode (LED) supply. The proposed power topology was first described by Ribas et al. in a recent work. In the present paper, the LED current is sensed and used to implement a feedback control loop instead of the simplified feedforward scheme used in this previous reference. To design the control, a novel, simplified small-signal model is presented. This model is used to analyze the converter behavior as a function of the output power. The proposed approximation is significantly simpler than the multifrequency averaging technique normally used to analyze resonant converters. The feedback control loop is designed to reduce the LED low frequency current ripple while providing dimming control. Both the model and the control are verified by simulation and laboratory experimentation and the results obtained are in good accordance with the expected values.

scholarly journals Closed-loop separation control over a sharp edge ramp using genetic programming

2016 ◽  
Vol 57 (3) ◽  
Author(s):  
Antoine Debien ◽  
Kai A. F. F. von Krbek ◽  
Nicolas Mazellier ◽  
Thomas Duriez ◽  
Laurent Cordier ◽  
...  
Keyword(s):  
Genetic Programming ◽  
Closed Loop ◽  
Sharp Edge ◽  
Separation Control
Sign in / Sign up

Export Citation Format

Share Document