基于改进的LS-SVM测量油气两相流空隙率

利用ECT电容传感器提供的电容测量值,基于改进的最小二乘支持向量机(LS-SVM),提出了一种油气两相流空隙率测量的新方法.运用该方法测量空隙率时,以ECT电容传感器获取的66个独立电容值作为空隙率模型的输入,计算即可得空隙率.建模阶段,针对LS-SVM使用中存在的问题,首先运用训练数据筛选技巧,对LS-SVM进行稀疏化改进,接着引入实数编码的遗传算法(RC-GA)优化LS-SVM参数,然后运用改进后的LS-SVM和基于RC-GA的参数选取方法建立空隙率测量模型.所提出的空隙率测量方法省去了常用ECT方法测量空隙率时的复杂耗时的图像重建过程,提高了空隙率测量的实时性.实验结果表明,提出的LS-SVM改进和参数优化方法是有效的,提出的空隙率测量方法具有实时性佳的优点,测量精度满足实际应用要求.

Improved LS-SVM Based Voidage Measurement of Oil-Gas Two-Phase Flow

Based on the capacitance information provided by the capacitance sensors in Electrical Capacitance Tomography (ECT) system, and by means of the improved Least Squares Support Vector Machine (LS-SVM), a new voidage measurement method for oil-gas two-phase flow was proposed. In the measurement process, the 66 capacitance values from ECT sensor are the input of the voidage model, and the output of the model is the voidage value. In the model establishing stage, in order to solve the problems existing in current LS-SVM, pruning skill was employed for training data set to make LS-SVM sparse and robust at first, then Real-Coded Genetic Algorithm (RC-GA) was introduced to solve the difficult problem of parameters selection in LS-SVM. These two aspects were applied to establish the voidage model. The proposed new voidage measurement method implements voidage measurement directly, and omits the complicated and time-consuming image reconstruction which is usually needed for voidage measurement by using currently used ECT method. Using the method proposed, the real-time performance of voidage measurement was greatly improved. Experimental results demonstrate that both the improvement of LS-SVM and the parameters optimization are effective. The results also show that the real-time performance of the proposed voidage measurement method is good, and the measurement precision can satisfy application requirement.