Prawito Prajitno
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Prawito Prajitno has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
D S Soejoko
Department of Physics, Faculty of Mathematics and Natural Science, Universitas Indonesia, Depok, 16424, Indonesia
4 shared publications- Djarwani Soeharso Soejoko
Universitas Indonesia
3 shared publications - Sastra Kusuma Wijaya
Department of Physics, Universitas Indonesia (UI), Depok, 16424, Indonesia
3 shared publications D Handoko
Department of Physics, Faculty of Mathematics and Natural Science, Universitas Indonesia, Depok, Indonesia
2 shared publicationsA Aida
Department of Physics, Faculty of Mathematics and Natural Science, Universitas Indonesia, Depok, 16424, Indonesia
1 shared publicationA Saputra
CTECH Laboratories, Alam Sutera, Tangerang, 15143, Banten, Indonesia
1 shared publication
Latest Publications
- 9 July 2026
- Published by Taylor & Francis in Nuclear Technology
- p. 1-26
A Review of Primary Activity Measurement Based on β-γ Coincidence Counting
I. The use of radionuclides for a wide range of applications, especially in the medical field, has increased significantly in recent years. Accurate measurement of medical radionuclide activity is ...
- 9 July 2026
- Published by Taylor & Francis in Nuclear Technology
- p. 1-26
A Review of Primary Activity Measurement Based on β-γ Coincidence Counting
I. The use of radionuclides for a wide range of applications, especially in the medical field, has increased significantly in recent years. Accurate measurement of medical radionuclide activity is ...
- Open Access
- 26 May 2026
- Published by Vilnius Gediminas Technical University in Aviation
- Vol. 30 (2) , 171-180
Investigation of C-rate and post-discharge rest effects on voltage recovery and capacity in Li-ion battery pack for LEO satellite applications
Lithium-ion batteries are extensively employed in LEO satellites because of their high energy density, lightweight design, and superior cycle efficiency; however, their on-orbit performance is strongly influenced by discharge rate and post-discharge voltage relaxation during repetitive eclipse-sunlight cycling. In this study, an experimental investigation was conducted on an 8s6p Li-ion battery pack with a nominal capacity of 19.2 Ah and nominal voltage of 28.8 V. The battery was tested under three different discharge rates – C/20, C/10, and C/5 – followed by rest periods of 30 minutes, 1 hour, and 2 hours. The results indicate that increasing the C-rate significantly enhances the voltage recovery amplitude, whereas extending the rest duration provides only marginal improvement in effective capacity. The voltage relaxation behaviors is well described by a double-exponential model. The proposed model (R2 > 0.99, RMSE < 0.03) confirms the coexistence of fast and slow relaxation processes driven by Li-ion redistribution and electrode structure relaxation. These findings demonstrate that orbit-aware management of discharge rates and rest phase can improve voltage stability and effective energy utilization without increasing system mass or volume, offering practical guidance for battery management strategies in power-constrained LEO missions, including CubeSats and nanosatellites.
- 26 May 2026
- Published by Vilnius Gediminas Technical University in Aviation
- Vol. 30 (2) , 171-180
Open AccessInvestigation of C-rate and post-discharge rest effects on voltage recovery and capacity in Li-ion battery pack for LEO satellite applications
Lithium-ion batteries are extensively employed in LEO satellites because of their high energy density, lightweight design, and superior cycle efficiency; however, their on-orbit performance is strongly influenced by discharge rate and post-discharge voltage relaxation during repetitive eclipse-sunlight cycling. In this study, an experimental investigation was conducted on an 8s6p Li-ion battery pack with a nominal capacity of 19.2 Ah and nominal voltage of 28.8 V. The battery was tested under three different discharge rates – C/20, C/10, and C/5 – followed by rest periods of 30 minutes, 1 hour, and 2 hours. The results indicate that increasing the C-rate significantly enhances the voltage recovery amplitude, whereas extending the rest duration provides only marginal improvement in effective capacity. The voltage relaxation behaviors is well described by a double-exponential model. The proposed model (R2 > 0.99, RMSE < 0.03) confirms the coexistence of fast and slow relaxation processes driven by Li-ion redistribution and electrode structure relaxation. These findings demonstrate that orbit-aware management of discharge rates and rest phase can improve voltage stability and effective energy utilization without increasing system mass or volume, offering practical guidance for battery management strategies in power-constrained LEO missions, including CubeSats and nanosatellites.
- Open Access
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 15 (2) , 2282-2290
Radionuclide identification system using convolution neural network for environmental radiation monitoring
Radionuclide identification is an important task for nuclear safety and security aspects, especially to environmental radiation monitoring systems. This study aims to build an automatic radionuclide identification system that can be applied in environmental radiation monitoring stations. The gamma energy spectrum was obtained by varying radionuclide types, measurement time and source distance using a scintillation detector. The dataset was collected by converting gamma energy spectrum into images, data pre-processing by removing background noise and normalizing the gamma spectrum. Automatic identification is demonstrated as a development method based on convolutional neural network (CNN) algorithm, where the images come from gamma-ray spectrum in the form of photoelectric peak characteristic. Three CNN architectures are used to train the model, which are VGG-16, AlexNet and Xception. The performance of each model is evaluated using accuracy, precision and recall to find the appropriate architecture. The most optimum results are shown by VGG-16 with an accuracy of 97.72%, a precision of 97.75% and a recall of 97.71%. The models are critically reviewed and it is concluded that the developed models can be further implemented on embedded devices utilizing the tiny machine learning (TinyML) platform in environmental radiation monitoring systems.
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 15 (2) , 2282-2290
Open AccessRadionuclide identification system using convolution neural network for environmental radiation monitoring
Radionuclide identification is an important task for nuclear safety and security aspects, especially to environmental radiation monitoring systems. This study aims to build an automatic radionuclide identification system that can be applied in environmental radiation monitoring stations. The gamma energy spectrum was obtained by varying radionuclide types, measurement time and source distance using a scintillation detector. The dataset was collected by converting gamma energy spectrum into images, data pre-processing by removing background noise and normalizing the gamma spectrum. Automatic identification is demonstrated as a development method based on convolutional neural network (CNN) algorithm, where the images come from gamma-ray spectrum in the form of photoelectric peak characteristic. Three CNN architectures are used to train the model, which are VGG-16, AlexNet and Xception. The performance of each model is evaluated using accuracy, precision and recall to find the appropriate architecture. The most optimum results are shown by VGG-16 with an accuracy of 97.72%, a precision of 97.75% and a recall of 97.71%. The models are critically reviewed and it is concluded that the developed models can be further implemented on embedded devices utilizing the tiny machine learning (TinyML) platform in environmental radiation monitoring systems.
- Open Access
- 31 July 2024
- Published by National Research and Innovation Agency in Journal of Mechatronics, Electrical Power, and Vehicular Technology
- Vol. 15 (1) , 68-81
ELM-based control system applications: A bibliometric analysis and review
This study conducts a bibliometric analysis of the extreme learning machine (ELM) research, with a particular emphasis on ELM-based control systems and applications. The objective of this study is to identify research trends, collaboration opportunities, and challenges in ELM applications. The analysis comprises the identification and retrieval of 3,174 articles from Scopus between 2018 and 2023. VOSviewer 1.6.20 is used for data interpretation, identifying six distinct keyword clusters and revealing both well-established research areas and emerging fields with significant potential for future exploration. Key research trends indicate a shift towards advanced or hybrid approaches, with recent interest in integrating optimization techniques. In the analysis, opportunities for collaboration with leading researchers are also highlighted. The findings emphasize the wide range of applications for ELM in improving the robustness of control systems while also highlighting important issues that need to be addressed. Finally, this study provides valuable insights into the current state and future directions of ELM research, especially ELM-based control systems.
- 31 July 2024
- Published by National Research and Innovation Agency in Journal of Mechatronics, Electrical Power, and Vehicular Technology
- Vol. 15 (1) , 68-81
Open AccessELM-based control system applications: A bibliometric analysis and review
This study conducts a bibliometric analysis of the extreme learning machine (ELM) research, with a particular emphasis on ELM-based control systems and applications. The objective of this study is to identify research trends, collaboration opportunities, and challenges in ELM applications. The analysis comprises the identification and retrieval of 3,174 articles from Scopus between 2018 and 2023. VOSviewer 1.6.20 is used for data interpretation, identifying six distinct keyword clusters and revealing both well-established research areas and emerging fields with significant potential for future exploration. Key research trends indicate a shift towards advanced or hybrid approaches, with recent interest in integrating optimization techniques. In the analysis, opportunities for collaboration with leading researchers are also highlighted. The findings emphasize the wide range of applications for ELM in improving the robustness of control systems while also highlighting important issues that need to be addressed. Finally, this study provides valuable insights into the current state and future directions of ELM research, especially ELM-based control systems.
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3163 (1) , 050003
Geothermal power plant performance estimation using ANFIS-PCA and ANFIS-GA
Monitoring the performance of the power plant is important to see the overall system efficiency. The performance in geothermal power plants can be viewed from t Monitoring the performance of the power plant is important to see the overall system efficiency. The performance in geothermal power plants can be viewed from t
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3163 (1) , 050003
Geothermal power plant performance estimation using ANFIS-PCA and ANFIS-GA
Monitoring the performance of the power plant is important to see the overall system efficiency. The performance in geothermal power plants can be viewed from t Monitoring the performance of the power plant is important to see the overall system efficiency. The performance in geothermal power plants can be viewed from t
- Open Access
- 1 March 2023
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 12 (1) , 171-179
Artificial neural network for cervical abnormalities detection on computed tomography images
Cervical cancer is the second deadliest after breast cancer in Indonesia. Sundry diagnostic imaging modalities had been used to decide the location and severity of cervical cancer, one among those is computed tomography (CT) Scan. This study handles a CT image dataset consisting of two categories, abnormal cervical images of cervical cancer patients and normal cervix images of patients with other diseases. It focuses on the ability of segmentation and classification programs to localize cervical cancer areas and classify images into normal and abnormal categories based on the features contained in them. We conferred a novel methodology for the contour detection round the cervical organ classified with artificial neural network (ANN) which was employed to categorize the image data. The segmentation algorithm used was a region-based snake model. The texture features of the cervical image area were arranged in the form of gray level co-occurrence matrix (GLCM). Support vector machine (SVM) had been added to determine which algorithm was better for comparison. Experimental results show that ANN model has better receiver operating characteristic (ROC) parameter values than SVM model’s and existing approach’s regarding 96.2% of sensitivity, 95.32% of specificity, and 95.75% of accuracy.
- 1 March 2023
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 12 (1) , 171-179
Open AccessArtificial neural network for cervical abnormalities detection on computed tomography images
Cervical cancer is the second deadliest after breast cancer in Indonesia. Sundry diagnostic imaging modalities had been used to decide the location and severity of cervical cancer, one among those is computed tomography (CT) Scan. This study handles a CT image dataset consisting of two categories, abnormal cervical images of cervical cancer patients and normal cervix images of patients with other diseases. It focuses on the ability of segmentation and classification programs to localize cervical cancer areas and classify images into normal and abnormal categories based on the features contained in them. We conferred a novel methodology for the contour detection round the cervical organ classified with artificial neural network (ANN) which was employed to categorize the image data. The segmentation algorithm used was a region-based snake model. The texture features of the cervical image area were arranged in the form of gray level co-occurrence matrix (GLCM). Support vector machine (SVM) had been added to determine which algorithm was better for comparison. Experimental results show that ANN model has better receiver operating characteristic (ROC) parameter values than SVM model’s and existing approach’s regarding 96.2% of sensitivity, 95.32% of specificity, and 95.75% of accuracy.
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2941 (1)
Improvement of the DDPG algorithm via twin delayed DDPG (TD3) on vertical rocket landing control system
The Double Deep Policy Gradient or DDPG Reinforcement Learning algorithm has a tendency to overestimate computations, which results in learning policies that are not optimum. When DDPG agent is applied in the rocket landing control system, high frequency chattering occurs at the main engine’s control command and this gives substantial drawbacks for the liquid propulsion life longevity. The Twin-Delayed Deep Deterministic Policy Gradient (TD3) Reinforcement Learning agent is proposed to overcame this chattering by delaying the landing rocket control command, as simulated in two dimensions. This TD3 controller is shown to be able to dampen the engine command as its achievements are contrasted with the outcomes of the DDPG one.
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2941 (1)
Open AccessImprovement of the DDPG algorithm via twin delayed DDPG (TD3) on vertical rocket landing control system
The Double Deep Policy Gradient or DDPG Reinforcement Learning algorithm has a tendency to overestimate computations, which results in learning policies that are not optimum. When DDPG agent is applied in the rocket landing control system, high frequency chattering occurs at the main engine’s control command and this gives substantial drawbacks for the liquid propulsion life longevity. The Twin-Delayed Deep Deterministic Policy Gradient (TD3) Reinforcement Learning agent is proposed to overcame this chattering by delaying the landing rocket control command, as simulated in two dimensions. This TD3 controller is shown to be able to dampen the engine command as its achievements are contrasted with the outcomes of the DDPG one.
- Open Access
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2858 (1)
Image fusion method for microwave tomography
Tomography is a process to obtain a two-dimensional image of a three-dimensional object separately between the object and other objects. Image depiction on microwave tomography has limitations at certain frequencies. Making information from the resulting image incomplete, especially on the type of material from more than one object in an experiment. In this study, using foam and plastic cylinders produced clear foam images at frequencies of 3 GHz and 5 GHz, but plastic cylinders were seen at frequencies of 9 GHz and 10 GHz. The result of image reconstruction requires an image fusion method so that both objects can be seen in one image. The image fusion method used is the wavelet, pyramid, and Gaussian-Laplacian methods. The value of the structural similarity index measure (ssim) generated after the image fusion method is applied has increased. In this study, it can be concluded that the image fusion method with the Gauss-Laplacian method gives better results than the wavelet and pyramid methods.
- 1 January 2023
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2858 (1)
Open AccessImage fusion method for microwave tomography
Tomography is a process to obtain a two-dimensional image of a three-dimensional object separately between the object and other objects. Image depiction on microwave tomography has limitations at certain frequencies. Making information from the resulting image incomplete, especially on the type of material from more than one object in an experiment. In this study, using foam and plastic cylinders produced clear foam images at frequencies of 3 GHz and 5 GHz, but plastic cylinders were seen at frequencies of 9 GHz and 10 GHz. The result of image reconstruction requires an image fusion method so that both objects can be seen in one image. The image fusion method used is the wavelet, pyramid, and Gaussian-Laplacian methods. The value of the structural similarity index measure (ssim) generated after the image fusion method is applied has increased. In this study, it can be concluded that the image fusion method with the Gauss-Laplacian method gives better results than the wavelet and pyramid methods.
- 23 July 2021
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2374 (1) , 020013
Temperature and level control in a water thermal mixing process by using neural network controller
In the industrial world, mixing operations are very widely used to process raw materials into products such as petroleum, chemicals, and various other types, so increasing the effectiveness of mixing operations is necessary. Usually, in industrial-scale plants, the conventional PID controller is used in the main controllers that are relied upon, but they often underperform, especially when dealing with non-linear systems. In this research, a neural network (NN)-based controller is proposed to overcome that problem. The plant model used for the simulation in this research is a mixing water plant, where the temperature and water level in the mixture tank will be controlled. This plant consists of 2 water flow inputs, namely cold water and hot water, which flows into the mixed tank where the temperature and water level of the mixture will be maintained and controlled according to the desired set points by regulating the flow of cold water and hot water. There are two types of NN-based controllers in this simulation study. The first is NN-based controller which has inputs in the form of the set point (SP), the current step (n) process variable (PV), and the previous step (n-1) process variable, while the second is a NN-based controller with set point (SP), error, and change of error for the inputs. Both of these NN-based controllers were developed using feed-forward neural networks, and the simulation was conducted by using MATLAB/SIMULINK. The simulation results show that the proposed NN-based controllers provide better performances when compared to conventional PID controllers. The best performance is obtained using the NN-based controller that has inputs in the form of set point, error, and change of error, with settling time of 472.7 s and rise time of 242.0 s in controlling the temperature, and settling time of 984.4 s, rise time about 447.6 s in controlling the level, that is faster than the PID controller which has 613.7 s and 393.9 s for settling time and rise time respectively in controlling the temperature, while gives settling time of 3216 s and rise time 412.8 s in controlling the level. Moreover, the NN-based controller produces a system response that has no overshoot at all.
- 23 July 2021
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2374 (1) , 020013
Temperature and level control in a water thermal mixing process by using neural network controller
In the industrial world, mixing operations are very widely used to process raw materials into products such as petroleum, chemicals, and various other types, so increasing the effectiveness of mixing operations is necessary. Usually, in industrial-scale plants, the conventional PID controller is used in the main controllers that are relied upon, but they often underperform, especially when dealing with non-linear systems. In this research, a neural network (NN)-based controller is proposed to overcome that problem. The plant model used for the simulation in this research is a mixing water plant, where the temperature and water level in the mixture tank will be controlled. This plant consists of 2 water flow inputs, namely cold water and hot water, which flows into the mixed tank where the temperature and water level of the mixture will be maintained and controlled according to the desired set points by regulating the flow of cold water and hot water. There are two types of NN-based controllers in this simulation study. The first is NN-based controller which has inputs in the form of the set point (SP), the current step (n) process variable (PV), and the previous step (n-1) process variable, while the second is a NN-based controller with set point (SP), error, and change of error for the inputs. Both of these NN-based controllers were developed using feed-forward neural networks, and the simulation was conducted by using MATLAB/SIMULINK. The simulation results show that the proposed NN-based controllers provide better performances when compared to conventional PID controllers. The best performance is obtained using the NN-based controller that has inputs in the form of set point, error, and change of error, with settling time of 472.7 s and rise time of 242.0 s in controlling the temperature, and settling time of 984.4 s, rise time about 447.6 s in controlling the level, that is faster than the PID controller which has 613.7 s and 393.9 s for settling time and rise time respectively in controlling the temperature, while gives settling time of 3216 s and rise time 412.8 s in controlling the level. Moreover, the NN-based controller produces a system response that has no overshoot at all.
- Open Access
- 30 April 2021
- Published by Institute of Research and Community Services Diponegoro University (LPPM UNDIP) in Journal of Physics and Its Applications
- Vol. 3 (2) , 136-141
Evaluation and Implementation of Otsu and Active Contour Segmentation in Contrast-Enhanced Cardiac CT Images
The CT cardiac acquisition process is usually conducted by using an additional image with contrast medium that is injected inside the body and reconstructed by a radiologist using an integrated CT Scan software with the aim to find the morphology and volume dimension of the heart and coronary arteries. In fact, the data obtained from the hospital are raw data without segmented contour from a radiologist. For the purpose of automation, dataset is needed to be used as input data for further program development. This study is focused on the evaluation of the segmentation results of CT cardiac images using Otsu threshold and active contour algorithm with the aim to make a dataset for the heart volume quantification that can be used interactively as an alternative to integrated CT scan software. 2D contrast enhanced cardiac CT from 6 patients using image processing techniques was run on Matlab software. Of the 689 slices that was used, as many as (73.75 ± 19.41)%of CT cardiac slices have been segmented properly, (19.15 ± 19.61)%of the slices that were segmented included the spine bone, (1.36 ± 0.98)%of the slices did not include all region of the heart, (16.58 ± 15.26)%of the slices included other organs with the consistency from the measurement proven from inter-observer variability to produce r = 0,9941.The result is due to the geometry influence from the diameter of the patient’s body thickness that tends to be thin.
- 30 April 2021
- Published by Institute of Research and Community Services Diponegoro University (LPPM UNDIP) in Journal of Physics and Its Applications
- Vol. 3 (2) , 136-141
Open AccessEvaluation and Implementation of Otsu and Active Contour Segmentation in Contrast-Enhanced Cardiac CT Images
The CT cardiac acquisition process is usually conducted by using an additional image with contrast medium that is injected inside the body and reconstructed by a radiologist using an integrated CT Scan software with the aim to find the morphology and volume dimension of the heart and coronary arteries. In fact, the data obtained from the hospital are raw data without segmented contour from a radiologist. For the purpose of automation, dataset is needed to be used as input data for further program development. This study is focused on the evaluation of the segmentation results of CT cardiac images using Otsu threshold and active contour algorithm with the aim to make a dataset for the heart volume quantification that can be used interactively as an alternative to integrated CT scan software. 2D contrast enhanced cardiac CT from 6 patients using image processing techniques was run on Matlab software. Of the 689 slices that was used, as many as (73.75 ± 19.41)%of CT cardiac slices have been segmented properly, (19.15 ± 19.61)%of the slices that were segmented included the spine bone, (1.36 ± 0.98)%of the slices did not include all region of the heart, (16.58 ± 15.26)%of the slices included other organs with the consistency from the measurement proven from inter-observer variability to produce r = 0,9941.The result is due to the geometry influence from the diameter of the patient’s body thickness that tends to be thin.