Journal of Measurement, Control, and Automation

Improving YOLOv8 Deep leaning model in rice disease detection by using Wise - IoU loss function

2024-11-27T18:14:20+07:00

This paper presents an improved method for a deep learning model applied to the detection of diseases in rice crops. Early detection and prevention of pests and diseases are essential to ensure effective crop productivity. The YOLOv8 deep learning model was employed to detect three common diseases in rice leaves: leaf folder, rice blast, and brown spot. To enhance the model's performance, we replaced the default CIoU loss function in YOLOv8 with WIoU, achieving an overall accuracy of 89.2%, with an improvement of 4.5% on mAP@50 and 4.4% on mAP@50-95. These results demonstrate promising potential for improving the performance and reliability of deep learning models in agricultural applications.

An improvement of the miniaturized microstrip patch antenna based on coupled microstrips for on-body sensors

2025-01-08T23:07:43+07:00

This paper presents a low-cost miniaturized microstrip patch antenna (MPA) designed for on-body applications at a frequency of 2.45 GHz. The antenna is based on the coupled-microstrip principle using FR-4 material. The microstrip transmission line feeding method is used to easily integrate the antenna into the sensor chip on the same plane. The antenna has a gain of 2.28 dBi, a bandwidth of 2.62%, an efficiency of 46.5%, and overall dimensions of 48.6×48.6×1.6 mm³. To evaluate the performance of the proposed antenna, it is compared with three different antenna versions, considering the impact of the human body on the antenna and the specific absorption rate (SAR) of the human body. The proposed antenna shows excellent compliance with the requirements for on-body sensor antennas.

Investigate model and reduction technique of common-mode noise for LLC resonant converter

2025-02-14T11:34:05+07:00

Switch Mode Power Supplies (SMPS) are extensively employed in fields such as telecommunications, information technology, and electric vehicles, where energy efficiency is increasingly demanded. Modern SMPS designs prioritize compact size, high power density, and optimal efficiency to meet these requirements. The LLC resonant topology, commonly utilized in the DC/DC conversion stage with planar transformers, enables operation at high frequencies (hundreds of kHz) and facilitates soft switching across the full load range. However, a significant drawback of this topology is electromagnetic interference (EMI), particularly common-mode (CM) noise, which arises from high switching frequencies and parasitic capacitance in the components. This challenge is further exacerbated by the use of planar transformers. This study presents a CM noise model for the Half-Bridge LLC (HB-LLC) converter based on a two-capacitor transformer representation, aiming to analyze CM noise behavior and investigate practical, cost-effective methods for its reduction.

Performance analysis of null-steering beamformers with phase-only and ampli-tude-only control via bat algorithm

2024-12-13T14:50:41+07:00

Recently, two types of BAT algorithm (BA)-based adaptive beamformers for pattern nulling in Uniformly Spaced Linear Arrays (ULA), utilizing either phase-only or amplitude-only control, have been proposed in previous studies. This paper evaluates the advantages and disadvantages of these approaches. Furthermore, to achieve higher performance, detailed recommendations for applying the optimized method in various scenarios are provided. To support this analysis, multiple ULA pattern scenarios with pre-set nulls were conducted. The simulation results are presented to validate the conclusions drawn.

Multimode space-vector overmodulation technique for enhancing voltage transfer ratio in matrix converters

2024-12-06T11:17:17+07:00

The space-vector modulation (SVM) method for matrix converters (MCs) inherently limits the voltage transfer ratio (VTR) to 0.866 in the linear modulation region, which constrains the broader application of MCs. This paper presents a detailed investigation into an overmodulation technique aimed at enhancing the VTR of MCs. The proposed method integrates the principles of inscribed circle and hexagon vectors, as well as hexagon and basic vectors, to extend the modulation range. Explicit equations for calculating the duty cycles of active and zero vectors are derived to achieve improved VTR. The effectiveness of the approach is validated through simulation results, demonstrating accurate output voltage fundamental components.

Building of short circuit ampere - second characteristics for a series of amorphous core transformers

2025-02-14T17:18:27+07:00

In this paper, the finite element method (FEM) with Ansys Maxwell software is used to simulate a series of ten amorphous core transformers (AMCT) power. These AMCT operate in the three-phase short-circuit fault mode on the low voltage winding. The study determines the maximum short-circuit current on the low voltage (LV) and high voltage (HV) winding; of 10 different AMCT power. From this result, the study builds the short-circuit ampere-second characteristic of AMCT. Thanks to that, the transformer operator determines the correct short-circuit cutting time to support the selection of the right fuse cut out (FCO). Thereby, the AMCT will avoid the harmful effects of electromagnetic force that causes the AMCT winding to break. This research result helps us to determine the correct short-circuit cutting time of the FCO to disconnect the transformer's electrical connection from the high-voltage grid. From there, we protect against short-circuit breaking and displacement of the winding caused by short-circuit electromagnetic force.

Empirical mode decomposition-based OS-ELM for short-term solar irradiance forecasting: A case study in Hanoi

2025-02-25T15:08:11+07:00

Solar irradiation forecasting is vital for the growth of renewable energy sources. In this paper, we propose a hybrid model that integrates Empirical Mode Decomposition (EMD) and an online sequential extreme learning machine (OS-ELM) for multiple steps ahead forecasting of solar irradiation. Initially, the solar irradiation dataset is processed and cleaned. Then, using the EMD model combined with the autocorrelation function, the cleaned dataset is decomposed into several Intrinsic Mode Functions (IMFs) and white noise, which is removed. Each IMF is subsequently predicted using OS-ELM. The final solar irradiation forecast is derived by aggregating the predictions from all Intrinsic Mode Functions (IMFs). The model's performance was assessed through forecasting solar irradiation in Hanoi, using weather data from 2018. The data was collected at 1-hour intervals and utilized for single-step, 12-step, and 24-step ahead forecasts. The forecasting accuracy of the proposed model was compared with four other models, including both single and hybrid approaches: Bidirectional Long Short-Term Memory network, ELM, OS-ELM, and EMD-ELM. Two evaluation metrics of RMSE and MAE were used to assess the forecasting performance of the models. The computational results show that when the multi-step ahead increases, accuracy decreases. In any case, the proposed method outperforms the others, achieving the lowest error rates at 18,01 W/m²for RMSE and 8,51 W/m² for MAE at 24-step.

An MIQP model for microgrid unit commitment incorporating grid-connected/islanded operating modes and power network

2025-01-22T13:25:33+07:00

This paper presents a mixed-integer quadratic programming (MIQP) model to address the day-ahead unit commitment (UC) problem for microgrids. The microgrid in this study includes conventional generators (CGs), renewable energy sources, and energy storage systems. The transition between grid-connected and islanded operating modes, as well as power grid constraints, is integrated into the proposed MIQP optimization model. The objective of the suggested formulation is to minimize the total operating costs of the microgrid, including the operating costs of conventional generating units and the expenses of purchasing electricity from the main grid. The MIQP model is derived from a mixed-integer nonlinear programming (MINLP) model by linearizing the product of two continuous variables and quadratic elements. The proposed MIQP model is implemented using the GAMS language with the commercial solver CPLEX and evaluated on a modified IEEE 33-node microgrid. The computational results demonstrate the effectiveness of the proposed optimization approach.

Optimization of sliding mode montroller parameters for planar parallel robot

2024-12-04T11:32:56+07:00

In this paper, we present a method for optimizing the parameters of a Sliding Mode Controller (SMC) using the Particle Swarm Optimization (PSO) algorithm. The SMC is well-known for its robust performance in handling system uncertainties and external disturbances. However, selecting the optimal parameters for the SMC plays a crucial role in achieving high performance in trajectory tracking tasks. Manual tuning often leads to suboptimal solutions, especially for complex nonlinear systems. To address this issue, we apply the PSO algorithm, a powerful parameter optimization technique inspired by the swarm behavior of birds or fish, to automatically search for the optimal control parameters. The optimized SMC parameters are then applied to the trajectory tracking control of a planar parallel robot. Simulation results, verified using Matlab Simulink, demonstrate high performance in trajectory tracking for the planar parallel robot.

Study on wind energy forecasting of Quang Tri region using artificial neural networks NARX

2025-02-13T11:00:29+07:00

Quang Tri Province has a rich source of wind energy, but effectively harnessing this valuable resource requires consideration of several factors. One of these is wind forecast databases, which are crucial to optimizing wind energy utilization. The purpose of this study is to develop a wind speed forecasting model based on artificial neural networks using the NARX structure. Our model was constructed using historical data on wind speed and meteorological factors in Quang Tri. A local monitoring station's wind speed is used as an input variable for the NARX model. The MAPE and MAE metrics indicate that the forecasting model is highly effective, even under changing weather conditions. The proposed model has also been compared with models that use additional correlated input data, including wind direction, temperature, pressure and humidity, at various altitudes of 40m, 60m, 80m. The study also shows that choosing appropriate NARX model parameters can potentially expand the early forecast range up to 7 days. As a result of this validation, not only is the accuracy of the model determined, but also how the model can be used in order to harness wind energy efficiently and sustainably. Local wind energy management systems can benefit from this forecasting model by improving resource utilization and minimizing risks.

Robust adaptive control for a class of uncertain affine nonlinear systems

2025-01-02T10:36:02+07:00

This paper presents the control synthesis method for a class of uncertain affine nonlinear systems based on adaptive and robust control techniques. The result is an identification law for the uncertain components, and a robust controller is constructed based on sliding mode control. The research results are simulated in Matlab Simulink with specific examples to demonstrate the correctness of the method. Moreover, the control system proposed in this paper operates stably and achieves high control quality under uncertain conditions of plants. These research results can be applied to the design and development of control systems for appropriate control targets in practical applications.