To solve the problem of electrical capacitance tomography(ECT ) system of flow regime identification，this paper presents a new method based on Elman neural network for ECT system of flow pattern identification，which is on the basis of the work principle of ECT system and the method of flow pattern identification. This method uses part of capacitance to extract its feature，and uses the extracted data to train the Elman neural network to achieve the aim of recognize the type of the flow.After simulation，the result shows that this method is better than BP neural network. This method has a simple structure and a fast convergence speed，at the same time;it does not have the network expansion problem of the unknown order. It is an effective method for ECT system of flow pattern identification.