作者：仝卫国, 李茂冉, 石宗锦, 寇德龙
作者单位：华北电力大学自动化系，河北 保定 071003
关键词：气液两相流;截面含气率;改进粒子群;elman神经网络;阵列电阻值
摘要：
为安全且非侵入式地测量气液两相流含气率，提出一种电阻层析成像（ert）陈列电阻与elman神经网络相结合的含气率测量方法。首先，为加快模型训练速度并避免数据冗余，使用主成分分析（pca）算法对120维的阵列电阻特征降维。然后，在粒子群（pso）算法中引入自适应惯性权重和非线性学习因子，并加入遗传算法（ga）的交叉和变异行为以加快算法收敛速度。最后，通过改进的粒子群（ipso）算法优化elman神经网络初始权值和阈值，并建立含气率测量模型。经对比实验发现，pca-ipso-elman含气率测量模型的平均绝对百分比误差为2.92%，且训练时间较ipso-elman模型减少68.8%。说明所提方法可以达到预期的测量效果。
void fraction measurement method of gas-liquid two-phase flow based on ipso-elman
tong weiguo, li maoran, shi zongjin, kou delong
department of automation, north china electric power university, baoding 071003, china
abstract: in order to obtain non-invasive measurement results of gas-liquid two-phase flow gas holdup safely, a method of measuring the void fraction based on electrical resistance tomography (ert) array resistance value and elman neural network was proposed. firstly, in order to accelerate the training speed of model and avoid redundancy of data, principal component analysis (pca) algorithm was used to reduce the dimension of the resistance feature of the 120-dimensional array. then, the adaptive inertia weights and nonlinear learning factors were introduced into the particle swarm optimization (pso) algorithm, and the crossover and mutation behaviors of genetic algorithm (ga) were added to improve the rate of convergence of the algorithm. finally, the initial weights and thresholds of elman neural network were optimized by improved particle swarm optimization (ipso) algorithm, and gas holdup measurement model was established. through the comparison experiment, it is found that the mean absolute percentage error of the gas holdup measurement model named pca-ipso-elman is 2.92% and the training time of the model is reduced by 68.8% compared with ipso-elman model, which manifests that the proposed method can achieve the expected measurement effect.
keywords: gas-liquid two-phase flow;void fraction;improved particle swarm optimization;elman neural network;array resistance value
2024, 50(7):26-32,62 收稿日期: 2022-11-02;收到修改稿日期: 2022-12-29
基金项目:
作者简介: 仝卫国（1967-），男，河北保定市人，副教授，博士，研究方向为流量参数测量与检测新技术、先进控制策略在大型电力机组的应用等。
参考文献
[1] 程洁, 郭亚军, 王腾, 等. γ射线法测量高压管束间气液两相流的截面含气率分布[j]. 化工学报, 2019, 70(4): 1375-1382.
cheng j, guo y j, wang t, et al. void fraction distribution of vapor-water two-phase flow in vertical tube bundles using gamma densitometer[j]. ciesc journal, 2019, 70(4): 1375-1382.
[2] 陈阳正, 王小鑫, 王博, 等. 电容法气液两相流相含率测量系统研究[j]. 仪表技术与传感器, 2020(2): 114-118.
chen y z, wang x x, wang b, et al. study on measurement system of phasevolume fraction in gas-liquid two-phase flow with capacitance method[j]. instrument technique and sensor, 2020(2): 114-118.
[3] li h j, ji h f, huang z y, et al. a new void fraction measurement method for gas-liquid two-phase flow in small channels[j]. sensors, 2016, 16(2): 159.
[4] zhao y, bi q c, hu r c. recognition and measurement in the flow pattern and void fraction of gas–liquid two-phase flow in vertical upward pipes using the gamma densitometer[j]. applied thermal engineering, 2013, 60(1-2): 398-410.
[5] 汪剑鸣, 李博, 王琦, 等. 基于提升小波时延估计的气液两相流流速测量[j]. 仪器仪表学报, 2017, 38(3): 653-663.
wang j m, li b, wang q, et al. measurement of gas/liquid two-phase flow velocity based on lifting wavelet time delay estimation[j]. chinese journal of scientific instrument, 2017, 38(3): 653-663.
[6] tan c, li f, lv s h, et al. gas-liquid two-phase stratified flow interface reconstruction with sparse batch normalization convolutional neural network[j]. ieee sensors journal, 2021, 21(15): 17076-17084.
[7] 彭珍瑞, 殷红, 祁文哲. 基于svm和ect的两相流离散相浓度测量[j]. 自动化与仪器仪表, 2008(2): 72-74.
peng z r, yin h, qi w z. ect and svm based concentration measurement of two-phase flow[j]. automation ＆ instrumentation, 2008(2): 72-74.
[8] 匡世才, 王武强, 周宏亮. 基于bp神经网络的集输-立管系统气液两相流流量测量[j]. 热能动力工程, 2019, 34(4): 79-85.
kuang s c, wang w q, zhou h l. gas-liquid two-phase flow metering based on bp artificial neural network in pipeline-riser system[j]. journal of engineering for thermal energy and power, 2019, 34(4): 79-85.
[9] 瞿红春, 许旺山, 郭龙飞, 等. 基于ibas-elman网络的滚动轴承故障诊断研究[j]. 机床与液压, 2020, 48(16): 201-205.
qu h c, xu w s, guo l f, et al. research on fault diagnosis of rolling bearing based on ibas-elman network[j]. machine tool ＆ hydraulics, 2020, 48(16): 201-205.
[10] 张建超, 单慧勇, 景向阳, 等. 基于elman神经网络的温室环境因子预测方法[j]. 中国农机化学报, 2021, 42(8): 203-208.
zhang j c, shan h y, jing x y, et al. prediction method of greenhouse environmental factors based on elman neural network[j]. journal of chinese agricultural mechanization, 2021, 42(8): 203-208.
[11] 林旭梅, 刘帅, 石智梁. 基于改进pso-fnn算法的钢筋混凝土腐蚀检测研究[j]. 中国测试, 2020, 46(12): 149-155.
lin x m, liu s, shi z l. research on reinforced concrete corrosion detection based on improved pso-fnn algorithm[j]. china measurement ＆ test, 2020, 46(12): 149-155.
[12] 任轩, 汪庆年, 尚宝, 等. 基于混合神经网络的短期电力负荷预测方法[j]. 电子测量技术, 2022, 45(14): 71-77.
ren x, wang q n, shang b, et al. short-term load forecasting method based on hybrid neural network[j]. electronic measurement technology, 2022, 45(14): 71-77.
[13] 仝卫国, 曾世超, 李芝翔, 等. 基于lstm的气液两相流液相流量测量方法[j]. 仪表技术与传感器, 2021(11): 94-98.
tong w g, zeng s c, li z x, et al. liquid phase flow measurement method of gas-liquid two-phase flow based on lstm[j]. instrument technique and sensor, 2021(11): 94-98.
[14] 朱菊香, 谷卫, 罗丹悦, 等. 基于pso优化bp神经网络的多传感器数据融合[j]. 中国测试, 2022, 48(8): 94-100.
zhu j x, gu w, luo d y, et al. multi-sensor data fusion based on pso optimized bp neural network[j]. china measurement ＆ test, 2022, 48(8): 94-100.
[15] 张立峰, 王化祥. 一种修正的电阻层析成像landweber迭代算法[j]. 计量学报, 2016, 37(3): 271-274.
zhang l f, wang h x. a modified landweber algorithm for electrical resistance tomography[j]. acta metrologica sinica, 2016, 37(3): 271-274.
