The temperature distribution in real-world industrial environments is often in a three-dimensional space, and developing a reliable\nmethod to predict such volumetric information is beneficial for the combustion diagnosis, the understandings of the complicated\nphysical and chemical mechanisms behind the combustion process, the increase of the system efficiency, and the reduction of\nthe pollutant emission. In accordance with the machine learning theory, in this paper, a new methodology is proposed to predict\nthree-dimensional temperature distribution from the limited number of the scattered measurement data. The proposed prediction\nmethod includes two key phases. In the first phase, traditional technologies are employed tomeasure the scattered temperature data\nin a large-scale three-dimensional area. In the second phase, the Gaussian process regression method, with obvious superiorities,\nincluding satisfactory generalization ability, high robustness, and low computational complexity, is developed to predict threedimensional\ntemperature distributions. Numerical simulations and experimental results from a real-world three-dimensional\ncombustion process indicate that the proposed prediction method is effective and robust, holds a good adaptability to cope with\ncomplicated, nonlinear, and high-dimensional problems, and can accurately predict three-dimensional temperature distributions\nunder a relatively lowsampling ratio.As a result, a practicable and effective method is introduced for three-dimensional temperature\ndistribution.
Loading....