Desmarita Leni

Ms. Desmarita Leni


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 February 2024 in Engineering Proceedings
Verified
Conference Paper
Open access
PDF file
DLDesmarita Leni
Desmarita Leni
DKDytchia Septi Kesuma
Dytchia Septi Kesuma
MAMaimuzar
Maimuzar
HAHaris
Haris
SASicilia Afriyani
Sicilia Afriyani
This study involves data augmentation modeling using Generative Adversarial Networks (GAN) on the tensile test data of austenitic stainless steel, which encompasses chemical compositions, heat treatments, and mechanical properties. The synthetic data generated by GAN is subsequently used as the training dataset for six different algorithm models. The best-performing algorithm is selected based on the best evaluation metric values. The results of the Kolmogorov–Smirnov (KS) test indicate that the distribution of synthetic data does not significantly differ from the distribution of experimental data. Furthermore, the training results of predictive models employing synthetic data with the six machine learning algorithms demonstrate that the gradient boosting model exhibits superior performance in predicting the mechanical properties of austenitic stainless steel.

14 December 2022 in Jurnal Engine: Energi, Manufaktur, dan Material
Verified
Journal Article
Open access
PDF file
DLDesmarita Leni
Desmarita Leni
This study designs and compares optimal machine learning models to predict the mechanical properties of aluminum, including Yield Strength (YS) and Tensile Strength (TS), based on the percentage composition of aluminum's chemical elements. The machine learning modeling in this study has nine input variables consisting of aluminum chemical elements such as Al, Mg, Zn, Ti, Cu, Mn, Cr, Fe, Si, and two output or target variables consisting of YS and TS. Additionally, Heatmap correlation is used to observe the correlation between chemical elements and the mechanical properties of aluminum. Three machine learning algorithms, namely Decision Tree (DT), Random Forest (RF), and Artificial Neural Network (ANN), are compared in this study. The comparison of these algorithms shows that Random Forest (RF) outperforms the other algorithms in predicting YS with MAE of 11.44, RMSE of 14.282, and R value of 0.93. On the other hand, ANN performs better in predicting TS with MAE of 19.593, RMSE of 22.005, and R value of 0.947.

15 November 2023 in Journal of Energy, Material, and Instrumentation Technology
Verified
Journal Article
Open access
PDF file
DLDesmarita Leni
Desmarita Leni
RSRuzita Sumiati
Ruzita Sumiati
ADAdriansyah
Adriansyah
NANike Angelia
Nike Angelia
ENElsa Nofriyanti
Elsa Nofriyanti
This study aims to evaluate the influence of heatmap correlation-based feature selection on predictive modeling of low alloy steel mechanical properties using an artificial neural network (ANN) algorithm. Heatmap correlation was used to determine the chemical elements most correlated to the low alloy steel mechanical properties, such as Yield strength (YS) and Tensile strength (TS). There were 15 input variables of chemical elements in this study, and after feature selection, 11 input variables were obtained for YS, and 13 input variables were obtained for TS. The ANN model was validated using K-fold 10 cross-validation and evaluated using loss metric, Mean Absolute Error (MAE), and Root Mean Square Error (RMSE). The results showed that modeling with feature selection was able to improve the YS prediction, with a decrease in value of 6.83% in MAE and 4.97% in RMSE, while the TS prediction decreased by 16.46% in MAE and 18.34% in RMSE after feature selection. These results indicate that the use of feature selection provides better performance compared to the model without feature selection, and heatmap correlation can be used as an alternative to improve model performance in predictive modeling of low alloy steel mechanical properties using the ANN algorithm.  

26 December 2023 in Jurnal Surya Teknika
Verified
Journal Article
Open access
PDF file
HMHendri Candra Mayana
Hendri Candra Mayana
DLDesmarita Leni
Desmarita Leni
The examination of tire damage can be categorized as part of vehicle maintenance with the aim of ensuring the tires are in good condition. Visual inspection using human observation has limitations, making it not always accurate and prone to errors in determining tire roadworthiness. This study designs a machine learning model using Convolutional Neural Network (CNN) with a ResNet-34 architecture to detect car tire damage. The parameters used in training this CNN model include the Adam optimizer, a learning rate of 0.0001, batch size 32, and 50 epochs. In this study, there are two predicted image classes: normal tires and damaged tires. The research results indicate that the CNN model with ResNet-34 architecture can predict both classes very well, as evidenced by the model evaluation results with an accuracy of 0.916, precision of 0.907, recall of 0.927, and an F1 score of 0.917. These results suggest that the CNN model with ResNet-34 architecture can be used as an effective tool for inspecting tire damage.

10 September 2024 in EUREKA: Physics and Engineering
Verified
Journal Article
Open access
PDF file
DLDesmarita Leni
Desmarita Leni
AKArwizet Karudin
Arwizet Karudin
MAMuhammad Rabiu Abbas
Muhammad Rabiu Abbas
JSJai Kumar Sharma
Jai Kumar Sharma
AAAdriansyah Adriansyah
Adriansyah Adriansyah
The use of Exploratory Data Analysis (EDA) and machine learning in material science has rapidly advanced in recent years. EDA enables researchers to thoroughly explore and analyze material datasets, while machine learning allows for the development of predictive models capable of understanding complex patterns within the data. This study aims to develop an optimization tool to enhance the analysis of tensile strength in stainless steel by leveraging integrated data exploration and machine learning approaches within the Streamlit framework. The developed tool consists of four main features: data visualization, correlation analysis, 3D visualization, and machine learning. The developed machine learning model has 14 input variables, including chemical elements and heat treatment temperatures. In this research, the machine learning features comprise three models: Decision Tree, Random Forest, and Artificial Neural Network. The research findings indicate that the optimization tool can automatically display stainless steel tensile strength data using available pandas profiling in the visualization feature. The correlation feature can illustrate the relationship between chemical elements and heat treatment temperatures concerning stainless steel tensile strength. The 3D visualization feature can be utilized to identify optimal values of chemical elements and heat treatment temperatures according to desired tensile strength. Meanwhile, the machine learning feature can accurately predict stainless steel tensile strength based on chemical composition and heat treatment temperatures. This is evident from the performance evaluation metrics of the Random Forest model, which achieved MAE of 10.36, RMSE of 14.44, and R-squared of 0.97

26 June 2023 in Jurnal Surya Teknika
Verified
Journal Article
Open access
PDF file
DLDesmarita Leni
Desmarita Leni
HYHelga Yermadona
Helga Yermadona
ABAde Usra Berli
Ade Usra Berli
RSRuzita Sumiati
Ruzita Sumiati
HHHaris Haris
Haris Haris
This research designs a machine learning model using an Artificial Neural Network (ANN) algorithm to predict the tensile strength of aluminum. This research produces a machine learning model that has 8 (eight) input data variables consisting of the percentage of aluminum chemical composition such as Mg, Zn, Ti, Cu, Mn, Cr, Fe, Si, and 1 output (output), namely aluminum tensile strength. This study makes changes to several variations of parameters, such as variations in the number of split data, training cycles, learning rates, and hidden neurons. This Artificial Neural Network (ANN) modeling produces an RMSE value of 15,383 with the best parameters being split into 60 training and 40 testing data, training cycle of 100, learning rate of 0.08, momentum 0.9, and hidden neuron 7.This research designs a machine learning model using an Artificial Neural Network (ANN) algorithm to predict the tensile strength of aluminum. This research produces a machine learning model that has 8 (eight) input data variables consisting of the percentage of aluminum chemical composition such as Mg, Zn, Ti, Cu, Mn, Cr, Fe, Si, and 1 output (output), namely aluminum tensile strength. This study makes changes to several variations of parameters, such as variations in the number of split data, training cycles, learning rates, and hidden neurons. This Artificial Neural Network (ANN) modeling produces an RMSE value of 15,383 with the best parameters being split into 60 training and 40 testing data, training cycle of 100, learning rate of 0.08, momentum 0.9, and hidden neuron 7.