Identification Method of Fluidized Bed's Gas-solid Two Phase Flow Regime Based on Images Processing and Genetic Neural Network~*

摘要

Gas-solid two-phase flow widely exists in modern industry process. Two-phase flow and heat transfer characters are extremely influenced by the flow regimes. Therefore, a flow regime identification method based on images statistical features of gray histogram and genetic neural network is proposed. Gas-solid fluidized bed flow images are captured by a high speed photography system in a self-designed and built fluidized bed device. The images statistical features of the gray histogram are extracted using image processing techniques. Then the images statistical eigenvectors of flow regime are established. The genetic neural network is trained using those eigenvectors as flow regime samples and the flow regime intelligent identification is realized. The test result shows after successful training the genetic neural network not only can effectively identify five typical flow regimes of gas-solid two-phase flow in fluidized bed, but also can solve the convergence problem in the network trains effectively. The whole identification accuracy is 99.72%, opening up a new avenue for the flow pattern recognition.