Performance Measurement Techniques for Adaptive Process Control.

Abstract

In adaptive process control systems, an essential requirement is the availability of on-line feedback data indicative of the actual process performance. The measurement of such data often poses difficult design problems because available sensors are capable of measuring only certain process variables which are related to true performance by complex functions. Determination of these functions requires extensive collection and analysis of experimental data for all types of process conditions. This report describes a project conducted to systematize and simplify the performance and measurement problem. The approach to this objective was based on the use of a digital computer to implement a trainable pattern recognition data processing technique. The approach was demonstrated by designing and fabricating a prototype performance measurement system and by evaluating the system operating in conjunction with three diverse manufacturing processes: a turning process (lathe), an electron-beam welding process, and a grinding process. (Author)

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1968
Accession Number
AD0844910

Entities

People

  • Frank A. Russo
  • Robert J. Valek

Tags

DTIC Thesaurus Topics

  • Computers
  • Control Systems
  • Data Processing
  • Digital Computers
  • Electron Beam Welding
  • Electron Beams
  • Experimental Data
  • Manufacturing
  • Measurement
  • Pattern Recognition
  • Recognition

Readers

  • Instructional Design and Training Evaluation.
  • Metallurgy
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • Directed Energy
  • Microelectronics
  • Microelectronics - Microelectromechanical Systems