Integrated evolutionary computation neural network quality controller for automated systems
- Texas Tech Univ., Lubbock, TX (United States). Dept. of Industrial Engineering
With increasing competition in the global market, more and more stringent quality standards and specifications are being demands at lower costs. Manufacturing applications of computing power are becoming more common. The application of neural networks to identification and control of dynamic processes has been discussed. The limitations of using neural networks for control purposes has been pointed out and a different technique, evolutionary computation, has been discussed. The results of identifying and controlling an unstable, dynamic process using evolutionary computation methods has been presented. A framework for an integrated system, using both neural networks and evolutionary computation, has been proposed to identify the process and then control the product quality, in a dynamic, multivariable system, in real-time.
- Research Organization:
- Amarillo National Resource Center for Plutonium, TX (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- FC04-95AL85832
- OSTI ID:
- 350895
- Report Number(s):
- ANRCP-99002639; ON: DE99002639; TRN: AHC29921%%80
- Resource Relation:
- Other Information: PBD: [1999]
- Country of Publication:
- United States
- Language:
- English
Similar Records
Examples of neural networks used for building system control and energy management
Dilute Combustion Control Using Spiking Neural Networks