Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow
| dc.contributor.author | Kadish, Shai | |
| dc.contributor.author | Schmid, David | |
| dc.contributor.author | Son, Jarryd | |
| dc.contributor.author | Boje, Edward | |
| dc.date.accessioned | 2022-04-07T10:28:27Z | |
| dc.date.available | 2022-04-07T10:28:27Z | |
| dc.date.issued | 2022-01-27 | |
| dc.date.updated | 2022-02-11T14:46:22Z | |
| dc.description.abstract | This paper presents a method to classify flow regime and vapor quality in vertical two-phase (vapor-liquid) flow, using a video of the flow as the input; this represents the first high-performing and entirely camera image-based method for the classification of a vertical flow regime (which is effective across a wide range of regimes) and the first image-based tool for estimating vapor quality. The approach makes use of computer vision techniques and deep learning to train a convolutional neural network (CNN), which is used for individual frame classification and image feature extraction, and a deep long short-term memory (LSTM) network, used to capture temporal information present in a sequence of image feature sets and to make a final vapor quality or flow regime classification. This novel architecture for two-phase flow studies achieves accurate flow regime and vapor quality classifications in a practical application to two-phase CO<sub>2</sub> flow in vertical tubes, based on offline data and an online prototype implementation, developed as a proof of concept for the use of these models within a feedback control loop. The use of automatically selected image features, produced by a CNN architecture in three distinct tasks comprising flow-image classification, flow-regime classification, and vapor quality prediction, confirms that these features are robust and useful, and offer a viable alternative to manually extracting image features for image-based flow studies. The successful application of the LSTM network reveals the significance of temporal information for image-based studies of two-phase flow. | |
| dc.identifier | doi: 10.3390/s22030996 | |
| dc.identifier.apacitation | Kadish, S., Schmid, D., Son, J., & Boje, E. (2022). Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow. <i>Sensors</i>, http://hdl.handle.net/11427/36287 | en_ZA |
| dc.identifier.chicagocitation | Kadish, Shai, David Schmid, Jarryd Son, and Edward Boje "Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow." <i>Sensors</i> (2022) http://hdl.handle.net/11427/36287 | en_ZA |
| dc.identifier.citation | Kadish, S., Schmid, D., Son, J. & Boje, E. 2022. Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow. <i>Sensors.</i> http://hdl.handle.net/11427/36287 | en_ZA |
| dc.identifier.ris | TY - Journal Article AU - Kadish, Shai AU - Schmid, David AU - Son, Jarryd AU - Boje, Edward AB - This paper presents a method to classify flow regime and vapor quality in vertical two-phase (vapor-liquid) flow, using a video of the flow as the input; this represents the first high-performing and entirely camera image-based method for the classification of a vertical flow regime (which is effective across a wide range of regimes) and the first image-based tool for estimating vapor quality. The approach makes use of computer vision techniques and deep learning to train a convolutional neural network (CNN), which is used for individual frame classification and image feature extraction, and a deep long short-term memory (LSTM) network, used to capture temporal information present in a sequence of image feature sets and to make a final vapor quality or flow regime classification. This novel architecture for two-phase flow studies achieves accurate flow regime and vapor quality classifications in a practical application to two-phase CO<sub>2</sub> flow in vertical tubes, based on offline data and an online prototype implementation, developed as a proof of concept for the use of these models within a feedback control loop. The use of automatically selected image features, produced by a CNN architecture in three distinct tasks comprising flow-image classification, flow-regime classification, and vapor quality prediction, confirms that these features are robust and useful, and offer a viable alternative to manually extracting image features for image-based flow studies. The successful application of the LSTM network reveals the significance of temporal information for image-based studies of two-phase flow. DA - 2022-01-27 DB - OpenUCT DP - University of Cape Town J1 - Sensors LK - https://open.uct.ac.za PY - 2022 T1 - Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow TI - Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow UR - http://hdl.handle.net/11427/36287 ER - | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/11427/36287 | |
| dc.identifier.uri | https://doi.org/10.3390/s22030996 | |
| dc.identifier.vancouvercitation | Kadish S, Schmid D, Son J, Boje E. Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow. Sensors. 2022; http://hdl.handle.net/11427/36287. | en_ZA |
| dc.publisher | Multidisciplinary Digital Publishing Institute | |
| dc.rights.license | https://creativecommons.org/licenses/by/4.0/ | |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | |
| dc.source | Sensors | |
| dc.source.journalissue | 3 | |
| dc.source.journalvolume | 22 | |
| dc.source.uri | https://www.mdpi.com/journal/sensors | |
| dc.subject | flow regime | |
| dc.subject | vapor quality | |
| dc.subject | computer vision | |
| dc.subject | machine learning | |
| dc.title | Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow | |
| dc.type | Journal Article |