International Journal on Advanced Science, Engineering and Information Technology
Not a member yet
2006 research outputs found
Sort by
Optimization of Soil Temperature and Humidity Measurement System at Climatology Stations with IoT-Based Equipment
Full text linkTemperature and humidity are important weather parameters that require close observation due to their importance across various fields, including agriculture. Apart from the use of automatic weather system (AWS), the station of meteorology and climatology also relies on conventional devices to observe these parameters, but they have been proven to be inefficient, imprecise, and prone to systematic errors. The alternative AWS consists of several sensors with different functions, allowing for more accurate measurements, but it also has one major limitation. This includes its inability to carry out measurements with the sensors when one of them is damaged. Therefore, this study aims to develop high-precision soil temperature and humidity (STH) monitoring equipment using the DHT11 sensor module. The equipment consisted of a box containing a series of device builder electronics. The building electronics circuit contained a DHT11 sensor, NodeMCU ESP8266 microcontroller, an on/off switch, and a reset button. The results of measurements of temperature and humidity often appeared on the smartphone. The DHT11 sensor detected the soil parameters, which were processed by the NodeMCU ESP8266. The data obtained were then sent to the Thingspeak server, where they could be accessed on a smartphone. The developed equipment showed good performance with accuracies of 98.201%, 97.330%, 98.982%, 98.973%, and 99.649% in measuring STH at each depth, while values of 98.487% and 98.587% were obtained for humidity measurement. Furthermore, precision values of 99.93% and 99.95 were recorded for the measurement of temperature and humidit
MiSREd: A Low Cost IoT-Enabled Platform Based on Heterogeneous Wireless Network for Flood Monitoring
Full text linkMotivated by an inherent difficulty of foreseeing the exact occurrence of disasters, attempts to rapidly detect and forecast associated information leading towards and in the aftermath of the disaster events can help minimize casualties and collateral damage, particularly in the rural and crowded urban environment. An emerging Internet of Things (IoT) technology is considered promising for these purposes due to Its inherent capability of capturing, sending and processing various types of environmental field data in real-time over a large geographical area. In this paper, the authors introduced MiSREd (Multi-input, Scalable, Reliable, and Easy-to-deploy) as the authors’ new low cost IoT platform envisioned to meet the needs of an integrated disaster management system. A key part of the MiSREd platform is the incorporation of heterogeneous wireless networks for improvement reliability and availability of message telemetry. Moreover, deployment of low-overhead protocols can improve the network traffic with a lower bandwidth load as a result of data reduction applied to the MQTT protocol. In order to evaluate the effectiveness of MiSREd, an IoT testbed was developed and evaluation was conducted at Western Flood Canal in Semarang, Indonesia. Data transmission testing in the backhaul using the MQTT protocol showed achievement of a transmission dela
An Improved Accuracy of Multiclass Random Forest Classifier with Continuous Attribute Transformation Using Random Percentile Generation
Full text linkThis study aims to improve classification accuracy by transforming continuous attributes into categories by randomly generating percentile values as categorization limits. Four algorithms were compared for the generation of percentile values and selected based on the small variability of the percentile values and the distribution of the highest revenue expectations. The distribution of testing and training data classification accuracy becomes the second consideration. Random forest (RF) classification is modeled from selected percentiles with three transformation variations. The results of the ANOVA test, the algorithm with three variations of the transformation, has a mean that is not significantly different from the best model and the original dataset model. However, in some variations of training data, RF classification with continuous attribute transformation was superior to the original dataset model. The effectiveness of this continuous attribute transformation algorithm was very well applied to the LR, MLP, and NB methods. In the tuition fee dataset, the application of the algorithm for the three methods each had an accuracy of 0.178, 0.204, and 0.318. The results of the attribute transformation give a significant increase in accuracy to 0.967, 0.949, and 0.594 for each method, respectively. In the date fruits dataset, the attribute transformation was effective in the MLP method with an accuracy of 0.193 (original attribute) to 0.690 (continuous attribute transformation). The transformation results are effectively applied to the LR, MPL, and NB methods for datasets with continuous and categorical mixed attributes
Precipitation Probability Prediction through NWP Bias Correction for South Korea Using Random Forest
Full text linkThis study presents the results of an effort to improve the forecast of precipitation (> 0.1 mm/hr or > 0.1 mm/3hr) in the Local Data Assimilation and Prediction System (LDAPS) and the Global Data Assimilation and Prediction System (GDAPS) by applying the Random Forest (RF) model in South Korea. LDAPS and GDAPS are Numerical Weather Prediction (NWP) models operated by the Korea Meteorological Administration (KMA) for weather forecasting. GDAPS operates the Unified Model (UM) and the Korean Integrated Model (KIM). This study used weather forecast data from LDAPS, GDAPS/KIM, and GDAPS/UM. Precipitation forecasts from LDAPS and GDAPS were corrected by RF training with rain gauge observations from about 685 stations. Approximately 35 selected NWP model output variables were used as inputs to the RF training. To reflect recent trends in biases between observations and NWP, the precipitation probability prediction model was designed for real-time learning using a sliding window technique. In addition, the precipitation data had a data imbalance problem with more precipitation cases than non-precipitation cases, so an under-sampling method was applied to solve this problem. Comparing the performance of the proposed method with NWP in predicting precipitation, the CSI was improved by 14.7-23.1% (LDAPS), 33.9% (GDAPS/KIM), and 6.7%-38% (GDAPS/UM) over NWP, and the accuracy was also better. In future research, automating the sampling rate selection to reflect recent weather trends when under-sampling is likely to improve forecast performance
Variable Precision Multiplier for CNN Accelerators Based on Booth Algorithm
Full text linkAs the utilization of CNN increases, many studies on lightweight, such as pruning, quantization, and compression, have been conducted to use CNN models in servers and edge devices. Studies have revealed that quantization greatly reduces the complexity of CNN models while lowering accuracy to a negligible level. CNN models with bit precision lowered from the existing 64/32 floating point to 16, 8, and 4 fixed points are being announced. Therefore, this paper proposes a variable precision multiplier that can select between 16 bits and 8 bits of precision. It consists of four 8-bit booth multipliers. When 16-bit multiplication is selected, the final product is calculated from four partial products, and when 8-bit multiplication is selected, four multiplications are possible simultaneously. The proposed multiplier was designed with Verilog HDL, and its function was verified in ModelSim. And it was synthesized for Altera Cyclone III EP3C16F484C6 using Quartus II 13.1.0 Web Edition. The proposed variable multiplier has increased combinational logic compared to general 8-bit/16-bit booth multipliers, and the clock speed is reduced by 65% and 82%, respectively. However, it can process four 8-bit multiplications within 1.68 times of normal 8-bit multiplication processing time and can process 16-bit multiplication within 75% of the normal 16-bit multiplication processing time. Therefore, the proposed multiplier is expected to increase speed and energy efficiency by selecting bit precision according to the layer in the CNN model
Research in Electronic Multi-Sensor Accuracy in the Implementation of Soil Fertility Monitoring System Using LoRA
Full text linkThe use of electronic sensors to track the nutrients in the soil is an interesting tool for farmers. This has led to the sale of many different kinds of electronic sensors with different levels of accuracy. The accuracy of this electronic sensor was figured out by comparing the results of the sensor's measurements with the results of lab tests done in different ways. This study compares the accuracy of electronic devices used to measure soil nutrients like nitrogen, phosphorus, potassium, electrical conductivity, water pH, and humidity to measurements made in the lab using the ICP-OES (Inductively coupled plasma-optical emission spectroscopy) method. We used three electronic sensors and a transmission system based on LoRA (Long Range) to measure the nutrients in the soil and put the results on our website. The similarities between electronic sensors and laboratory test parameters include the standard deviation, accuracy value, and correlation test between sensors and from the sensors to laboratory test results. The standard deviation parameter test showed a big value between the electronic sensor and the lab test results. However, none of the three used electronic sensors had a standard deviation number that differed greatly from the others. Except for the pH value of the soil, the electronic sensor's accuracy tests for the other five parameters were not very good compared to the lab tests. Also, the sensor correlation test showed a high correlation, while the correlation test between sensor data and lab test results showed a low correlation
Optimization and Analysis of Polyhydroxyalkanoate (PHA) by Bacillus sp. Strain CL33 and Bacillus flexus Strain S5a from Palm Oil Mill Waste
Full text linkPolyhydroxyalkanoate (PHA) is a biodegradable polymer that microorganisms can synthesize amidst non-optimal growth conditions with excess carbon sources. Palm oil, rich in fatty acids, serves as a carbon source for PHA synthesis. The bacterial PHA production can be influenced by carbon concentration in the growth medium. Therefore, determining the optimal concentration of palm oil as a carbon source is crucial for PHA production. Additionally, it is possible to determine the type of PHA generated by bacteria, which can then be utilized as information when processing utilizing the PHA. The experiment employed palm oil concentrations of 0.5%, 1%, and 2% and was carried out for periods of 48, 72, 96 hours. It was discovered that Bacillus sp. strain CL33 and Bacillus flexus strain S5a produced the most effective PHA at a concentration of 25 with an incubation period of 96 hours. The PHA generated by these bacteria was quantitatively analyzed through measurements of total bacterial growth, cell dry weight, and the levels of crotonic acid. PHA types were also analyzed using GC-MS, with monomers including 2-hydroxybutyrate(-2HB), 2-hydroxy-3-phenylpropionate (2H3PhP), 3-Hydroxyhexanoate (3HHx), 3-hydroxyoctanoate (3H2O), and 3-hydroxydecanoate (3HD). The Bacillus sp. strain CL33 yielded a PHA level of 92.23%. Meanwhile, Bacillus flexus strain S5a synthesized a polyhydroxyalkanoate comprising mostly 3-hydroxyhexanoate (3HHx) and polydimethylsiloxane (PDMS). The monomers used were decamethyltetrasiloxane, dodecamethylpentasiloxane, hexamethylcyclotrisiloxane, octamethylpentasiloxane, and dodecamethylcyclohexasiloxane. The type of PHA produced accounted for 85.93% of the total
Activated Carbon from Palmyra Palm Peel as an Alternative Adsorbent for Removing Heavy Metal Ions Fe(III) and Cr(VI) in Industrial Waste
Full text linkPalmyra palm peel served as raw material for preparing activated carbon. In addition to its high cellulose content, palmyra palm shells are also easily found in Gresik and Tuban, East Java. Palmyra palm shell is also an abundant solid waste with low economic value, so using palmyra palm shells as raw material for activated carbon production is low cost to reduce the contaminant in liquid waste. This experiment aims to determine the effectiveness of palmyra palm peel as a bio-adsorbent for heavy metal ions Fe(III) and Cr(VI) in industrial waste. This research was conducted through 3 processes: chemical activation, carbonization, and adsorption. The methods used in this study consisted of pre-treatment, activation of raw materials, manufacture of standard solutions, calibration of standard solutions, and adsorption of heavy metals from textile waste. The carbons' activation was conducted at 600 and 700oC in the presence of KOH as the activating agent. The results are a water content of 17.50% and an ash content of 8.37%. The moisture content and ash produced results comply with the SII and SNI 06-3730-1995 standards. The carbon produced at 700oC has a better adsorption performance than that produced at 600oC. The maximum removal efficiency for Fe(III) was 95.25, and Cr(VI) was 89.7%. Two well-known equations, Langmuir and Freundlich, were used to correlate the experimental adsorption data. Langmuir equation could represent the data better than Freundlich with an R2 value close to unity
Recognition of Agricultural Land-Use Change with Machine Learning-Based for Regional Food Security Assessment in Kulon Progo Plains Area
Full text linkHigh conversion of agricultural land in Kulon Progo Regency, as such the construction of the Yogyakarta International Airport (YIA) and the Bedah Menoreh road, has resulted in food production and impacted food security, including Kulon Progo plains area. This study aimed to calculate the conversion rate of agricultural land and analyze its impact on food security in the Kulon Progo plains area from 2005 to 2020. The primary materials needed are Kulon Progo administrative maps, Landsat 7 and 8 images, land productivity data, population data, and consumption per capita data. With tools used is Google Earth Engine (GEE), SPSS 25, Google Earth Pro, and ArcGIS 10.3. The method used is calculating the Normalized Difference Vegetation Index (NDVI) and machine learning-based classification through GEE to identify land-use change and analyze the state of food security. The study proved that between 2015 and 2020, there was a conversion of paddy fields, with an average rate of 126 ha/year. The existence of new paddy fields influenced this land increase. However, in 2020 there is still food insecurity in Pengasih District, thus caused by the new paddy fields not being optimally used for rice growth. The productivity of the land produced is not optimal. With the availability of agricultural land in 2020 (1382.85 ha), food self-sufficiency will be limited for the next 24.75 years if there is no effort to increase paddy fields
Wire Extensometer Based on Optical Encoder for Translational Landslide Measurement
Full text linkA landslide is a natural disaster mostly accompanied by heavy rains, earthquakes, or volcanic eruptions. Due to its significant incurred losses, several studies have been conducted to develop a landslide monitoring system. In this report, we built and implemented optical-based wire-extensometers to measure and monitor a translational landslide in a prone area. This extensometer was built of an optical rotary encoder (whose shaft bonded to a spiral spring and sling rope) interfaced to a low-cost microcontroller as a principal component and subsequently linked to a GSM-based wireless network. The working principle of the employed sensor described in this paperwork is to count optical pulse signal and convert it into a length unit. This sensor can provide much better signal stability and show high resolution for a wide-range measurement than voltage- or current-based sensors. The specification of the engaged optical encoder provides 2000 pulses per rotation, leading to a length resolution of 0.011 ± 0.0083 mm with a speed limit of about 36 mm/s. Furthermore, the wire extensometer was examined in a remote place near a double-track train road to assess its performance in an actual field. A solar cell system was applied as its main power supply. An example of transmitted data shows a land shift from 12 mm to 150 mm, which is mainly triggered by high rainwater infiltration. This result demonstrates that the developed extensometer is deserved to be promoted for landslide monitoring in the geological research-work area