Thailand Electrical Engineering Journal (TEEJ)

A Half-Bridge-Based Inverter Topology for Multi-Transmitter Coil Wireless Power Transfer System

Ekkachai Chaidee — 2026-06-02

This paper presents a half-bridge–based inverter topology for multi-transmitter wireless power transfer (WPT) systems. The proposed topology reduces the number of power switches compared with conventional full-bridge multi-transmitter systems while allowing independent control of each transmitter coil using a shared DC power source. An analytical model of the multi-transmitter WPT system is developed and simplified into an equivalent two-coil model to analyze output power and transfer efficiency. The operating modes of the inverter and a hierarchical transmitter-side control scheme with outer voltage and inner current control loops are presented to regulate output voltage and control individual transmitter coil currents. The proposed system is experimentally validated using a prototype operating at 67 kHz. Experimental results demonstrate stable inverter operation, effective voltage regulation under load variation, good dynamic response under step load changes, and improved transfer efficiency under receiver coil displacement compared with a conventional single-transmitter system. The results confirm that the proposed inverter topology reduces system complexity while improving misalignment tolerance and extending the effective operating area of multi-transmitter wireless power transfer systems.

Bit Error Probability Analysis for Massive MIMO Wireless Communications utilizing Zero Forcing Precoding

Ditsapon Chumchewkul — 2026-06-02

This article derives bit error probability (BEP) for massive multiple-input, multiple-output (M-MIMO) wireless communication, utilizing zero-forcing precoding. Frequency-flat fading channel is focused on this research work. The author utilizes the first-order Neumann series to analyze a closed-form expression for the normalization factor of the M-MIMO system, and proves that the probability distribution function (PDF) for the effective noise approaches the Gaussian distribution. A closed-form expression for the variance is additionally analyzed, and the derived PDF is then chosen to analyze the BEP for the M-MIMO system, utilizing B/QPSK (Binary/Quadrature phase shift keying) and 𝑀-QAM (𝑀-ary Quadrature amplitude modulation), with Gray-coded mapping. The analytical results are validated through Monte-Carlo simulation, and the empirical results confirmed that the derived BEP significantly matched the exact results. Focusing on a system with 512 transmit antennas and 20 users, the derived BEP for 16-QAM at 𝐸𝑏/𝑁0 = -5 dB was 4 × 10-3, and the deviation between the analysis and the exact BEP was only 4.78 × 10-6. Thus, the proposed BEP analysis can be utilized for analyzing performance of M-MIMO system efficiently.

Development of an IoT-Based Smart Farming System Powered by Solar Energy with Supplemental LED Lighting for Green Cos Lettuce Cultivation

Piyapat Poungsri — 2026-06-02

This paper presents the design and implementation of a smart farming system for Green Cos lettuce cultivation using solar energy as the primary power source. The objective is to optimize resource usage and enhance plant growth through an automated control system based on the ESP8266 microcontroller. The system monitors and controls environmental factors, including soil moisture, ambient temperature, and light intensity, accessible via a web application. Additionally, LED grow lights were employed to extend the photoperiod by 2 hours daily. The experimental results compared the growth of Green Cos lettuce between the proposed smart farm system and a conventional method over 35 days. The results revealed that the smart farm system significantly improved growth rates; the average height of lettuce in the smart farm was 51 cm, compared to 41.5 cm in the conventional method. Furthermore, the smart farm system reduced the harvest time from 42 days to 35 days. These findings demonstrate the efficiency of integrating IoT and renewable energy in precision agriculture.

Optimal Allocation of Renewable Distributed Generation Considering Harmonic Distortion and Hosting Capacity in Active Distribution Networks

Trieu Ngoc Ton — 2026-06-02

The increasing penetration of renewable distributed generation (DG) has significantly transformed the operation of modern power distribution networks. While integrating renewable energy sources such as photovoltaic and wind systems improves energy sustainability and reduces transmission losses, it also introduces several operational challenges, including voltage fluctuations, increased power losses, reverse power flow, and harmonic distortion caused by inverter-based technologies. These issues can degrade power quality and limit the hosting capacity (HC) of distribution networks. To address these challenges, this paper proposes a multi-objective optimization framework for the optimal placement and sizing of renewable DG units while considering harmonic distortion constraints and hosting capacity enhancement. The proposed model simultaneously minimizes active power losses, voltage deviation, and total harmonic distortion (THD), while maximizing DG hosting capacity under network operational constraints. An improved metaheuristic optimization algorithm is employed to determine the optimal DG locations and capacities. The effectiveness of the proposed approach is validated using the IEEE 33-bus and 69-bus radial distribution systems. Simulation results demonstrate that the optimal DG integration significantly reduces power losses, improves voltage profiles, mitigates harmonic distortion, and enhances the hosting capacity of the distribution network.

Real-Time Drowsiness Detection System for Driver Safety Based on Computer Vision

Natthariya Laopracha — 2026-06-02

This study proposes a lightweight real-time driver drowsiness detection system based on traditional computer vision and machine learning techniques for resource-constrained environments. The proposed framework integrates Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) for face detection, followed by facial landmark analysis and Eye Aspect Ratio (EAR) computation using Euclidean distance to estimate eye closure and detect drowsiness. The system was evaluated using real-world driving video datasets collected from multiple drivers under practical driving conditions, including variations in illumination, head pose, and driver behavior. Experimental results demonstrate that the proposed method achieves an average accuracy of 82% with a low False Negative Rate (FNR) of 0.099 while maintaining low computational complexity. The proposed approach provides a practical balance between detection performance and hardware efficiency, making it suitable for real-time deployment on CPU-based and low-cost embedded systems.