A Neural Network Strategy for Learning of Nonlinearities Toward Feed-Forward Control of Pressure-Compensated Hydraulic Valves With a Significant Dead Zone

2018 ◽  
Author(s):  
Jarmo Nurmi ◽  
Mohammad M. Aref ◽  
Jouni Mattila
Keyword(s):  
Adaptive Learning ◽  
Dead Zone ◽  
Effective Means ◽  
Previous Method ◽  
Deep Convolutional Neural Networks ◽  
Feed Forward ◽  
Feed Forward Control ◽  
Highly Nonlinear ◽  
Hydraulic Manipulator ◽  
Hydraulic Valves

A velocity feed-forward-based strategy is an effective means for controlling a heavy-duty hydraulic manipulator; in particular, a typical valve-controlled hydraulic manipulator, to compensate for valve dead-zone and other hydraulic valve nonlinearities. Based on our previous work on the adaptive learning of valve velocity feed-forwards, manually labelling and identifying the dead-zones and the other nonlinearities in the velocity feed-forward curves of pressure-compensated hydraulic valves can be avoided. Nevertheless, it may take two to three minutes or more per actuator to identify a pressure-compensated valve’s highly nonlinear velocity feed-forward in real-time with an adaptive approach, which should be reduced for realistic applications. In this paper, inspired by brain signal analysis technologies, we propose a new method based on deep convolutional neural networks comparing with the previous method to significantly reduce this online learning process with the strong nonlinearities of pressure-compensated hydraulic valves. We present simulation results to demonstrate the effectiveness of the deep learning-based learning method compared to the previous results with an adaptive control-based learning.

Automated Feed-Forward Learning for Pressure-Compensated Mobile Hydraulic Valves With Significant Dead-Zone

2017 ◽  
Author(s):  
Jarmo Nurmi ◽  
Jouni Mattila
Keyword(s):  
Motion Control ◽  
Dead Zone ◽  
Heavy Duty ◽  
Automated Identification ◽  
Forward Models ◽  
Feed Forward ◽  
Loop Control ◽  
Hydraulic Manipulators ◽  
Hydraulic Manipulator ◽  
Hydraulic Valves

Hydraulic manipulators on mobile machines, whose hydraulic actuators are usually controlled by mobile hydraulic valves, are being considered for robotic closed-loop control. A feed-forward-based strategy combining position and velocity feedback has been found to be an effective method for the motion control of pressure-compensated mobile hydraulic valves that have a significant dead zone. The feed-forward can be manually identified. However, manually identifying the feed-forward models for each valve-actuator pair is often very time-consuming and error-prone. For this practical reason, we propose an automated feed-forward learning method based on velocity and position feedback. We present experimental results for a heavy-duty hydraulic manipulator on a forest forwarder to demonstrate the effectiveness of the proposed method. These results motivate the automated identification of velocity feed-forward models for motion control of heavy-duty hydraulic manipulators controlled by pressure-compensated mobile hydraulic valves that have a significant input dead zone.

Feed Forward Control for Pneumatic Forearm Prosthesis, Manipulator

2012 ◽  
Author(s):  
Antoine Andrieu ◽  
Nobutaka Tsujiuchi ◽  
Takayuki Koizumi ◽  
Sho Maeda ◽  
Yoichiro Nakamura
Keyword(s):  
Dead Zone ◽  
Artificial Muscle ◽  
Great Accuracy ◽  
Human Hand ◽  
Hysteresis Cycle ◽  
Human Beings ◽  
Feed Forward ◽  
Pneumatic Actuators ◽  
Feed Forward Control ◽  
Rubber Balloon

In order to reach a better productivity in numerous fields, robots have been wildly used for automatic tasks. The main issue of a robot is his lack of adaptability, which is one of the most important ability of human beings. The best tool for adaptability is the human hand. Using pneumatic actuators to drive a robot forearm grants the possibility to move light articles with accuracy and without harming it. It is then necessary to develop accurate model of the different types of actuators. As every human muscle are simulated by those actuators, the size of these must vary and so their properties as well. These artificial muscle-type pneumatic actuators are composed of a rubber balloon, a net, a feeding channel and finally two anchors at both end of the net. Starting with the simplest linear model, we increased the complexity of the model to finally obtain a powerful control of the plastic muscle. To obtain the data we needed for the model, we used a test bench for pneumatic actuators and adjust it to fit the new size of the current actuator. The main difficulty to control the actuator is to find a way to overcome its hysteresis cycle. Then once the model is ready we need to use it to control the different part of the pneumatic forearm. The previous wrist and hand control is enhanced by feed forward control and can perform motion with a great accuracy. The hand can therefore be used without any electronic devices on it and remains lighter and safer. Previous starting dead-zone has been understood and avoided.

On the Dynamic Response of Actuation Devices in Open and Closed-Loop Mechanisms With Nonlinear Dynamics

2014 ◽  
Author(s):  
Jahangir Rastegar ◽  
Dake Feng
Keyword(s):  
Nonlinear Dynamics ◽  
Dynamic Response ◽  
Closed Loop ◽  
Mechanical Systems ◽  
Open Loop ◽  
Feed Forward ◽  
Feed Forward Control ◽  
Actuation System ◽  
Highly Nonlinear ◽  
Actuation Devices

This paper presents a study of the dynamic response of actuation devices used in mechanical systems with open and closed-loop linkage mechanisms and highly nonlinear dynamics such as robot manipulators. The study shows that the actuation forces/torques provided by actuation devices can be divided into two basic groups. The first group corresponds to the components of each actuator force/torque that is “actuator motion independent”. The dynamic response of this group is relatively high and limited only by the dynamic response limitations — for the case of electrically driven actuation systems — of the driving power amplifiers, electronics, computational and signal processing devices and components. The second group corresponds to those components of the actuator forces/torques that is “actuator motion dependent”. The dynamic response of this group is relatively low and dependent on the actuator effective inertial load and actuation speed. In all mechanical systems that are properly designed, the dynamic response of the first group is significantly higher than those of the second group. By separating the required actuating forces/torques into the above two groups, the dynamic response of such nonlinear dynamics systems may be determined for a given synthesized trajectory. The information can also be used to significantly increase the performance of mechanical systems. When a feed-forward control signal is used, the performance of the system is shown to be significantly improved by generating each one of the group of actuation components separately considering the dynamic response of the actuation system to each group of components. A method for separating the actuation forces/torques into the said “actuator motion independent” and “actuator motion dependent” groups for mechanical systems with open-loop and closed-loop linkage mechanisms is provided. Provided examples include an open-loop manipulators with feed-forward trajectory control and a closed-loop mechanism, both with highly nonlinear dynamics. Practical methods for implementing the proposed feed-forward control for nonlinear dynamics systems are discussed.

scholarly journals An Improved Frequency Dead Zone with Feed-Forward Control for Hydropower Units: Performance Evaluation of Primary Frequency Control

Energies ◽  
2019 ◽  
Vol 12 (8) ◽  
pp. 1497 ◽  
Author(s):  
Hao An ◽  
Jiandong Yang ◽  
Weijia Yang ◽  
Yuanchu Cheng ◽  
Yumin Peng
Keyword(s):  
Frequency Control ◽  
Dead Zone ◽  
Hydropower Plant ◽  
Feed Forward ◽  
Feed Forward Control ◽  
Wear And Tear ◽  
Integral Quantity ◽  
Frequency Fluctuations ◽  
Primary Frequency ◽  
Primary Frequency Control

Due to the integration of more intermittent renewable energy into the power grid, the demand for frequency control in power systems has been on the rise, and primary frequency control of hydropower units plays an increasingly important role. This paper proposes an improved frequency dead zone with feed-forward control. The aim is to achieve a comprehensive performance of regulating rapidity, an assessment of integral quantity of electricity, and the wear and tear of hydropower units during primary frequency control, especially the unqualified performance of integral quantity of electricity assessment under frequency fluctuations with small amplitude. Based on a real hydropower plant with Kaplan units in China, this paper establishes the simulation model, which is verified by comparing experimental results. After that, based on the simulation of real power grid frequency fluctuations and a real hydropower plant case, the dynamic process of primary frequency control is evaluated for three aspects, which include speed, integral quantity of electricity, and wear and tear. The evaluation also includes the implementations of the three types of dead zones: common frequency dead zone, the enhanced frequency dead zone, and the improved frequency dead zone. The results of the study show that the improved frequency dead zone with feed-forward control increases the active power output under small frequency fluctuations. Additionally, it alleviates the wear and tear problem of the enhanced frequency dead zone in the premise of guaranteeing regulation speed and integral quantity of electricity. Therefore, the improved frequency dead zone proposed in this paper can improve the economic benefit of hydropower plants and reduce their maintenance cost. Accordingly, it has been successfully implemented in practical hydropower plants in China.

Parallel Computing for Tire Simulations

2011 ◽  
Vol 39 (3) ◽  
pp. 193-209 ◽  
Author(s):  
H. Surendranath ◽  
M. Dunbar
Keyword(s):  
High Performance ◽  
Effective Means ◽  
Cost Effective ◽  
Turnaround Time ◽  
Careful Consideration ◽  
Rolling Contact ◽  
Element Analysis ◽  
Parallel Solvers ◽  
Design Changes ◽  
Highly Nonlinear

Abstract Over the last few decades, finite element analysis has become an integral part of the overall tire design process. Engineers need to perform a number of different simulations to evaluate new designs and study the effect of proposed design changes. However, tires pose formidable simulation challenges due to the presence of highly nonlinear rubber compounds, embedded reinforcements, complex tread geometries, rolling contact, and large deformations. Accurate simulation requires careful consideration of these factors, resulting in the extensive turnaround time, often times prolonging the design cycle. Therefore, it is extremely critical to explore means to reduce the turnaround time while producing reliable results. Compute clusters have recently become a cost effective means to perform high performance computing (HPC). Distributed memory parallel solvers designed to take advantage of compute clusters have become increasingly popular. In this paper, we examine the use of HPC for various tire simulations and demonstrate how it can significantly reduce simulation turnaround time. Abaqus/Standard is used for routine tire simulations like footprint and steady state rolling. Abaqus/Explicit is used for transient rolling and hydroplaning simulations. The run times and scaling data corresponding to models of various sizes and complexity are presented.

scholarly journals Model-based feed-forward control for time-varying systems with an example for SRF cavities

2020 ◽  
Vol 53 (2) ◽  
pp. 1331-1336
Author(s):  
Sven Pfeiffer ◽  
Annika Eichler ◽  
Holger Schlarb
Keyword(s):  
Time Varying ◽  
Feed Forward ◽  
Feed Forward Control ◽  
Model Based ◽  
Time Varying Systems ◽  
Srf Cavities

Flatness Feed-Forward Control Model Based on Neural Networks

2014 ◽  
Vol 989-994 ◽  
pp. 3386-3389
Author(s):  
Zhu Wen Yan ◽  
Hen An Bu ◽  
Dian Hua Zhang ◽  
Jie Sun
Keyword(s):  
Neural Networks ◽  
Rolling Mill ◽  
Rolling Force ◽  
Work Roll ◽  
Control Model ◽  
Good Effect ◽  
Feed Forward ◽  
Force Fluctuations ◽  
Feed Forward Control ◽  
Roll Bending

The influence on the shape of the strip from rolling force fluctuations has been analyzed. The combination of intermediate roll bending and work roll bending has been adopted. The principle of rolling force feed-forward control has been analyzed. The feed-forward control model has been established on the basis of neural networks. The model has been successfully applied to a rolling mill and a good effect has been achieved.

scholarly journals Feed-forward control of preshaping in the rat is mediated by the corticospinal tract

2010 ◽  
Vol 32 (10) ◽  
pp. 1678-1685 ◽  
Author(s):  
Jason B. Carmel ◽  
Sangsoo Kim ◽  
Marcel Brus-Ramer ◽  
John H. Martin
Keyword(s):  
Corticospinal Tract ◽  
Feed Forward ◽  
Feed Forward Control
Sign in / Sign up

Export Citation Format

Share Document