PNU Journal of Applied Science, Engineering & Technology

Efficiency and Innovation in Liquefied Petroleum Gas (LPG) Conservation for Cooking: An Integrated Analysis

2026-05-14T16:06:59+07:00

This academic article presents an integrative analysis of innovations and strategies for enhancing the efficiency of liquefied petroleum gas (LPG) utilization in cooking. The study is structured into three main sections: (i) an explanation of the fundamental scientific principles of combustion and heat transfer, (ii) an analysis of technological innovations and user behavior, and (iii) an assessment of the socio-economic and policy contexts. The findings indicate that advanced gas burner technologies, such as porous radiant burners and infrared burners, can significantly improve thermal efficiency (up to 71%) while substantially reducing harmful emissions, including carbon monoxide and nitrogen oxides. In parallel, innovations in cooking utensils and modifications in user practices also play a crucial role in energy conservation. Furthermore, a case study in Thailand highlights that social barriers particularly high investment costs and limited access to information—remain critical obstacles to technology adoption. This underscores the necessity of adopting holistic policy strategies and bottom-up approaches. Accordingly, this article proposes comprehensive pathways for technology developers, consumers, and policymakers to advance sustainable energy use and improve overall quality of life.

Fall Detection System for Elderly Using Open CV and MediaPipe

2026-05-14T14:14:24+07:00

This research aims to develop a fall detection system for the elderly using the MediaPipe framework. A webcam is mounted on the ceiling, and processing is conducted on a Raspberry Pi to minimize privacy intrusions. The displayed images are represented as human skeletons. The fall detection process consists of four steps: (1) capturing images from the webcam, (2) pose estimation using the MediaPipe framework, (3) storing image data in a database, and (4) creating conditions for fall detection and sending alerts to caregivers via Twilio. The experimental results are divided into three parts: (1) the system accurately identifies falls in various positions within the test area, achieving a maximum accuracy of 99.67% in the center of the test area, (2) the system detects side falls more accurately than forward falls due to camera angle limitations, and (3) the fall alert accuracy reaches 97%. However, errors in fall detection occur when sitting postures resemble falling postures.

Automatic Vehicle License Plate Recognition System Using Computer Vision

2026-05-14T14:40:14+07:00

This research presents the development of an Automatic License Plate Recognition (ALPR) system for Thai license plates using YOLOv8 model for license plate detection and Tesseract OCR for character recognition. The developed system consists of five main components: data preparation, image pre-processing, license plate detection, character recognition and user interface and software development. The system was tested using a dataset of 791 Thai license plate images. The experimental findings indicated a license plate detection accuracy of 92%, accompanied by character recognition accuracies of 80% for white-background plates and 50% for graphic-background plates. The system can be applied in traffic management, automated parking systems, and security systems. However, the system's performance is dependent on environmental factors such as lighting conditions, camera angles, and license plate conditions.

Compressive Strength and Water Absorption of Fly Ash Based Geopolymer Containing Para Wood Ash

2026-05-14T15:03:30+07:00

This study aimed to evaluate the effects of the alkaline solution–to–binder ratio and partial replacement of fly ash with Para wood ash on the properties of geopolymer mortar. The mix variables included alkaline-to-binder ratios of 0.4, 0.6, and 0.8; a sodium silicate–to–sodium hydroxide ratio of 2.5; Para wood ash contents of 0%, 20%, 40%, and 60%; water-to-binder ratios of 0.2 and 0.3; and sand-to-binder ratios of 2.75, 3.00, and 3.25 by mass across all mixtures. Specimens were ambient-cured at 30 ± 2 °C for 24 h, with compressive strength tested at 1, 7, and 28 days and water absorption measured at 28 days. The 40P0.6A mixture exhibited the highest compressive strength of 35.93 MPa and a 24‑h water absorption of 0.88%. Increasing the Para wood ash and sand contents increased water absorption, whereas increasing the alkaline-to-binder ratio reduced it. Environmentally, this research is beneficial for lowering cement consumption and upcycling biomass waste to produce durable, high‑strength construction materials.

Comparative Heat Transfer Characteristics for a Channel with Plain Fins and Bending Fin Baffles

2026-05-14T15:44:10+07:00

This article presents an experimental investigation into the heat transfer characteristics on the surface of Plain Fins and Bending Fin Baffles, both having a total length of 2H and a thickness of 2 mm. Here, H represents the fin height, which is equal to the channel height (32 mm). The fins were designed with equal length and height dimensions, while the curved fin was bent at an angle of 120°. The fins were installed with longitudinal spacings of W = 64 mm and W = 32 mm between the upstream and downstream edges, respectively. A staggered arrangement of the fins was employed. The experiments were conducted within a Reynolds number range of Re_H = 10,000–24,000. In this study, the heat transfer distribution on the surface was measured using Thermo Liquid Crystals (TLCs) to visualise and determine the local Nusselt number (Nu) distribution. Pressure taps were also installed along the sidewalls of the channel to measure the pressure drop, which was subsequently used to calculate the friction factor and the thermal performance factor (TPF). The results indicated that the curved fin configuration enhanced the Nusselt number compared with the plain fin arrangement, and the overall thermal performance factor was found to be greater than 1 in all cases.