PropNet-R: A Custom CNN Architecture for Quantitative Estimation of Propane Gas Concentration Based on Thermal Images for Sustainable Safety Monitoring

Luis Alberto Holgado-Apaza (), Jaime Cesar Prieto-Luna, Edgar E. Carpio-Vargas, Nelly Jacqueline Ulloa-Gallardo, Yban Vilchez-Navarro, José Miguel Barrón-Adame, José Alfredo Aguirre-Puente, Dalmiro Ramos Enciso, Danger David Castellon-Apaza and Danny Jesus Saman-Pacamia
Additional contact information
Luis Alberto Holgado-Apaza: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru
Jaime Cesar Prieto-Luna: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru
Edgar E. Carpio-Vargas: Departamento Académico de Ingeniería Estadística e Informática, Universidad Nacional del Altiplano-Puno, Puno 21001, Peru
Nelly Jacqueline Ulloa-Gallardo: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru
Yban Vilchez-Navarro: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru
José Miguel Barrón-Adame: Departamento de Mantenimiento Industrial, Universidad Tecnológica del Suroeste de Guanajuato, Valle de Santiago 38407, Mexico
José Alfredo Aguirre-Puente: Departamento de Mantenimiento Industrial, Universidad Tecnológica del Suroeste de Guanajuato, Valle de Santiago 38407, Mexico
Dalmiro Ramos Enciso: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru
Danger David Castellon-Apaza: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru
Danny Jesus Saman-Pacamia: Departamento Académico de Ingeniería de Sistemas e Informática, Facultad de Ingeniería, Universidad Nacional Amazónica de Madre de Dios, Puerto Maldonado 17001, Peru

Sustainability, 2025, vol. 17, issue 21, 1-24

Abstract: Liquefied petroleum gas (LPG), composed mainly of propane and butane, is widely used as an energy source in residential, commercial, and industrial sectors; however, its high flammability poses a critical risk in the event of accidental leaks. In Peru, where LPG constitutes the main domestic energy source, leakage emergencies affect thousands of households each year. This pattern is replicated in developing countries with limited energy infrastructure. Early quantitative detection of propane, the predominant component of Peruvian LPG (~60%), is essential to prevent explosions, poisoning, and greenhouse gas emissions that hinder climate change mitigation strategies. This study presents PropNet-R, a convolutional neural network (CNN) designed to estimate propane concentrations (ppm) from thermal images. A dataset of 3574 thermal images synchronized with concentration measurements was collected under controlled conditions. PropNet-R, composed of four progressive convolutional blocks, was compared with SqueezeNet, VGG19, and ResNet50, all fine-tuned for regression tasks. On the test set, PropNet-R achieved MSE = 0.240, R 2 = 0.614, MAE = 0.333, and Pearson’s r = 0.786, outperforming SqueezeNet (MSE = 0.374, R 2 = 0.397), VGG19 (MSE = 0.447, R 2 = 0.280), and ResNet50 (MSE = 0.474, R 2 = 0.236). These findings provide empirical evidence that task-specific CNN architectures outperform generic transfer learning models in thermal image-based regression. By enabling continuous and quantitative monitoring of gas leaks, PropNet-R enhances safety in industrial and urban environments, complementing conventional chemical sensors. The proposed model contributes to the development of sustainable infrastructure by reducing gas-related risks, promoting energy security, and strengthening resilient, safe, and environmentally responsible urban systems.

Keywords: propane detection; thermal imaging; convolutional neural networks; regression models; gas concentration estimation; industrial safety; sustainable safety; environmental monitoring (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/17/21/9801/pdf (application/pdf)
https://www.mdpi.com/2071-1050/17/21/9801/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:17:y:2025:i:21:p:9801-:d:1786769

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().