Engineering Headway Vol. 27

Abstract: The need for energy sources, particularly electrical energy, continues to increase every year. Currently, the entire world, including Indonesia, is engaged in energy planning with a steadfast commitment to expedite the transition towards clean and renewable energy sources. Currently, to support the energy transition, the planning of renewable energy power plants, especially PV PP, is increasing significantly. In current planning, PV PP is designed in both large and small scales. To facilitate the cost-effective integration of PV PP into the power system, a techno-economic comparison of PV PP scales is necessary. This paper focuses on optimizing photovoltaic (PV) generators in the generation planning process to enhance PV penetration and the integration of distributed generators. The optimization process will consider both fully distributed PV power plants and centralized PV power plants at several points. The contribution of this paper lies in providing an insight into the optimal capacity scale schemes of PV power plants in planning to enhance the integration of distributed generators. The research findings indicate that Centralized PV PP can produce 7% more energy compared to Distributed PV PP. With electricity generation reaching 166,440 MWh per year, it can provide a lower LCOE of $0.48/kWh compared to distributed PV PP, which reaches $0.78/kWh.

Abstract: In recent decades, relative humidity has become a research topic that has received increasing attention due to its important role in climate change and global warming. One of the most typical issues with relative humidity is data loss due to instrument deterioration. This research attempts to apply feature selection and hyperparameter tuning methods as an approach to optimizing the reliability of the multilayer perceptron (MLP) model to predict relative humidity values designed into the MLP-CV framework. The coefficient of determination (R²), root mean squared error (RMSE), and absolute error (MAE) are used to determine the model's correctness. The results showed that the MLP-CV model had better accuracy compared to the MLP model for predicting relative humidity missing values, with R² = 0.788, RMSE = 1.838, and MAE = 1.431.

Abstract: Identifying nonlinear dynamic systems that often exhibit chaotic behavior can be a complex task. Recognizing chaotic behavior in data can provide valuable insights for its utilization. Mathematically, the identification of chaotic behavior in data typically requires the consideration of multiple parameters. However, nonlinear dynamic systems can be readily identified using machine learning. In this study, a machine learning model was constructed using a deterministic dataset generated from logistic map equations, employing the Long Short-Term Memory (LSTM) architecture. The outcomes of the machine learning model take the form of data classification, distinguishing between predictable and unpredictable data with quite high validity.

Abstract: Thermal management plays a vital role in ensuring the overall performance. SSD (Solid State Drive) NVMe (Non-Volatile Memory Express) is latest generation of data storage that periodically generates unwanted heat. The present paper presents simulation result of heatsink orientation as presented in different models. The selected techniques include forced convection from air flow which will dissipate amount of heat from base area. We select M.2 NVME as study case which coupled with heatsink. The result was higher velocity resulting in lower gap temperature. Case VIII has the lowest temperature gap (3 Kelvin) while the highest is Case III (10.63 Kelvin). Then, the optimum model based on temperature and mass parameter is model B.

Abstract: The frequent use of railway tracks in railway operations can cause damage or wear that can disrupt comfort and resulted vibrations on the trains. There are various types of damage that can occur to railway tracks, one of which is longitudinal level damage. Machine learning can be employed to predict the damage. However, it is quite difficult to predict based on real data with a high amount of data. Therefore, a railway miniature is fabricated with a controlled damage. Therefore, this study has purpose to predict the damage using the produced data from railway miniature. The vibrations was measured using an accelerometer device that available on smartphones with the Phypox application, and it will be mounted on a miniature railway track with three different track conditions: one normal and two abnormal, with each track condition has 50 data points. With the assistance of machine learning as the main brain behind the vibration detection program, vibration data can be classified based on the track conditions experienced. The data was processed into frequency domain using Fast Fourier Transform (FFT) algorithm, filtered using SG-Filter, and Power Spectral Density (PSD) will be used to assess the strength of the vibration signal. The vibration data processing was carried out using Jupyter Notebook software with Python programming language. Classification was performed by applying supervised machine learning using the classification method of Support Vector Machine (SVM). In classification process, results obtained show an accuracy of 88.19% for training model and an accuracy of 82.61% for testing model, computed using 85% of total data for training model and 15% of total data for testing model. The produced data and built machine learning can be further applied for checking the rail damage at uncontrollable environment.

Abstract: To know the quality of drinking water in Indonesia, conventional method, which must be done by testing water samples in a water testing laboratory which not everyone knows where it is and takes quite a long time to get the results. For this reason, it is necessary to design a drinking water monitoring system prototype in a water tank that is easy to use using IoT technology. Sensors are used to retrieve water quality parameter data in real time and then the ESP32 microcontroller which is equipped with a WIFI module will send data to a database server which can then be downloaded and displayed via an application. The parameters monitored in this work are the pH value, electric conductivity, temperature and turbidity, following the standards of the Indonesian government for drinking water. The test results show that the prototype is working according to the design.

Abstract: Subsurface photoacoustic imaging with high resolution is utilized to image the teeth layer due to periodontal tissue inflammation. The aim of this imaging method is to determine the difference between healthy teeth and teeth affected by inflammation based on the acoustic signals obtained. An 808 nm diode laser is used as a radiation source, a condenser microphone is used as a detector, and computer numerical control (CNC) is used for the sample scanning process. The samples were healthy and inflamed Sprague-Dawley rat teeth. The average acoustic intensity of healthy teeth layers was 60.200.9 dB for enamel and 52.210.9 dB for dentin, while the average acoustic intensity of inflammation-affected teeth layers was 93.140.4 dB for enamel and 84.840.4 dB for dentin. Based on the histogram results obtained, the generative adversarial network (GAN) method can be used to improve photoacoustic image resolution to become more detailed. The study shows that high-resolution subsurface photoacoustic imaging can be utilized to image teeth layers.

Abstract: Health monitoring systems for industry workers are needed to maintain their safety, health, productivity, and to prevent accidents using technologies to measure workers' physiological and environmental variables could predict and prevent potential human risk in industry. This study aimed to review several health monitoring systems to get information about their system designs, methods, requirements, and performances. Scoping keywords of industrial subjects, actions, health, and devices, along with their synonyms, are used to retrieve articles from the Scopus database from 2009 to June 2023. The screening results in 18 papers. The health monitoring system comprised of several types of personal health and environmental sensors, comprised of EEG, ECG, EMG, PPG, IMU, camera, FMCW, PIR, USR, and sensors of heart rate, body and environment temperature, respiratory rate, relative humidity, dust or particulate matters, noise, hazardous gases, air pressure, and UV. The supporting systems comprised processors, network infrastructures, servers, databases, software and algorithms, actuators, displays and websites, validators, and surveys. Those studies are done either by field or laboratory experiments, software simulations, secondary data analysis, or concept designs. The requirement insights are grouped into ten aspects: validity, effectivity, connectivity, functionality, reliability, safety and security, compatibility, economy, user-friendliness, and supportiveness. The system results and performances varied through the objective and sensor data used, from monitoring purposes to fatigue and health issue detection such as drowsiness, falling, stress depression, and distress. Fatigue and other health issues could be detected by camera image analysis, EMG, IMU, and HRV signals, not by HR or %HRR.

Abstract: This research aims to analyze the reliability of a non-invasive blood glucose detection device (glucometer) based on semi-cylindrical capacitive sensor (SCCS) with ATmega328P microcontroller. The device is designed to measure blood glucose levels non-invasively, without requiring blood sampling. This research uses SCCS to measure the dielectric constant of the finger containing glucose. This dielectric constant is then linked to blood glucose levels through a calibration equation. The calibration equation is obtained from experimental data that links the dielectric constant to blood glucose levels. The ATmega328P microcontroller is used to process signals from the SCCS and calculate blood glucose levels based on the calibration equation. The blood glucose measurement results are displayed on an I2C LCD. The reliability test of this device is carried out by comparing the measurement results with the measurement results of a standard invasive glucometer. The results of the study show that the device has an accuracy of 88.45% close to Invasive glucometer.

Abstract: A tissue processor is a device that automatically processes tissue into several steps for histological studies in medical or biological laboratories. One of the stages of this device is embedding, where the tissue is inserted into a vessel that is heated to a stable temperature of 60 to 80°C so that the paraffin remains liquid during this process. The handling of tissue specimens is a crucial aspect of ensuring an accurate diagnosis of the tissue. However, due to its high cost, it can affect scientific development that requires tissue processing. Reverse engineering is one of the methods commonly used to develop a tool at a lower price; one way is to change the essential components used while still maintaining the device's primary function. This research aims to develop a heater controller for a tissue processor using STM8 and LM35 analog sensors equipped with a Kalman filter. The design of the Kalman filter will be determined based on its component equations. The measurement and variance constant values will be simulated and analysed in MATLAB for the most suitable parameter value. Afterwards, the proposed design will be conducted using the Arduino Integrated Development Environment (IDE) to reduce the noise in the LM35 sensor reading and then saved using ArduSpreadsheet. The simulation and implementation results provide evidence that the proposed Kalman filter effectively filters signals contaminated with noise, even in the presence of high data variations that may occur in sensor readings. In testing, it is necessary to pay attention to the ratio that can reduce noise while keeping the characteristics of the sensor itself. Numerous investigations have explored various ratios, recommending an appropriate ratio of 100. Based on the implementation of this ratio with measurement constant (R) 10 and variance constant (Q) 0.1, resulted in a 0.29°C mean error and stable sensor readings.