| Related electromagnetic flow measurement sensors are widely used in oil wells measurement, and related decisions electromagnetic flow measurement accuracy of key parts of the sensor is an electromagnetic sensor associated electrode structure, and therefore the study of electromagnetic sensor electrode structure optimization method is necessary. However, the conventional electromagnetic flow measurement sensor associated electrode structure design only through experience in general design, electromagnetic sensor electrode structure there are significant flaws. To solve this problem in order to develop a higher precision electromagnetic flow measurement sensor correlation, this paper presents a sensor electrode structure design a new method to measure electromagnetic relevant traffic.Design related electromagnetic flow sensor electrode structure provides a new solutions and approaches to build neural network model electrode structure parameters and use one kind of swallow swarm optimization algorithm to solve the optimal solution of the model of this project is. This research project is the development of new gas-water multiphase flow downhole flow measurement instrumentation theoretical basis, flow measurement technology research in other fields also has a good reference value.Firstly, electromagnetic flow measurement sensor related FLUENT simulation models. Electromagnetic sensor electrode in accordance with the relevant physical structure, FLUNT finite element analysis software to build simulation model of the sensor electrode structure, and accordingly study when the upstream and downstream from the pipe diameter and associated electromagnetic sensor electrode changes, fluid flow through the sensor and the associated actual flow rate velocity error showing changes, on this basis, using the fluid velocity cloud and MATLAB software metering the fluid flow rate sensor data calculation and analysis of electromagnetic relevant traffic.Secondly, a neural network model associated electromagnetic sensor electrode structure. By FLUNT sample data and simulation MATLAB calculated input to the RBF neural network, the training of the neural network has been amended relevant electromagnetic sensor electrode structure model overcomes the experimental conditions of limited resources and the environment in real drawback is difficult to obtain actual data.Finally, the optimization methods associated electromagnetic sensor electrode structures based yan swarm optimization. The electromagnetic sensor electrode structure associated neural network model to build a value distribution rights, the use of optimization fitness function verification swallow swarm optimization algorithm, using algorithms to search for optimal solutions of nonlinear parametric model the actual flow rate and velocity errors associated swallow swarm optimization, complete optimization of the associated electromagnetic sensor electrode structure. |
PDF Full Text Request