| Author: | Tang, Siu-wai |
| Title: | Combined PID controller and neural controller for process control |
| Degree: | M.Sc. |
| Year: | 1994 |
| Subject: | Process control Automatic control PID controllers Neural networks (Computer science) Hong Kong Polytechnic -- Dissertations |
| Department: | Department of Electrical Engineering |
| Pages: | 80, [9] leaves : ill. ; 30 cm |
| Language: | English |
| Abstract: | In this project, an Artificial-Neural-Network-based controller coupled with PID controller is developed for compensating the change in process plant dynamic. The task has been split into: software simulation and hardware implementation. The software work demonstrates an Artificial-Neural-Network-based controller coupled with a PID controller which provides compensation for the change in process plant dynamics. An Artificial-Neural-Network-based controller consists of a well trained Artificial-Neural-Network (ANN), which as its input some past information; such as system input, system output and some system state variable. Combining the output of the ANN with the output of the PID controller produces the necessary control action for process plant such that the output of process plant performs according to the desirable response. The configuration of ANN is simple. It consists of several input nodes, and one output node. There is a number of intermediate node in between the input ports and output port. All the input and intermediate nodes, the intermediate and output nodes are interconnected by some weighted links. The ANN-based-controller should be trained before engaging on the actual system; hence, the ANN employs the back propagation training algorithm to update the weighted links and to learn the dynamics of system such that the system response can be improved. During the training session of the ANN, the weighted values of the weighted links are adjusted until a pre-defined criteria has been reached. Using state feedback information and the learning capability of the ANN, the controller is expected to produce reasonable performance even with unpredictable variations on the system parameters and disturbances. From the simulation result, it can be seen that the ANN can learn the system dynamics with acceptable accuracy and give satisfactorily results on the system performance. Furthermore, the parameters of process plant has been modified during the simulation. This simulated the actual plant parameter changes due to environmental and characteristic changes. It can be seen that the output response of the process plant is still acceptable owing to the learning capability of the ANN-based-controller. For the hardware work, the trained ANN-based-controller is implemented on a PC equipped with an A/D card to control a simple process plant running on an analog computer so as to investigate the performance of the ANN-based-controller coupled with PID controller on actual process control. Both software simulation and hardware experimental results are compared to analyse the performance of the proposed control scheme. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b11684446.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 2.33 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/4445