Investigation into Adaptive Control of a Slip-Cast, Reaction-Bonded Silicon-Nitride Process via Adaptive Learning Network Modeling

Abstract

A program was conducted to model the modulus of rupture (MOR) strength using Adaptive Learning Networks (ALN's) for aircraft engine components produced by a slip-cast, reaction-bonded, silicon-nitride production process. The primary objectives of the work were to identify key process variables and to predict optimum values for those variables as a guide for further experimentation. Nonlinear models have been synthesized that predict MOR with an average error of about 4 ksi over a range from 18.6 to 47.8. The manufacturing and analysis work done to date has demonstrated the feasibility of modeling the slip-cast RBSN process with the Adaptive Learning Network methodology and is viewed as the first iteration in the optimization procedure which is ultimately aimed at finding those manufacturing conditions which will produce the strongest, most consistent material strengths.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 30, 1979
Accession Number
ADA083730

Entities

People

  • Anthony N. Mucciardi
  • Basil A. Decina
  • Dixon Cleveland
  • Peter M. Garafola

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Air Force
  • Aircraft Engines
  • Aircrafts
  • Algorithms
  • Body Weight
  • Ceramic Materials
  • Contracts
  • Databases
  • Engine Components
  • Engines
  • Learning
  • Manufacturing
  • Materials
  • Materials Laboratories
  • Optimization
  • Particle Size
  • Production

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Powder metallurgy of Titanium alloys.
  • Software Engineering