City Research Online

Abstract

This thesis is concerned with the mathematical modelling and identification of industrial temperature sensors. Instruments of concern are the most common: resistance thermometers and thermocouples.

Both distributed and lumped parameter models are developed and validated for a common industrial thermometer assembly. The former are based on the finite-element method which is particularly suitable forindustria1 thermometers because of the different materials contained in them. These distributed models enabled (i) sensitivity of response to material dimensions and heat transfer coefficient and (ii) reduced (lumped-parameter) models to be determined. The models developed were fully validated with experimental results.

The models were then applied to investigate the most sensitive parameters which contribute to the dynamics of a typical industrial thermometer. The results indicated that air gap tolerances and variations in heat transfer coeifficients lead to substantial variations in time constants.

Steady-state models which assume both radial heat flow to the sensor and axial heat flow along the well and/or sheath wall allow estimates of stem correction errors to be made. Both the reduced (dynamic) and steady-state models were simulated on MEDIEM, an interactive modelling package developed at the City University.

Design can be based on models and a methodology is presented which could form a basis for a future program for computer- aided-design. Because the dynamic and steady-state mode 1 s can be simulated interactively using MEDIEM, they can conveniently be used for design, which is an iterative process.

System identification is used here to predict the response of a fluid temperature transient when response data are obtained by applying a large step in the electrical current through the sensing element leading to deliberate self heating.

Problems associated with identification of the electrical step transient data and the subsequent use of this data for prediction of fluid transients is discussed. It is concluded that the technique maybe used in the future for practical in-situ testing of industrial thermometers in, for example,
the nuclear industry.

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments:	School of Science & Technology > Computer Science School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads