The Research Of Temperature Field Reconstruction In Acoustic Temperature Tomography

Temperature field measurement in industrial combustion or heating process is of great significance. Because of the complexity of combustion, the large size of combustion equipment and the high temperature up to 1000 degrees centigrade and such disadvantageous factors, it is very difficult to achieve accurate measurements of the whole temperature filed by using the traditional technology of thermocouple or radiation pyrometer. As a new temperature measurement technology, acoustic method has many advantages such as wide temperature range measurement (0～1900 degrees centigrade), large space of measuring object (up to a few decameters), high-precision, non-contact, real-time, continuous measurement, easy maintenance and so on. It will be more and more useful in industrial boiler temperature field measuring. Some key issues to temperature field measurement by acoustic tomography are discussed in this thesis.The principle and the development status of temperature field measurement by acoustic tomography are introduced in this thesis. Temperature field reconstruction algorithm based on least-square method is expatiated. The influences of the sensor location and the measurement space division on the temperature reconstruction precision are investigated. Computer simulations using a single-peak symmetrical round temperature field as the temperature model to test above two factors' influence on the reconstruction precision are implemented. The results show that the sensor location and the space division should ensure the sound wave path distribution density as uniform as possible.The influence of the bended sound wave paths on the reconstruction precision is researched in this paper. When a sound wave passes through non-uniform temperature fields, it will be refracted and its travel path will be bended. The research result shows that viewing the sound wave path between the sound emitter and receiver to be beeline willcause large errors, especially for the temperature field with steep temperature gradient. In this thesis, sound wave path equation is derived from Fermat's theorem and the calculus of variations, and then the temperature field is reconstructed by the combination least-square method and the modified sound wave path.The thesis also makes another computer simulation which takes the effect of the bended sound wave paths into account. The single-peak symmetrical temperature field, the single-peak deflective temperature field and the double-peak symmetrical temperature field are chosen as the model temperature field respectively, and the sensors are evenly positioned around a round measurement space. The reconstruction results are satisfying. The simulation research confirms that the reconstruction accuracy will be increased when we take the effect of the bended sound wave paths into account, particularly for the temperature fields with great temperature gradient, although the method has disadvantage in the reconstruction velocity.