scholarly journals Verification of Closed-loop Systems with Neural Network Controllers

10.29007/btv1 ◽  
2019 ◽  
Author(s):  
Diego Manzanas Lopez ◽  
Patrick Musau ◽  
Hoang-Dung Tran ◽  
Taylor T. Johnson
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Control Systems ◽  
Closed Loop ◽  
Closed Loop Control ◽  
Feed Forward ◽  
Loop Control ◽  
Closed Loop Systems ◽  
Feed Forward Neural Networks ◽  
Neural Network Controllers

This benchmark suite presents a detailed description of a series of closed-loop control systems with artificial neural network controllers. In many applications, feed-forward neural networks are heavily involved in the implementation of controllers by learning and representing control laws through several methods such as model predictive control (MPC) and reinforcement learning (RL). The type of networks that we consider in this manuscript are feed-forward neural networks consisting of multiple hidden layers with ReLU activation functions and a linear activation function in the output layer. While neural network con- trollers have been able to achieve desirable performance in many contexts, they also present a unique challenge in that it is difficult to provide any guarantees about the correctness of their behavior or reason about the stability a system that employs their use. Thus, from a controls perspective, it is necessary to verify them in conjunction with their corresponding plants in closed-loop. While there have been a handful of works proposed towards the verification of closed-loop systems with feed-forward neural network controllers, this area still lacks attention and a unified set of benchmark examples on which verification techniques can be evaluated and compared. Thus, to this end, we present a range of closed-loop control systems ranging from two to six state variables, and a range of controllers with sizes in the range of eleven neurons to a few hundred neurons in more complex systems.

scholarly journals Fault Identification for a Closed-Loop Control System Based on an Improved Deep Neural Network

Sensors ◽  
2019 ◽  
Vol 19 (9) ◽  
pp. 2131 ◽  
Author(s):  
Bowen Sun ◽  
Jiongqi Wang ◽  
Zhangming He ◽  
Haiyin Zhou ◽  
Fengshou Gu
Keyword(s):  
Neural Network ◽  
Control System ◽  
Control Systems ◽  
Deep Neural Network ◽  
Closed Loop ◽  
Fault Identification ◽  
Closed Loop Control ◽  
Loop Control ◽  
Identification Methods ◽  
Loop Control System

Fault identification for closed-loop control systems is a future trend in the field of fault diagnosis. Due to the inherent feedback adjustment mechanism, a closed-loop control system is generally very robust to external disturbances and internal noises. Closed-loop control systems often encourage faults to propagate inside the systems, which may lead to the consequence that faults amplitude becomes smaller and fault characteristics difference becomes more inapparent. Hence, it has been challenging to achieve fault identification for such systems. Traditional fault identification methods are not particularly designed for closed-loop control systems and thus cannot be applied directly. In this work, a new fault identification method is proposed, which is based on the deep neural network for closed-loop control systems. Firstly, the fault propagation mechanism in closed-loop control systems is theoretically derived, and the influence of fault propagation on system variables is analyzed. Then deep neural network is applied to find fault characteristics difference between different data modes, and a sliding window is used to amplify the fault-to-noise ratio and characteristics difference, with an aim to increase the identification performance. To verify this method, the simulations that are based on a numerical simulation model, the Tennessee industrial system and the satellite attitude control system are conducted. The results show that the proposed method is more feasible and more effective in fault identification for closed-loop control systems compared with traditional data-driven identification methods, including distance-based and angle-based identification methods.

EEG-FES-Force-MMG closed-loop control systems of a volunteer with paraplegia considering motor imagery with fatigue recognition and automatic shut-off

2021 ◽  
Vol 68 ◽  
pp. 102662
Author(s):  
Paulo Broniera Junior ◽  
Daniel Prado Campos ◽  
André Eugenio Lazzaretti ◽  
Percy Nohama ◽  
Aparecido Augusto Carvalho ◽  
...  
Keyword(s):  
Control Systems ◽  
Motor Imagery ◽  
Closed Loop ◽  
Closed Loop Control ◽  
Loop Control

A Control Strategy for Intelligent Stamp Forming Tooling

2011 ◽  
Vol 133 (6) ◽  
Author(s):  
William J. Emblom ◽  
Klaus J. Weinmann
Keyword(s):  
Fuzzy Logic ◽  
Control Systems ◽  
Control Strategy ◽  
Pid Controller ◽  
Closed Loop ◽  
Closed Loop Control ◽  
Linear Quadratic ◽  
Blank Holder ◽  
Loop Control ◽  
Stamp Forming

This paper describes the development and implementation of closed-loop control for oval stamp forming tooling using MATLAB®’s SIMULINK® and the dSPACE®CONTROLDESK®. A traditional PID controller was used for the blank holder pressure and an advanced controller utilizing fuzzy logic combining a linear quadratic gauss controller and a bang–bang controller was used to control draw bead position. The draw beads were used to control local forces near the draw beads. The blank holder pressures were used to control both wrinkling and local forces during forming. It was shown that a complex, advanced controller could be modeled using MATLAB’s SIMULINK and implemented in DSPACE CONTROLDESK. The resulting control systems for blank holder pressures and draw beads were used to control simultaneously local punch forces and wrinkling during the forming operation thereby resulting in a complex control strategy that could be used to improve the robustness of the stamp forming processes.

Heat Release Experimental Analysis for RCCI Combustion Optimization

2018 ◽  
Author(s):  
V. Ravaglioli ◽  
F. Ponti ◽  
F. Carra ◽  
M. De Cesare
Keyword(s):  
Low Temperature ◽  
Experimental Analysis ◽  
Closed Loop ◽  
Combustion Stability ◽  
Closed Loop Control ◽  
Combustion Control ◽  
Transient Conditions ◽  
Feed Forward ◽  
Loop Control ◽  
Injection Parameters

Over the past years, the increasingly stringent emission regulations for Internal Combustion Engines (ICE) spawned a great amount of research in the field of combustion control optimization. Nowadays, optimal combustion control has become crucial, especially to properly manage innovative Low Temperature Combustion (LTC) strategies, usually characterized by high instability, cycle-to-cycle variability and sensitivity to slight variations of injection parameters and thermal conditions. Many works demonstrate that stability and maximum efficiency of LTC strategies can be guaranteed using closed-loop control strategies that vary the standard injection parameters (mapped during the base calibration activity) to keep engine torque and center of combustion (CA50) approximately equal to their target values. However, the combination of standard base calibration and closed-loop control is usually not sufficient to accurately control Low Temperature Combustions in transient conditions. As a matter of fact, to properly manage LTC strategies in transient conditions it is usually necessary to investigate the combustion methodology of interest and implement specific functions that provide an accurate feed-forward contribution to the closed-loop controller. This work presents the experimental analysis performed running a light-duty compression ignited engine in dual-fuel RCCI mode, the goal being to highlight the way injection parameters and charge temperature affect combustion stability and ignition delay. Finally, the paper describes how the obtained results can be used to define the optimal injections strategy in the analyzed operating points, i.e. the combination of injection parameters to be used as a feed-forward for a closed-loop combustion control strategy.

Tangible Programming for Basic Control System New Frameworks to Engineering Education for Children

2014 ◽  
Vol 931-932 ◽  
pp. 1298-1302
Author(s):  
Thiang Meadthaisong ◽  
Siwaporn Meadthaisong ◽  
Sarawut Chaowaskoo
Keyword(s):  
Control System ◽  
Engineering Education ◽  
Control Systems ◽  
Primary Schools ◽  
Closed Loop ◽  
Open Loop ◽  
Closed Loop Control ◽  
College Level ◽  
Loop Control ◽  
Tangible Programming

Programming control in industrial design is by its nature expert upon an example being Programmable Logic Controller (PLC). Such programmes are unsuitable for children or novices as they cannot understand how to use the programme. This research seeks to present tangible programming for a basic control system in new frameworks in engineering education for children. Such programmes could be for use in kindergartens, primary schools or general teaching where knowledge about basic control is required. Normally open-loop and closed-loop control system programming is taught at university and college level. This may be late as far as acquiring knowledge of basic control systems is concerned. Using tangible programming without a computer but instructions and interface, relay and motor could result in children in kindergartens and primary schools being able to programme open-looped control systems which mix chemicals or closed-loop control systems which control conveyor belts. However, the children would not be able to undertake programming using programmable control in a similar scenario.

Sign in / Sign up

Export Citation Format

Share Document