Thanawut Thanavanich - Academia.edu (original) (raw)

Uploads

Papers by Thanawut Thanavanich

IAES International Journal of Artificial Intelligence (IJ-AI)

Air pollution has become one of the most significant problems impacting human health. Particulate... more Air pollution has become one of the most significant problems impacting human health. Particulate matter (PM) 2.5 is usually used as an identifier of the intensity of the pollution. The PM2.5 forecasting is essential and gainful for reducing health risks. The efficient model for forecasting PM2.5 concentration can be used in determining the period of outdoor activities, thereby reducing the impact on health. In addition, the government sector can use the forecasting model as a tool for laying down measures a burning control. In this work, the hybrid forecasting algorithms for improving accuracy are presented. The hybrid forecasting algorithms combine neural network models with historical predictive data for improving the accuracy of forecasting. The experimental results show that the proposed algorithms can reduce the mean absolute error and root mean square error of forecasting at 36% and 45%. Therefore, the proposed algorithms are not only effectively used to forecast the PM2.5 co...

2022 Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering (ECTI DAMT & NCON), 2022

ECTI Transactions on Computer and Information Technology, 1970

An inefficient energy consumption of computing resources in a large cloud datacenter is a very im... more An inefficient energy consumption of computing resources in a large cloud datacenter is a very important issue since the energy cost is now a major part of the operating expense. In this paper, the challenge of scheduling a parallel application on a cloud platform to achieve both time and energy efficiency is addressed by two new proposed algorithms Enhancing Heterogonous Earliest Finish Time (EHEFT) and Enhancing Critical Path on a Processor (ECPOP). The objective of these two algorithms is to reduce the energy consumption while achieving the best execution makespan. The algorithms use a metric that identifies and turns off the inefficient processors to reduce energy consumption. Then, the application tasks are rescheduled on fewer processors to obtain better energy efficiency. The experimental results from the simulation using real-world application workload show that the proposed algorithms not only reduce the energy consumption, but also maintain an acceptable scheduling quality. Thus, these algorithms can be employed to substantially reduce the operating cost in a large cloud computing system.

ABSTRACT: Grid system is an emerging infrastructure that promises to deliver an extremely high co... more ABSTRACT: Grid system is an emerging infrastructure that promises to deliver an extremely high computing power for many important classes of application. One of the most popular and standard methodology for grid programming is GridRPC. In this paper, the model of GridRPC performance is proposed. Then, the factors involved are measured from a real Grid system using GridRPC runtime. Finally, test applications are developed to study the behavior of GridRPC mechanism related to the performance factors defined. The understanding obtained from this work is very useful for the improvement of process assignment, and application development on the Grid system.

2021 Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunication Engineering, 2021

The problem of substandard air quality in northern Thailand has been found over the past decades.... more The problem of substandard air quality in northern Thailand has been found over the past decades. One of the current methods used by the government agencies to mitigate the problem is to release the burning control measures. A forecasting technique has been widely used as an effective approach for supporting government agencies to set the period for the measures. In this work, we present a forecasting model using hybrid neural networks consisting of two-steps of forecasting outputs. The proposed method uses a neural network model to forecast PM2.5PM_{2.5}PM2.5 concentrations and also improves the accuracy of forecasting with the linear regression technique. The experimental results show that the proposed method is able to reduce the mean absolute error of forecasting PM2.5PM_{2.5}PM2.5 concentrations from 1.95 to 0.41. Therefore, the proposed method not only effectively uses to forecast the PM2.5PM_{2.5}PM2.5 concentrations but also apply to achieve a time-series prediction in a similar context.

2019 16th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)

This paper proposes an applying to use a neural network for a pineapple sorting machine based on ... more This paper proposes an applying to use a neural network for a pineapple sorting machine based on the skin color to classification the level of the pineapple’s ripeness by dividing it into unripe pineapples, partial ripe pineapple, and fully ripe pineapples. The using Convolutional Neural Network (CNN) to help classify made the machines used for classification without many labor worker. The result after prediction from confusion metrics is 79% accuracy for unripe pineapples, 82% for partial ripe pineapples and 100% for fully ripe pineapples after training has test accuracy 98.38% and 90.77% when evaluating with the test set.

In the last decade, the consumption of coffee in Thailand is rapidly growth, because of the taste... more In the last decade, the consumption of coffee in Thailand is rapidly growth, because of the taste of coffee roasting that needs to pass with high quality standards. Normally, roasting processes to obtain a quality of roasted coffee bean requires an expert with experiences in verifying colors of the roasted coffee beans. Previously, a roasting plant showed that roasting requires meticulous processes to deliver the quantity and quality of bean to meet the needs of consumers. Therefore, the researcher proposes a technique of image processing method in order to analyze the colors of roasted coffee beans with 8 standard colors. The histogram matching technique is used to identify the degree of roasted coffee colors. The experimental results show that the technique reveals an accuracy average of 92.5 percent, compared to the standard colors of roasting coffee beans. The error occurring from the similarity of coffee bean colors brings about the implementation of a physical technique to nor...

2016 13th International Joint Conference on Computer Science and Software Engineering (JCSSE)

2018 22nd International Computer Science and Engineering Conference (ICSEC)

Grid Computing is a novel technology for the new generation of High Performance Computing. This n... more Grid Computing is a novel technology for the new generation of High Performance Computing. This new technology has the performance even greater than Super Computer-class computer in the past generation. However, such a new computing platform needs a better simulation model for the performance prediction. In this paper, we proposed a model for application development using Grid Service. The model

Grid system is an emerging infrastructure that promises to deliver an extremely high computing po... more Grid system is an emerging infrastructure that promises to deliver an extremely high computing power for many important classes of application. One of the most popular and standard methodology for grid programming is GridRPC. In this paper, the model of GridRPC performance is proposed. Then, the factors involved are measured from a real Grid system using GridRPC runtime. Finally, test applications are developed to study the behavior of GridRPC mechanism related to the performance factors defined. The understanding obtained from this work is very useful for the improvement of process assignment, and application development on the Grid system. KEYWORD: Grid system, Grid Programming, GridRPC In this paper, the model of GridRPC performance factors that measure from a real grid system is proposed. The factors are usefully to develop test application to study the behavior of GridRPC mechanism related to the performance factors defined. Furthermore, we will present a speedup of GridRPC with our test application. The understanding obtained from this work is useful for the improvement of process assignment.