Temporal Data and Short-Time Averages Improve Multiphase Mass Flow Metering

TL;DR

This study shows that using short-time averages and temporal data improves machine learning-based corrections for multiphase flow measurements with Coriolis meters, achieving high accuracy.

Contribution

It introduces a method that preserves temporal information through short-time averaging, significantly enhancing ML model performance over traditional single-averaged approaches.

Findings

CNN achieved approximately 95% of relative errors below 13%

Short-time averaging outperformed single-averaged models

Results were consistent across data splits and random seeds

Abstract

Reliable flow measurements are essential in many industries, but current instruments often fail to accurately estimate multiphase flows, which are frequently encountered in real-world operations. Combining machine learning (ML) algorithms with accurate single-phase flowmeters has therefore received extensive research attention in recent years. The Coriolis mass flowmeter is a widely used single-phase meter that provides direct mass flow measurements, which ML models can be trained to correct, thereby reducing measurement errors in multiphase conditions. This paper demonstrates that preserving temporal information significantly improves model performance in such scenarios. We compare a multilayer perceptron, a windowed multilayer perceptron, and a convolutional neural network (CNN) on three-phase air-water-oil flow data from 342 experiments. Whereas prior work typically compresses each experiment into a single averaged sample, we instead compute short-time averages from within each experiment and train models that preserve temporal information at several downsampling intervals. The CNN performed best at 0.25 Hz with approximately 95 % of relative errors below 13 %, a normalized root mean squared error of 0.03, and a mean absolute percentage error of approximately 4.3 %, clearly outperforming the best single-averaged model and demonstrating that short-time averaging within individual experiments is preferable. Results are consistent across multiple data splits and random seeds, demonstrating robustness.