Prototipe Perangkat untuk Pemantauan dan Pengendalian Berbasis Web Diiintegrasikan ke Smarthome System

The design and construction of a device prototype for a water level measurement system in a tank and controlling a number of garden light analogies has been carried-out and the prototype can be integrated into smarthome system. Three topics are discussed in this paper, including the manufacture, programming, and performance measurement of device prototypes. The formation of prototype of the device is done through wiring integration between electronic devices, in order to obtain the hardware handshacking. Programming the prototype of device is done through the creation of algorithms and preparation of syntax, in order to obtain the software handshacking. The performance of the prototype of device is measured when integrated into the Smarthome system, in order to obtain the hardware and software handshacking. The performance of prototype of the device when monitoring in the form of information about the water level in the water tank with 3 (three) conditions, namely the criteria of "empty", "medium", and "full", while the control in the form of information about the operation of ON/OFF of the LED as an analogy to the lamp garden are done for 3 (three) positions, namely position #1, #2, and #3. The manufactured subsystem prototype can be integrated into the smarthome system when a validation test is performed. Prototype of the device for monitoring and control based-on web that can be integrated into the smarthome system.

Kendali Arah pada Brain Computer Interface Berbasis Steady State Visual Evoked Potentials

Various studies regarding to Steady State Visual Evoked Potentials (SSVEP) based Brain Computer Interface (BCI) system with Electroencephalogram (EEG) signal has developed as BCI implementation on directional control, however lackness found on those studies which are long time on classification duration, to many electrode channels used and the electrode channels located on special area. This study we developed the SSVEP based BCI system with one second classification duration, four active channels used and electrode channels located based on The International 10-20 System. Stimulus used are red colored object with 11 Hz frequency rate represents as left directional control class, blue colored object with 13 Hz frequency rate represents as right directional control class and white colored background represents as relax class. Filter bank with eight frequency range (11 Hz, 22 Hz, 33 Hz, 13 Hz, 26 Hz, 39 Hz, 12-29 Hz dan 30-50 Hz) followed by Root Mean Square (RMS) used as feature extraction for every second of data. Artificial Neural Network (ANN) classification and 5-Fold Cross Validation are used to knowing the performance of the developed system. Developed BCI system resulted accuracy 78,20% with True Positive Rate (TPR) 86,00% and False Discovery Rate (FDR) 23,21%.

Sintesis Taganing Adaptif Menggunakan Metode Pitch Shifting by Delay-Line Based untuk Standardisasi Gondang Batak Toba

This research using pitch shifting by delay line based method which consist of two main stage. The first stage is called analysis stage (framing, windowing, pre-emphasis and de-emphasis and FFT) that can detect the value of fundamental frequency of each taganing’s gendang. Then, this fundamental frequncy from each gendang will be classified into keyboard tones. The second one is called synthesis stage that will process the fundamental frequency become a new desire signal by creat an upward pitch change or a downward pitch change by delay line based method. Result of this research is created new signals as standard tones of each taganing’s gendang. The evaluation of synthesis output is using comparation method between fudamnetal frequency value of signal output as result of synthetis stage and the fundamental frequency value of keyboard standard’s tone. From the results of the system, it can be concluded that taganing synthesis tone have 98.87% accuration rate.

Rancang Bangun Platform Intel Edison Sebagai IoT Gateway Berbasis Protokol MQTT

The role of IoT gateway on IoT network is vital, like for strengthen the sensor management. Accordingly, in this research designed an IoT gateway based on MQTT Protocol which implemented in Intel Edison platform and using ten node sensors with star topology.Results from this research is an IoT gateway system who capable to receive data packet from node sensor, saves the data in database, and publishes the data to MQTT broker. User able to do data acquisition by subscribe method, using MQTT Dasboard application. The system tested by doing the calculation of time response of gateway with variation of data length and variation of distance, life time of node sensor testing by the variation of data length. The conclusions are the time average of from data acquisition by node sensor until gateway saves the data in database and user subscribes the data is twelve seconds. The shorter the data length makes the response time of gateway faster and makes the life time of node sensor shorter because the measured current is bigger. The distance is not gives any impact to the response time of gateway.

Fabrikasi dan Karakterisasi Sensor Elektrokimia untuk Mendeteksi Kadmium Berbasis Teknologi Film Tebal

Electrochemical sensor has been fabrication and characterization electrode to detect heavy metals of cadmium with thick film technology. Fabrication electrodes electrochemical sensors start with the design stage of electrodes, screen maker, screen printing, and substrate cutting. The fabrication process produces a ten-pieces of sensors to be used to sensors characterization of SEM, XRD, and Cyclic Voltammetry (CV). The process of characterization produces morphology and the structure of the composites N/IL/G, and linieritas 0,9623, precision ranges between 4,11 and 10,35%, sensitivity of 0,04112 µA.L-1/mg.mm2, limit detection 0,18 mg/L, and accuracy between 79,5 and 126,8%.

Performance Analysis of Data Transmission on a Wireless Sensor Network Using the XBee Pro Series 2B RF Module

Wireless Sensor Network (WSN) is a wireless network technology that is capable of scanning and can be used to monitor environmental conditions, both indoors and outdoors. One of the devices used in data communication at WSN is XBee. XBee is a device in WSN that uses radio frequency as a data transmission path from one device to another. In sending data, Xbee can be affected by the distance and signal strength between devices. In this research, Xbee communication distance analysis is done outdoors and indoors so as to get data in the form of maximum distance vulnerable from Xbee. The results of this study indicate that the Xbee Pro Series 2B that is measured outside the room is capable of sending data up to a distance of 110 meters, while the Xbee Pro Series 2B device that is measured indoors is capable of sending data up to a distance of 20 meters.

Klasifikasi Tingkat Kekakuan Dinding Beton Terhadap Getaran Dengan Metode K-Nearest Neighbor

The low level of wall stiffness can cause damage to buildings during large-scale earthquakes. There are many systems for measuring the level of stiffness in buildings, but they have not yet reached the classification stage. Therefore, a system that can classify stiffness is needed to determine the impact of vibrations on the wall to minimize the losses incurred.This study creates a system that can classify the level of wall stiffness using the K-Nearest Neighbor (KNN) method into several categories (safe, vulnerable, dangerous, and destroyed). The data taken at the acquisition stage are ground acceleration, inclination angle, displacement, drift ratio, and peak value. The KNN input is a peak ground acceleration value, which causes a drift ratio of 1%. The resulting output is a category of wall stiffness based on the Earthquake Intensity Scale by BMKG.Functionally, the system designed can classify wall stiffness with non-linear data input using the K-Nearest Neighbor (KNN) method. The success rate of KNN reaches a value of 100%. Based on the PGA drift ratio reading, it is assumed that the wall can withstand the maximum vibration with a PGA drift ratio value of 0.34 g without causing damage to the wall even though it has a low level of stiffness. Testing on the walls has a less high degree of precision. That may be due to factors other than PGA. That can affect the drift ratio on the walls, which have not been considered in this study.

Enhanced Active Filter for Single-Phase Controlled Rectifier Applications

Active filter can be used to increase the performance of single-phase controlled rectifier by reducing the generated ripple and harmonic distortion. Active filter can also increase power factor and efficiency of thyristor. In this work, active filter was designed by using NPN Bipolar Junction Transistor (BJT) compared with active filter created by utilizing NPN Metal Oxide Semiconductor Field Effect Transistor (MOSFET). The obtained efficiency of BJT based active filter with firing angles of 0° to 90° were 87.01% - 96.67%. Whereas, MOSFET based active filter produced comparable efficiency between 90.62% - 96.07%. It can be seen that MOSFET based active filter produced higher efficiency at firing angle of 0° which was 90.62% compared to BJT based active filter that has produced efficiency of 87.01% at the same firing angle. Both active filters obtained similar power factors within the range of 0.57 lagging – 0.92 lagging. It is expected that this work can be useful for the design of active filters for various power electronics applications including the design of High Voltage Direct Current (HVDC) power transmission system.

Survei Penggunaan Tensorflow pada Machine Learning untuk Identifikasi Ikan Kawasan Lahan Basah

Wetlands are habitats commonly used for fish cultivation. South Kalimantan is one of the provinces that has a wetland area, which is 11,707,400ha, there are 67 rivers and an estimated 200 species of fish. This shows the abundant wealth of fish treasures and economic value. The study of fish identification is an important subject for the preservation of wetland fish. In the field of artificial intelligence, identification can be done using Machine Learning (ML). There are many libraries, a collection of functions that can be used in ML development, one of which is Tensorflow. In this paper, we survey a variety of literature on the use of Tensorflow, as well as datasets, algorithms, and methods that can be used in developing wetland area fish image identification applications.The results of the literature survey show that Tensorflow can be used for the development of fish character identification applications. There are many datasets that can be used such as MNIST, Oxford-I7, Oxford-102, LHI-Animal-Faces, Taiwan marine fish, KTH-Animal, NASNet, ResNet, and MobileNet. Classification methods that can be used to classify fish images include CNN, R-CNN, DCNN, Fast R-CNN, kNN, PNN, Faster R-CNN, SVM, LR, RF, PCA and KFA. Tensorflow provides many models that can be used for image classification, including Inception-v3 and MobileNets, and supports models such as CNN, RNN, RBM, and DBN. To speed up the classification process, image dimensions can be reduced using the MDS, LLE, Isomap, and SE algorithms.

Penerapan Algoritma Feature Selection Chi-Square (X2) Dalam Sistem Monitoring Tumbuh Kembang Balita

The healthy family program is a priority program of the ministry of health which in principle refers to the 5th nawacita program. In line with this program, the Lampung provincial government has launched development in the health sector in all cities or districts by allocating 11.52% of the special budget for the health sector. This is inseparable from the condition of health development, especially the city of Bandar Lampung, where the infant mortality rate, under-five mortality rate and maternal mortality rate are decreasing. One of the conditions of health development is the health service posts. Health service posts have limited medical personnel and health post data searches still use conventional models so that the results obtained are not optimal. Based on these descriptions, the authors propose to design a software engineering system in the field of medical health in the form of a toddler growth monitoring system with the concept of the Chi-Square or X2 feature selection algorithm. The results of this study are able to provide a web-based monitoring system model that can be accessed anywhere and anytime, able to provide a system model that can directly monitor the condition of toddler health growth and is able to provide a system model that can be used by health workers, both medical and midwives to monitor village health post points.