摘要
This paper presents a study on measuring rotation speed of moving glass beads with an average diameter of 500 p.m in a pilot-scale circulating fluidized bed （CFB） riser with a high-speed digital imaging system. Two methods have been developed to calculate particle rotation speed from the particle images. The first method consists of a fully automated algorithm based on cross-correlation of gray distribution of particle images for particles whose rotation axes are （nearly） perpendicular to the imaging plane, and the second method calculates the speed of particle rotation by identifying its rotation axis using two or more characteristic points on its surface. The reliability of the two methods is verified by using a small sphere with known speed and direction of rotation. The first method is shown to be capable of measuring accurately the rotation speed for the particle with a rotation axis （nearly） perpendicular to the imaging plane and filtering off other particles using an appropriate threshold of correlation coefficient. The second method is shown to be capable of yielding both the speed and direction of particle rotation, with a measurement error of less than 10%. Results of both methods on real glass beads in a CFB riser are compared against each other.
This paper presents a study on measuring rotation speed of moving glass beads with an average diameter of 500 p.m in a pilot-scale circulating fluidized bed （CFB） riser with a high-speed digital imaging system. Two methods have been developed to calculate particle rotation speed from the particle images. The first method consists of a fully automated algorithm based on cross-correlation of gray distribution of particle images for particles whose rotation axes are （nearly） perpendicular to the imaging plane, and the second method calculates the speed of particle rotation by identifying its rotation axis using two or more characteristic points on its surface. The reliability of the two methods is verified by using a small sphere with known speed and direction of rotation. The first method is shown to be capable of measuring accurately the rotation speed for the particle with a rotation axis （nearly） perpendicular to the imaging plane and filtering off other particles using an appropriate threshold of correlation coefficient. The second method is shown to be capable of yielding both the speed and direction of particle rotation, with a measurement error of less than 10%. Results of both methods on real glass beads in a CFB riser are compared against each other.
基金
supports from the Special Funds for China Major State Basic Research Projects(Grant 2005CB221201)
the National Natural Science Foundation Projects (Grant 50806067)
the Program of Introducing Talents of Discipline to University (B08026)
