Real-time heat flux estimation using filter based solutions for inverse heat conduction problems

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
University of Alabama Libraries

Real-time heat flux measurement has great significance in numerous applications such as metal heat treating, quenching, fire safety tests, furnace operation, thermal therapy and more. Direct measurement of heat flux is not always possible and therefore, developing methods for accurate heat flux estimation using temperature data is desirable for several applications. An Inverse Heat Conduction Problem (IHCP) has to be solved to estimate the unknown heat flux at the surface, knowing the temperature measurements in the interior points of the medium. This problem is mathematically ill-posed and a small error in temperature data result in large errors in estimated heat fluxes. The focus of this dissertation is developing solutions for IHCP’s using digital filter approach. A series of IHCP’s are closely investigated and filter based solutions are developed for each problem which allows real-time heat flux estimation. The characteristics of the filter form solutions are discussed in detail for each case. A series of articles are written to discuss these problems. The articles in this dissertation provide real-time solutions for a series of IHCP’s with wide applications in industries. In particular, a filter based solution technique is developed for heat flux estimation in multi-layer mediums with temperature dependent material properties. While one application of this solution technique in using Directional Flame Thermometers is discussed in detail, there are many other applications for this solution method, e.g. in thermal therapy and tissue engineering (Articles 1-5). Furthermore, the concept of using artificial neural networks as digital filters for real-time heat flux estimation is investigated in Article 6. Several test cases are demonstrated to study the applicability of ANN’s for solving linear and non-linear IHCP’s. Finally, developing filter form solutions for two-dimensional IHCP’s with multiple unknown heat fluxes is presented in Article 7. This study not only provides a basis for heat flux estimation in heat treating processes in industries, but also proves the applicability of the filter solution for multi-dimensional mediums with more complexities and demonstrates the characteristics of this technique.

Electronic Thesis or Dissertation
Mechanical engineering