当前位置: X-MOL 学术Measurement › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Prediction of flow by linear parameter varying model under disturbance
Measurement ( IF 5.6 ) Pub Date : 2021-09-14 , DOI: 10.1016/j.measurement.2021.110124
Vemulapalli Sravani 1 , Santhosh Krishnan Venkata 1
Affiliation  

Flowrate being one of the most measured process parameters, thus it is essential to produce higher accuracy. The objective of the paper is to design an estimator using the Linear Parameter Varying (LPV) model which would estimate the flow rate using the data from the orifice meter for varying parameters like, the beta ratio of an orifice, and liquid density. For the development of an estimator, a process model is used, which is designed with the help of a system identification technique using a data-driven method. Data for system identification is obtained by the process model designed using Computational Fluid Dynamics (CFD). The output of the CFD model is compared with experimental results, there is a good agreement in results obtained with an average of 5.97% and 3.19% error in terms of differential pressure and discharge coefficient respectively. The estimated output from the LPV model is compared with that of results obtained from the neural network model and experimental setup, which are also in good agreement with an average error of 2.14%. Thus can be used to estimate flow, when orifice design or density of fluid change intentionally or unintentionally.



中文翻译:

扰动下线性参数变化模型的流量预测

流量是测量最多的过程参数之一,因此必须产生更高的精度。本文的目的是设计一个使用线性参数变化 (LPV) 模型的估计器,该模型将使用来自孔板流量计的数据来估计流量,这些数据用于改变参数,例如孔板的 Beta 比和液体密度。对于估算器的开发,使用过程模型,该模型是在使用数据驱动方法的系统识别技术的帮助下设计的。系统识别数据是通过使用计算流体动力学 (CFD) 设计的过程模型获得的。将 CFD 模型的输出与实验结果进行比较,获得的结果具有良好的一致性,平均为 5.97% 和 3。差压和流量系数误差分别为 19%。将 LPV 模型的估计输出与神经网络模型和实验装置获得的结果进行比较,它们也具有良好的一致性,平均误差为 2.14%。因此,当孔口设计或流体密度有意或无意发生变化时,可用于估算流量。

更新日期:2021-09-17
down
wechat
bug