A NUMERICAL EVALUATION OF A SIMPLE PROCEDURE FOR USING TRANSIENT SURFACE TEMPERATURE MEASUREMENTS TO DETERMINE LOCAL CONVECTIVE HEAT TRANSFER RATES

摘要

A transient method, based on an inverse heat conductionsolution, for experimentally determining the distribution of local heattransfer rates on the surface of a body has been numerically evaluated. Theparticular interest is in situations in which the heat transfer coefficients arerelatively low and in which there are relatively large changes in the heattransfer coefficient over the surface of the body being considered. In themethod, a solid body of the shape being investigated, constructed from a lowconductivity material, is heated to a uniform temperature and then exposedto a test flow. Using a layer of temperature sensitive crystal placed over thesurface of this model or by other means, the time taken for the temperature ata relatively small number of selected points on the surface to reach a selectedvalue is determined. The surface heat flux rate distribution is then foundfrom these measured times using a simple inverse heat conduction method.The feasibility of this method has been evaluated by considering relativelylow Reynolds number flow over a square cylinder and natural convectiveflow over a circular cylinder. Known local heat transfer coefficientdistributions for these situation have been applied as boundary conditions inthe numerical solution of the transient cooling of a the "experimental"models. These solutions are used to generate "measured" data i.e. to generatesimulated experimental data. The inverse heat transfer method has then beenused to predict the local heat transfer coefficient distribution over the surfaceand the predicted and input distributions have been compared. The effect ofuncertainties in the experimental measurements on this comparison has thenbeen evaluated using various assumed uncertainty values. The results of thestudy indicate that the proposed method of measuring local heat transfercoefficients is capable of giving results of good accuracy.