International Society for Industrial Process Tomography

ECT flow regime identification technology based on 2­D Maximum Entropy Threshold Image Segmentation with Chaos­Particle Swarm optimization algorithm

Zhengyuan Luo1; Hongli Hu1; Tongmo Xu2; Houzhang Tan2,*

1 State Key Laboratory of Electrical Insulation for Power Equipment, Xi’an Jiaotong University, Xi’an710049, China

2 State Key Laboratory of Multi­phase Flow for Power Engineering, Xi’an Jiaotong University, Xi’an710049, China

Abstract: Electrical capacitance tomography technology is widely used in the field of flow regime identification. And image segmentation is one of the key technologies of flow regime identification. 2­D Maximum Entropy Threshold Image Segmentation is a newly developed method to segment images. A method of 2­D Maximum Entropy Threshold Image Segmentation with Chaos­Particle Swarm optimization algorithm is presented to segment the images collected by an ECT system. Thus, more spatial information of an image is sufficiently used by the method; better effect is obtained to identify the flow regime. A laboratory­level experiments prototype is built, and a gas­solid two­phase flow regime identification experiment is carried out on the prototype to check the image segmentation method. The result shows the flow regime identification of ECT system is satisfactory.

Key words: Flow regime identification; Electrical capacitance tomography;

Image segmentation; 2­D Maximum Entropy Threshold Image Segmentation; Chaos­Particle Swarm optimization algorithm.